Jump to content

Talk:Rorschach test: Difference between revisions

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Content deleted Content added
Line 576: Line 576:


::That said, again my primary argument was about the release of test materials (including tests, answers (if applicable), explanations of design, principles used for evaluation) to the general public after a certain period of time. If you note, I did argue that this information first be released in a more limited way (for example, in professional circles, then more generally), but 75 years seems a rather long lifespan to argue (as some of the professional associations apparently do) against widespread dissemination. Norming statistics and similar information is not enough, since statistics can easily be used incorrectly or even deliberately used to manipulate trends in data. Do I really mean to argue that professional psychologists are incompetent? Of course not. But the argument of "leave it to the professionals, who understand what's going on" sounds strikingly to me like the same arguments used to restrict access to state secrets. Everything should be open to more widespread oversight after a period of time; it serves as a check against the occasional myopia of the professionals. I admit that my concern is more about testing in general (whether it be IQ testing, personality testing, or SATs and GREs), but the argument for release of and easy access to detailed information after a reasonable amount of time is just as applicable here.[[Special:Contributions/65.96.161.79|65.96.161.79]] ([[User talk:65.96.161.79|talk]]) 18:58, 16 July 2009 (UTC)
::That said, again my primary argument was about the release of test materials (including tests, answers (if applicable), explanations of design, principles used for evaluation) to the general public after a certain period of time. If you note, I did argue that this information first be released in a more limited way (for example, in professional circles, then more generally), but 75 years seems a rather long lifespan to argue (as some of the professional associations apparently do) against widespread dissemination. Norming statistics and similar information is not enough, since statistics can easily be used incorrectly or even deliberately used to manipulate trends in data. Do I really mean to argue that professional psychologists are incompetent? Of course not. But the argument of "leave it to the professionals, who understand what's going on" sounds strikingly to me like the same arguments used to restrict access to state secrets. Everything should be open to more widespread oversight after a period of time; it serves as a check against the occasional myopia of the professionals. I admit that my concern is more about testing in general (whether it be IQ testing, personality testing, or SATs and GREs), but the argument for release of and easy access to detailed information after a reasonable amount of time is just as applicable here.[[Special:Contributions/65.96.161.79|65.96.161.79]] ([[User talk:65.96.161.79|talk]]) 18:58, 16 July 2009 (UTC)

===Recognition===
The first image in particular has a somewhat iconic quality. I saw it some years ago - I don't remember the exact context, but I believe I was reading about Rorschach. Now that image in particular is associated in my mind with Rorschach and symmetric inkblots. Someone with a worse memory than me might see Rorschach mentioned somewhere, not remember who he was or why he was important, look it up here, and see the image and remember "oh, that's who he was". This is one reason to include images in general - not just to show people new to the subject what the subject looks like, but also to jog the memory of those with a vague recollection of the subject. --[[User talk:NE2|NE2]] 04:57, 17 July 2009 (UTC)


== Inline citation for inkblots needed ==
== Inline citation for inkblots needed ==

Revision as of 04:57, 17 July 2009

WikiProject iconPsychology Start‑class Mid‑importance
WikiProject iconThis article is within the scope of WikiProject Psychology, a collaborative effort to improve the coverage of Psychology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
StartThis article has been rated as Start-class on Wikipedia's content assessment scale.
MidThis article has been rated as Mid-importance on the project's importance scale.
WikiProject iconMedicine B‑class Mid‑importance
WikiProject iconThis article is within the scope of WikiProject Medicine, which recommends that medicine-related articles follow the Manual of Style for medicine-related articles and that biomedical information in any article use high-quality medical sources. Please visit the project page for details or ask questions at Wikipedia talk:WikiProject Medicine.
BThis article has been rated as B-class on Wikipedia's content assessment scale.
MidThis article has been rated as Mid-importance on the project's importance scale.

All 10 images

Archived to Talk:Rorschach test/Archive 7#All 10 images
The ten inkblots of the Rorschach inkblot test

Arguments Pro

#01 - The cat's out of the bag

#02 - No evidence of harm

From what I have read here over the last hour or so I see three "parties" that are being argued about potentially being "harmed".

  1. General public (readers)
  2. Inkblot copyright holder
  3. Psychological professionals who use the test.

To address each point:

1. Is a member of the general public directly harmed by viewing the images? No. Even a user with a mental health issue for which the test may prove beneficial will not be directly harmed by viewing the images.
1.A. Is a member of the general public indirectly harmed by viewing the images? Possibly, because their exposure to the images may remove a potentially useful tool from the hands of a psychologist or psychiatrist attempting to assist them with a later mental health issue.
2. Is the copyright holder of the images harmed by their display? No, because the images are in the public domain and no longer subject to any copyright restrictions.
3. Are psychological professionals harmed by the display of the images? Possibly, because a potentially useful tool may be rendered useless if a subject has already been exposed to the images.

All of that said, the question remains as to whether the possible harm of presenting the images outweighs the value of having the images appear in the encyclopedic and information sharing venue of Wikipedia.

At this point the only "evidence" presented of harm have been statements issued by psychological professional organizations. While these statements can be viewed as having significant weight due to their pedigree and preparation by those professional bodies, one must also recognize that those bodies have a vested monetary and temporal interest in keeping the images from the public. (Nothing sinister is implied in this statement, simply that the investment of time and money to develop new images for the test would be extensive.)

Without further documented empirical evidence I don't see a viable argument to declare that the presentation of the images constitutes harm.


Note that the above conclusion ignores the fact that the images are also available and presented elsewhere in the public sphere, which also lead me to the argument that placing the images in context with documented and vetted information about the test is, in fact, a more responsible way to present the images than the way the may be found elsewhere. --Raukodraug (talk) 19:00, 15 July 2009 (UTC)[reply]

Raukodraug, point 1A pretty much trumps your later points. Does Wikipedia's moral responsiblity end at direct harm? Morally, I would imagine the answer is no. In practice, see the David S Rohde argument below, in which the answer is also no. Indirect harm is still harm and should still tend to be avoided. Dark Nexus (talk) 21:58, 15 July 2009 (UTC)[reply]

While I agree that indirect harm is still harm and should be avoided, point 1A also notes that the images only present a potential for indirect harm. Doing a little research last night I was unable to find any documented evidence that harm occurs from the presentation of the pictures. And if research has been done on the subject on the loss of efficacy of the Rorschach test due to previous exposure to the images, I also haven't been able to find it, nor have I seen it presented anywhere in the discussions here. (That said, I no longer have access to all of the research websites from my university days so it's very possible that something buried in a psych journal, if so I would love to peruse the articles, so citations would be appreciated.)
In the end, if Wikipedia works on principals of verifiability of research and information, then while harm can be inferred and even anecdotally reported, unless there is documented evidence that the presentation of images causes harm, I don't see any direction the removal argument can be taken. --Raukodraug (talk) 16:41, 16 July 2009 (UTC)[reply]

#03 - Adds to the page

#04 - Removing the images amounts to censorship

#05 - Reputable organizations have been showing the images, too

But reputable organizations have also chosen to NOT show the images. See 2001 Scientific American Article Clearly, Poundstone, in his book Big Secrets thought he was doing something very scandalous. So if the title of this argument is trying to say that it has become acceptable to show the images, I think it's a flawed argument. Danglingdiagnosis (talk) 03:10, 10 July 2009 (UTC)[reply]

This point is related to argument #1, that the cat is out of the bag. Whether the cat is out of the bag and whether others have been displaying the image is not at issue. It also isn't very relevant. Wikipedia has a unique place in the English speaking / Internet accessing zeitgeist; it is a very visible public source of information. Due to its high page rank, this is where many people end up from search engines. Placing the images here (or not placing them) has a large impact on whether a casual non-savvy person will see them. Wikipedians need to accept their role in the zeitgeist and realize that the decision here should be independent of the decisions of other sources of information. Dark Nexus (talk) 22:04, 15 July 2009 (UTC)[reply]

#06 - It doesn't matter if we publish. The test is nearly worthless anyway.

There has often been expressed on this talk page a valid opinion about the poor reliability of the test. But this opinion has no place when deciding whether to show the images or not. This thinking is similar to that of someone who vandalizes another's car saying, "Oh well, they won't care. It's a worthless pile of scrap metal, anyway." It's one thing to point at someone else's car and say it's worthless. It's quite another to then go and sabotage it. (See owner's 2009 letter of complaint and owner's 2006 letter of complaint) I don't think that ethical and we can all agree on that then maybe we can come to some kind of consensus. If we need a policy to help us, I'm all for finding or creating one. But please, may we first bring our ethics to the table? I think it's important that we all show up for the meeting bringing our whole selves. Even soldiers may be permitted to speak freely. In another context, this is sometimes called, "speaking truth to power." for example, standing in the oval office and telling the President of the US what you really think. Danglingdiagnosis (talk) 02:34, 10 July 2009 (UTC)[reply]

I disagree. This is a valid arguement. If the test is not very useful or it´s use possibly harmful than that would refute the arguements of possible harm in showing these images.--Doc James (talk · contribs · email) 03:03, 10 July 2009 (UTC)[reply]
If being the operative word. Consensus in the field, reflected in the fact that 80% of graduate programs teach it and 80% of clinical psychoilogists who are in a position to use it, do so, is that it is a useful test. Actually even its harshest critics (Lillenfeld, Wood, et al) - a minority within the field - don't go so far as to claim it is useless. See this article: [1] "Controversy has surrounded the Rorschach throughout most of its history, not because it is worthless, but because it has so often been used for the wrong purposes." The claim that it is "worthless" is basically nothng more than a fringe belief.Faustian (talk) 03:12, 10 July 2009 (UTC)[reply]
Let me add this to what Faustian wrote: Even articles that are highly critical of the test [See http://www.psychologicalscience.org/newsresearch/publications/journals/sa1_2.pdf 2001 Scientific American] admit that it has some utility, especially in diagnosing Schizophrenia and bi-polarizm. You wouldn't want that functionality to be harmed, would you? So to use my analogy, if we're walking along the street and we see a car, as long as the car is owned and being used by somebody to some effect, (whatever the degree), I don't think it's right to go vandalizing it. Can we agree on that? Danglingdiagnosis (talk) 03:29, 10 July 2009 (UTC)[reply]
Discussing a test does not invalidate it. Wikipedia discusses mathematics. This does not invalidate math tests. Discussing and showing what the Rorschach test is is not in any way like vandalising a car. As another example we have tests ( exams ) we use to determine if someone is having a pseudoseizure or a real seizure. If people know about them then they could better fake a seizure and thus might be put on dangerous drugs as a result. They may than die from side effects from this drugs. I do not in any way see this as justification for hiding this information in medical books or keeping it off wikipedia.
The Rorscharch is not the only potential health care peice of information that may cause harm if used inappropriately. If you research depression to determine how to avoid being detected all the power to you. I have not found this to be a real concern with depression. With anorexia however that is a different matter. Here you have a predominantly young female population who work very hard to foil their caregivers and escape treatment. Techniques for tricking you physician are distributed widely on the internet. Health care profesionals need to know about these techniques.
Those who use this test need to know that the people they give it to may have seen it before. That they may lie and say they havn´t. And what better way to emphasis this is keep the images on wikipedia.--Doc James (talk · contribs · email) 03:50, 10 July 2009 (UTC)[reply]
Better way to emphasise what? You think the choice we must give the testeee is to be truthful or to lie? And you see the use of this article primarily as the means by which a person can "trick his physician"? Martinevans123 (talk) 06:51, 10 July 2009 (UTC)[reply]
The difference here is that the harm is not voluntary, requiring a person to make a choice, but automatic. It happens when someone merely sees the image. This places the responsibility for the harm not on the reader but on the editor, or on wikipedia (indeed, it robs the viewer of the choice to see the image or not to see it). With respect to your examples, would you feel the same way if, theoretically, the act of coming across a particular image of a seizure increased the chances of involuntarily getting one by the viewer? Would you still insist on putting that image into the lead?Faustian (talk) 04:07, 10 July 2009 (UTC)[reply]
Anoxia?? Martinevans123 (talk) 07:00, 10 July 2009 (UTC)[reply]
So Doc, if you and I can agree that the test has some value, (say in cases of diagnosing schizophrenia and bi-polarism) then can you agree with me that harming that value would be unethical. Should Wikipedia do anything that might forcibly "substitute for the advice of a medical professional?" Danglingdiagnosis (talk) 07:20, 10 July 2009 (UTC)[reply]

Discussing this test does not do it harm. Most people who come here I would guess are actually health care providers and others in related feilds not those the test will be used up. Discussing it may actually do the test some good. We discuss the rest of health care honestly and I see no reason why this test is anything special. So no I do not see anything unethical with this page.--Doc James (talk · contribs · email) 22:47, 10 July 2009 (UTC)[reply]

Discussing the test is not the same thing as showing the test materials. One can discuss any test, such as the SAT or GMAT, and not compromise or spoil it by posting actual items from those tests. It's interesting that your claim that there is nothing, in your opinion, unethical about this page contradicts what the field itself states about doing this sort of thing. Any reason why you are right and the field, collectively, is wrong?Faustian (talk) 22:52, 10 July 2009 (UTC)[reply]
While I am a medical preactitioner I am not a Clinical Psychologist, and claim no expertise in the use of the Rorschach test images. But it does seem to me, and please correct me if I am wrong, that in this whole discussion one factor is being overlooked. If we stipulate that the interpreatation of the images is subjective (and the entire value of the test depends on this aspect) and the assessment of the subject by the psychologist depends on this interpretation, then surely this assessment can still be made whether the subject has seen the images before or not. All that is required is the the psychologist be aware of the presence or absence of previous exposure. Given that the assessment of a subject depends not simply on his/her bald indentification of an image, but more particularly how, and in what words and in what manner this identification is made, I would expect a trained psychologist to be able to make this assessment irrespective of past exposure by the subject to the images, given only that the fact of exposure is known. If this were not the case then a subject could be Rorschach tested only once in their whole life; is this in fact the situation? --Anthony.bradbury"talk" 08:10, 11 July 2009 (UTC)[reply]
A good question. But I think that your single factor may be (at least) two separate but related factors - a) that the test is subjective i.e. no right or wrong answers, and b) that it is the pattern of interpretation that is analysed not the pictorial subjects identified. It seems from earlier contributions that the test may certainly be used more than once with the same subject, although I think it was suggested that self-reported pre-test or pre-exposure might lead the practionner to choose an alternative test. But it has not been made clear what are the critera for such a choice. I'm unclear how you "would expect a trained psychologist to be able to make this assessment irrespective of past exposure" since some of those very clinicians have argued here that they can't, for whatever reason(s) - that's the problem, that's the basis for most of this discussion.
I also thought it ironic that this comment was added in this section. I think we may have missed a certain circularity. If, as the practioners advise us, pre-exposure weakens the utility of the test, continued exposure will provide further argument, in the longer term, that the test "doesn't work", i.e. no longer works. There will then be even less reason for protecting the images. I realise that this is not really the fault of wikipedia. The Rorschach's own popularity or notariety in the internet age may be the very cause of its demise? Martinevans123 (talk) 09:29, 11 July 2009 (UTC)[reply]
To Anthony: previous exposure is adjusted for. Nevertheless, it weakens the utility of the test. The norms for the test responses were developed using thousands of people with no previous exposure, across hundreds of studies. The more different that the person being tested is from the normative sample, the less applicable the results based on the normative sample are to the person being tested. Preexposure to any test alters one's responses. Now, someone may argue that norms ought to be developed using preexposed people. Perhaps - but creating norms involves thousands of people in hundreds of studies. It is simply not feasible to recreate this massive body of research involving multiple times to account to every possible type of preexposure - someone exposed once, then again someone exposed twice, or three years ago, or one year ago, etc. Instead, we base norms on a "pure" sample and the psychologist adjusts according to the level of exposure (as well as other fatctors that are different than in the normative sample, such testing setting, etc.). As I mentioned earlier, a good psychologist can adjust to such contamination, just as an experienced driver can adjust to a snow storm. But despite the adjustment the results won't be as good, just as the experienced driver still probably won't get to his destination as quickly or smoothly in a blizzard as he would if the roads were clear. Those placing the image heres are essentially creating blizzards - negatively impacting others' work in a way that negatively impacts others' care.12:19, 11 July 2009 (UTC)
Also to Anthony: I'm trained in the administration and interpretation of the Rorschach according to the Exner Comprehensive System, so please let me explain briefly. The responses that an individual gives to the blots are subjective - whatever you see is whatever you see, and the professional just wants you to honestly engage with the task. The way those responses are analyzed under the CS is much, much more objective than most people realize, and interrater reliability is actually quite good. Both the content and the presentation of the responses are analyzed, for a dizzying array of features. There is also an element of subjectivity in the interpretation, and all interpretations have to be considered in light of everything else we know about a person from their history, presentation, etc, but the subjective stuff is considered to be secondary to the stuff with the validity and reliability data.
Also, there's quite a large difference between an exposure to the test blots multiple times, a week or a few months or years apart, under test conditions, versus a chance to sit down and study the blots for hours or weeks, to try to figure out "right answers" or to determine what one should say in order to fool the evaluator, say, during a custody evaluation or other forensic procedure. Now, I agree that with the current level of published information (on WP and on other sites that explicitly claim to provide information) about how the test is scored and interpreted, someone trying to do this would probably just make a fool of himself, and it may be that this concern is not of sufficient weight to warrant removing or hiding the blots on WP, but I just want to make clear that this is a nontrivial part of what the professionals are concerned about. Mirafra (talk) 06:00, 15 July 2009 (UTC)[reply]
I supposed it's okay to talk amongst ourselves, but the following outside sources support the above articulated claims of loss of utility (to some degree or another) and the need for security:
  1. The British Psychological Society
  2. the designers of Rorschach test materials in 2009
  3. the designers of Rorschach test materials in 2006
  4. The American Psychological Association Code of Conduct (see also preamble)
  5. American Psychological Association, Statement on the Disclosure of Test Data, 1996.
So it's certain that recognized authorities ask for help in maintaining the security of the images. Are there any sources that claim that releasing the images to the general public could actually help? Not that I'm aware of. I don't like to engage in speculation, but let's assume that there are sources that we can cite saying that publishing the images can help. Now, let's ask ourselves the following question: If there was even a remote chance that familiarity with the Rorschach images were beneficial, wouldn't it be more likely that the APA and the British Psychological Society would advocate that the images be advertised on television. This would undoubtedly increase interest in the test, and bring in more business. More people would receive early diagnoses of bi-polarism and schizophrenia. I don't know about schizophrenia, but I have a friend who tells me that if doctors had known earlier that she was bi-polar, the chemical damage in her brain would be less severe than it is today. Even if the test images served only to attract more people into the door and some other method of assessment were utilized, don't you think the APA and the BPS would leap at the chance to do this? Of course they would. But this is only speculation. Clearly no one is advocating this. So I don't think we should either. Danglingdiagnosis (talk) 04:39, 12 July 2009 (UTC)[reply]
Discussing how the test works does not require showing the standard images; a new image can be used when discussing how it works or is used. -- SEWilco (talk) 21:59, 14 July 2009 (UTC)[reply]

Came across this debate once it made headlines on slashdot. Here is my point of view:

The images themselves are not readily available to patients unless:

a. They specifically research them beforehand, or
b. They bring their laptop into the shrink's office and study before the test

Anyone willing to dig the images up from a wikipedia article is likely willing to do a google search for them anyway.

Ethical concerns aside, I question whether such "security through obscurity" is even effective. The fact that wikipedia is but one of many sources for the images in question, combined with the implied tenacity of anyone looking up said images on wikipedia, in my mind works against any argument for their removal.

Shentino (talk) 01:59, 15 July 2009 (UTC)[reply]

It seems to me that it's best to publish the images and let them be abandoned for clinical diagnosis. Example ink blotches similar to the originals have been published for decades, and that doesn't seem to affect the clinical interpretation for people who are exposed to the Official Blotches for the first time. This implies that any set of new, unseen ink blotches would be just as effective as the old ones, once there are enough published studies to compare the results against. It will take time for the same quality of research to build up around new blotches, but in the end I think it's the only way to recreate the same effectiveness that the original botches had. I wouldn't be surprised if there was already research into other suitably secret ink blotches that could be used as a starting point. In short, the clinical community probably should have foreseen the consequences of the blotches entering the public domain and started testing replacement blotches quite a while ago. 216.67.57.174 (talk) 03:38, 15 July 2009 (UTC)[reply]

I'm not exactly clear on why it is that WP editors, who are mostly not apparently operating from a position of knowledge about the test, its administration, or its interpretation, should be making that decision on behalf of the professional community. Other posters have mentioned the immense cost and effort of "just making a new test" or designing some hypothetical self-evolving mechanism that would magically replace ninety years of existing research. It's also unclear to me why this standard of "We can find the information, therefore we should publish it," is being applied to the Rorschach, when there does not seem to be a similar effort applied to other psychological and cognitive tests. Mirafra (talk) 05:33, 15 July 2009 (UTC)[reply]
We're not telling clinical psychologists what they should do, we're telling them what they must inevitably do given that these images are in the public domain. — PhilHibbs | talk 07:56, 15 July 2009 (UTC)[reply]
I apologize if this is an inappropriate place to post these points but, as has been mentioned by others, this page is a labyrinth.
1. The term “test” is broad – much more broad than “experiment” or “measure.” Comparison to other “tests” such as the SAT can be grossly inappropriate. Much of the insight gleaned from the Rorschach test is contextual: the test giver is collecting data from previous sessions, the test taker’s behavior and responses before and after as well as during the test, and so on. These “subjective response” tests are not at all like “quick fire response” or “objective” tests in that the test taker’s responses to the individual items are secondary to their responses over time, their affect and body language during testing, etc. It’s impossible to “study” for a subjective response test because the test taker can read into your responses – if you answer too quickly, or with too much detail, too much conviction, too much recognition, too much “remembrance” and so on, that tells just as much as the actual words that come out of your mouth.
2. What about retesting? I may have missed a comment or two in this thread, but no one seems to have mentioned the fact that many test takes (myself included) have been exposed to and tested with the Rorschach multiple times – the test giver may or may not remain constant, the location may or may not become constant, and, of course, the test taker’s personal circumstances, state of mind, concern, and other traits may or may not be similar to the previous testing date. The Rorschach is in this way similar to the Tarot. You may memorize the entire deck and several books on interpreting the cards, etc, but the interaction between the person read and the person doing the reading, the randomizing mechanism (either by shuffling the cards or simply living a complex life), the order and context in which the cards are presented – all of these factors make every reading “unique.” (I am not claiming that the Tarot is as scientific or standardized as the Rorschach but then again, I’m sure that with appropriate development it could be adapted for much the same purposes by having the subject ‘interpret’ their own cards.)
3. Multiple Exposure In any case, this is not to say that previous exposure isn’t going to affect future exposures. But considering that many people who are being tested will be tested again or have been tested in the past, this is a confounding variable that should not only be accounted for by the tester but which can be as useful and revealing as a lack of exposure.
4. Whose Ethics? As far as Wikipedia and the so-called ethical quandary presented by this issue, the encyclopedia cannot concern itself with the professional concerns of any group, including psychologists. While I am hesitant to propose a slippery slope argument here or anywhere, the fact remains that we cannot cherry-pick our facts. Many of the photos, quotes, and facts published on Wikipedia are inconvenient for some group or another, and quite frankly, just because many psychologist have a professional investment in this particular set of images does not grant them a monopoly on those images (now that the images are public domain, at least).
5. Not The Biggest Fish To Fry Wikipedia is far from the least of your concerns as regards the publication of these photos, but let’s face it – even if this debate does not become generally known, magazines, newspapers, and websites across the globe will pick this up (and the more public this debate becomes, the more this will be an issue). WebMD causes similar problems with self-diagnosis and, worse, self-treatment. And while I find many people’s use and abuse of WebMD annoying, I do not see it as an ethical issue – more information is always better, despite as much as because of the consequences.
6. Wikipedia affecting the "real world" As far as “I'm not exactly clear on why it is that WP editors, who are mostly not apparently operating from a position of knowledge about the test, its administration, or its interpretation, should be making that decision on behalf of the professional community.” We’re not. We’re making a decision on behalf of Wikipedia. Your concerns are not our concerns. I couldn’t even begin to list the thousands of contentious and controversial pages on Wikipedia. I’m sure none of the corporations, politicians, celebrities, or political organizations whose scandals, mistakes, arrest records, and other matters of fact are published here want to see them here. But the fact remains that it is not only our job but our responsibility to publish relevant facts about anything notable. While I understand that it is the responsibility of psychologists to diagnose and treat their clients, you must understand that it is our responsibility to write an encyclopedia. You would never dream of modifying, let alone abandon, your responsibilities for the sake of an online encyclopedia. What makes you think for a second that we’re going to modify, let alone abandon, our responsibilities for you? Again, this is not an ethical issue – we aren’t hiding the truth, we aren’t telling lies, we aren’t misleading anyone, we aren’t publishing anything that isn’t in the public domain, and we aren’t presenting facts that haven’t been verified. Those are our ethical responsibilities. In fact, doing this is our ethical duty.
7. Whose Ethics? Part II Whether or not psychologists, individually or as a profession, will need to design a new test or not based on the public’s newfound access to the Rorschach is irrelevant to this discussion. Even if that is true, that is beyond the scope of Wikipedia’s concern. Indeed, the immediate response to such a concern is simple: if this is such a problem, and psychologists as a whole knew that the test was going to become public domain sooner or later, why wait until AFTER the test became public domain to start doing anything about it? 173.170.88.238 (talk) 12:18, 15 July 2009 (UTC) Richard Paez [redacted email][reply]

Arguments Con

The below discussions are transcluded from /2009-06 Arguments Con

#1 - It may harm a psychologist's ability to protect the welfare of his/her patient.

This conversation can be viewed and restarted at the transcluded subpage /2009-06 Arguments Con#1

These are no longer copyrighted images, and to suggest that they comprise a valid psychological assessment is both inane and immaterial. Immaterial because everyone in the general public is quite aware of the inkblot test, and very few administrations of it are objective. The subject invariably knows what kind of assessment is being given to him, and responds not as he truly feels, but with the answer he wants to give, which he thinks will best serve his purposes. Inane because this test is not only subjective on the part of the subject, but also entirely subjective on the part of the administrator. The administrator brings his own sets of ideas, prejudices, preconceived notions and desires to the test, and applies them subjectively as he sees fit. —Preceding unsigned comment added by 38.103.167.122 (talkcontribs) 02:56, 30 July 2009

... and you were trained in Rorschach administration where? And you've administered how many Rorschachs?Daveandmicasmom (talk) 22:00, 29 October 2011 (UTC)[reply]

#2 - It violates Wikipedia policy

This conversation can be viewed and restarted at the transcluded subpage /2009-06 Arguments Con#2

(no comments since 25 June 2009 - remove this line when restarting conversation)

#3 - It interferes with the workings of another profession

This conversation can be viewed and restarted at the transcluded subpage /2009-06 Arguments Con#3

See response at Talk:Rorschach_test/images/2009-06_Arguments_Pro#.2301_-_The_cat.27s_out_of_the_bag Danglingdiagnosis (talk) 23:05, 31 July 2009 (UTC)[reply]

#4 - It violates Wikipedia policy on neutrality.

Portions of this conversation older than the first comment may be viewed at the transcluded subpage /2009-06 Arguments Con#4

I believe that neutrality dictates that we take a less-active posture, one that does not "reproduce" results favorable to any particular point of view concerning the utility of the Rorschach test. (See definition of neutral.) That psychological tests are so easily vulnerable to vandalism (see 1996 APA statement is unfortunate, but does not take away from our responsibility to be neutral. If you pick up a carton of eggs, you have to watch your step more carefully. Danglingdiagnosis (talk) 00:58, 10 August 2009 (UTC)[reply]

But we didn't pick up a carton of eggs. A carton of eggs was placed in the middle of the street (the public domain), only to later complain that cars are crushing it. True, it wasn't the psychologists who placed it in the stre- public domain, but the law, but that doesn't change the fact the psychologists have known for decades that was going to happen. --LjL (talk) 13:11, 10 August 2009 (UTC)[reply]
I don't accept the premise that we are not responsible for our actions. The problem with your metaphor is that we have spent quite a bit of time next to this carton of eggs. We're not driving 60 miles an hour. We're going quite slow, and our actions are quite deliberate. So, we have to question our neutrality. Danglingdiagnosis (talk) 14:50, 10 August 2009 (UTC)[reply]
That is true. However, I'm pretty sure we have questioned it, for quite a long time (length of these discussions testify it), and the prevailing consensus is that we should go ahead and crush the eggs. --LjL (talk) 18:05, 10 August 2009 (UTC)[reply]
The definition of neutral is something that does not produce (or reproduce) a desired effect, even if that effect is desired by consensus. Neutrality trumps consensus. It's better to give zero information than non-neutral information. Danglingdiagnosis (talk) 18:31, 19 August 2009 (UTC)[reply]
So wrong it verges on ridiculous. Ever read WP:NPOV at all? NPOV is about representing multiple points of view, not suppressing them. --LjL (talk) 19:26, 19 August 2009 (UTC)[reply]
I hear you, but I don't think it's ridiculous. Neither do at least five other Wikipedians. Wikipedia's neutrality should not be compromised by information that cannot be presented in a non-neutral manner. By providing the questions and answers to a psychological test, we are, in effect, invalidating the test and validating the point of view of people, like Garb and Lilienfeld, who question the validity of the test and have called for a moratorium on its use. [2] [3] Garb is, of course, welcome to start a Wikipedia account and contribute, but he must not produce results that reproduce his desired outcome. That would be a violation of our neutrality policy. So if you wouldn't mind, some of us would appreciate it if you took this comment seriously. Danglingdiagnosis (talk) 04:42, 20 August 2009 (UTC)[reply]
Oh I take the comment seriously, I simply find it appalling that people can be Wikipedia editors and have such absurd censorship ideas all while keeping a straight face. I'm dropping the debate at this point, for reasons including apparent legal threats against James. The consensus is very clear and I'm not willing to discuss it anymore, especially with "five other" editors among which are some who used legal threats, which are now apparently put into action. --LjL (talk) 11:48, 20 August 2009 (UTC)[reply]
I hear you, but I don't think the debate we're having here is absurd. I don't think that I am advocating that we violate the WP:NOTCENSORED policy against offensive, profane, or obscene content. Instead, I think we're violating our Wikipedia:Neutral point of view. Neutrality is a core concern of Wikipedia, and to force our readers into a position that limits their ability to choose is non-neutral at the very least. I look forward to discussing possible ways to neutralize this effect on the reader, but I fear the best option is to remove the information. Better to have zero information than information that is non-neutral. Danglingdiagnosis (talk) 16:00, 20 August 2009 (UTC)[reply]
By that logic we should delete half the encyclopedia. Scary censoreship logic DD. Garycompugeek (talk) 19:19, 20 August 2009 (UTC)[reply]
I think I see your point. Allow me to rephrase: Better to have a blank article than an article that is non-neutral. If, as in most articles, it were possible to neutralize the effect of the information and allow the reader to choose, I'd be for it. Danglingdiagnosis (talk) 01:26, 21 August 2009 (UTC)[reply]

#5 - It violates Wikipedia policy on indiscriminate collection of information.

Portions of this conversation older than the first comment may be viewed at the transcluded subpage /2009-06 Arguments Con#5


They are pictures of the subject itself, and there are only 10 of them: the information is important and small in amount, therefore not indiscriminate, as it is highly relevant to the subject. There might be a case here if there were 300 pictures, each individually pictured. But the small number of them shows that this particular bit of information is limited, not excessive and indiscriminate. --Mysidia (talk) 02:27, 15 July 2009 (UTC)[reply]
The problem is that you've published the entire test. I don't see similar publications of content on any other cognitive or projective test pages (TAT, WISC, etc), nor a push to do so. A description of the test and a sample non-test blot would be quite sufficient to communicate "what an inkblot is."Mirafra (talk) 06:41, 15 July 2009 (UTC)[reply]
The subject of this article though is the test though, and it is not just an article about inkblots. Thus showing just a single inkblot as an example could be considered insufficient. As for the thought that other tests are not included in their entirety, that has more to do with the status of the inkblots used in this test now being in the public domain, as opposed to most other mental health or childhood development tests which are still under some form of copyright protection. --Raukodraug (talk) 17:41, 15 July 2009 (UTC)[reply]
Sorry, I wasn't clear -- "what an inkblot is" was a bit flip on my part. I'll be more clear. On the many other pages covering the many other cognitive, academic, and projective tests in common use by psychologists, in every case where I have checked, test security was not apparently violated in any way. I didn't do an exhaustive search, but I checked a whole bunch of the obvious ones. The descriptions of test structure and item type were well within the boundaries of what is generally disclosed by test publishers or professionals who write reports interpreting test data, and would not raise any hackles under the "need to protect test security" requirements of the APA code of ethics. Whether by laziness or accident or intention, WP editors have thus far respected the boundaries of the psychological profession, rather than setting themselves directly in contradiction.
I understand that the specific legal status of the Rorschach (what with the mess that is international copyright law and treaties) is a complicating factor, but there exist copyright-legal and internet-feasible ways to trash the security of other tests in common use today as well, and I don't see any push to do that. In this case, by contrast, the entire Rorschach test has been published -- the fact that it's a test with only ten items doesn't mean that this is not a potential violation of WP:IINFO, any more than the publication of the complete lyrics of a song is dependent upon the number of verses in the song. This is a different standard of disclosure than is applied to other protected psychological tests on WP.
The purpose of explaining the history and structure of the Rorschach and giving an encyclopedia-relevant level of basic information about how the test is administered and interpreted could be just as easily served with an explicitly fake blot. In fact, it might be easier for the expert editors to give *more* useful information by using a fake example, because then we would be less worried about the problem of extensive exposure to the specific blots themselves and would feel more comfortable in clearly explaining what we *can* explain under our ethical guidelines. Mirafra (talk) 19:51, 15 July 2009 (UTC)[reply]
I don't believe the Rorschach test has any test security; the pictures are definitely available. Any member of the public can actually buy them and possibly look them up in the library. The fact that other articles don't include pictures of the test elements does not mean it's Wikipedia rules not to publish pictures. Other articles are in different stages of development. Take a look at Talk:Thematic_Apperception_Test, note the questions from editors about whether the images are copyrighted. If they are not, I fully expect in the future, the other articles about projection tests to include images, however I wouldn't expect all 30 cards to be pictured, it would be far in excess of what is useful to explain the subject. (5 to 10 pictures seems plenty). If a popular song were very short, for example: 4 or 5 lines long, it would not be indiscriminate information for the article to simply include its entire contents. --Mysidia (talk) 12:35, 16 July 2009 (UTC)[reply]
Actually, not just any member of the public can purchase the Rorschach images. There are security features in place. Psychological Assessment Resources sells the plates for $100/each to only people with certain qualifications. Qualifications level C requires that you have an
"advanced professional degree that provides appropriate training in the administration and interpretation of psychological tests; OR license or certification from an agency that requires appropriate training and experience in the ethical and competent use of psychological tests."
So we can say that the images are protected. You can't just buy them. Danglingdiagnosis (talk) 02:16, 17 July 2009 (UTC)[reply]
Just because one seller imposes restrictions doesn't mean others do. Try a google search for 9783456826059. When I performed that search, hit #1 was an online bookseller[4] that has copies of the book listed for sale. --Mysidia (talk) 20:23, 18 July 2009 (UTC)[reply]
Those are relatively unusual situations, that a set of plates (it's not a book) is available in that way. What generally happens to create these situations is that a psychologist dies without leaving a professional will, so that the office and such are cleaned out by people who don't know what they've got, and are unaware that these are not supposed to be sold on the open market. There is ongoing concern in the professional community about this problem, because it affects all sorts of tests. They tend to show up on eBay. Now, eBay does have a policy restricting the sale of "teachers editions" of textbooks, but they have not thus far explicitly extended that policy to psychological tests as far as I know -- the APA and the test publishers were talking to them about it. In general, if a professional lists a test for resale on eBay, they check credentials before completing the sale (that's how I bought my own plates, actually) -- many will list the need to check credentials explicitly in the sales listings. So I think Danglingdiagnosis's point stands -- just because someone can get a random copy of something on what amounts to a black (or at least gray) market does not mean that this is a general "sales policy." For all tests covered under section 9.11 of the APA code of ethics, publishers and licensed distributors of the tests check credentials before making a sale. Mirafra (talk) 01:15, 19 July 2009 (UTC)[reply]
The qualifications required to purchase psychological test materials come awfully close to the language of things listed in WP:NONFREE. Not for commercial use is a complicated legal status that I don't pretend to understand. But since Wikipedia is already imposing stricter standards on itself than mere copyrights, I wonder if there isn't some form of status that Wikipedia might recognize and be written into this section of policy? Danglingdiagnosis (talk) 08:45, 19 July 2009 (UTC)[reply]
I think that's a very good point. The policy says, "free as in cost and free as in freedom are two entirely different things, images freely available on the internet may still be inappropriate for Wikipedia." Perhaps the better thing to do is to propose an edit of the existing WP:NONFREE policy to mention that there is a class of information (restricted-access psychological tests) which should not be posted? Mirafra (talk) 02:34, 21 July 2009 (UTC)[reply]
I find this line of argument totally ridiculous. First First-sale doctrine would shoot down any efforts by anybody to control books once they are sold. Second the Rorschach test images have been published before "The Nuremberg Mind" (1975) and Poundstone's "Big Secrets" (1983) Finally nevermind the dozens of websites and online news articles the images are on including the web site the images came from. I even found one that has had them on the internet since 2003. This horse has been out of the barn for a long time.--BruceGrubb (talk) 12:36, 3 August 2009 (UTC)[reply]

#6 - It violates Wikipedia policy on non-free content.

This conversation can be viewed and restarted at the transcluded subpage /2009-06 Arguments Con#6

Something that someone else said got me thinking. And I looked back at the page about WP:NONFREE... and noticed the following prominently-placed sentence. "On Wikipedia, our goal is to be a free content encyclopedia, with free content defined as any content granting the right to redistribute, study, modify and improve, and otherwise use the works for any purpose in any medium, even commercially." Protected psychological tests are not freely permitted to be redistributed or used for any purpose in any medium. You have to have an advanced degree with specific training in the use and interpretation of protected tests in order to purchase them. If you are not appropriately licensed, you can do jail time for trying to use them. Even copies of outdated tests are supposed to be kept at least somewhat protected, because there are not infrequently reuses of items or item formats between editions. We're not talking just about the Rorschach here, we're talking about all protected tests. WP accepts that some information is not just happy hippie-dippy information dancing with infinite abandon around the intertubes. We're asserting that protected psychological tests fall within that realm, and we've backed that up with strong sources: policy statements from relevant well-recognized professional societies. That's not censorship, that's WP following its own self-description and holding itself to its own standards. (originally posted elsewhere by Mirafra (talk) 03:31, 21 July 2009)Faustian (talk) 04:30, 21 July 2009 (UTC)[reply]

Indeed, Wikipedia's employs standards that are intentionally stricter than just copyright laws. See WP:NONFREE. Danglingdiagnosis (talk) 15:27, 21 July 2009 (UTC)[reply]
I think there may be a distinction between "distributing a test" and "distributing the images used in a test". The images are free, although the test, presented in a usable form, may carry restrictions in some jurisdictions under laws other than copyright. — PhilHibbs | talk 19:35, 21 July 2009 (UTC)[reply]

#7 - Such vandalism defeats the purpose of an encyclopedia.

I suggest we rename this to "Showing the images defeat the purpose of an encyclopedia"--Dela Rabadilla (talk) 01:30, 6 August 2009 (UTC)[reply]

A long time ago, someone defined the purpose of an encyclopedia.

Indeed, the purpose of an encyclopedia is to collect knowledge disseminated around the globe; to set forth its general system to the men with whom we live, and transmit it to those who will come after us, so that the work of preceding centuries will not become useless to the centuries to come; and so that our offspring, becoming better instructed, will at the same time become more virtuous and happy, and that we should not die without having rendered a service to the human race in the future years to come.

— Diderot, editor and publisher of Encyclopédie 1751 - 1772

Three secondary sources, position statements by major health organizations (i.e. the best kind of sources according to WP:MEDRS), have stated that the security of test material is important to the work of psychologists and should be maintained. It follows that Wikipedia would be vandalizing their work by breaching that security. This is the same conclusion reached by Scientific American in its 2001 article on the subject. Such vandalism is contrary to the purpose of preserving "the work of preceding centuries." I want our children to receive the benefit of the Rorschach test. Too many of them may need it. ( 0.7% Schizophrenia, 1-2% bi-polar disorder, etc.)

Is the test perfect? Are there valid criticism of the test? Yes. And we can and should report those criticisms, for to hide such criticism would be censorship. As Denis Diderot said about his encyclopedia, "All things must be examined, debated, investigated without exception and without regard for anyone's feelings." But even he, I'm sure, would draw the line at vandalism. Just because a car parked in front of someone's house isn't the quality vehicle you may wish, that doesn't give you cause to slash its tires.

Many of us have seen and fought against vandalism on Wikipedia. So you should be sympathetic when the tables are turned and Wikipedia becomes the source of vandalism to the work of others. Danglingdiagnosis (talk) 16:31, 24 July 2009 (UTC)[reply]

The use of the word vandalism to describe publishing a thing is a flawwed analogy, there is an extreme dissimilarity between the things being compared. Publishing now-freely available images that were used in a test is no more vandalism than investigative reporting, or other publication of information in the public interest. It's not vandalism if other parties publish things about Wikipedia within their own publications, even if they are harmful, even if Wikipedia doesn't want them published, and that applies to any public object: it's not vandalism for Wikipedia to print images like the ones attached to the article. --Mysidia (talk) 04:39, 31 July 2009 (UTC)[reply]
Inclusion of the images in fact aids in the collection and dissemination of knowledge, and serves to preserve information about the test. It does not defeat the purpose of Wikipedia. Perfect test security would ensure that no information about the test is published, and the result is, knowledge of the test is lost of the centuries.
Therefore based on the definition above, publishing all details of the test would in effect "transmit it to those who will come after us", and be more inline with that definition than not publishing.
The "definition" of an encyclopedia above contains rationale that is not accepted as part of the definition of an encyclopedia.
There is no robust test security anymore. The images have already been available to the public freely for long enough that even if Wikipedia removes them, they will be freely available at numerous online sources, including public archives that exist of older versions of Wikipedia image files. --Mysidia (talk) 04:39, 31 July 2009 (UTC)[reply]
Not really. To the extent that previous norms are damaged wikipedia has destroyed that knowledge. For example, if previous research has shown that only 30% of depressed people see a mouse on the first card (this is not true, I'm making up a much less complex hypothetical here) and wikipedia puts this up, it may be that now 40% of depressed people see the mouse. Wikipedia has rendered the previous findings and whatever benefit they could have given, useless. This is the equivalent of describing a house by tearing it apart in order to catalogue where all the pieces are. By the time you're finished, there is no more house and the description is of course obsolete and false.
To the extent that a test becomes less useful due to its publication, wikipedia destroys the ability of the test to add to more knowledge and renders the previously obtained knowledge meaningless.Faustian (talk) 05:17, 31 July 2009 (UTC)[reply]
The average layperson doesn't even recognize the name "Rorschach test". It's an extremely unlikely proposition that a statistically significant percentage of the general public actually reads the article and actually looks at the images; so long as Wikipedia doesn't publish the full details of the test, the methods of interepreting answers, or expected answers for each image, the test has not been rendered useless, and the impact is minimal, except maybe for a small number of individuals who have read the article and studied the images in great detail, so they can make contrived answers. I would expect there to be extroardinary proof that the publication of just the images invalidates the results of the test to justify retracting the images fron an article, and those 3 sources are not extroardinary evidence. As always, professionals utilize multiple tests, not just one, and more information about one test being published can be utilized to assist in developing other tests based on it, a net benefit to the collection of human knowledge. --Mysidia (talk) 22:54, 31 July 2009 (UTC)[reply]
Mysidia, I don't see the flaw in my use of the word "vandalism". You're looking at the word through the definition of Wikipedia:Vandalism. I'm using the general definition found in wiktionary:vandalism Vandalism is the impairment of property, even shared property which is in the public domain (such as the Rorschach inkblots), such that it does not look or function as it was intended.
Psychological tests are not designed to be generally published. Wikipedia is. They are designed differently. If you put a boat in the lake, it floats; you haven't harmed it any. If you put a car in the lake, that's vandalism.
I understand you have a personal belief that information is best published for everyone to see. But that's just a general rule-of-thumb, isn't it? I have 3 sources that say otherwise. [5] [6] [7] They're good quality sources that say that psychological tests need to be kept reasonably secure. Publishing in Wikipedia defeats that purpose and is, thus, vandalism. Worse, it vandalizes a base of knowledge, something that Wikipedians, such as yourself, should appreciate. We should be guardians of knowledge, not vandals of it. Like museum curators, we should put as much as we can out for public display, but not in a way that harms or destroys the item. Because we want our children to reap the same benefit as we have. Danglingdiagnosis (talk) 20:13, 31 July 2009 (UTC)[reply]
Point one: I see that your definition of "vandalism" is quite valid, but I wonder why some people here (and in other related talk page) insist on using word "not in their Wikipedia-specific definition". I though we were on Wikipedia.
Point two: given that we are on Wikipedia, I am ready and willing to rely on reliable sources to write articles. But nowhere on Wikipedia does it say that any sources should be using for deciding what's ethical for Wikipedia, or what's vandalism. That's up to the community to decide, not sources.
--LjL (talk) 00:24, 1 August 2009 (UTC)[reply]
Point one: I agree that some of us are finding the word confusing because we normally use it with reference to Wikipedia policy and not general usage. However, the other options are not very good. I just opened up my thesaurus and found "sabotage." I'm happy to substitute this word, but I don't wish to imply that anyone is intentionally damaging the test. I think what we have here is more a matter of "Oops. Sorry about the problem I caused, but I have other more important things to consider." That sounds to me like vandals who are just doing their thing. Which kind of makes me the old man yelling out the window, "Hey, you kids, get off my lawn!" I swore it would never happen, but it did: I've become that guy down the street.
Point two: If we don't use reliable sources for our information, who do we use? The community? You and me? I think that's just wrong. You can try and make a distinction between information we use solely in our discussion pages and how it trickles up to our articles, but I worry about going down that road. You're saying that policy only applies to what we do on our articles and not to how we discuss them. I think there are plenty of policies for how we are supposed to act in our discussions. Why some and not others? I also think the better we organize our talk pages, the better will be our articles, because one derives much from the other. Think about it... If we don't apply some pretty basic rules to our discussions, like fact checking and attribution, then I shudder to think what will happen as the results trickle up to our articles. At the very least, it will result in articles that change drastically over time, depending on who's doing the talking, which is what is happening here at the Rorschach test.
Which brings me back to the argument in this thread. By going down this road, we've begun to compromise basic principles to the operation of an encyclopedia. I think we need to take a step back and really consider what it is that we are trying to do. I think we should stick to writing an encyclopedia with general articles that inform us about real-world context, (i.e. information that is relevant but not exhaustive) and that does not destroy knowledge or utility. Because, really, what is knowledge without utility? Danglingdiagnosis (talk) 02:56, 1 August 2009 (UTC)[reply]
"Why some and not others" - because that is what the guidelines and policies say. You may perhaps like guidelines, policies and talk pages to be reliably sources,[citation needed] but then you're thinking about a different encyclopedia,[citation needed], because that is not how this one works.[citation needed] Is there any reliable source that says that "consensus" is the Most Important Thing Ever? Maybe, but in any case that's not what was used to decide that consensus was important on Wikipedia. Policy here has never been built based on what "sources" said about policy. --LjL (talk) 14:26, 1 August 2009 (UTC)[reply]
Talk pages, according to Wikipedia:Talk page are not for discussion personal views. They are for discussing content on the article. In doing so, we are informed by outside sources. This must be so because we are not supposed to rely on our own expertise. Therefore, if a reliable source says that our article is damaging a test, you are not free to disregard this information. You must accept this as verifiable fact, find another source that refutes this, or perform original research. You may hold this fact in dynamic tension with other considerations and apply judgment about weight, but you may not disregard it. Danglingdiagnosis (talk) 00:16, 2 August 2009 (UTC)[reply]
Ideally, article talk pages aren't about discussing policy, either. If policy says that something should or should not be in an article, then there is nothing to discuss on a talk page; at best, you discuss on the policy page.
I'm not necessarily saying that we should adhere to the above strictly... but the point is that, while we should "not" disregard infrmation such as a source saying that our article is damaging a test, we should, if anything, use that information to put it in the appropriate article, not to dictate policy. I repeat, not to dictate policy. This seems so obvious to me I never thought I'd have to repeat it... If you want, though, we can bring it to a more appropriate venue. --LjL (talk) 00:25, 2 August 2009 (UTC)[reply]
Ah, I see, now, where we are thinking differently: I think policy gives us the freedom to use our discretion. You think you are bound by policy to act in a certain way. Of course, you're free to consider policy or anything else when considering the facts. But Let me free you. From Wikipedia:What "Ignore all rules" means:
"The spirit of the rule trumps the letter of the rule. The common purpose of building a free encyclopedia trumps both. If this common purpose is better served by ignoring the letter of a particular rule, then that rule should be ignored."
So, I wasn't aware that I was discussing policy. I think I'm discussing content and using some pretty basic concepts about how one should speak and use (or not use) references. Regarding content, let me ask you, how do you regard the information that we are damaging the test? Facts are facts, and I don't wish you to ignore them. Are you okay with the idea that Wikipedia is disclosing information that may harm or destroy the utility of the test? ...because I'm not. Who are we to interfere? We are encyclopedists. We're supposed to observe and report, not influence, and certainly not destroy the utility of information. Danglingdiagnosis (talk) 15:18, 2 August 2009 (UTC)[reply]
How many times do I have to say it? Yes, yes, YES. It's not Wikipedia's problem whether the information it provides somehow damages the test. Wikipedia isn't sponsored by the APA, it doesn't owe anything to the test. It's not interfering; actually, psychologists are currently interfering with Wikipedia by bringing their conflict of interest with their code of ethics here, and even forcing administrators to full-protect the article.
Wikipedia's purpose is to report, but who said reporting won't have effects on society? I'm pretty damn sure Wikipedia has already had quite an important effect on a lot of people, simply by bringing so much knowledge and information to easy access. Of course that's perfectly fine.
--LjL (talk) 14:08, 3 August 2009 (UTC)[reply]
It's your arguments not how many times you say anything. The objective of any encyclopedia is to gather knowledge for the benefit of humankind. It's been shown many times that dissemination of knowledge is a good thing. That is our default notion. What you fail to realize is that even simple ideas like the one I just stated have exceptions. History has also shown that our most basic assumptions sometimes have exceptions. Newton's law was fine for most human experiences for thousands of years, today we know now that at high speeds bizarre things happen. There has never been an encyclopedia like the Wikipedia and the images on this article challenges the basic assumption that all knowledge is good for human kind. Because it hampers the ability of our physicians to take care of us. It's a weird case there are few cases like these.--Dela Rabadilla (talk) 01:30, 6 August 2009 (UTC)[reply]
No excuse me, if the question is "are you okay with the idea...?", the answer is "yes, I am". I am being asked for an opinion; and I had given it a number of times.
As to the content of your objections, I'll repeat things I've already said: 1) not all parties seem to agree there's harm in the accidental viewing of the images (more so in deliberate use of the image to "cheat" the test, but it's not up to us to prevent that) 2) no one has demonstrated that disseminating test materials won't encourage scientists to successfully develop other tests that are resilient to dissemination.
What do where do when there are disagreements and incognita? We jump back our "default" stance, which is well documented and which you aren't in general disagreeing with. --LjL (talk) 13:10, 6 August 2009 (UTC)[reply]
But Wikipedia does owe something to the test. Every time I write for an article, my responsibility is to the subject. I take great care so that I may pass on knowledge that is correct and useful. I consider my reader and ask, "how will they use the information." But if 5 years from now, nobody can use a Rorschach test, then we will have failed our purpose. Any why? Because we failed to recognize our circumstances. Do you know the fable of The Scorpion and the Frog? You seem to be saying "It's my nature to act a certain way." But there arises special situations in which it is in everyone's best interest if he or she reassesses his nature. The human mind is the most adaptable of any on the planet. Adaptability and ingenuity is the reason for our success as a species and for the failure of other hominid species. But we still get into trouble when we let custom dictate our actions. As the poet Shelly wrote, "But custom maketh blind and obdurate the loftiest hearts." We have a unique situation here. It's best to recognize that and make allowances. Danglingdiagnosis (talk) 07:01, 4 August 2009 (UTC)[reply]
That requires believing the test is useful, the test will be unrepairably broken by publishing information about it, and the test cannot be replaced by a more robust test (the typical CS argument against security by obscurity, and people can keep saying it just doesn't apply to psychology, but I have yet to hear a good reason why). That's a lot of things to believe on their face. I don't necessarily believe they are true. --LjL (talk) 22:13, 4 August 2009 (UTC)[reply]
I don't think you or I have the authority to force medical professionals to replace the test with another. That you would dare suggest otherwise is interesting, but such opinions don't belong on Wikipedia. Perhaps you could write an article as a commentary and submit it to an on-line magazine. Or if you have a medical license, you could submit it to a professional journal. Danglingdiagnosis (talk) 15:35, 5 August 2009 (UTC)[reply]
Everyone who can read can see that is not what I said. --LjL (talk) 20:19, 5 August 2009 (UTC)[reply]
What you said was that you have trouble believing that a)the test is useful, b) the test will be unrepairably broken by publishing information about it, and c) the test cannot be replaced by a more robust test. Here's my response: Your understanding about these things is not required. You simply need to accept a source (in this case, a statement by a national health organization) and stop relying on your own understanding or your own original research. We live in a big world; even people with encyclopedic knowledge can be forgiven if they must sometimes rely on others for their information. In this case the information tells us that a) the Rorschach test is considered to be useful, b) we are vandalizing psychological tests, including the Rorschach, by disclosing test data, and c) that doing so may invalidate the results and require the substitution of other tests. I believe that such vandalism defeats the purpose of an encyclopedia and should stop. If you continue to have doubts about the appropriate use of the Rorschach test, then I encourage you to present those arguments in a peer-reviewed psychological journal. Danglingdiagnosis (talk) 18:51, 6 August 2009 (UTC)[reply]
Problem is that as much as you may insist otherwise, I still don't believe that it's expected (or beneficial) to use sources to decide what to include on Wikipedia (or otherwise what policies to have). Wikipedia uses sources for its content; it uses editor consensus for its policies. You may not like that but I do, and I'm pretty sure a vast majority of Wikipedians do, as well. --LjL (talk) 20:46, 6 August 2009 (UTC)[reply]
I'm using sources to inform my decision about whether content advances the mission of providing a free encyclopedia. If consensus shows that we are harming the subject of the article, then I believe Wikipedia:Ignore all rules applies. Danglingdiagnosis (talk) 22:38, 9 August 2009 (UTC)[reply]
You're most certainly absolutely free to base your opinion (which counts, like all opinions do) on. You just can't expect others to do the same, because that's not required. I believe, additionally, that the current RfC quite clearly shows that consensus wants the images to stay. --LjL (talk) 22:47, 9 August 2009 (UTC)[reply]
Consensus can change Danglingdiagnosis (talk) 16:40, 11 August 2009 (UTC)[reply]
Oh please! The RfC was just created, WP:CCC doesn't exist for you to be able to claim that the consensus of 10 minutes ago isn't valid. --LjL (talk) 17:30, 11 August 2009 (UTC)[reply]
Let me back up to something you said earlier. You said you don't believe that it's expected (or beneficial) to use sources to decide what to include on Wikipedia. But I think you're mistaken. We make content decisions on article discussion pages; that is, in fact, the primary purpose of a discussion page. And according the the guidelines for discussion pages, editors are supposed to utilize rules of evidence and verification. "The policies that apply to articles apply also (if not to the same extent) to talk pages, including Wikipedia's verification, neutral point of view and no original research policies." Therefore, it's entirely valid for me to cite a source to support an argument I make on a talk page. Just as it is valid for you to cite policy. Doing either does not end the debate. On the contrary, the debate can then begin. So let us begin, recognizing the both of us have verifiable sources to back up our statements and that neither of us are engaging in personal opinions or viewpoints when we cite our sources. Danglingdiagnosis (talk) 10:20, 12 August 2009 (UTC)[reply]
Actually, the primary purpose of a talk page would be to collaborate in the improvement of the article. And the issue here is we're not really discussing that; you could say we're all at fault. What we're really discussing is policy. I think I've been pretty careful to always say "policy" rather than "talk page", and there's a reason for that - you, yourself, have started a policy proposal about this, so you clearly realize yourself that this is really a matter of policy. It shouldn't even belong to this talk page. After all, the very guideline that you cite does also say "[...] it is usually a misuse of a talk page to continue to argue any point that has not met policy requirements". This one clearly hasn't. --LjL (talk) 13:12, 12 August 2009 (UTC)[reply]
I believe I am trying to improve the article and bring it more in line with our mission, which is the preservation of knowledge; not the vandalism of public property. Danglingdiagnosis (talk) 17:06, 12 August 2009 (UTC)[reply]

#8 - How Dare we Not?

I'm sympathetic to my fellow Wikipedians who find the very idea of withholding information to be distasteful and contrary to the spirit of the free flow of information. I listened to you and heard the visceral disgust you expressed: How dare we decide who should and should not have access to information? How dare we, indeed! I respect the place where that reaction comes from.

Unfortunately, I don't think we have the luxury of avoiding the reciprocal question: How dare we not take responsibility? If not us, then who? Can the reader take responsibility? No, because a) we can't assume they know what a Rorschach test is prior to reading the article, b) the consequences are involuntary, and c) we don't warn the reader about the consequences. Can the psychologists take responsibility? No, because they also have no choice but to substitute an alternative procedure. And parenthetically, whether they should've, could've, or would've anticipated this problem and created a copyrighted, alternative test, one that doesn't rely on the good graces of reporting agencies like Wikipedia, is not something many of us are qualified to second guess and is "water under the bridge" at this point. This means that the people who are most accountable for their actions is us, Wikipedians. We can't shirk the responsibility. Other sites may be publishing the images, but our readers are our readers and no one else's. Let other sites do what they will. We gotta do what we think is right.

I know many of you fear doing anything that seems like censorship. The twentieth century saw enough of that with certain despotic rulers. I argue that withholding images is not censorship because we aren't withholding either criticism or praise. It's simply information: the questions and answers to a diagnostic psychological test. "How dare we withhold information?" you say. To which I reply, "How dare we not." Danglingdiagnosis (talk) 16:40, 11 August 2009 (UTC)[reply]

"The twentieth century saw enough of that with certain despotic rulers." I call reductio ad Hitlerum (and Godwin's law). Axl ¤ [Talk] 23:02, 21 August 2009 (UTC)[reply]
I wasn't trying to accuse anyone of anything, and I did not mean that sentence in the way that you took it. Please feel free to disregard it. The rest of the post is, hopefully, more to your liking. Danglingdiagnosis (talk) 04:59, 22 August 2009 (UTC)[reply]

Recently archived discussions

Page protected
Archived to /Archive 7#Page protected
Question
Archived to /Archive 7#Question - if necessary, continue discussion below
Use show/hide mode for Rorschach Test inkblots?
Archived to /Archive 7#Use show/hide mode for Rorschach Test inkblots? - if necessary, continue discussion below
Including all 10 Argument #5
Archived to /Archive 7#Including all 10 Argument #5 - if necessary, continue discussion below
New expert sign-up section
Archived to /Archive 7#New expert sign-up section - if necessary, continue discussion below
Consensus
Archived to /Archive 7#Consensus - if necessary, continue discussion below
Old debate
Archived to /Archive 7#Old debate - if necessary, continue discussion below
Reverting, status quo, and survey re inkblots
Archived to /Archive 7#Reverting, status quo, and survey re inkblots - if necessary, continue discussion below
An informal survey for the inclusion of the 10 inkblots
Archived to /Archive 7#An informal survey for the inclusion of the 10 inkblots - if necessary, continue discussion below
Harm
Archived to /Archive 7#Harm - if necessary, continue discussion below
To be honest...
Archived to /Archive 7#To be honest... - if necessary, continue discussion below
Are all 10 inkblots really necessary?
Archived to /Archive 8#Are all 10 inkblots really necessary? - if necessary, continue discussion below
list of reliable sources mentioned here
Archived to /Archive 7#list of reliable sources mentioned here - if necessary, continue discussion below
We need an image of the test being administered

I think the best way to compromise and partially satisfy the concerns of all involved is to obtain a more descriptive image for the lead. I think that most would agree that a high-quality image of the test being administered would serve that purpose. Such an image would effectively describe (at least the administration of) the test. The inkblot should be visible in the picture, but since it won't be the main focus of the whole image, it should at least give the reader an opportunity to read our newly minted final lead paragraph before observing the inkblot(s) in detail. –xenotalk 15:35, 23 June 2009 (UTC)[reply]

Comments archived to /Archive 7#We need an image of the test being administered - if necessary, continue discussion below
External links
Archived to /Archive 8#External links - if necessary, continue discussion below
False inkblot as the lead image
Archived to /Archive 8#False inkblot as the lead image - if necessary, continue discussion below
Just a suggestion
Archived to /Archive 8#Just a suggestion - if necessary, continue discussion below
Is source for inkblots reliable?
Archived to /Archive 8#Is source for inkblots reliable? - if necessary, continue discussion below

A request to the regulars regarding the below RFC

To the regulars: Let's try and limit our participation as much as possible in the below RFC I've just filed. I think we all know very well where each other stands - and let's see what the wider community thinks. Feel free to respond to questions from new faces, but at the very least, please do not ping-pong back and forth with other regulars. –xenotalk 02:00, 5 July 2009 (UTC)[reply]

RFC: Should the potential for harm to result inform our editorial decisions regarding encyclopedic content?

Should the potential for harm to result inform our editorial decisions regarding encyclopedic content? –xenotalk 02:00, 5 July 2009 (UTC)[reply]

The debate ongoing above centers on a single issue: harm.

It is posited that disseminating Rorschach inkblot images may result in harm, due to pre-exposing the reader to the images and possibly tainting the results of a Rorschach test they may take in the future. Because one can't "unsee" an image, having the image in the lead of the article does not give the reader the opportunity to read about potential harm that may result before seeing it. It is also suggested that we not display all ten images in the article to further reduce potential harm.

Both the American Psychological Association and the British Psychological Society have gone on record that harm may result to the general public as a result of dissemination of test materials (in general). It should also be noted that the (potential) harm is "passively transmitted", i.e., it's not the same type of harm that might result from explaining how pipe bombs are constructed. This is also distinct from a 'content spoiler' - in that while readers should be reasonably expect to find a detailed plot summary about a work of fiction, in this case they may not know ahead of time that pre-exposure to the image(s) may impact potential future test results.

Iff we are willing to accept harm as a possibility, even if slight, should we then heed these concerns and compromise to limit or otherwise restrict our content to reduce the potential for harm? –xenotalk 02:00, 5 July 2009 (UTC)[reply]

I think we are talking about a few different forms of harm. Let me give two that concern me:
* Given that the Rorschach is often used in forensic settings, where people might have specific desires to present false information, either to malinger (fitness to stand trial, etc) or to fake-good (custody, etc), giving people the opportunity to carefully mull over the blots has to be considered as a possibility for harm. I don't like raising scare-tactics, but the point is that some of what psychologists do with this test is evaluate potential "bad guys" in order to serve the public good.
* Contrary to popular belief, the test is both a cognitive and a projective test. The cognitive aspects of the test assume that the processing of the information require a certain amount of effort, which is why we want the stimuli to be relatively novel. They don't have to be completely novel, which is why people can be retested without serious loss of validity, but if someone could pore over them carefully for hours, this would remove the novelty and change the nature of the test task for them, thus reducing validity. There are other neuropsychological tests that are vulnerable to this kind of exposure and would suffer similarly in their usefulness as well. Mirafra (talk) 20:16, 15 July 2009 (UTC)[reply]

Points of order

As far as I know the APA has not made a statement about harm resulting from showing these images, I am not sure about the BPS. The APA source only prohibits psychologists from making test material available to the public. It does not mention harm or go into motive at all, nor does it mention the Rorschach test. Beyond these important corrections, I support this debate. Chillum 02:05, 5 July 2009 (UTC)[reply]
Sorry, I amended it to say "test materials (in general)" [8]. You are right in that we haven't been presented with any statements about Rorschach images specifically. –xenotalk 02:08, 5 July 2009 (UTC)[reply]
It was more my point that the sources presented for the APA have not made mention of harm, or any mention of the motive of this rule. Chillum 02:10, 5 July 2009 (UTC)[reply]
The final lead paragraph does reliably source assertions of harm from dissemination of test materials in general: The APA states that the dissemination of test materials "imposes very concrete harm to the general public" as well, in that "there are a limited number of standardized psychological tests considered appropriate for a given purpose"xenotalk 02:12, 5 July 2009 (UTC)[reply]
Which source is this? The one I am thinking of is much more limited. Chillum 02:15, 5 July 2009 (UTC)[reply]
"^ American Psychological Association, Statement on the Disclosure of Test Data, 1996." quoted in [9]xenotalk 02:19, 5 July 2009 (UTC)[reply]
I see. In that case point conceded. Chillum 02:20, 5 July 2009 (UTC)[reply]
Superceded?

Does anyone agree this RFC has been superceded by the one running at WT:Involuntary health consequences ? –xenotalk 15:10, 13 July 2009 (UTC)[reply]

Someone has taken the RFC tag down. I think Dangling should incorporate the opinions of the people who respond to this RFC and try to make a more refined policy proposal after this one closes. –xenotalk 04:02, 14 July 2009 (UTC)[reply]

Comments

  • I have thought about this and my answer is no. Wikipedia is not censored directly applies here. We should not hold back relevant, informative and verifiable information from people seeking because we feel it is for their own good. The images in the the form of small thumbnails and the very first section describes the concerns about previewing the images. People don't have to stare at them and interpret them, they don't need to click and see the larger version, but if they want to they should be able to. By the logic given here we should not explain the weaknesses in a polygraph test because a criminal may get loose. We should not explain how Cold reading works because it may prevent a psychic from counseling a customer. The fact is that pretty much any test is invalidated if you look up the questions or the answers ahead of time. We should not reduce our content to avoid this. Chillum 14:46, 5 July 2009 (UTC)[reply]
Pretty much any test is invalidated if the answers are shown to you whether you asked to see them or not. Martinevans123 (talk) 19:00, 5 July 2009 (UTC)[reply]
Which has nothing to do with this article, as it doesn't show any answers. There is no right answer to inkblot tests anyway. DreamGuy (talk) 19:12, 5 July 2009 (UTC)[reply]
Pretty much any test is invalidated if the questions are shown to you whether you asked to see them or not. Martinevans123 (talk) 19:40, 5 July 2009 (UTC)[reply]
They're not questions either, even if they were that is why examination boards take a lot of effort to create new standardised questions every exam cycle. It seems either laziness or cargo-cultish to rely on these images when others could easily be produced to serve the claimed purpose, and copyright would then undoubtedly be claimed on those. 80.177.9.239 (talk) 00:27, 15 July 2009 (UTC)[reply]
This is ridiculously inaccurate. The current scientific quality of the Rorschach came about as a result of nine decades of research, which today, with a new set of inkblots, would cost hundreds of millions of dollars to produce and would require decades to complete even if every psychology research program in the English speaking world allocated most of its resources to such a project. And even then, there's no guarantee that the end result would be as good or better than the current Rorschach inkblots. This comment epitomizes the profound ignorance of non-expert opinions here. Ward3001 (talk) 00:44, 15 July 2009 (UTC)[reply]
In that case, how was the efficacy of these magic pictures ever established in the first place? Please do enlighten me. 80.177.9.239 (talk) 06:03, 15 July 2009 (UTC)[reply]
Through 90 years of research. Ward3001 (talk) 14:40, 15 July 2009 (UTC)[reply]
"would cost hundreds of millions of dollars to produce and would require decades to complete even if every psychology research program in the English speaking world allocated most of its resources to such a project", that is a rather impressive claim. Is that your opinion or did you get that information from a reliable source? Chillum 01:20, 15 July 2009 (UTC)[reply]
How much do you think it would cost to repeat nine decades of research, conduct approximately 2000 studies (which requires getting permission from research policy committees, finding thousands of subjects in the age range of 5 to 90 years old willing to participate, paying each of these research subjects approximately $100 each, get the studies accepted in peer review journals, and then synthesize all of the research into a coherent whole). Then, you must deal with the extremely difficult-to-find research subjects (such as several dozen patients who have suicided and also had a Rorschach administered prior to the suicide), reformulate hundreds of possible variables (and you can't use the old variables because they're from a completely different set of inkblots). How much do you think that would cost, and how long would it take? I said decades. One decade would be an underestimate. Two decades might be possible, with unlimited resources dedicated to the project. Do you suggest that it could be done in a year or two? If so, you are even more naive than the anon who glibly proclaimed that "It seems either laziness or cargo-cultish to rely on these images when others could easily be produced to serve the claimed purpose". And please tell me how you can find reliable sources for hypothetical future events? Do we look in the Journal of Psychological Research That Will Occur? Where could I find that reliable source. The reliable sources are the thousands of previous studies that must be repeated. Ward3001 (talk) 02:43, 15 July 2009 (UTC)[reply]
You see I don't know Ward, because I am not here to engage in original research. Do you have any sources for your impressive claim? Chillum 04:50, 15 July 2009 (UTC)[reply]
Read the thousands of existing studies accumulated over the past 90 years. Those are the sources. As you have done in the past, you ask for impossible information in insisting on a source for hypothetical future events. John Exner was the most prolific researcher on the Rorschach. He devoted almost all of the last thirty years of his professional life to researching the Rorschach. He completed several hundred studies and wrote a few books. Do you actually think a handful of researchers are going to duplicate that, plus all the previous 50 years of research, plus the research by dozens of other researchers during Exner's lifetime, in a few years at a cost of a few thousand dollars? And you're quite correct: You don't know. No one knows anything about the future. What I'm not sure about is do you know anything about the past when it comes to Rorschach research. I'm basing this on past research. If you want sources I can post the entire bibliography of Rorschach research on your talk page, along with average salaries of the dozens of university professors who will conduct the research, their research assistants, and university budgets that will oversee the research. Ward3001 (talk) 14:40, 15 July 2009 (UTC)[reply]
I only ask that you support what you claim. If what you are claiming cannot be verified then it is unverifiable. You are asking me to take several sources and join them into a new idea, that is original research by synthesis. Chillum 15:16, 15 July 2009 (UTC)[reply]
No, you ask for the impossible. The only way to estimate the future is based on the past. How do you think anyone prepares a budget? Do they look into the future with their crystal ball and come up with exact figures? It is perfectly legitimate, including on a Wikipedia talk page, to make a reasonable estimate about future events based on accurate interpretation of past events. When an editor (anon 80.177.9.239) makes an outrageous claim about the future ("others could easily be produced") on a Wikipedia talk page, it perfectly acceptable to rebut that argument with an educated estimate that reflects a knowledge of past events. If anon 80.177.9.239 had made his absurd statement in the article, it would have been reverted instantly, and rightly so. If I made my estimate of the future in the article, that also would be reverted. But this talk page is an acceptable place to separate wild speculation from educated estimates, and that's what I did. I have given accurate data about past research, so let's see if others besides you think it can be done in less than "decades" rather "months" or "years", and what they base this estimate on. Otherwise you and I could disagree endlessly about the future. Ward3001 (talk) 16:03, 15 July 2009 (UTC)[reply]
As you didn't even attempt to answer my question, this hyperbole looks increasingly like an argument from authority with no actual proof. I see no evidence of commons tragedy as claimed. 80.177.9.239 (talk) 22:31, 15 July 2009 (UTC)[reply]
As you apparently cannot understand simple English, this discussion is pointless. Your question was answerered quite sufficiently; you simply cannot or choose not to understand it: ninety years of research. A "year" means 365 days. "Research" means scientific study. Ward3001 (talk) 00:31, 16 July 2009 (UTC)[reply]
  • This RFC is pointless, mainly because it's based upon the idea that seeing inkblots somehow harm you, which is complete and utter nonsense. Framing the RFC this way plays right into the hands of some amazingly unrepentant and calculated POV-pushers. Should we consider harm? Yes, but only if that harm can be reliably proven to be a genuine and immediate result, not just asserted and insisted upon in a rather POV way. Their argument goes like this: "Invalidating this test by seeing the images" (unproven and highly unlikely- you still see what you see even if you've seen them before) "means that no other test is usable" (unproven and frankly ridiculous -- these people are acting like inkblots are the end all and be all of clinical psychology, which is absurd) "which means that maybe there'll be some case where someone won't otherwise be able to be treated for a harmful condition" (highly unlikely theoretical situation which flies in the face of contrary POV that the tests are useless and sometimes even harmful -- we could equally argue that invalidating these, even if we could, we be a GOOD thing) "and therefore we should censor this information" (conclusion based upon a series of completely dubious, unproven arguments). DreamGuy (talk) 19:12, 5 July 2009 (UTC)[reply]
Please show us the evidence that this claim is "complete and utter nonsense". Martinevans123 (talk) 19:41, 5 July 2009 (UTC)[reply]
I don't know about nonsense, but it is all rather unconvincing. Chillum 03:54, 6 July 2009 (UTC)[reply]
That was the thrust of the entire post. Refusing to acknowledge it does not mean it wasn't already given. DreamGuy (talk) 19:00, 6 July 2009 (UTC)[reply]
And please show us the evidence that "the tests are useless and sometimes even harmful". Martinevans123 (talk) 06:28, 6 July 2009 (UTC)[reply]
Why do you keep asking us to present evidence to disprove your claims? It is your claim, it is your burden to convince us. I for one am not convinced. Chillum 13:33, 6 July 2009 (UTC)[reply]
Because I am still trying (after over a year, now) to draw your attention to the fact that it is not possible to provide scientific evidence for EITHER side of this argument. I have yet to see your response. Martinevans123 (talk) 19:26, 6 July 2009 (UTC)[reply]
Then the point really is moot; content will not be removed from a Wikipedia article because of unproven and unprovable concerns, when such content is encyclopedic and extremely relevant to the article's subject. We know already that Wikipedia is not censored, please don't choose to ignore that. --LjL (talk) 23:42, 16 July 2009 (UTC)[reply]
If you were paying attention that was not my argument, it was my argument that that's a valid POV, and that has already been sourced and is (or was -- maybe I should see if a POV pusher afraid to have criticism in the article took it out) in the article itself. The fact that multiple POVs exist mean that we can't follow the opinion of just one side. This is VERY basic NPOV stuff here. DreamGuy (talk) 19:00, 6 July 2009 (UTC)[reply]
I'm sorry DreamGuy, I have no idea what you're actually saying. If it's just my level of attention that is to blame, then I apologise. Martinevans123 (talk) 19:31, 6 July 2009 (UTC)[reply]
  • I think the answer is a definite maybe. There are currently about the same number of argument Pro as Con, many of them new. Each one is separate. Ethical arguments are valid, as well. If everyone can agree on the ethic then, why not consider it? We shouldn't dismiss all ethical arguments as categorically subjective. That's just not true. Some ethics are nearly universal. I get nervous when people, who are free to express themselves in a forum that encourages collaboration, limit their comments to issues of policy only. Many of us have not responded or provided their opinion about Argument Con #1. I'm not sure why, but I'm guessing it's because they couldn't find a basis in policy. It seems odd to me. People shouldn't allow themselves to disengage so completely from their own ethics. Even soldiers, prior to following orders, can be permitted to speak freely or to lodge a protest. Instead, what's so unworkable by first presenting an ethical argument, and then, if not everyone can agree, falling back on policy to direct the course of consensus? That'll work and it also allows people the freedom to express their ethics. I also cite the following sources:
    [10] [11] [12] [13]
    I'd like to get comments on Argument Con #4, I think a WP:NPOV violation is worse than a WP:NOTCENSORED one. But I can be patient and wait until this RfC is finished. One argument at a time, right? Danglingdiagnosis (talk) 00:09, 7 July 2009 (UTC)[reply]
  • Yes, the principle of "harm" should be considered in ordinary editorial decisions like this one. However, that principle is not an absolute. It is held in dynamic tension with other core principles such as NOTCENSORED. The degree of potential harm must be balanced against the potential benefit and the fit with Wikipedia's educational mission. It must considered in the full light of all the surrounding circumstances. In this regard, the analogy to BLP is apt. Disclosing the Social Security Number of a private citizen exposes the person to identity theft and is clearly inappropriate but publishing Todd Davis' is fine. (He's the guy in the LifeLock commercials.) Disclosing other personal details may or may not be appropriate depending on how widely the information has already been disseminated, how easily the information could be exploited, etc. BLP is actually quite nuanced in this regard.
    To apply the principle in this case, an allegation of harm was made. Counter-evidence was presented that the image is already widely disclosed. Considerable discussion drew out the nature, degree and probability of incremental harm. Reading the past discussions, it seems clear to me that the community has already given due weight to the principle of harm and reached a rough consensus that it is negligible in this case. (Perfect consensus is, of course, an unattainable goal at Wikipedia.) Rossami (talk) 23:55, 7 July 2009 (UTC)[reply]
Thank you for your wonderful comment. I loved how you seem to appreciate the distinction between laying out all arguments for everyone to see, and then once that's done, assigning weight to them. That's exactly what we need to do. It engaged my thinking very much, and was very welcome indeed. And yet, I might have come to a different conclusion if I were to have attempted to add the following issues to the dynamic tension of arguments of which you wrote. I wonder if your opinion might change after considering the following:
1. Neutrality of the article. This is a core principle of Wikipedia, more important than WP:NOTCENSORED How can we serve the poor reader who says:
"Where? Oh, where can I go to find a balanced article that allows me to make up my own mind about prominent issues surrounding the Rorschach test -- issues such as the one about why psychologists are advocating for the removal of images from the internet. What's that issue about? I see it on so many web-pages; but those pages seem so biased. Where can I go to find the unbiased version -- one that doesn't demonstrate bias by showing me the images. "
It was only 11 days ago that the allegation was made and countered that this important discussion was even occurring outside of Wikipedia, and we're still coming to grips with the idea that we need to show our readers that we are neutral on the subject and not engaging in the controversy or advancing a point of view. (See talk:Rorschach_test#Argument Con #4 - It violates Wikipedia policy on neutrality.)
2. The discussion so far, has also lacked any mention of the flaw in the thinking of many here that readers have already made their choice by coming to this page to see information about a medical subject. By clicking on "What links here," we can see that there are over 350 pages that link to the article, including one from a #1 song called Crazy and another about a fictional character in a major motion picture called The Watchmen and the corresponding graphic novel that Time Magazine has called "one of the 100 best novels ever written." Idly curious readers may have no idea what's coming when they click on the Rorschach wikilink. The same goes for fans of these things who will come to our article using search engines such as Google. All the discussions about harm have made the assumption that the reader knows what they are doing when they come to the article. This thinking is flawed, but no one has yet pointed out the flaw. Perhaps no one here has seen the movie.
3. Plus, I see a disturbing influence that has yet to be challenged. There has often been expressed a valid, but negative opinion about the utility of the test. But this opinion has no place when deciding whether to show the images or not. This thinking is similar to that of someone who vandalizes another's car saying, "Oh well, they won't care. It's a worthless pile of scrap metal, anyway." It's one thing to point at someone else's car and say it's worthless. It's quite another to then go and sabotage it. (See owner's 2009 letter of complaint and owner's 2006 letter of complaint) It's possible that once this is pointed out, consensus may change. (See consensus can change.)
4. I'm optimistic that consensus is movable about the interpretation of the APA and The BPS statement. It hasn't been pointed out, yet, how unreasonable it is to read a code of conduct and expect the authors to list specific tests when general categories will suffice. The Rorschach is a psychological test. Both codes ask that test materials and stimuli be kept secure. The Rorschach, by definition, falls under the category of test material and is thus covered under Section 9.11 of the APA Code. Moreover, it is highly popular test material. In a 1995 survey of 412 randomly selected clinical psychologists, 45 percent said that they use the Rorschach frequently and 89 percent said they use it occasionally. To interpret the APA's and BPS's code as not specific enough to include such a popular test is unnecessarily obtuse. Therefore, I think we can use the APA and the BPS as reliable sources showing that experts feel the need to maintain the security of Rorschach for, as is stated both codes, the purpose of protecting the welfare of the patient. I can't help but think that consensus will change on this. It has yet to be explained to me how any other position can be tenable. This will undoubtedly have the effect of changing the dynamic tension and weight of the arguments in "the final analysis."
So speaking as the new guy to this discussion page, (22 days and counting) I'm wondering if you agree that it's a little premature to draw any conclusions, yet. I think there may be a few things yet to discuss. Danglingdiagnosis (talk) 10:27, 8 July 2009 (UTC)[reply]

Comments from Dlabot

Why would I need to 'expand' on it? Dlabtot (talk) 17:30, 6 July 2009 (UTC)[reply]
Because you're claiming I filed a disingenuously framed RFC. That's a fairly strong claim, I would like you to elucidate what part of the RFC you feel was disingenuous and how I could've better framed it. –xenotalk 17:34, 6 July 2009 (UTC)[reply]
I'm not going to be drawn into a repeat of one the pointless arguments that I see on this talk page. You filed a Request for Comments; I commented. I don't think there is anything vague or unclear about my comment. How could you have better framed the question? Let's look at another example question: Should dlabtot stop beating his wife? Certainly this question is disingenuously framed, but does that does not mean a change in phrasing will improve it. Rather it should simply not be asked. Similarly, I see your RfC as a disruptive unwillingness to accept consensus. Though no doubt meant in good faith and with the best of intentions. Dlabtot (talk) 18:02, 6 July 2009 (UTC)[reply]
You'll note I did use the construction "Iff" - "If and only if" - we are willing to accept the possibility of harm (...). I have also been trying to remain neutral in this, so I filed, to the best of my ability, a neutral RFC that takes into consideration both sides of the debate. I'm sorry that you feel I was "disingenuous" and that, I think you are confused as to my role here. –xenotalk 18:08, 6 July 2009 (UTC) striking/superscript at 18:27, 6 July 2009 (UTC)[reply]
I'm certainly not confused. Please refrain from making any further personal comments about me or what you presume to be my mental state. Dlabtot (talk) 18:11, 6 July 2009 (UTC)[reply]
Then I ask the same of you. Perhaps you would like to strike your above comments about disingenuity and disruptive unwillingness. –xenotalk 18:14, 6 July 2009 (UTC)[reply]
I certainly will refrain from making personal comments about you or what I presume to be your mental state. Just as I asked you to. Dlabtot (talk) 18:22, 6 July 2009 (UTC)[reply]
I've amended my comment. I am still of the opinion that you are confused, since you're trying to say that I am disruptively unwilling to accept consensus. –xenotalk 18:27, 6 July 2009 (UTC)[reply]
I'm not at all confused. If I recollect correctly, some weeks or months ago I answered an RfC about whether this article should include an inkblot image. Looking at the talk page, history, and archives, it is apparent that a clear consensus has been formed since then. Despite this, the arguments continue. Dlabtot (talk) 18:34, 6 July 2009 (UTC)[reply]
The purpose of this RFC is to determine whether Wikipedians think Rorschach images should be suppressed iff they accept there is a potential for harm. Perhaps you can answer that questions directly rather than complaining about the framing of the same. If the answer is no (i.e. "No, we shall not limit or otherwise restrict encyclopedic content even if it can be demonstrably shown to cause harm"), and the community agrees with you, then (in theory) no further debate on images will need to take place and the talk page can be free of the never-ending debate. My goal here is to bring an end to this long-running dispute. Nothing more, nothing less. –xenotalk 18:42, 6 July 2009 (UTC) (FWIW, you recall incorrectly. You haven't visited this page before today. [14] It may have been an RFC at another forum though)[reply]
I can accept that you are unhappy with my response to your Request for Comment. Nevertheless, it is my response. I again repeat that I have no doubt that your filing of the RfC was meant in good faith and with the best of intentions. Dlabtot (talk) 18:52, 6 July 2009 (UTC)[reply]
Thanks for your comments. –xenotalk 18:53, 6 July 2009 (UTC)[reply]

More comments

potential for harm is a nebuleuos phrase. In this specific instance, prior exposure can result in inaccurate/misleading scores. In this specific instance, the more one looks at the images, the more one sees in them. Unfortunately for test takers, there is a correlation between what one describes, and the degree of pathology that this test claims one has. That is where the harm, for this specific test comes. (Obviously I'm assuming that the test is valid in the first place, and the person who administers it, is competent to do so. On second thoughts, even if only one, or neither of those applied, but it was being given in an "official" capacity, the harm is still present, albeit magnified by the incomptence of the person administring the test, and/or the degree to which the test is invalid.)jonathon (talk) 06:41, 7 July 2009 (UTC)[reply]

In short, I don't think potential for harm should be policy, but it might be a factor to consider, in determining the appropriateness of including/excluding specific data.jonathon (talk) 06:41, 7 July 2009 (UTC)[reply]

How many kicks at the can?

I have seen something very similar at the Muhammad depiction debates. After a few months of being told that their basis for the removal of the images(it offended their religion) was not a valid reason to remove them people began to try each and every policy they could find. They would explain how NPOV and OR requires that we remove the images, then they tried copyright, then they starting bringing up obscure manuals of style. All the while the true reason for them wanting the image removed was the same as it always was, it offended their religion. In the end such a tactic failed as because a) the policies did not support their arguments, and b) it was seen as a transparent attempt to get what they wanted.

I am seeing a bit of the same thing happening here. The reason to want these images removed remains the same, but due to this reason being rejected a variety of replacements arguments come to take its place. In the last couple days I have seen history repeat itself here. I have no objection to policy based arguments, I prefer them in fact. However I do ask that they be sincere. Perhaps every one of these new arguments was a sincere attempt to bring Wikipedia in line with its own policies, perhaps it was an attempt to use policies as a tool to get what one wanted. It is possible there is very real and honest concern about the public domain status of the images we use, or if they violate neutrality, or if they are properly verified, or if they are derivative works, or if they are undue weight, or perhaps some people just wants the images removed for the original reasons given.

In the interest of creating a neutral point of view I request each one of you to ask yourself: "What is more important to you, advancing the interests of Wikipedia, or advancing an outside interest?". Where advancing outside interests is more important to an editor than advancing the aims of Wikipedia, that editor stands in a conflict of interest. I am sorry if I sound like I am accusing anyone of something, but it does seem very familiar to me. Perhaps I am wrong about this? Chillum 13:30, 6 July 2009 (UTC)[reply]

This is why I initiated the above RFC, based solely on the supposed actual motivation for removing the image(s). –xenotalk 13:37, 6 July 2009 (UTC)[reply]

A productive step. It sort of reminds me of a poll we had a few weeks back, and another RFC a few months ago. Sometimes Wikipedia is like watching re-runs of an old show. I wonder, if this RFC finds that we should not remove the images for these reasons, will it satisfy those who wish to remove them or will this just go in circles forever? Chillum 13:47, 6 July 2009 (UTC)[reply]

I agree it does seem like a broken record after a while. It seems like we have settled this issue except for the disagreement of a small vocal group of editors.--Doc James (talk · contribs · email) 19:42, 6 July 2009 (UTC)[reply]
If the "outside interest" is preventing harm to others, then yes Chillum, for me that wins every time over "the aims of Wikipedia". While I don't really mind being compared to a religious fanatic, I think others might - I honestly believe that the concerns here are different to those at Muhammad. And yes, I remain very disenchanted. Martinevans123 (talk) 19:53, 6 July 2009 (UTC)[reply]

I have already stated that preventing harm is not congruent to our goals of creating an encyclopedia. Information may be used to help or harm and even that varies from perspective. Garycompugeek (talk) 20:05, 6 July 2009 (UTC)[reply]

The "use" of information, to help or harm, would seem to usually depend on conscious decision, which, in this particular case, may not be an available option. Martinevans123 (talk) 20:09, 6 July 2009 (UTC)[reply]

Yes, but back onto my original topic. How many kicks at the can? If the same arguments are made and rejected by the community for say another 2 months will we accept consensus then? How about after a year? How about 5 years? Will Ward and Faust just keep arguing as they have since early 2008 until those disagreeing with them simply give up? Chillum 13:22, 9 July 2009 (UTC)[reply]

Here you go, repeating the falsehood that it is only 2 editors opposing what you want, even after your claim was shown to be false here: [15] and you even admitted it. I guess some habits are hard to break.Faustian (talk) 02:50, 10 July 2009 (UTC)[reply]

I said no such thing. I asked how long we must entertain the same notions that have been rejected time and time again. I mentioned you are Ward because you are the two who have been pushing this the longest and the most. This is demonstrable by the archives which I am very familiar with. Chillum 03:05, 10 July 2009 (UTC)[reply]

Thank you for the clarification and I retract my last comment although given the history you ought to be careful about not commenting in a way that could interpreted as emaning it is only two people. FYI, since your arrival onto this topic you have been at least as active, if not more so, than either Ward or I or any other editor. As for "rejected", that is irrelevent. PEople do disagree and I never denied that in this case the expeerts are outnumbered and in the minority. The point is what we do with this disagreement. Faustian (talk) 03:53, 10 July 2009 (UTC)[reply]

Are there any new arguments?

Will a continuation of this debate be an endless repetition of the same rejected arguments over and over? Or is there anything new to add?

It is a terrible waste of time to tell the same people that the same arguments still are not compelling month after month. I think we should stop responding to arguments that have already been given plenty of consideration and rejected. We have already established a stable consensus on this matter and I think we have given this one issue enough attention and would like consider it settled until such a time as arguments that have not already been rejected by the community are presented.

Surely there are other aspects of this article we can work on. Chillum 21:04, 6 July 2009 (UTC)[reply]

The sad t hing is that you can archive everything above your comment, and in six months time, it will all be repeated again. — Preceding unsigned comment added by Pseudo daoist (talkcontribs) 06:18, 7 July 2009 (UTC)[reply]


If you're concerned that people might engage in endless repetition of the same old arguments, you might consider fully developing the ones we have now

  1. 1.3 Arguments Pro
     #01 - The cat's out of the bag
     #02 - No evidence of harm
     #03 - Adds to the page
     #04 - removing the images amounts to censorship
     #05 - Reputable organizations have been showing the images, too
     #06  It doesn't matter if we publish. The test is nearly worthless anyway.  
  1. 1.4 Arguments Con
   * 1.4.1 Argument Con #1 - It may harm a psychologist's ability to protect the welfare of his/her patient. 
   * 1.4.2 Argument Con #2 - It violates Wikipedia policy
   * 1.4.3 Argument Con #3 - It interferes with the workings of another profession
   * 1.4.4 Argument Con #4 - It violates Wikipedia policy on neutrality.
   * 1.4.5 Argument Con #5 - It violates Wikipedia policy on indiscriminate collection of information.
   * 1.4.6 Argument Con #6 - It violates Wikipedia policy on non-free content.

They are well organized and easily found when someone goes looking back in the archives. However, some of these arguments still have questions left hanging. Future readers might question, as I did, whether the issues that they have had ever been fully explored before.

If you're thinking about the future, you might also consider what Steven Colbert in his popular television show, The Colbert Report, might say when he visits this discussion a few months from now. I think we've raised some issues that would get his attention, namely the role of ethics in Wikipedia. What will he say? Will he excoriate us for ignoring ethics in favor of policy, and compare us to Islamic or Christian fundamentalists: calling us "Fundawikians" for taking our policy "Bible" as the complete and total word of authority and in our zeal, promoting the cause of evil in the world? Or would he applaud us for not deviating from "our goals" in the face of "outside interests." I think he'll portray both extremes. Or maybe he'll run a sketch portraying us like the guards in the Stanford Prison Experiment. We'll get lost into our assigned roles as guards until suddenly someone calls a halt to the experiment and we are all exposed as mere college students, sitting in our underwear in our darkened dorm rooms, deciding the fate of others. I know Steven Colbert. He'll do something funny. But I think we should pay better attention to how we want ourselves to be portrayed. Danglingdiagnosis (talk) 07:52, 7 July 2009 (UTC)[reply]

We don't base our content on what comedy talk show folk have to say about us. Every last one of those arguments have been given plenty of attention over the last several weeks. Do you have any new arguments? You know, ones that have not already been rejected? Chillum 14:23, 7 July 2009 (UTC)[reply]
This doesn't even sound like a serious argument. To even think that Stephen Colbert will even be here in a few months (where did THAT come from?) or that he would take your side (your kind of ignorant arguments are exactly the sorts of things he loves to skewer mercilessly), you're just deluded. And none of it has anything to do with how we conduct ourselves on an encyclopedia. DreamGuy (talk) 15:40, 7 July 2009 (UTC)[reply]
What a surpise, more personal attacks from dreamguy ("Your kind of ignorant arguments").Faustian (talk) 03:42, 8 July 2009 (UTC)[reply]
Does Wikipedia not have any policy on publishing test materials? Shouldn't it? Martinevans123 (talk) 10:42, 7 July 2009 (UTC)[reply]
Not that I know of. Chillum 14:23, 7 July 2009 (UTC)[reply]
No (other than copyright laws) and no. DreamGuy (talk) 15:40, 7 July 2009 (UTC)[reply]

Einstein says the definition of insanity is doing the some thing over and over again yet expecting a different result. These are old arguments. Just because questions have not been answered, or answers fall short of your expectations, means nothing to the community as a whole. Broad community consensus has been achieved regarding the images therefore the burden of proof to change it lies with anyone wishing to change it. None of these arguments has apparently changed anyone's mind. Repeating them will not either. Trying to find any possible way to remove the images is becoming disruptive. Garycompugeek (talk) 14:28, 7 July 2009 (UTC)[reply]

Does he, I've not seen his talk page. If there really is no policy then perhaps Wikipedia would be improved by having one and the debate here would be a useful test-case in the path to achieving this. Questions deserve considered answers and not accusations. How else can debate proceed? Martinevans123 (talk) 14:41, 7 July 2009 (UTC)[reply]
WP:VPP is a good place to propose a new policy. We have been answering your questions for months, I can't imagine anybody who would say we have not given them due attention. This has been weeks of endless repetition, you ask how else can a debate proceed? It can proceed by presenting an argument that has not already been repeatedly rejected. Or you can simply stop beating a dead horse. Chillum 14:44, 7 July 2009 (UTC)[reply]
(edit conflict) ::Ever the sarcastic one. Martin debate is good but not the same refuted debate. If you wish to create policy try the WP:Village pump. Garycompugeek (talk) 14:48, 7 July 2009 (UTC)[reply]
Thanks for the tip, Chillum and Gary. But can you remind us when the question above about policy (five hours ago), which you have kindly answered, was ever previously asked (or answered) in this debate? But please stop throwing your wiki UNCOOL STICKS. And the point I was making was that Albert is not a wikipedia editor. Martinevans123 (talk) 14:54, 7 July 2009 (UTC)[reply]
That was more of a question than an argument. I think it has come up in the debate before, or perhaps not. There is no such policy as far as I can tell. If you think this needs further discussion you can create a thread about it. I did understand that you were referring to Einstein. Chillum 15:04, 7 July 2009 (UTC)[reply]
And where is the debate on harm refuted? Thanks. Martinevans123 (talk) 14:56, 7 July 2009 (UTC)[reply]
This section is to present new arguments, not to rehash old ones. There are sections above dedicated to the question you just posed about harm and there has been plenty of response and consideration given above. People have remained unconvinced. Please, does anyone have any new arguments? Chillum 15:00, 7 July 2009 (UTC)[reply]
Where is the "section above" dedicated to the question of wikipedia policy on publishing test materials? Martinevans123 (talk) 15:05, 7 July 2009 (UTC)[reply]
There is a section above about harm. Regarding the question about the non-existent policy I just said you can create a section if there is more to talk about. I suggest you create it at WP:VPP because creation of policy is beyond the scope of this article. Regardless, asking if a policy exists that would support your position and being told it does not exist is not really an argument in favor of your position. Chillum 15:08, 7 July 2009 (UTC)[reply]
I am suggesting that the lack of a clear policy to prevent it is the reason why the images have been displayed here in the first place. But thanks for the clear, non-accusatory, advice (please note my non-sarcastic and concilliatory tone). Martinevans123 (talk) 15:15, 7 July 2009 (UTC)[reply]
(in a soothing deep and friendly tone) I suggest that such a policy proposal is likely to come to the same conclusion the consensus here did. However, if you propose a policy then I will participate in the debate. It is possible I am wrong about my prediction that the community would reject it. Chillum 15:20, 7 July 2009 (UTC)[reply]
The policy for prohibiting test materials doesn't exist because we don't need one... and in fact it'd be actively opposed to other policies we do have. If you disagree with our other policies on this matter you can't just up and create a new policy that will somehow overrule all the others. Making policy is a long process that requires an extremely broad and active consensus from the wider project and not something you just write up in a dusty corner of the site and expect others to follow. Seeing as how other longstanding rules such as WP:EL aren't even policy yet despite years of being here and supported, you have exactly zero chance of having any policy preventing test materials from being included here. But if that's the only avenue you have left to try to get what you want to do over the strong consensus against you, good luck with that. (That ought to keep you busy for the next ten years.) DreamGuy (talk) 15:40, 7 July 2009 (UTC)[reply]
Thank you Chillum for your friendly encouragement. But ten years in a dusty corner with "zero chance of success" doesn't sound so attractive, DreamGuy. So maybe I'd better stay here after all, like all the other "stubborn" editors. Martinevans123 (talk) 15:56, 7 July 2009 (UTC)[reply]
It's only been 7 days since the discussion at Argument#1 - It may harm... began. In that time, new sources where discovered and questions left hanging. I answered Xeno's call to come to this discussion 22 days ago. It makes no sense to invite someone and then ignore their questions. It's not very WP:CCC of you. Danglingdiagnosis (talk) 00:57, 8 July 2009 (UTC)[reply]
Not true Danglingdiagnosis. That argument was first brought up in early 2008 and again numerous times before this most recent turning of the wheel. I can see how someone relatively new to this very long debate would think that this has not been given due consideration, the archives are very long and tedious to read but they do tell the story of this endless repetition I speak of. Chillum 01:03, 8 July 2009 (UTC)[reply]
Correction it was even earlier See Talk:Rorschach test#Which image (From July 2007). You can see that Ward and Faustian were busy stating the exact same arguments that are still being argued today. Just compare this poll from February 2008 with this poll from June 2009 and you can see that this debate was settled ages ago. Look at this index of the archives I have created to see a full history of this debate: User:Chillum/Rorschach test talk page archive. Chillum 02:22, 8 July 2009 (UTC)[reply]
Correction before I disappear again: the debate was never settled because consensus, defined as compromise, was never achieved. Instead a 2/3 majority had their way with no input, in terms of the article's construction, from the minority. So the only thing that wa settled, I suppose, is that experts are outnumbered and that 2/3 want something done a particular way. As long as this sad state of affairs exists these things will keep coming up again. This is exactly what happens when you don't have consensus and when the debate isn't settled.Faustian (talk) 03:42, 8 July 2009 (UTC)[reply]
Yet another argument that has repeatedly been refuted. Consensus does not mean everyone is satisfied, there has been consensus for showing the images for over a year and arguments to the contrary have not changed that. This debate was settled last year, you lost. This idea has been given more than enough consideration. Do you have any arguments that have not already been repeatedly rejected? Chillum 04:00, 8 July 2009 (UTC)[reply]
Translation: The majority choose to ignore the policy concerning consensus and still win by majority vote. This is how the debate has been "settled" and arguments "refuted.". The reality is that the problems keep coming up again because indeed there is no consensus. Indeed, this is a perfect example of the wisdom of consensus policy.Faustian (talk) 04:32, 8 July 2009 (UTC)[reply]
How come you did not object to consensus in February 2008 but you do now? How is this consensus not to suppress the images different than the consensus not to suppress the images back then? Chillum 04:12, 8 July 2009 (UTC)[reply]
Because over a year ago I had't read the consensus policy and naively took others' word for it when they claimed that votes equals consensus. Thanks for showing an example of me trying to forge some sort of compromise, though.Faustian (talk) 04:32, 8 July 2009 (UTC)[reply]
I think the thread on "what is consensus" got archived. You can restart it if your like and we can talk about it some more. This section is for any new arguments that have not already been rejected by the community. Chillum 04:46, 8 July 2009 (UTC)[reply]
Translation: a majority of editors involved on this topic have chosen to reject wikipedia consensus policy and to steamroll their version through, claiming that their majority equals "consensus."
Here it is: Talk:Rorschach test/2009 consensus review, 92kb of discussion on the subject. Chillum 04:51, 8 July 2009 (UTC)[reply]
That just shows how many editors prefer what. Consensus isn't majority preference: [16]"Consensus is a group discussion where everyone's opinions are heard and understood, and a solution is created that respects those opinions. Consensus is not what everyone agrees to, nor is it the preference of the majority. Consensus results in the best solution that the group can achieve at the time. Remember, the root of "consensus" is "consent". This means that even if parties disagree, there is still overall consent to move forward in order to settle the issue. This requires co-operation among editors with different interests and opinions." The current "solution" does not respect the the opinions of the minority (indeed it does not take them into account), it is merely the preference of the majority, and there is no consent to move forward. Thus there is no consensus. I initiated a discussion and an rfc on the wikipedia"consensus discussion page: [17] and [18]. Very few people responded, and their opinions, though not consistant, seemed to indicate that consensus is indeed compromise, provided that such compromise doesn't violate wikipedia policy (NPOV, use of reliable sources, censorship, etc.).
When atempt at discussing policy is made, you conveniently chose to ignore the policy points: [19], instead choosing to add a false statement about how many people are in the minority, incorrectly (disohonestly?) claiming it was just two people. Then you used a straw man argument that I claimed that consensus policy calls for 100% agreement even t hough I never claimed that it did. You then cherry picked from policy rather than using whole quotes to produce a skewed interpretation. When you were called on this misbehavior, you chose to revert to ignoring the policy guidelines. The bottom line is that here we have an example of a majority choosing to ignore wikipedia policy and having their way because there is no mechanism in place to enforce that policy. The result: no consensus.Faustian (talk) 12:56, 9 July 2009 (UTC)[reply]
If you want to create a section to talk about my behavior go ahead, my user talk page would be a good place for that. The consensus was reviewed by multiple outside sources including a post at ANI. How can you act like your viewpoint is being disregarded when it has been given so many months of consideration? It simply has not been convincing. We all know if the numbers were reversed and the majority wanted to remove the images that you would call it a consensus(if I am wrong then deny it). This section is for new arguments, not rehashing old ones. Chillum 13:05, 9 July 2009 (UTC)[reply]
Does anyone have any new arguments that have not already been discussed ad nauseum? Chillum 13:09, 9 July 2009 (UTC)[reply]
But Faustian's points are perfectly valid, whether written here in this section or anywhere else. They are the very reasons that prompted me earlier to take objection to the proposal made by one editor to "hammer out agreement" and to suggest that such an approach was redolent of the Secret Police. Apparently those whose views cannot be accommodated can simply be accused of "wasting time" and of using the same old arguments "for months on end". Martinevans123 (talk) 13:13, 9 July 2009 (UTC)[reply]
Secret police? A bit over dramatic perhaps? We just don't agree with(and are thus not obeying) your arguments, we are not kicking in your door in the middle of the night. Have you read the archive of this debate I have posted at the bottom of this page? What we are facing is nothing more than proof by assertion which is no proof at all. These are the same arguments for months on end, there is no accusation about it, it is all in the archives. A huge amount of consideration was given to the arguments, it is not accurate to imply that we are trying to unduly dismiss them. I am just asking for an argument that has not already been rejected. Chillum 13:17, 9 July 2009 (UTC)[reply]
I also read it all when it was going on, thanks. I just think that hammers may be the wrong tools with which to try and reach "agreement". I'm keeping Adolf in reserve, of course. Martinevans123 (talk) 13:24, 9 July 2009 (UTC)[reply]
What are you talking about with "hammers", and "Adolf"? We already have an agreement, it does not include 100% of the people here, but it includes enough people to be a strong consensus. We don't obey every minority viewpoint, we do give them all consideration, but in this case they have just not been convincing. If the numbers were reversed and most people wanted the images removed, would you not consider that a consensus to remove them? Chillum 13:30, 9 July 2009 (UTC)[reply]
You accused me of a "not so subtle variation of Godwin's law", didn't you? And I am objecting to the idea that agreements can always be "hammered out". That's what I'm talking about. But I also don't see that consensus is compatible with the notion of "obedience". There was a time when the debate was considering changes in position or display mechanism of images as a way of reaching agreement. But now we the choice seems to be "images vs no images". Hang on, don't tell me... I'm not allowed to make this point, as it's not new? Martinevans123 (talk) 14:02, 9 July 2009 (UTC)[reply]

I see that, predictably, Chillum has chosen to sidestep a discussion on actual wikipedia policy points as usual. His strategy seems to be: ignore policy, go with majority rule because the majority can get away with ignoring policy, and then label the dissenting minority whose views are not accomodated at all (in violation of the consensus policy whose points Chillum refuses to discuss) as troublemakers repeating the same arguments. He complains about my drawing attention to his behavior, then makes up claims about my own ("We all know if the numbers were reversed and the majority wanted to remove the images that you would call it a consensus(if I am wrong then deny it). "). For the record, I would accomodate a minority view that disagrees with me and have stated so. I'm not Chillum, after all. And this is indeed the appropriate place to highlight Chillum's behavior because it reflects the process that has occurred on this page. He also claimed that consensus was discussed. Yes, it has been - and he derailed that discussion as demonstrated.Faustian (talk) 14:25, 9 July 2009 (UTC)[reply]

Your logic is weak Faustian. If we let minority view rule then we would basically have chaos. The wiki in practice is very democratic even though we state its not a democracy. That's because we weigh arguments based on logic and common sense. It is possible to have 10 editors on an article abandon those principals and have a minority win consensus with the help of an RFC or outside mediation because their arguments were more logical and given more weight. Attacking Chillum is not going to change things. If you have a problem with his behavior and cannot solve it on your talk pages you can place a wikiquette or a request for comment on a user. This page is for article construction of the Rorschach test. If you do not wish to accept consensus then follow the dispute resolution process. Garycompugeek (talk) 14:57, 9 July 2009 (UTC)[reply]
Your statement is based on a straw man - I do not propose that minority rule - that would be a violation of consensus policy (as is, of course, "majority rule"). I only point out that per policy the minority viewpoint needs to be taken into account in the layout of the article, otherwise there is no consensus.Faustian (talk) 21:10, 9 July 2009 (UTC)[reply]
Not every viewpoint needs to be reflected in the article, I don't see anywhere in the consensus policy that says that. Chillum 21:18, 9 July 2009 (UTC)[reply]
Consensus is not unanimity - so viewpoints by scattered individuals indeed do not have to be taken into account. On the other hand 1/3 of involved editors are not scattered individuals.
  • Wikipedia:What is consensus? "Disputes on Wikipedia are settled by editing and discussion, not voting. Discussion should aim towards building a WP:consensus. Consensus is a group discussion where everyone's opinions are heard and understood, and a solution is created that respects those opinions. Consensus is not what everyone agrees to, nor is it the preference of the majority. Consensus results in the best solution that the group can achieve at the time. Remember, the root of "consensus" is "consent". This means that even if parties disagree, there is still overall consent to move forward in order to settle the issue. This requires co-operation among editors with different interests and opinions."
  • "A consensus can be found by looking for common ground and synthesizing the best solution that the group can achieve at that time.""
  • "[20] "Consensus is a broader process where specific points of article content are consiered in terms of the article as a whole, and in terms of the article's place in the encyclopedia, in the hope that editors will negotiate a reasonable balance between competing views, as well as with the practical necessities of writing an encyclopedia and legal restrictions."
So is the solution you ram through with your majority respecting the minority's opinions? It is not. Is there overall consent to move forward? There is not. Is there common ground? There is not. IS there a synthesis of the two viewpoints? There is not. Is there a balance betweent he competing views? There is not. It is, actually, 100% what the majority wants. So thre is not consensus per consensus policy.Faustian (talk) 21:24, 9 July 2009 (UTC)[reply]

Somewhere, we have a definition of the rough consensus which is the actual standard at Wikipedia. Perfect consensus is impossible in any group larger than about 8 (according to some study I read in a leadership class more years ago than I'd care to admit). In any sufficiently large group, there will always be a minority unable or unwilling to concede to any position but their own. No functioning group can allow itself to be held hostage by an uncompromising minority (the so-called tyranny of the minority problem). While rough consensus does not mean that majority rules or even that supermajority rules, it also does not mean unanimity or even unanimous consent. Does anyone know where that "rough consensus" discussion got moved to? I think it might be useful reading. Rossami (talk) 23:31, 9 July 2009 (UTC)[reply]

I agree with what you have written. That is why wikipedia consensus policy wisely involves the principle that [21] "Consensus is a group discussion where everyone's opinions are heard and understood, and a solution is created that respects those opinions. Consensus is not what everyone agrees to, nor is it the preference of the majority. Consensus results in the best solution that the group can achieve at the time. Remember, the root of "consensus" is "consent". This means that even if parties disagree, there is still overall consent to move forward in order to settle the issue. This requires co-operation among editors with different interests and opinions."
So basically, consensus means an article that no one is 100% satisfied with but that, with the exception of a couple of isolated editors (inevitable), people can live with. Compromise means giving something up. Given the ratio of preferences the minority ought to give up more than the majority when the compromise is forged - article ought to reflect the majority's wishes more than those of the minority - the minority which at 1/3 of involved editors is substantial cannot be ignored for consensus to be achieved. However, to repeat myself, in this articles' case the majority refuses to compromise and seems to demand that the article reflects 100% what they want. As the majority they get to outvote any attempt at compomise. Every attempt at compromise has been shot down by this majority. So what do we do when a majority opposes policy?Faustian (talk) 02:17, 10 July 2009 (UTC)[reply]
Ratios? This is not math. You simply have not convinced the community of your point of view. If you think there is a violation of consensus on this talk page then seek outside scrutiny, just don't just keep insisting it is the case when so many people have told you this is not so. Post at WT:Consensus and ask the people there what they think, that has worked well in the past. I remember a posting at WP:ANI where several people agreed that consensus has formed. How is it that the only people who think consensus is not being served are those who are not getting what they want? Chillum 03:16, 10 July 2009 (UTC)[reply]
Why do you insist on using this section I created to talk about new arguments to repeat things that have already been discussed to death in the past? There are pages of discussion in this thread and none of it presents a single new point. It is all just a rehashing of the same old thing. I have had to start yet another sub-thread just to attempt to give someone the opportunity to breath fresh air into this debate. Chillum 03:17, 10 July 2009 (UTC)[reply]
Since when am I and the other 1/3 who disagree with the other 2/3 not part of the community? We are all the community, and we ought to work together to forge a compromise. I did post to the consensus article and as I stated only a few responded and they seemed to indicate that consensus was indeed compromise as long as no policies such as NPOV, Reliable Sources etc. were not violated. I will note that once again you have chosen to avoid discussing the policy points, Chillum, but to change subjects.
When you make incorrect claims about consensus or the debate being settled I have the right to correct you no matter what section those false statemtns are in.Faustian (talk) 04:29, 10 July 2009 (UTC)[reply]
But this is a NPOV issue. But you knew that because we have already had this discussion. Why do you keep using this off all threads to repeat conversations? The whole spirit of this thread is new arguments. Chillum 04:59, 10 July 2009 (UTC)[reply]
As I said, when you make incorrect claims about consensus or the debate being settled I have the right to correct you no matter what section those false statements are in.Faustian (talk) 13:47, 10 July 2009 (UTC)[reply]

New arguments go here

Since the above section got filled with off-topic content unrelated to presenting new arguments I am roping off this section for people to present new arguments to the debate. Please keep this sub-section on topic by sticking to new arguments here, if you have anything else to say there are plenty of other threads. Chillum 13:20, 9 July 2009 (UTC)[reply]

David S Rohde argument

Okay, I admit, I saw the Slashdot article, but I'm not a newbie here.

There was a recent controversy over removing material at David S. Rohde because it would harm a person. While the possible death of a person is more serious harm than the harm caused by showing Rorschach inkblots, it raises some of the same issues. The Rohde case was also different because the New York Times complained and it was one of their own reporters (which I personally think of as a conflict of interest), and Jimbo personally intervened. Suppressing the information there and not suppressing information here has uncomfortable implications about whether we suppress information based on the importance of the person who complains about it. If we're going to be consistent, we should also suppress the inkblots. If we don't want to, perhaps we should rethink our policies on suppressing anything at all.

I've also complained about how policies seem to be written in absolutist ways that don't seem to allow for suppressing information. WP:IAR says it should only be used to improve the encyclopedia (and I can't honestly call protecting outsiders "improving the encyclopedia") and WP:NOTCENSORED doesn't seem to allow us to be censored in exceptional circumstances. Ken Arromdee (talk) 21:10, 14 July 2009 (UTC)[reply]

Slashdot debate [22]. Garycompugeek (talk) 21:48, 14 July 2009 (UTC)[reply]
That's a very interesting comparison case. Perhaps the argument is refected here in terms of the "importance of the organisation" which complains. As far as I know neither APA nor BPS have directly complained (or even if they had, that the ordinary wikipedia editor would necessarily ever get to hear about it?). But I can't help thinking that the APA and BPS might find it hard to flex legal muscles in the same way that, say, a multinational corporation might. In contast, however, some of the argument here has been about protecting Wikipedia FROM unwanted "external forces" i.e. the expert clinicains who use the test and/or those professional bodies who represent them. Martinevans123 (talk) 21:26, 14 July 2009 (UTC)[reply]
My impression is that the vast majority of clinicians won't use this test anymore because it is considered to be so unreliable and unscientific. So the APA or the BPS won't comment because they by and large don't care. (Incidentally is there any evidence that either the APA or the BPS were complaining about this test in particular and not other tests?) JoshuaZ (talk) 21:33, 14 July 2009 (UTC)[reply]
Your impression is quite understandable if your knowledge is limited to what is put out by the small number of Rorschach critics within the field (an example is this article:[23] )However, as noted in this wikipedia article, in surveys 80% of clinical psychologists engaging in assessment services utilize the Rorschach, and 80% of psychology graduate programs teach it. The Rorschach has recently been used by forensic psychologists in 8,000 court cases, during which the appropriateness of the instrument was challenged only six times, and the testimony was ruled inadmissible in only one of those cases. The consensus within the field is clear that the test is useful and indeed it is used and taught frequently.Faustian (talk) 02:59, 15 July 2009 (UTC)[reply]
The APA and BPS probably have very little clue about Wikipedia. Plus, they've already published ethical codes that are really quite clear on the matter.Mirafra (talk) 06:58, 15 July 2009 (UTC)[reply]


I am sorry, but I don't think looking at a smudge of ink is anywhere near as dangerous as being kidnapped by the Taliban. Chillum 01:39, 15 July 2009 (UTC)[reply]
Of course it is not. But the principle is the same: causing harm.Faustian (talk) 03:01, 15 July 2009 (UTC)[reply]
Name the individual in harm by showing a PD image. Resolute 03:09, 15 July 2009 (UTC)[reply]
I will point out that the Rorschach is used commonly in forensic contexts (custody, dangerousness, fitness to stand trial, etc).Mirafra (talk) 06:58, 15 July 2009 (UTC)[reply]

Argument for general disclosure of testing materials (after an appropriate time)

I've read through most of the previous discussions, but they are so voluminous that I may have missed this. But has the Pro argument ever been considered for the general disclosure of testing materials -- in ANY field where a test is used to evaluate a significant segment of the public -- after a reasonable time has elapsed? Of course tests must be kept secret for a reasonable period of time to preserve their norming and keep results consistent. But I fully believe that no field should have information that is kept secret indefinitely if it can have significant effects on the general public. This goes for SAT tests and IQ tests, as well as state secrets. Yes, keep the materials secret for a few years, perhaps even a few decades. But at some point, that information should be opened up and made easily accessible to all. It's the only way to allow checks and balances that prevent abuses of power, whether it's by a few high-level politicians or by a profession that controls major decisions made about people's lives through psychological evaluations.

I do not mean to imply that there is anything wrong with the Rorschach test. It isn't relevant to me whether it's complete pseudoscience or whether it's the best psychological test ever created. The fact is that the more people who are able to examine something, the greater likelihood that potential flaws may be detected and discussed. Also -- again, this is not at all a criticism of psychology -- but the history of science has shown that experts can often overlook, disregard, or even be completely blind to significant problems with research in their own field. I'd be making the same argument if civil engineers had some secret algorithm to compute bridge designs or if a plumbers' union had a secret set of rules they used to create the most efficient plumbing in a house. The fact is that most of the time the experts do know what they're doing, but if the public as a whole can't have access to evaluate it, there might be some serious flaws that have been overlooked or even deliberately ignored. I think this is definitely true in standardized test materials I've seen, where questions are often severely flawed and can have ambiguous answers. However, if the methodology behind the creation and evaluation of those questions is not available to the general public, those flaws will never be corrected. I wouldn't be surprised if the same were true of most standardized IQ tests, professional personality tests, etc.

In an ideal world, I'd argue for a two-stage release of materials (first publishing them in a specialized location with less restricted access, and then in a completely public venue), since the rise of the internet creates special problems in information management. The first stage in the past would allow for an extended useful lifespan for test materials in most circumstances. But in the internet age, perhaps this is no longer possible. In any case, the Rorschach test is more than 75 years old. Sorry, but I simply can't see any good reason to keep ANY stuff secret for that long, let alone believing that any test could be kept valid and relevant over such a long time span. You want a properly normed test? Think ahead and norm a new test before the old one goes into the public domain. There are probably a hundred times as many psychologists active now as were in Rorschach's day -- how hard is it for someone to make up a new test that does a similar kind of thing (and in fact, somebody should be able to make one that does things BETTER)? People who use patent laws and copyright laws do the same thing (at least they used to before Disney corrupted the whole enterprise) -- they keep creating and progressing if they want to keep making money and being productive. Why should psychologists be any different? And don't give me any crap about "science" being privileged here; the fact is that secret information allow its users special power. Our trust in that power needs to be periodically examined and justified.

Lastly, for those who would argue that if the general public knows how a test works, its results are no longer valid -- so what? People learn and adapt. Maybe if the general public knew more about how psychological tests worked, they might actually improve themselves psychologically in some way. I don't know. Do psychologists know? In just about every other field, new generations require new tests to evaluate the new expectations, knowledge, and culture of the people who are being tested. Do psychologists really think that their ideas or tests have universal applicability for all time in a way that no other field does? Even periodic renorming cannot make a test valid forever; eventually the questions it answers become less relevant, and the field should move on to other tests or methodologies.

Apologies if some of this repeats what others have said. I don't think my main point has been made before in the abstract, however. 65.96.161.79 (talk) 00:10, 16 July 2009 (UTC)[reply]

The information that would allow the general public to view and debate the scientific validity of the test is already readily available, in the scientific literature. A quick shot I just tossed at the major EBSCO databases (PsycInfo, etc) with the search "subject = Rorschach" yielded over 6000 hits. I appreciate your personal desire to create a set of standards for the use, revision, and renorming of test materials, but the professional world already does that as part of its dynamic peer-reviewed process, and from a far more well-researched point of view than "I can't believe that they still use that old thing." By the way, the current norms for the Comprehensive System aren't 70 years old -- the most recent revisions are in Exner, JE (2002). The Rorschach: Basic principles and foundations of interpretation, Vol. 1. NY:Wiley. Mirafra (talk) 03:35, 16 July 2009 (UTC)[reply]
Thanks for your response. I do note that you seem to assume that I am not familiar with scientific literature in the field of psychology. While it is not my primary discipline, I assure you that I am well-versed in some subdisciplines that overlap with my research. But that is irrelevant. Far from saying, "I can't believe that they still use that old thing," my argument is based not only on trends I have seen in psychology but in many other disciplines throughout the history of science. I am familiar with a number of general disciplines and even some subdisciplines within psychology itself where peer-review does not do its job effectively (for various reasons that I don't need to get into here). And at no point did I imply that there wasn't significant scientific literature on the subject, nor did I imply that there haven't been many renorming attempts over the decades for the Rorschach. I merely stated that the test *materials* themselves are old, and maybe after 75 or 80 years it would be better to design new materials to suit better and/or more specific purposes (as some have; I'm aware of that too) rather than simply collecting new data with old (and potentially flawed) stimuli. There aren't many other examples of specific materials remaining useful for over 75 years in the history of science in the last few centuries, that I'm aware of. Generally, over that span of time in a discipline, the specific interpretation and understanding of the details of how things work changes so much that significant modifications to previous materials are motivated, if for nothing else than because new understanding means that better methods and materials can be developed.
That said, again my primary argument was about the release of test materials (including tests, answers (if applicable), explanations of design, principles used for evaluation) to the general public after a certain period of time. If you note, I did argue that this information first be released in a more limited way (for example, in professional circles, then more generally), but 75 years seems a rather long lifespan to argue (as some of the professional associations apparently do) against widespread dissemination. Norming statistics and similar information is not enough, since statistics can easily be used incorrectly or even deliberately used to manipulate trends in data. Do I really mean to argue that professional psychologists are incompetent? Of course not. But the argument of "leave it to the professionals, who understand what's going on" sounds strikingly to me like the same arguments used to restrict access to state secrets. Everything should be open to more widespread oversight after a period of time; it serves as a check against the occasional myopia of the professionals. I admit that my concern is more about testing in general (whether it be IQ testing, personality testing, or SATs and GREs), but the argument for release of and easy access to detailed information after a reasonable amount of time is just as applicable here.65.96.161.79 (talk) 18:58, 16 July 2009 (UTC)[reply]

Recognition

The first image in particular has a somewhat iconic quality. I saw it some years ago - I don't remember the exact context, but I believe I was reading about Rorschach. Now that image in particular is associated in my mind with Rorschach and symmetric inkblots. Someone with a worse memory than me might see Rorschach mentioned somewhere, not remember who he was or why he was important, look it up here, and see the image and remember "oh, that's who he was". This is one reason to include images in general - not just to show people new to the subject what the subject looks like, but also to jog the memory of those with a vague recollection of the subject. --NE2 04:57, 17 July 2009 (UTC)[reply]

Inline citation for inkblots needed

I put up a {{fact}} in the Inkblots section because I thought an inline citation should be there, and the template was deleted! Please note that when the reader comes upon the section with the inkblots, he won't know what the RS for them is without a citation there. Isn't that why we have inline citations in Wikipedia, instead of just a list of references at the end? Gee whiz, don't know what the problem is about putting the citation there. As I mentioned in my edit summary, I can't do it because I haven't seen the inkblots in the RS that should be in the inline citation. I don't dispute that they are probably in the RS, like they are in the wiki. I just think that someone who has seen them in the RS that should be cited in the inkblots section should put up a proper inline citation there! Thanks. --Bob K31416 (talk) 15:56, 7 July 2009 (UTC)[reply]

I think someone (hopefully someone other than me) should write some text to go with the pictures and then an inline citation can go there. Otherwise, a floating ref just looks peculiar. The references are almost immediately below the images, so I don't see it as that big of a deal either way. –xenotalk 15:59, 7 July 2009 (UTC)[reply]
What's the problem with attaching the inline citation to the section title, The ten inkblots, so that it would be useful for the reader like inline citations normally are? That's why I put up the {{fact}}, and it should be there to encourage editors to do that. Thanks. --Bob K31416 (talk) 16:17, 7 July 2009 (UTC)[reply]
Before anyone writes that text, maybe we ought to be clear on which are the "real images" and which are the "real test images" (see the[24] section above)? The source in Commons for these ten images is a website in Spanish (English translation available) written by someone in Argentina. Curiously it ends with the following advice:
"I find that following these tips for prepairing your own responses is much more useful than indicating specifically what to answer: psychologists are not stupid, no matter how stupid their science can be. And the mistakes that could get ourselves busted are not only the answers we give but also how to act in general." Martinevans123 (talk) 16:40, 7 July 2009 (UTC)[reply]
(edit conflict) I've never seen an inline citation in a header like that. Let me look into it a bit. I still think the text is a better way to go about it. –xenotalk 16:43, 7 July 2009 (UTC)[reply]
Xeno, We don't know when there will ever be text in that section. So until then, it seems best to attach the inline citation to the section title. After all, it's the reader who is the main consideration, and an inline citation there would benefit the reader. A {{fact}} there would encourage editors to do that. --Bob K31416 (talk) 16:55, 7 July 2009 (UTC)[reply]
Bob I have never seen a section referenced. The WP:MOS#Section headings does not discuss it. Common sense would seem to dictate that TOC/indexes help navigate a document while sourcing is done to relevant text in the article. I'm sure clarification can be made at the MOS talk page if neseccary. Garycompugeek (talk) 17:06, 7 July 2009 (UTC)[reply]
This should do for now. Wasn't there a source that listed common responses? That might be useful text to go with the section. –xenotalk 17:20, 7 July 2009 (UTC)[reply]
Ah, now in good conscience I can presume that you have checked that the inkblots in the book match, and I can improve that and replace it with a proper inline citation to the actual book. OK? --Bob K31416 (talk) 17:25, 7 July 2009 (UTC)[reply]
Sure, but it seems redundant, no? Inline citations are not a requirement when the text is explicit. An inline citation here would be like writing "Kurt Vonnegut's [[Cat's Cradle]] involves fictional substance known as ice-nine.<ref>Kurt Vonnegut, Cat's Cradle</ref>". And, no, I haven't confirmed. I think DreamGuy above said he did, though. –xenotalk 17:31, 7 July 2009 (UTC)[reply]
Actually, I had something better in mind. But in the process of looking up the 13 digit ISBN at amazon.com, I found this. It looks like the images that are in the wiki have different backgrounds than the images that are in the item at the ISBN. The official images appear to have plain white backgrounds, whereas the images in the wiki don't. --Bob K31416 (talk) 17:51, 7 July 2009 (UTC)[reply]
Nice find! So what do we make of the ones we're using now? Poor scans? Background of paper tarnished with age? –xenotalk 17:56, 7 July 2009 (UTC)[reply]
My guess is water damage, which may have also altered the inkblots themselves. This seems more apparent for the multicolored ones. --Bob K31416 (talk) 18:03, 7 July 2009 (UTC)[reply]

← Yes, that would certainly explain the running colours. –xenotalk 18:04, 7 July 2009 (UTC)[reply]

This speculation seems silly. Ward and Faustian claim to administrate the test as practicing psychologist and they have never spoke up that these are not the correct ink blots and a google search almost always seems to show them this way. Garycompugeek (talk) 18:27, 7 July 2009 (UTC)[reply]
(edit conflict)So where do we go from here? I don't think we can display probably water damaged images and claim they are the official inkblots. I don't even think we can display them in a qualified manner by saying that they are nearly the same as the official inkblots but have been water damaged or altered in some other way. We don't know how they compare. And putting in the wiki any comparison of the altered images and the official images by an editor would be a violation of WP:NOR. --Bob K31416 (talk) 18:30, 7 July 2009 (UTC)[reply]
If the image quality is not perfect that is not a verifiability issue, that is a quality issue. We can show a grainy black a white image of a historical figure and still say it is them, even if they are not really black and white and overexposed. We use the best quality images we can find. The images have been cited to a reliable source, and this source seems to be verifiable. If it is shown as a reference or an inline citation is of little importance, I would support either. If someone can provide higher quality scans that is wonderful, we can always use better copies of public domain images. Chillum 18:34, 7 July 2009 (UTC)[reply]
This doesn't look like just a scan problem. More likely water damage, as I mentioned. As Xeno noticed, the colors are running. We don't know how much it has altered the images. Apparently, no one has even looked at the actual inkblot plates in the RS. --Bob K31416 (talk) 18:44, 7 July 2009 (UTC)[reply]
More speculation. The only way challenge the existing inkblots is to bring verifiable proof that they are incorrect. (and in that case we would use those other ink blots that discredit the previous ones) Garycompugeek (talk) 19:44, 7 July 2009 (UTC)[reply]
Gary, Have you ever seen the actual inkblots in an RS that can be cited here? --Bob K31416 (talk) 20:03, 7 July 2009 (UTC)[reply]
More than one source has confirmed we have the correct images for our reference. This is a quality issue, not a verifiability or original research issue. If the images are less than perfect then we should seek to improve them when we are able to, but currently they are the best representation we have. Chillum 19:47, 7 July 2009 (UTC)[reply]
re "More than one source has confirmed" - Are you referring the the mini B&W images in an advertisement for "blanks" and some small low res ones that had different backgrounds than the ones in the wiki? Those aren't confirmation in my opinion. All you can say is that the general appearance is similar. --Bob K31416 (talk) 20:03, 7 July 2009 (UTC)[reply]
these ones have a yellowy background, lol.

On second thought, I think I'll leave it to you folks to decide what you want to do and pull out now. I really don't have a dog in this show. Also, I thought you folks were pretty good to work with. Good luck. --Bob K31416 (talk) 20:45, 7 July 2009 (UTC)[reply]

aw, don't leave so soon! =) re your over-written question [25], is File:Rorschach inkblots.jpg the one? –xenotalk 20:49, 7 July 2009 (UTC)[reply]
Xeno, You've been very kind and charming, but it's time to move on. I can repeat what I said when I came here originally, I appreciate your invitation. : ) --Bob K31416 (talk) 20:59, 7 July 2009 (UTC)[reply]
Thanks for your input here, especially the "list of reliable sources" section that gave us a little push to...well, work on the actual article, once in a while, rather than just arguing about it ;p –xenotalk 21:00, 7 July 2009 (UTC)[reply]

One more time: I have seen the original images and can confirm the ones here are accurate. The images to me look largely like poor photos/scanning. Most reproductions add color casts invariably, and the existence of tonal shifts does not in any way invalidate these images. Some may also have some water damage, but from a quick glance at a couple of the scans at larger size I see nothing of any substantive difference for our purposes. Considering that at one point people were arguing that we could completely make up our own inkblots entirely out of thin air, it seems odd know to be raising quibbles about nonexact matches in color. Web images and computer screens are never going to be exact matches anyway. DreamGuy (talk) 17:59, 9 July 2009 (UTC)[reply]

A note to the newly arrived editors

It seems the article has attracted the attention of some new editors. I would ask that they review the very lengthy history of this debate, familiarize themselves with relevant Wikipedia policies and guidelines, and participate in the discussions ongoing above to gather consensus before unilaterally removing images from the page. Thanks! –xenotalk 22:44, 7 July 2009 (UTC)[reply]

I welcome the new users to the debate, however Zeitgest and Psychology12345 ‎are both obvious sock puppets of the same person. I had really hoped this debate would not break down into edit warring and sock puppetry but it has now started. Chillum 00:07, 8 July 2009 (UTC)[reply]
Well, there's also the possibility they were alerted to this dispute from some web forum somewhere. I would like to think the regulars here know better than to engage in sockpuppetry. –xenotalk 00:09, 8 July 2009 (UTC)[reply]
No, a look at the timeline:
22:10, 7 July 2009 Psychology12345 creates a new account
22:11, 7 July 2009 Psychology12345 removes images
22:25, 7 July 2009 Psychology12345 removes images
22:32, 7 July 2009 Zeitgest creates a new account
22:36, 7 July 2009 Zeitgest removes images
22:43, 7 July 2009 Zeitgest removes images
And it is very clear this is the same person. Perhaps it is a form a meat puppetry, but that is the same animal. I would like to think better of the regulars too, but even reasonable people will become unreasonable will they will not give up something they cannot have. Chillum 00:12, 8 July 2009 (UTC)[reply]
I am going to find a wholly uninvolved admin to look into this. Chillum 00:18, 8 July 2009 (UTC)[reply]
Perhaps you are right. But there is also the possibility that it was a non regular (perhaps even a non-Wikipedian) who came across this debate and then raised the issue at some off-site forum. The regulars here, while persistent, have for-the-most-part acted within policies and guidelines. –xenotalk 00:19, 8 July 2009 (UTC)[reply]
I have asked Bongwarrior to look into this. I looked at the block logs and found he is active at this time, and as far as I know has not involved himself in this debate. Chillum 00:30, 8 July 2009 (UTC)[reply]
It looks like Bongwarrior has come to a similar conclusion as me. I think as long as we keep an eye on these accounts and they remain inactive that there is not much more to worry about. Chillum 00:59, 8 July 2009 (UTC)[reply]

Summary of archives

I understand that the archives for this debate are so incredibly huge that most people will simply not read it. Due to this I have dusted off an old piece of perl code I wrote that creates an index of a page by going through the history and watching when sections are removed. Here is an index of links to the last revision of each section that has ever been on this page before it was removed: User:Chillum/Rorschach test talk page archive. Chillum 02:24, 8 July 2009 (UTC)[reply]

Wow, what a repetitive read. Like watching a TV show during a writers strike. Chillum 13:13, 9 July 2009 (UTC)[reply]

Well, if they don't read the archives, I guess we'll get a few more repetitions? Martinevans123 (talk) 20:46, 10 July 2009 (UTC)[reply]

That is why I indexed the archives, so that we don't have to reject all the same arguments for the nth time. Chillum 00:14, 11 July 2009 (UTC)[reply]

Oouy Veyy. This may go round and round forever. Perhaps we should stub image talk out like we do for Muhammad? Garycompugeek (talk) 23:17, 10 July 2009 (UTC)[reply]

Forever is quite a long time. That's what we do is it? Like a used cigarette, you mean? Martinevans123 (talk) 23:44, 10 July 2009 (UTC)[reply]

Image of Hermann Rorschach

Glad to see that the image of Hermann Rorschach has now been restored to the article(s) by Chillum's addition of "Hermann Rorschach, psychiatrist died in 1922, so this image is public domain." to the Licensing field. This seems to have satisifed Garion96. But there still seems to be no source information, as was pointed out to the original image uploader on 2 Jan 2007.

Surely, regardless of the subject matter, the copyright of a photograph is held by the photogapher (and up to 70 years after his death) or by a publisher. I think it seems unlikely that less than 70 years have passed since the death of the photographer in this case, even if his ir her identity is known. But were the rights holder suddenly to come forward, I think they might be entitled to a compensation payment for the infringing use of the image. I wonder if this photograph has ever been published in a book about Rorschach? Martinevans123 (talk) 17:28, 9 July 2009 (UTC)[reply]

You don't seem to get US copyright law. If it was from before 1923, which it clearly was in this case because it's not a photo of a skeleton of Rorschach, then it's automatically public domain. The 70 years from death of the photographer does not apply in this country for the time period in question. Source is irrelevant, as whereever it came from it's pubilc domain and the photographer would never "be entitled to compensation payment". DreamGuy (talk) 17:34, 9 July 2009 (UTC)[reply]
Very sorry, it seems I "don't get" a lot of things. So the country in which the photograph was originaly published is irrelevant? Did wikipedia ask for source information just for RS? Thanks. Martinevans123 (talk) 17:43, 9 July 2009 (UTC)[reply]
Wikipedia always asks for source info these days because it's easier to demand it even in cases where it's irrelevant then to let the typical image uploaders who don't know anything about copyright upload whatever they want. Unfortunately some of the people going around tagging images as unsourced don't use some common sense before doing so. DreamGuy (talk) 17:49, 9 July 2009 (UTC)[reply]
I was asking: "So the country in which the photograph was originaly published is irrelevant?" Thanks. Martinevans123 (talk) 18:00, 9 July 2009 (UTC)[reply]
Good question. Wikipedia:Public domain seems to cover the topic. I don't have time to read it now, but will later. Chillum 21:37, 9 July 2009 (UTC)[reply]
(note) If we look here [26] we see this "The picture of Hermann Rorschach on this page is from H. Ellenberger, The Discovery of the Unconscious (Basic Books 1970). Reproduced with kind permission from Institut Henri Ellenberger, Paris."
Wondering if the original photograph (original being the one Ellenberger used since Rorschach would have been dead for almost 50 years by the time The Discovery of the Unconscious was published) is owned with rights by The Hermann Rorschach Archives and Museum 76.217.156.168 (talk) 21:52, 14 July 2009 (UTC)[reply]
Well, that answers that, emailed Rita Signer, contact for The Hermann Rorschach Archives and Museum who says and I quote "This photo of Hermann Rorschach is kept at the Hermann Rorschach Archives and Museum. The photographer is unknown. The copyrigt is held by the International Society of the Rorschach (ISR)." (typo included) 76.217.156.168 (talk) 14:17, 15 July 2009 (UTC)[reply]

US Copyright law

Dreamguy, re your change [27], does the Copyright Term Extension Act not apply? Thanks. Martinevans123 (talk) 17:33, 9 July 2009 (UTC)[reply]

Per that article: "Under this Act, additional works made in 1923 or afterwards that were still copyrighted in 1998 will not enter the public domain until 2019 or afterward" The inkblots were made prior to 1923. US copyright laws over the years have been a mess of conflicting periods of coverage and other standards. The original US copyright laws (ones in effect for old, public domain works such as these inkblots) had nothing to do with the death of the creator but only with the publication date. It's not safe to lump the US in with other countries on the 70 years thing, as most of the time it doesn't match up. DreamGuy (talk) 17:40, 9 July 2009 (UTC)[reply]
Perhaps it depends on which inkblots we are discussing? As far as I can see the ones on the test cards are being printed even today and so are "works made after 1923". Psychologists aren't using the actual ones printed by Hermann Rorshach before he died, are they? Martinevans123 (talk) 17:51, 9 July 2009 (UTC)[reply]
They are using the original images he made before he died, yes. Merely copying old images and printing new versions faithfully reproducing the originals does not give a new copyright. See Bridgeman Art Library v. Corel Corp.. DreamGuy (talk) 15:31, 10 July 2009 (UTC)[reply]
Did those originals have white backgrounds or coloured? Do we know which are being used now? Does that matter? Martinevans123 (talk) 16:15, 10 July 2009 (UTC)[reply]
See my comment at the bottom of #Inline citation for inkblots needed; this may explain where the different versions of the images come from. –xenotalk 16:17, 10 July 2009 (UTC)[reply]
That's a fascinating citataton, xeno, that suggests all sorts of variety in the original printing. I wonder how it continues and concludes. Perhaps it explains away the "water damage theory"? But I am still left wondering whether what are used today on the cards are really the same as Rorshcach himself produced in 1921, or even the same as those published by Verlag in 1927. Martinevans123 (talk) 16:37, 10 July 2009 (UTC)[reply]
Same images. There are photos of Rorschach displaying some of the inkblots in question (one even was the cover of a book), and they are the same ones. Variations in background colors make no difference. DreamGuy (talk) 17:18, 10 July 2009 (UTC)[reply]
I see. Which book was that? But "makes no difference" to who? Was that Rorscharch's own view (and that of his proponents) or is it just the view of the US legislature? How different does an image have to become to be seen, legally, as a different image? Martinevans123 (talk) 20:43, 10 July 2009 (UTC)[reply]
If you don't understand copyright law and sincerely want to know, please educate yourself; don't insist other people do it for you. DreamGuy (talk) 23:06, 11 July 2009 (UTC)[reply]
I have insisted nothing. By simply asking some questions I had hoped that my lack of knowledge might enable you, or any other editor, to educate the community here. If you understand copyright law and sincerely want to explain it to us, please just answer. Thanks. Martinevans123 (talk) 08:16, 12 July 2009 (UTC)[reply]

Sockpuppet investigation

At what point do we file a sockpuppet investigation into single purpose accounts such as User:Danglingdiagnosis, User:Zeitgest, User:Psychology12345 and User:Dolphinfin? It's pretty clear that these accounts exist to try to give the appearance of more people supporting a position than actually exist. It wouldn't surprise me if other socks were at work as well. Maybe if we figure out who is responsible they can all be banned (with the sockmaster) and the constant whining will die down or go away completely. DreamGuy (talk) 17:45, 9 July 2009 (UTC)[reply]

You'll still get my "whining", I can assure you, constant or not. Martinevans123 (talk) 17:57, 9 July 2009 (UTC)[reply]
AS long as it's not disruptive and you limit yourself to one account, that's fine. DreamGuy (talk) 15:57, 10 July 2009 (UTC)[reply]

I am neither sockpuppet, nor impressed with the arguments for keeping the images up on the page. The issue is quiet clear to me. It is a matter of ethical conduct. A large, internationally recognized society, the American Psychological Association, includes within its ethical guidelines to protect test material. These test materials are potentially being spoiled by being placed on this page. Although I am not a "sockpuppet" as you call them, I would support their use as part of protecting the test's images. This would be keeping with the ethical principles of the APA. Dolphinfin (talk) 17:56, 9 July 2009 (UTC)[reply]

Confirmed sockpuppet of Psychology12345 and blocked indef. Garycompugeek (talk) 17:13, 10 July 2009 (UTC)[reply]
I don't think it is fair to lump Danglingdiagnosis in with these other blatant single purpose accounts. He created his account in 2006 and has contributed to a variety of medical articles over the years. I am going file a checkuser request on the new SPAs that have been showing up to remove the image after work. Chillum 18:07, 9 July 2009 (UTC)[reply]
Dolphinfin, may I ask how you came to find this debate? Were you contacted either in private or through a public forum? Chillum 18:08, 9 July 2009 (UTC)[reply]
I'm happy to answer. I was bumping around wikipedia today, procrastinating as I often do, and decided to see what was put together for the Rorschach test. My interest is professional, and I recognize that this particular test often generates a great deal of controversy, both within and without the clinical psychology profession. However, the writing is now on the wall, this test is valid for quite a few purposes, and much of the complaints leveled at it in the 1990's and early 2000's have been resolved through empirical work and responses. I see it as an enormously useful tool that has achieved its utility through a great deal of work from a lot of well meaning people. That isn't to say that like ALL technology, it can be used to harm someone, but the preponderance of the use is for helping others. I see that utility of it being threatened and degraded by its posting here. When I saw it I was quite surprised that wikipedia would allow it to be posted. So, I created a name so that I could edit it off; isn't that how this is supposed to work? Only then did I realize that there was so much controversy going on behind the scenes. In general, I am a huge fan of wikipedia. I trust it more than most sources for information, recognizing its limitations. Yet this is one time that the process has failed. There are times when just because you can do something, doesn't mean you should do it. Just because the information is available, doesn't mean it should be broadcast. I see this as one of those cases. I think that much of the controversy around the Rorschach stems from fear of it. The actual practice and implementation of it is not nearly as nefarious as those who would see it debunked imagine it to be. In fact, it is not nefarious at all. I see the hard work of my colleagues being destroyed (however minutely) through the posting of these cards. I don't see why everyone feels that they are entitled to this information. It is not being kept secret out of some sort of "masonic" or "secret society" ritualism; it is more useful when someone can be exposed for the first time in the testing scenario. I guess I am disappointed that wikipedia has gone the route it has gone with this. Although it may not be copyrighted, it should be protected information. To use the old cliche, I hope those who insist on it being posted on here don't need it at some point, because it may have been worn away as a tool by then. I see many acceptable alternatives, such as a inkblot that is not part of the 10 cards that could serve as an exemplar. Or, to a lesser degree, an outline or two. Please remove them for the sake of clients who seek services of clinical psychologists. That is all I would ask. Dolphinfin (talk) 01:55, 10 July 2009 (UTC)[reply]
Yet another expert opposed to showing the images.Faustian (talk) 02:19, 10 July 2009 (UTC)[reply]
Yet another person who is clearly not a real account, you mean. One vote per person, please. DreamGuy (talk) 15:57, 10 July 2009 (UTC)[reply]
Could you explain to us how you know why Dolphinfin is not a "real account"? I can't believe that your statement was just meant to be plain insult. Martinevans123 (talk) 16:20, 10 July 2009 (UTC)[reply]
Martin, Dolphinfin is an obvious sock puppet and has been indefinitely blocked as such along with his/her other two accounts. Sock puppets are not "real accounts" because they are just one person pretending to be many. It is a way people push their opinion when their arguments fail. Chillum 16:29, 10 July 2009 (UTC)[reply]
Yes, I see they are now banned. But how is that sock puppetry proved exactly? And how do we know that the sock puppet accounts weren't created to discredit the proponets of image removal? If genuine, it seems to be another example of the ethics of professionals at odds with the rules of wikipedia. If that editor now creates a new single account to continue debate will they be allowed to contribute? Martinevans123 (talk) 16:47, 10 July 2009 (UTC)[reply]
If the "Sockpuppet Investigation" achieves nothing else it has solicited one of the most eloquent and reasonable statements yet added to the whole discussion so far. Thank you, Dolphinfin. Martinevans123 (talk) 12:31, 10 July 2009 (UTC)[reply]
Martin are you encouraging sockpuppets? Garycompugeek (talk) 17:13, 10 July 2009 (UTC)[reply]
As you can see< Gary, I made this comment before the investigation concluded. The statement still seems genuine to me, partly for the reasons suggested by Faustian. But I'm even less sure now how we would ever decide. Martinevans123 (talk) 17:25, 10 July 2009 (UTC)[reply]
  • I think it's more likely that these users were canvassed here from some off-wiki forum. I do hope that none of the regulars are responsible for this, and remind that meatpuppetry carries the same risk of sanction as sockpuppetry. –xenotalk 18:15, 9 July 2009 (UTC)[reply]
I have filed for an investigation: Wikipedia:Sockpuppet investigations/Psychology12345. Perhaps check user results will shed some light on the matter. Chillum 20:36, 9 July 2009 (UTC)[reply]
More like yet another sock puppet opposed to showing the images Faustian. The investigation came back positive. Chillum 03:48, 10 July 2009 (UTC)[reply]
Question: what if one person sees the images, is horrified, and then tells his colleagues who then also chip in, without any instructions from the first guy. Is this meat puppetry? Do people have a right to tell others what they see? and if someone is told by a colleague s their right to contribute forfeited? This seems different from canvassing for support to screw with the system. The talk pages of one of the two "sockpuppets" seemed to suggest it was two guys in the same office who came across the image. I also wonder about the "likeliness" of the third, dolphinfin. His case showed different geography from the other two but was based on "similar behavior" and the fact that it can be accessed remotely (is that rare or something?) Faustian (talk) 03:56, 10 July 2009 (UTC)[reply]
An interesting possibility. The other possibility is that someone here is using proxies. Both a check user and an uninvolved administrator came to the same conclusion as me. Chillum 04:58, 10 July 2009 (UTC)[reply]
Having now read Faustian's comment I am very unconvinced that this sockpuppety has been "proved" at all. But I am guessing, not being an administrator, I have no say in that process? Martinevans123 (talk) 16:51, 10 July 2009 (UTC)[reply]
Being that you have confessed no knowledge of how we flush these things out in the open do you believe your qualified to discredit the process? Garycompugeek (talk) 17:13, 10 July 2009 (UTC)[reply]
It was more of question Gary. But I'm certainly sorry I asked now. So that's how "we" do it, is it? My belief in my own utility here grows smaller by the hour. But thanks for all the encouragement. Martinevans123 (talk) 17:21, 10 July 2009 (UTC)[reply]
The best thing to do, Martin, would be to ask the blocking administrator. You could also comment at Wikipedia:AN#Uninvolved administrator needed for sock puppets. FWIW, checkusers are intentionally vague about how they come to their conclusions, to prevent sockpuppeteers from gleaning much information about how they are identified by technical evidence. That being said, if the CU said "confirmed" then that probable means they were editing from the same IP/browser. "Likely" probably had more to do with locational evidence, which would lend credibility to Faustian's comment above about colleagues discussing and unknowingly becoming meatpuppets of eachother (no doubt this distinction isn't immediately apparent to non-Wikipedians and actually pretty much breaks down as a term if they weren't previously editors). –xenotalk 17:23, 10 July 2009 (UTC)[reply]
Thank you once again, xeno, for taking time to explain and not to make personal judgements. Martinevans123 (talk) 17:30, 10 July 2009 (UTC)[reply]
Considering they showed up near each other in time, acted the same way, gave the same excuses(compare User talk:Zeitgest with the reasoning Dolphinfin gave above), wrote in the same style, and had technical evidence linking them, I think we can assume it is just someone trying to bypass 3RR and unduly influence the debate with multiple accounts. It happens all the time on Wikipedia. Chillum 19:55, 10 July 2009 (UTC)[reply]
Thanks for the explanation Chillum. Martinevans123 (talk) 20:37, 10 July 2009 (UTC)[reply]
According to the checkuser, zerttgeist and psychology12345 were "confirmed" sockpuppets. However, the explanation they/he gave makes the confirmation problematic. If indeed it's a case of two people in the same practice or the same graduate department finding out about this page and acting on what they found, wouldn't a checkuser result in a "confirmed" categrization? According to checkuser, Dolphinfin was not a "confirmed" sockpuppet but a "likely" one, based on "Likely IPs are geographically distinct, but are those that can easily be accessed remotely, other technical evidence is similar, and behavioral evidence is similar." This doesn't seem strong enough for something as extreme as a block, in dolphinfin's particular case, and given his explanation for his actions which cover the "behavioral evidence". Moreover, his edit warring stopped after he was informed of it, so at the time of the block he wasn't doing anything wrong. It might very well be a well-intentioned novice wikipedia user who has now been driven off from further contributions, which would be a sad thing. It ought to be noted that the same person calling for the block for sockpuppetry also unfairly accused danglingdiagnosis of being one of the sockpuppets too. We should avoid witchhunts here.Faustian (talk) 21:12, 10 July 2009 (UTC)[reply]
You can always ask for further review at Wikipedia:AN#Uninvolved administrator needed for sock puppets. I suspect however that most administrators, like me, will see this as an obvious sock puppet. I wonder how you would feel about these three brand new carbon copy users if they all were supporting the showing of the images? Chillum 00:24, 11 July 2009 (UTC)[reply]
I certainly suspected a well intentioned novice and to drive them away would indeed be a sad thing. Perhaps dolphinfin will request an unblock and provide further explanation. It seems unfair to expect novices to always know and apply all the wikipedia rules. I hope danglingdiagnosis has received an apology. Martinevans123 (talk) 21:50, 10 July 2009 (UTC)[reply]
Apology? For what? He's clearly not a new uninvolved editor either. DreamGuy (talk) 22:49, 11 July 2009 (UTC)[reply]
Don't worry, I suspect the person who created those three accounts is still using Wikipedia, and possibly even still using this very talk page. Lots of people know how to use proxies, what most people don't know is that some of them forward your IP and if one uses such a proxy they will be revealed by a checkuser. Chillum 00:07, 11 July 2009 (UTC)[reply]
It appears that Dolphinfin did in fact requested an unblock just a three hours after you predicted it. You sure called that one. Chillum 20:30, 11 July 2009 (UTC)[reply]
Hopefully their shenanigans will lead to the sockmaster soon. DreamGuy (talk) 22:49, 11 July 2009 (UTC)[reply]
And the unblock has been declined. Chillum 23:10, 11 July 2009 (UTC)[reply]
By whom and where? Looks like a nasty witchhunt, given the facts. I guess we've come to blocking/banning new users, trying to hide the debate on a subpage, etc. as a way of forcing "consensus" and silencing the experts.Faustian (talk) 23:57, 11 July 2009 (UTC)[reply]
And, of course, you would be equally appalled if sockpuppets supporting the inclusion of the images were blocked, right? Or does your failure to assume good faith only apply to the blocking of proxy accounts that are on the same side of your little obsession here? Resolute 23:38, 13 July 2009 (UTC)[reply]
And I suppose we'll never know why the unblock was declined. Let's hope the next new editor here has more success. Martinevans123 (talk) 08:21, 12 July 2009 (UTC)[reply]
Those were not new editors, they were sock puppets. Chillum 13:13, 12 July 2009 (UTC)[reply]
  • I've confirmed via email that Psychology12345 and Zeitgest are distinct users, and still have strong doubts that they are socks of any of the regulars here. I've also verified their assertion that they are indeed both students of the same graduate-level psychology programme. They've been unblocked with the understanding they should participate in the discussion rather than unilaterally remove images. –xenotalk 17:02, 15 July 2009 (UTC)[reply]

Lead para about ethics

Why is the last lead para about ethics included? It's not about the ink blots or the test at all. I think it should be removed. Verbal chat 19:39, 9 July 2009 (UTC)[reply]

I think it's a good compromise to balance the concerns of showing the images and an image in the lead with the argument that this may cause harm. It's not a disclaimer per se, but it does at least highlight the concern in an encyclopedic fashion. –xenotalk 19:47, 9 July 2009 (UTC)[reply]
The professional bodies responsible for teaching the test and overseeing its use have very strong concerns about the public display of test materials, of which the inkblots are prime examples. These concerns are only made more serious by the fact that the Rorschach may be used to diagnose suicidality. So the para is certainly "about" the inbklots, and if it doesn't seem to be so, perhaps it ought to be improved so that it does. Martinevans123 (talk) 19:54, 9 July 2009 (UTC)[reply]
So the reason we have this discussion in the lead is because of an argument on wikipedia, rather than for encyclopaedic concerns? It really looks like a disclaimer, and I really don't think that is a good enough justification. (Just so you don't think I'm dumb, I've studied psychology at Uni level (not my main subject, though) and I know many who work in the field.) Verbal chat 20:01, 9 July 2009 (UTC)[reply]
Nor do I. Garycompugeek (talk) 20:06, 9 July 2009 (UTC)[reply]
I think it reads well, is encyclopedic, and expounds on the final sentence (or "final fragment" seperated by ;'s... what do you call that?) of the 3rd lead paragraph. –xenotalk 20:11, 9 July 2009 (UTC)[reply]
The users of the test, as well as the publisher don't see the ethical concerns as "an argument on wikipedia". If you want to read about the inkblots as if they were pretty patterns in a coffee-table book, then you wouldn't want any ethical stuff. But if you're more concerned about the USE of this test, you need to know. Martinevans123 (talk) 20:27, 9 July 2009 (UTC)[reply]
I supported that lead paragraph as a compromise, however it seems the compromise did not satisfy anyone because they still argue for the removal of images. It seems like it has accomplished nothing. I agree there are synthesis issues, it is a group of related statements by different sources lumped together to advance a point of view. This seems to be present more for the sake of the debate than the article. Chillum 20:32, 9 July 2009 (UTC)[reply]
It had accomplished quite a lot before people started taking it out without discussion and arguing that it should not be there. I have to agree that the two issues have become entwined - the insistence on the inkblot at the top led to the ethics being put near the top also, where you might not have expected to see them. But as with the later introduction of the ten inkblots, I too have become wary of compromise, as I feel once again that the rug is about to be pulled from under those who want the inkblots not to appear or to appear less prominently. Martinevans123 (talk) 21:11, 9 July 2009 (UTC)[reply]
When we took the most recent poll nobody sided with the idea of not showing them at all, and most people decided they should be shown as they are. Chillum 21:20, 9 July 2009 (UTC)[reply]
When exactly was that "most recent poll" and what was it's title? Not this one then:[28]? Martinevans123 (talk) 21:47, 9 July 2009 (UTC)[reply]
It is still on this page: #An informal survey for the inclusion of the 10 inkblots. The consensus was clearly in favour of not hiding the images, and nobody supported the idea that they should not be shown at all. Chillum 21:55, 9 July 2009 (UTC)[reply]
Oh, I see now that your position was against the showing of the images, you just did not count yourself. My mistake. My point regarding consensus stands despite this. Chillum 21:57, 9 July 2009 (UTC)[reply]
I'm sorry, I "did not count myself"? Martinevans123 (talk) 22:01, 9 July 2009 (UTC)[reply]
So, voting = concensus? Martinevans123 (talk) 22:09, 9 July 2009 (UTC)[reply]
I was simply pointing out why I did not notice your stated position in the poll, you did not number your position like everyone else did. I never said voting = consensus, there was plenty of discussion to go along with it. Chillum 22:17, 9 July 2009 (UTC)[reply]
"I am not a number, I am a free man". Hmmm, yes plenty. Martinevans123 (talk) 22:23, 9 July 2009 (UTC)[reply]

I'd just like to add that it is rather obvious that the images should be included, and if I knew of the debate beforehand I would have said that. That doesn't change the apparent irrelevance of the paragraph in the lead. Verbal chat 21:28, 9 July 2009 (UTC)[reply]

I have to agree with verbal. If not for the image debate this section would not have been included because it is not really relevant to the subject of the article. Chillum 21:34, 9 July 2009 (UTC)[reply]
I am truly astounded that you can suggest that the ethics of psychometric testing are "not really relevant" to the subject of this article. Why is it "rather obvious" that the images should be included? And is it really the task of an encyclopedia just to point out what is obvious? Martinevans123 (talk) 21:59, 9 July 2009 (UTC)[reply]
It is not relevant because the sources are talking about test material in general not the Rorschach test. Perhaps it would be better suited in an article on medical ethics. There seems to be undue interpretation of the sources, and the way they are presented implies that they are referring to this test. Chillum 22:18, 9 July 2009 (UTC)[reply]
Um, they're actually refering to ALL tests, this one included. But which other article shows the test materials quite so openly? Martinevans123 (talk) 22:26, 9 July 2009 (UTC)[reply]
Welcome Verbal. The conflict is real and not just limited to the English Wikipedia. It is found on the French, the Polish, the Norwegian, the Italian, and it is found prominently on (at least) the first three web-sites listed on Google. Some of these sites are showing the images with the expressed purpose of helping people "beat the test," -- one in circumstances surrounding a custody dispute of children. We can't ignore this dispute. Instead, we're attempting to stand out from the fray and provide a neutral point of view of both sides of the conflict. This is complicated by the fact that the dispute centers around the idea of "pre-exposure" to the images, so we can't expose our readers to the image without showing bias. (e.g. a TV news reporter wearing a fur coat reporting on camera at a street riot between people throwing paint and arguing about animal rights versus the right to wear a fur coat.) We have reputable sources that "categorically" say that the lack of security of these test results can harm the ability of a psychologist to protect the welfare of his patient, and we have letters from the the designers of the test that make the same claim "specifically" to the Rorschach. We also have an article from Scientific American that while highly critical of the Rorschach test (while granting that it seems to do well with diagnosing Schizophrenia and bi-polarism) chose not to publish the photos saying simply "the images cannot be published." Here at Wikipedia, the contention seems to center around whether or how we are to be informed by these sources and weigh them against our goal of providing information. So I'm thinking we have a neutrality WP:NPOVissue up against and WP:NOTCENSORED problem. The question is how much should we let this conflict (this "outside influence" as Chillum calls it) move us in either direction. I think that since
  1. WP:NPOV is so very important, and
  2. Wikipedia can and has provided restrictions on information before, (see WP:IINFO) And
  3. since censorship is not simply an either/or binary proposition, (book burning being one extreme, a movie rating system, Rated PG-13, on the other)
I'm thinking we need to inform our readers about the issue BEFORE they see the images so they can make up their own mind. The spirit of most wikipedia policies is to let the reader make up their own mind, so anything that helps them do that is, I think, very appropriate. Clearly, any compromise is going to hurt each of us Wikipedians in some way, and that's okay with me. I think we need to explore the dynamic tension of this conflict and find a consensus. You're welcome to help us do so. Let's see, did I forget to mention something? Just read the index and find the 6 arguments PRO and the 6 arguments CON, and you can get a good perspective. And welcome. Danglingdiagnosis (talk) 01:29, 10 July 2009 (UTC)[reply]
Having a discussion about ethics in the lead is a compromise. I am happy to see it stay.--Doc James (talk · contribs · email) 03:12, 10 July 2009 (UTC)[reply]
I think we should have something, but not what we had for the reasons I gave above. Chillum 04:19, 10 July 2009 (UTC)[reply]

Not showing the images is not an option. They are relevant, add to the encyclopaedic content of the article, and are PD. Ethics should be mentioned in the article, but only related to this test and not in the lead. WP should not bow to special interest groups in suppressing information, and I'm glad it hasn't. This over the top disclaimer needs significant trimming to meet our policies. I'm afraid whenever I see an ink blot test I either see bats and Kim Basinger, or a Far Side cartoon. Verbal chat 07:59, 10 July 2009 (UTC)[reply]

Whenever I see an ink blot test I see a Wikipedia article. Martinevans123 (talk) 09:29, 10 July 2009 (UTC)[reply]
I agree with Verbal. Having the information in the article at all is a compromise but having it in the lead is WP:UNDUE. Garycompugeek (talk) 13:17, 10 July 2009 (UTC)[reply]
I think our standard of inclusion for the subtopic "ethics" in the Rorschach test article should be sources talking about ethics and Rorschach tests. Taking sources that talk about testing in general and then applying it to this test is original research. Where are the sources about ethics as related to the topic of this article? Chillum 14:20, 10 July 2009 (UTC)[reply]
It is no more OR to include the Roschach when a source mentions tests in general than to include elephants when a source mentions all mammals just because that specific source did not mention any mammal species specifically. The APA Code doesn't mention any specific tests and the APA statement about the harmfulness of compromising tests doesn't mention any specific tests either. To claim this statement doesn't apply to the Rorschach because the Rorschach isn't specifically named is like claiming that claiming that laws against thefts don't apply when the specific victim isn't listed by name in the law itself. Nice try. Faustian (talk) 14:32, 10 July 2009 (UTC)[reply]
By that logic we could include any source that has general relativity to topic regardless of specifics. Garycompugeek (talk) 14:44, 10 July 2009 (UTC)[reply]

(edit conflict)

All tests huh? So this would apply to a reflex test too? Is it unethical to let the public know your going to use a little rubber hammer? Repeatedly sources are being brought up that do not directly apply to the subject and they are being creatively interpreted to support a position they do not support. That is original research. I ask once again, why are there no sources discussing the ethics as related to the subject of the article? If somebody would provide that then we could have a section on ethics that is not original research. Chillum 14:40, 10 July 2009 (UTC)[reply]
Ah, hammers again, little rubber ones this time. Sorry to have confused you Chillum, but no I meant all psychological tests. I had assumed the context of APA and BPS might have provided a clue. Martinevans123 (talk) 16:00, 10 July 2009 (UTC)[reply]
I don't think it's a great leap to induct that the APA concerns speak to tests where reliable goes down with pre-exposure, and we've sourced that this is the case with the Rorschach. I still think the paragraph is entirely appropriate and encyclopedic. Sure, if we can find another source to attach to it, that's great, but for now I think that in the interests of balancing the concerns we should leave it where it is. –xenotalk 14:48, 10 July 2009 (UTC)[reply]
From WP:OR: "Drawing conclusions not evident in the reference is original research regardless of the type of source. It is important that references be cited in context and on topic."
This is exactly what is being done here. All we need are sources that actually talk about this huge controversy being described by a few Wikipedians, and then to limit our commentary to a faithful representation of those sources. Our original research policy makes it clear that A + B does not equal C unless a reliable source has already made that connection for us. We should not be combining sources in this matter. Chillum 14:50, 10 July 2009 (UTC)[reply]
I must admit I'm surprised we haven't been able to find anything like this. –xenotalk 14:53, 10 July 2009 (UTC)[reply]
Ethics codes do not mention any test specifically just like laws don't mention any potential victim of a crime specifically. There seems to be nothing inherently surprising about this. BTW, the ethics code also doesn't mention by name every person whom the psychologist is forbidden to cheat or steal from. The APA has concluded quite clearly that "The APA states that the dissemination of test materials "imposes very concrete harm to the general public" as well, in that "there are a limited number of standardized psychological tests considered appropriate for a given purpose". Those who implictly claim that the Rorschach inkblots do not constitute "test materials" when they claim that the APA code doesn't apply to the Rorschach becasue the ROrschach isn't listed by name are the ones engaging in OR. Faustian (talk) 14:54, 10 July 2009 (UTC)[reply]
Yes, but we can find reliable sources on the ethics of stem cell research, and of cloning, and of numerous other medical topics. Where are the reliable sources on the ethics of showing Rorschach test images? It is not the place of an encyclopedia to document a topic that is not already being covered by reliable sources. Our original research policy says that too. Chillum 14:59, 10 July 2009 (UTC)[reply]
Is there something about the ethics, specificlly, of every kind of species of animal to be cloned? Is a source that speaks about cloning "animals" not applicable when mentioning the cloning fo Dolly specifically?Faustian (talk) 15:03, 10 July 2009 (UTC)[reply]
Indeed. I'm not sure why people want this paragraph removed. It doesn't strike me as OR or SYNTH at all. Then again, I may be biased being the one who penned it. –xenotalk 14:57, 10 July 2009 (UTC)[reply]
I don't want it removed, I want it replaced with something more in line with policy. There must be something related to ethics and Rorschach tests out there, and we should accurately reflect what that says. Chillum 15:00, 10 July 2009 (UTC)[reply]
Fair enough. –xenotalk 15:04, 10 July 2009 (UTC)[reply]
I am also not sure why a sub-topic is in the lead(other than to act as a disclaimer). The lead is to summarize the topic, this content refers to nothing in the article, rather it is a self contained glob of loosely related sources. I think it should be a section in the controversies section(once we have a source that established that there is such a controversy related to Rorschach tests). Chillum 15:11, 10 July 2009 (UTC)[reply]
"Ethics in psychology" By Gerald P. Koocher, Patricia Keith-Spiegel , pp 159-160 http://books.google.ca/books?id=KwatUOmKCrUC&pg=PA159&lpg=PA159&dq=psychologists+outraged+at+rorschach+test&source=bl&ots=N_GPpPM4P9&sig=QzqAQTmkhG4rySHGq8kw9t8rBKI&hl=en&ei=BlpXStvcBo7CNq_puJ0I&sa=X&oi=book_result&ct=result&resnum=8. –xenotalk 15:16, 10 July 2009 (UTC)[reply]
Now that is the start of a good section on ethics. Good find. Chillum 15:19, 10 July 2009 (UTC)[reply]
But they are exactly the same ethics. And don't appear to be wholly "subjective" either. So, if we knew that black labradors are dogs and that dogs can bite, to state "black labradors can bite" without a verbatim reliable source would be WP:OR, yes? Martinevans123 (talk) 16:09, 10 July 2009 (UTC)[reply]

Both from previous discussion and the above, there was clearly no consensus to add the major WP:NOR and WP:NPOV violating text to the lead, and no consensus exists now, so it will be removed per WP:BRD. If someone would like to propose a new version that is fully sourced to reliable sources and not slanted in such a way as to give WP:UNDUEWEIGHT, by all means write up something to suggest here, and if you get consensus to add it through hammering out the wording, then it will be added, but not before then. DreamGuy (talk) 15:59, 10 July 2009 (UTC)[reply]

It's already at BRRRD, so why would you do that before we finish discussing and looking for better sources? –xenotalk 16:04, 10 July 2009 (UTC)[reply]
Simple, because we should not be writing the content first then looking for sources, we should find the sources first then write the content. We have at least one very good source now, thanks Xeno, so instead of putting that old stuff back lets write something up from that. Chillum 16:31, 10 July 2009 (UTC)[reply]

Propose moving "removed the inkblot" debate to a sub page

This has been dominating the article talk page for too long. I suggest we set up a sub-page to keep this page manageable. There has been a pitiful level of non-inkblot removal related discussion on this talk page and the article is suffering because of it. I saw the same thing on the Muhammad image removal debates, the article talk page got so blocked up with one issue that it would not function as an article talk page anymore. Once they moved the debate to a sub-page the regular talk page once again began to be used for constructive purposes. It is my hopes that moving this one issue to a sub-page will restore productivity to this page, and benefit the article. Opinions? Chillum 00:17, 11 July 2009 (UTC)[reply]

The problem is that the editors responsible for most of the actual non-inkblot related content of the article are the ones objecting to the inkblot. As long as no consensus re: the inkblots is achieved, the experts will have little else to discuss. I suppose exiling the inkblot conversation off to some subpage is your way of ending the debate.Faustian (talk) 05:13, 11 July 2009 (UTC)[reply]
Again I note that you are falsely trying to claim that all the experts support the removal. If you have nothing else to discuss here, you have no point to even being on the page anymore, because you already lost decisively on the one thing you apparently give a damn about. DreamGuy (talk) 22:39, 11 July 2009 (UTC)[reply]
Wow, a lot of false claims all at once. I never claimed that "all experts support the removal." Actually I don't support complete removal from wikipedia of the inkblots. Secondly, name one expert involved here who does not support some sort of suppression of the inkblots. With regards to "the one thing I care about", well, when there is a wound the focus is on healing that wound before moving ontop to other things. I may have contributed more referenced content to this article than all of the one pushing to include the inmages combined. i would like to contribute more, but cannot if the article stands the way it is, because doing so - contributing to an article that harms people - is unethical and immoral. That's where I stand. And most experts would do likewise.Faustian (talk) 23:48, 11 July 2009 (UTC)[reply]
Again with the speaking for "experts" (switching from all to "most" doesn't change anything). As far as naming an expert "here" goes, I'm clearly more an expert on this topic than MartinEvans123 is as he doesn't seem to know when the inkblots were made, if the ones being used now are the same ones Rorschach made, and other extremely basic facts about this topic. And as far as the real world goes, plenty of experts think the Rorschach is inherently flawed, so obviously they don't care if the images are here. And you'd have to be pretty addled to think this article could at all harm anyone. DreamGuy (talk) 01:27, 12 July 2009 (UTC)[reply]
Actually it is completely accurate to state that most experts involved in this debate urge some sort of suppression. The only experts on this topic are psychologists, and almost every identified psychologist involved in this debate supports suppression. This can be easily verifed thanks to Xeno's monumental work: [29]. Sorry if the facts don't agree with you. As for your own self-claimed "expertise", it is contradicted by obviously bizarre statements about almost nobody using the Rorschach or that its supporters are a "tiny minority." The facts, included in the article, are that 80% of psychologists who do work that could potentially involve the Rorschach (clinical psychologists performing asseszsment services) use it and that 80% of graduate programs teach it. Thanks for sharing that you, personally, as a nonexpert apparantly holding a fringe belief concerning the Rorchach's worrthlessness, claim that that "you'd have to be pretty addled to think this article could at all harm anyone." I guess the expeerts whose collective statements about the harmfulness of distributing test materials and stimuli are "addled", right, according to Dreamguy? As for the comment on Martinevens, there you go atttempting to reframe the debate by pretending I said something I didn't say. I said that almost all experts support some sort of suppression. Not that everyone who supports suppression is an expert. Nice try, though.Faustian (talk) 15:46, 12 July 2009 (UTC)[reply]
DreamGuy, where's that reference that tells us the inkblots created by Rorschach in 1921 are exactly the same as those now printed on the cards used by practionners today? There seem to be quite a wide variety of images, like this one: [30] which is described as having been "Bought to Rorschach Institute in the 80's and digitally restored by me." I have never claimed to be any kind of expert, so I'm afraid I can't be in any competition with your good self. Martinevans123 (talk) 17:02, 12 July 2009 (UTC)[reply]
Yes, they're the same ones. If they weren't the same ones, you'd have a US copyright law stopping you from printing them, remember? There has been a lot of work to preserve the images as is. In fact, some folks even still use actual plates that were printed in 1921. There were also concerns (that I can't remember where they were in the discussion) about the different background colors: those are just scanning artifacts. The plates are white. Mirafra (talk) 06:34, 15 July 2009 (UTC)[reply]
Surely any editor has the right to be on this page, unless they have been banned. But are experts considered "external forces" and thus not be trusted? I didn't realise it was about winning and losing. I thought it was an encyclopedia. Martinevans123 (talk) 22:57, 11 July 2009 (UTC)[reply]
Again, you guys have to stop pretending that "experts" means people who agree with you. This is an encyclopedia. You seem to want it to be the mouthpiece of some tiny minority of the psychology profession. DreamGuy (talk) 01:27, 12 July 2009 (UTC)[reply]
Where are the psychology profession experts who agree with you DreamGuy? Where's the evidence that "some tiny minority of the psychology profession" would think that showing the inkblot images is wrong (if asked)? Or even, where's the evidence that of those psychology professionals asked only a tiny minority think that? In fact - how many have we even asked? I thought it was being argued that experts such as these are seen as unwanted "outside forces" in wikipedia. Martinevans123 (talk) 17:11, 12 July 2009 (UTC)[reply]

There is a consensus right now actually, a strong consensus, that is why the state of the images has been so stable for the last couple weeks. While ending this debate sure would be productive at this point, that is not my goal. Don't worry, once the Muhammad image removal debate moved to its own sub-page it continued for several months and several mega-bytes of text. Moving to a sub-page is not going to give either side any sort of advantage and I am not sure why you would think that. No, the reasons I gave are above, this article talk page is basically being held hostage by this all but settled debate and it is disrupting other work. I don't accept the dichotomy that the page will not be improved until we remove the images. Chillum 13:16, 11 July 2009 (UTC)[reply]

There you go again with your false claims of consensus. There is no consensus. If there was, there wouldn't be a big debate.Faustian (talk) 23:48, 11 July 2009 (UTC)[reply]
There isn't a big debate, just two people whining about some things that aren't even real concerns (and if they were wouldn't matter as far as our policies are concerned), and some sockpuppets. DreamGuy (talk) 01:27, 12 July 2009 (UTC)[reply]
Here you go, apparently falsely accusing everyone but two people of being sockpuppets.Faustian (talk) 15:46, 12 July 2009 (UTC)[reply]

You may not accept it, but there is a consensus, it is plain to anyone without a strong bias. The debate ended ages ago when new arguments stopped being brought up, this is nothing but arguing. Chillum 23:57, 11 July 2009 (UTC)[reply]

It is plain to anyone who hasn't read consensus policy or who seeks refuge in majority dictatorship. The fact is that 1/3 are ignored (20 people last time this was counted), in terms of the article layout. No compromise regarding the images has been brokered. All attempts at compromise have been shot down and argued against by you, with support from the majority. Thus, no consensus. No matter how many times you falsely claim there is.Faustian (talk) 00:02, 12 July 2009 (UTC)[reply]

Why don't you seek outside scrutiny of this consensus? Oh wait that was already done and the consensus still held up. The consensus policy does not say that every opinion should be represented in the article, especially when it is completely contrary and incompatible with the majority of reasoned opinions. You did not win this debate, just accept it. Pretty much every compromise offered involves suppressing the images which is directly against consensus. A compromise should result in more people being satisfied, not less. Chillum 00:11, 12 July 2009 (UTC)[reply]

There you go again with your straw man argument, once again falsely claiming that I stated that consensus means 100% agreement. It seems that you are willing to do anything other than actually address the policy points. We have a majority that refuses to compromise, thus torpedoing consensus (which is compromise). This majority is exploiting an inherent flaw in the system in that there is no mechanism to enforce consensus policy when the majority refuses to follow policy.Faustian (talk) 15:46, 12 July 2009 (UTC)[reply]
Surely this aspect of debate has been "dominating the article talk page" for so long only because people think it's worth debating? If people wish to debate other topics they should just start a new thread. All but settled? We are still arguing whether or not concensus can even exist unless the minority view has been accommodated instead of being voted out. Why should this debate disrupt anything? If editors wish to improve the article in others ways that is entirely their choice. Improvements of all kinds are worthwhile. But if they choose to debate image display, because they consider it the most important and fundamental question, that too is their choice, whichever side of the debate they support. I agree with Faustian that some people may well see the 11 images of inkblots and deem the whole article unworthy of their efforts. But Faustian should be thanked for at least improving this article throughout the debate despite his insistence that keeping the images here might prevent some others from doing so. Martinevans123 (talk) 16:28, 11 July 2009 (UTC)[reply]

Only the few people who did not get what they wanted are arguing that there is not a consensus. Thank you Faustian for improving this article despite the presence of the images. This debate clearly is disrupting the regular editing of the article, a quick look at the archives shows that. Why do you think that moving to a sub-page would be disadvantageous to your side of the debate? It seems like a neutral move with the best interests of the article in mind. Chillum 16:47, 11 July 2009 (UTC)[reply]

Yes, and many people who got what they wanted are saying there is "strong concensus". Is that a surprise? We seem to have different views on what concensus is. The archives are full of debate on the display of images because that's what people wanted to debate. I don't think moving to a sub-page would necessarily be disadvantageous to either side. But I still think it's unnecessary and that the reasons you are giving for doing it are not robust. Let's see what others think. Martinevans123 (talk) 18:22, 11 July 2009 (UTC)[reply]
There's consensus on the inkblots, consensus on the consensus and consensus on the meaning of the word consensus. What next, we have to demonstrate consensus that there's consensus of the meaning of consensus? IT'd be beating a dead horse at thios point to point out that you're beating a dead horse. Enough. DreamGuy (talk) 22:39, 11 July 2009 (UTC)[reply]
All horses still quite healthy, thanks. Chillum was trying to discuss moving image discussions to a sub page? Not a stub page. And you already at the glue factory. Martinevans123 (talk) 22:51, 11 July 2009 (UTC)[reply]

What is that even supposed to mean Martin? Chillum 23:59, 11 July 2009 (UTC)[reply]

That is supposed to mean that I disagree with DreamGuy. In my view the horses are not dead, let alone being beaten. To me DreamGuy sounds here like a Knacker, taking away these horses for boiling down into glue. And it was supposed to mean that your thread here, Chillum, was created to discuss moving to a sub page, not concensus about images. And I still don't know what a stub page is. Martinevans123 (talk) 08:35, 12 July 2009 (UTC)[reply]
Dreamguy can't seem to make any accurate statements today.Faustian (talk) 23:48, 11 July 2009 (UTC)[reply]
Your opinion of what is accurate and isn't has been shown to be completely worthless. DreamGuy (talk) 01:27, 12 July 2009 (UTC)[reply]
your opinion, as accurate as all your others.Faustian (talk) 15:46, 12 July 2009 (UTC)[reply]
I don't think Faustian's opinion of what is accurate has been shown to be worthless at all. And he has made some very useful contributions. I suggest that you apologise for such a derogatory remark, DreamGuy. Chillum - if a sub page was produced for image discussion, would someone need to copy all those places where discussion cropped up about images in other threads (as this certanly tends to happen)? Martinevans123 (talk) 08:48, 12 July 2009 (UTC)[reply]

Yes, lets see what others think. Chillum 18:24, 11 July 2009 (UTC)[reply]

  • Support the move of this discussion to a subpage.--Doc James (talk · contribs · email) 23:12, 11 July 2009 (UTC)[reply]
  • Support as well. Lets be honest here, this debate is being maintained by five or six editors at this point: Gary, Chillum and DreamGuy in support of them, Faustian, Martin and Ward (though he hasn't edited in about a month). You six have combined for nearly 2200 edits to this talk page, which is well over half of all edits, and I would be very confident in betting that at least 80% of those edits are related to the image issue. This little battle royale you have going here is completely overshadowing any topic that focusses on the article itself. Moving this private war to a sub page will allow you to continue rehashing your debates while also allowing other discussions to breathe. Resolute 00:08, 14 July 2009 (UTC)[reply]
That does sound like a fair appraisal, Resolute. All other discussions should certainly have room to breathe. For some editors I think the images have been the central issue - if removal or display can't be agreed, any other contributions are seen as a waste of time and/or unethical, or even reflecting poorly on Wikipedia as a whole. Martinevans123 (talk) 22:24, 14 July 2009 (UTC)[reply]
  • As mentioned in the immediately below thread, I'm testing a subpage method with /2009-06 Arguments Con. As far as I can tell, it's been intuitive enough thus far as several IPs and new users have managed to comment properly there. So for the next image-related thread that comes up, I will make a subpage for it. In the meantime I think we should let everything foregoing archive in the usual way. –xenotalk 16:50, 15 July 2009 (UTC)[reply]

Further Discussions

I agree with Chillum that the use of sub-pages could be helpful. I'd like to see separate sub-pages for each argument: the 6 PRO and the 6 CON. Each sub-page could have a watch function that we could subscribe to. A better organized talk page (or sub-talk page) means a better organized article.

I'd like to point out that there are a number of discussions that are still open.

talk:Rorschach_test##05 - Reputable organizations have been showing the images, too

talk:Rorschach_test##06 Argument Pro - It doesn't matter if we publish. The test is nearly worthless anyway.

talk:Rorschach_test#Argument Con #1 - It may harm a psychologist's ability to protect the welfare of his/her patient.

talk:Rorschach_test#Argument Con #3 - It interferes with the workings of another profession

talk:Rorschach_test#Argument Con #4 - It violates Wikipedia policy on neutrality.

talk:Rorschach_test#Argument Con #5 - It violates Wikipedia policy on indiscriminate collection of information.

Also be aware that I'm proposing a new policy called Wikipedia:Involuntary_Health_Consequences. Please come and discuss it. Danglingdiagnosis (talk) 03:08, 13 July 2009 (UTC)[reply]

Great idea proposing a policy. We can get some input from the community as to what they think of these ideas. Chillum 23:44, 13 July 2009 (UTC)[reply]

re the subapge idea: I think it might work. we could transclude all the active ones and even "noinclude" the upper portions of the argument as time passed (that everyone had time to absorb already). This way someone new could go to the transcluded subpage to see all that had come before in that particular debate or "line of argument", as it were. It might get a little messy, but it can't hurt to try. –xenotalk 03:19, 14 July 2009 (UTC)[reply]

As you may have noticed, I've taken the liberty of giving this method a test run with the subpage /2009-06 Arguments Con. Please let me know if this is amenable and I may refactor future image discussion into subpages of this type. –xenotalk 04:45, 14 July 2009 (UTC)[reply]

http://yro.slashdot.org/story/09/07/14/1829231/Wikipedia-Debates-Rorschach-Censorshipxenotalk 21:34, 14 July 2009 (UTC)[reply]

yup...that's how I got here. Now I'm tainted for life by seeing the images. Oh noes! Thomas Dzubin (talk) 22:20, 14 July 2009 (UTC)[reply]
I got here from Reddit. I'm going to get a set of the blots, have them framed & hang them in the lobby of my home. I think they're quite pretty. Stroller (talk) 05:28, 15 July 2009 (UTC)[reply]
Here's the reddit thread http://www.reddit.com/r/WTF/comments/917us/editors_on_wikipedia_are_engaged_in_an_epic/ --Skofo (talk) 05:34, 15 July 2009 (UTC)[reply]

File:Rorschach.png

Why all the Bickering Pals? —Preceding unsigned comment added by 159.251.6.76 (talk) 15:51, 15 July 2009 (UTC)[reply]

Looks like the discussion is becoming infamously. They might just start showing the ten inkblots on the evening news yet if we all keep this up. :-) --Doc James (talk · contribs · email) 16:44, 15 July 2009 (UTC)[reply]

data on scientific status

1) The article has a lot of words about the validity and scientific status of the test, but nothign really concrete. are there any numbers or other objective data available ? —Preceding unsigned comment added by 65.220.64.105 (talk) 21:59, 14 July 2009 (UTC)[reply]

2) Although critique against anything sold under the robe science (e,g, the Rorschach test) it is important to try to give a more comprehensive picture of the scientific status. The main reference used in a major part of the "Controversy" section of the article is Wood, James M, M Teresa Nezworski, Acott O. Lilienfeld, and Howard N. Garb. "What's Wrong with the Rorschach?". San Francisco, CA: John Wiley & Sones, Inc., 2003. ISBN 0-7879-6056-X. It is weak to criticize with one voice - please improve the article by adding good sources. There must be more.

I agree we are lacking objective analysis on this. The Controversy Over Exner’s Comprehensive System for the Rorschach: The Critics Speak is a document that may yield some hard numbers. It directly references studies done on the effectiveness of this test. Chillum 23:42, 14 July 2009 (UTC)[reply]
It's the same critics that are a minority within the field...Faustian (talk) 23:45, 14 July 2009 (UTC)[reply]
How significant a minority? Being completely honest on this question, do you see it as a significant minority, or only a fringe theory? Resolute 00:10, 15 July 2009 (UTC)[reply]
It's more of a small, but significant minority, than a fringe theory. Kind of like, global warming is caused by natural processes rather than by humans (versus global warming is caused by aliens).Faustian (talk) 02:14, 15 July 2009 (UTC)[reply]
Can you support that statement? Has their work been debunked by any other studies you can refer to? The primary author Dr. James M. Wood is the professor of psychology at the University of Texas and has worked in the field since 1987, that sounds like a reliable source. Chillum 23:49, 14 July 2009 (UTC)[reply]
Does the study "Rorschach Comprehensive System Data for 100 Nonpatient Children From the United States in Two Age Groups" by Hamel, Shaffer, & Erdberg 2000 also suffer from the same minority issues in your opinion? If not then do you have any references to studies on the accuracy of this test? Chillum 23:58, 14 July 2009 (UTC)[reply]
Multiple studies have shown that in their writings Wood, Lilenfeld etc. play games with studies, such as describing 2 studies that support their view (i.e., poor reliability) and ignoring 10 that support the Rorschach. An example: [31]. Another example, from an article by J. Reid Meloy, just one part of Wood et al's "What's Wrong with the Rorschach" (Journal of Personality Assessment, volume 83, pp.344-346) addresses their claim that (p.251 of Lilienfeld's book on the Rorschach) "ten replication studies examined the relationship between reflection responses and psychopathy. Nine of the ten found no significant relationship."
This may sound impressive. However, eight of the ten studies were doctoral dissertations that had never been peer reviewed and had never been published in scientific journals. Meloy wasn't familiar with all of those unpublished dissertations, but was on the committee of one of the ones cited by Wood et al as showing no significasnt relationship between reflection responses and psychopathy. The dissertation did not include enough participants who were psychopathic to make this comparison. It therefore wasn't a valid study to use by Lilienfeld. It is thus mistaken (at best) or misleading (at worst)to use it as a counterexample.
The only study that was peer reviewed and published in a journal was the one that found a significant difference. The tenth study, one of only two published ones, appeared in a book chapter. It did indeed fail to find a significant difference between psychopaths and nonpsychopaths on reflective responses. Although the psychopaths produced three times as many reflective responses, this did not quite meet significance (p value was .07, just a bit higher than the .05 necessary for significance). Of course, none of these details were mentioned by Wood et al. Their absence confirms the conensus within the field that Lilienfeld are more interested in scoring polemic points than examining evidence objectively.
Those are just two studies debunking Wood's and Lilenfeld's work; there are many more.
Dr. James Wood's CV, btw, is not all that impressive: [32] (you have to click on it and download the MS Word program). Compare that to his critic J. Reid Meloy, considered a foremost expert in the field in terms of forensic work: [33].
This stuff is all confirmed by the fact that the the Rorschach continues to be very commonly used and commonly taught in the field.Faustian (talk) 02:14, 15 July 2009 (UTC)[reply]

Here, this is brief, but will at least give you someplace to start, and it is extensively referenced. (2005). The status of the Rorschach in clinical and forensic practice: An official statement by the Board of Trustees of the Society for Personality Assessment. Journal of Personality Assessment 85:219-237. [34] Mirafra (talk) 06:25, 15 July 2009 (UTC)[reply]

You are wasting your time. As long as copyright laws are followed, Wikipedia has no reason to censor these images. Besides, many, if not most, of these images have appeared in psych textbooks for many years. OhNoitsJamie Talk 06:38, 15 July 2009 (UTC)[reply]

A quick reality check

As usual with calls for censorship all the pro arguments are utter shit. There is *always* someone claiming damages but it's not the world's responsibility to take care of them. Besides, it's clear that the only useful longterm solution is to develop an open-ended test where new images can be created rather than rely on a system of secrets. It's clear not because I'm writing this from a position of such great understanding but because the problem is so obvious - secret images that are available to every professional aren't secret. Pretending they are is not only unprofessionalism of the highest degree but just plain bad security design.

But reality though... You can't censor the images. As has been noted, this discussion has brought them to the attention of more people than if this hadn't come up. Had the article just quietly included the images only a few people would have seen them, now the images are forever available in many torrents of banned images - amusingly, to the very people the medical community would least prefer had them. Anyone who honestly believes these should be secret had best shutup now before they draw even more attention to this.

Do the sensible thing, mention the dilemma to the reader above the link and make the thumbnail small enough that people will have to click-through to see anything meaningful.

But finally, you don't have the right, assholes, to make this choice for me or anyone else. I appreciate spoiler warning where they aren't obvious, but I absolutely protest any form of restriction on this or any other content. Like deletionists, censorers are a plague upon wiki and world. I am an adult, and *must* if I am to function, decide issues like this for myself. If you take away my ability to read about this here, and that there, pretty soon I can't find anything but lawyer-approved market speak even in a "user"-edited site.

How am I supposed to tell the difference between an E-meter and a Inkblot test if the article on either has been neutered to please a special interest.

I now return you to your debate already in progress, but I'd like to remind you that you're talking about DENYING THE RIGHT TO READ to people. If you can do this and talk about their own good with a straight face you're a sick person.

74.198.110.39 (talk) 03:58, 15 July 2009 (UTC)[reply]

The importance of having a leg to stand on

Thank you Chillum for reorganizing the page, but I did not put my comment (A quick reality check) at the top because I was mistaken about convention. The top is the place for the most important issues, and this one is it.

"Who the fuck do you (collectively) think you are?"

Wikipedia policy my ass, and potential damage to the medical field regardless, this is a call for censorship and should not be entered into lightly as it is the first step to tyranny.

By making a Yea vote in here you are stating that your opinion is more righteous than someone else's ability to inform themselves. While it is all well and good that you have an opinion, as respectfully as this can be said, it is all you are entitled to have - certainly not an effect.

We all rightfully decry China's policy of censorship, not because they don't have good arguments but because they simply don't have the moral right to make these choices for other people even if it was objectively "for the best". (A point they make asses of themselves, as are many of you, trying and failing to prove.)

The only reasonable solution, other than dropping this issue, is putting the images behind a small thumbnail and giving the choice of self-censorship to the user. Not only is this all that is our right to do, but as the images are out there it is ultimately an issue of individual choice anyways.

Frankly nobody needs your sorry-assed help, all you do-gooders. Simply coming to this article is proof enough of a user's intent to see, or at least have the option of seeing, the images. They (people) are the reason for Wikipedia, currently you're just deadweight arguing about which portions of wiki you can keep others from using...

Nobody legitimately gets a vote except the people being censored. If you disagree say so, here, but leave this comment atop the trivia and under the images.

74.198.102.17 (talk) 07:42, 15 July 2009 (UTC)[reply]

You. And Doc James, Chillum, Faustian, and everyone in here (pro or con) who thinks that their attitude matters. The only one whose does is the reader. You're all trying to make this decision for someone else or validating those who do.
YOU DO NOT HAVE THE RIGHT TO DO THIS. Asking this question is akin to saying you have the right to choose an answer. How fucking clear can I make myself? Your only qualification for being here are your ever-running mouths. No argument is good enough, you may point them out for users to read if they wish, but nobody asked you to get up in the morning and censor anything.
That you're in this conversation IS the problem. Your only contribution here is in taking options away from others. You ARE the problem.
I don't care about your thoughts. Not only are they likely stupid, but they weren't asked for. None of yours, thanks, even those on the 'show' side. It's not like it's the type of question that's open for debate. There is only one issue - you're arguing to keep information from others.
It doesn't matter that the images are technically here meanwhile, what matters is that many of you wish to make this decision for people who don't want you to be their protectors.
You're right that this isn't an issue for a vote, or majority rule. One single person wishing to see these images is enough reason to leave them up. That standard is obviously met, yet you meddlers keep circling like vultures, looking for an option to force your consensus on others for whom a black and white answer is the only reasonable solution. Censorship? None, please.
We don't do research into the physical sciences by vote because it's ridiculous. But when you're trying to decide how others can run their lives suddenly these methods (poll the pool of retards) becomes accepted policy?
I've been told out-of-band twice since this started that this is how WP works, and I and others must adjust to it. No. Hell no. Fuck no. This is how the world works and YOU must adjust to it. If you try to tell me what to do, or limit my actions, you're the intrusive asshole. You have NO CALL to be here. You could just as easily go edit some article on Care Bears, as be here, acting like you have a clue let alone a say.
The only reason WP exists to have all these rules, like CIVIL, which I am breaking, is because it operates on the assumption of basic civility. One can't rightly expect the lamb to remain polite to the wolves debating its future, nor can you expect niceties like posting in the proper place and placating the meddlers when an issue like censorship is on the table. You're proposing very uncivil outcomes and need to understand that you're not going to be treated very civilly.
Pack up, go home. You aren't wanted here. This page, this site if you keep acting like this, or the world at all if you think other people are yours to control. 74.198.104.110 (talk) 20:13, 15 July 2009 (UTC)[reply]
WP::NPA doesn't apply, this isn't a personal attack, it's an answer to "To whom is vitriol directed?" It's not personal because the answer is everyone, and it's not an attack because it's not aimed to hurt. It is merely an impolite comment to those who would be my captors. 74.198.127.61 (talk) 21:50, 15 July 2009 (UTC)[reply]
Don't worry, we don't plan on removing the images regardless of what slashdot says. What they call an "epic battle" is practically over, the images are staying. Chillum 15:29, 15 July 2009 (UTC)[reply]
One almost suspects the slashdot story was intended to bring random IP editors here to remove the images. The "battle" was over before I got here. Verbal chat 15:38, 15 July 2009 (UTC)[reply]
That was not the intention at all. Gigs (talk) 16:18, 15 July 2009 (UTC)[reply]
The IP dose make some good points even if in less than a freindly way. I think if I am correct there is something call freedom of speech in the USA. And attempted sensorship of these images is atttempting to controvein this. I agree that we do not want to become intelletually restrictive like China.--Doc James (talk · contribs · email) 16:35, 15 July 2009 (UTC)[reply]
Slashdot doesn't seem to support the removal. More comments on the story appear to support the display of the images than not, though both points of view are there. -- Consumed Crustacean (talk) 17:20, 15 July 2009 (UTC)[reply]
I'm not surprised. Most slashdotters are 'hackers' at heart. (i.e. Information wants to be free). –xenotalk 17:26, 15 July 2009 (UTC)[reply]

What would actually be wrong with including the gallery of all the images in a collapse box with some kind of a brief comment about why it's in one? I see this was proposed by someone above, but it never got consensus because it "violated" the no spoilers policy, or was "censorship". I seem to recall in the distant past Elonka, or somebody, got their head handed to them on a platter for suggesting something similar in the Mohammed article, so I do this with some trepidation, but I'm just curious why such an easy compromise is a bad idea. Not from a "it's against policy" point of view, or references to "no spoilers" or "no censorship", but from an actual "here's how it harms the encyclopedia" point of view. Probably neither side would be thrilled, but wouldn't it give everyone a chance to go do something else? Those arguing against inclusion would no longer be concerned that people could be exposed unwittingly. Those arguing for inclusion would only need to click "show". It stretches the definition of "censorship" to the breaking point to say that putting them in a collapse box is censorship. The information would be easily available to everyone who wanted to see it. Aside from being unwilling to cave in to the demands from the other side, why is this wrong? Compromise is at the heart of consensus. --Floquenbeam (talk) 18:21, 15 July 2009 (UTC)[reply]

I think of the main unresolved argument used against this was the issue of content in drop-down boxes not printing properly. It's been suggested several times in the history. –xenotalk 18:31, 15 July 2009 (UTC)[reply]
Ah, that's one problem I hadn't seen up there (I admit to not having the intestinal fortitude to look thru all of the archives). That's a shame, otherwise it seemed like a good solution to me. I think I can think of a similar solution, but it would no longer be as elegant. --Floquenbeam (talk) 18:37, 15 July 2009 (UTC)[reply]
I am not convinced it will solve the debate. If one is searching for this test by name they expect to see the images.--Doc James (talk · contribs · email) 18:40, 15 July 2009 (UTC)[reply]
Well, it appears moot now, but the idea was not to hide them so only clever people could find them. The idea was that people make a conscious decision to view them. I'm fully supportive of people being able to see them if they so choose; I'm only thinking that it might be nice if people understood that looking at them would evidently alter the efficacy of the test before they did. Anyway, I'm aware that printability is a big WP:Usability thing, so the collapse box probably won't work. --Floquenbeam (talk) 18:50, 15 July 2009 (UTC)[reply]
Some (probably masochistic) person made an effort to summarize most of the positions here. –xenotalk 18:42, 15 July 2009 (UTC)[reply]
The willingness of that person to wade through all of the previous discussion in order to summarize it says more about their precarious mental health than a Rorschach test ever could... I salute you and pity you, in equal measure. --Floquenbeam (talk) 18:50, 15 July 2009 (UTC)[reply]
Wish I'd seen that before reading for over an hour... oh, well... killed a boring lunch hour anyway. --Raukodraug (talk) 19:11, 15 July 2009 (UTC)[reply]

redacted post

  • Hm, if we don't have the right to remove them (because we'd be deciding for people), what gave us the right to put them there? For that matter, what gave us the right to make an encyclopedia in the first place? Maybe people don't want a free online encyclopedia, why should we force it upon the world? You're talking complete nonsense, 74.x.x.x. –xenotalk 20:18, 15 July 2009 (UTC)[reply]
Listen to the tremendous arrogance in your words, as if YOU had anything to do with building any of this. But even if you did, would that give you the right to kick it down later on the basis of seniority?
How can you ask these questions and expect to be treated like an adult. People of the world show they want a free encyclopedia by coming to Wikipedia. They show they want THIS information by coming to this page. If there is an issue with seeing these images we need to inform them, lest they merely look elsewhere, and having done that, have no reason to hide them.
This couldn't be clearer and that you're fighting it says something that I'm sure everyone who knows you would strongly agree with. You are deadweight. If you ever cared about improving wiki then you should realize why you must leave. The world does not need more censors and that's all you've become - try to not delete any articles on the way out, thug. 205.205.172.20 (talk) 22:36, 15 July 2009 (UTC)[reply]
I'm actually not fighting, nor trying to censor, anything. I'm asking questions. I don't think 74's method of argument is helpful here. All sides are acting in good faith. The strong language being directed at various parties (on both sides) isn't going to bring a peaceful end to the debate. Telling them to "Pack up, go home. You aren't wanted here." is especially peculiar because if these folks hadn't built this article, there wouldn't be a "here" in the first place. The articles don't write themselves, you know. In fact, until a short while ago, it only had one picture [35]. –xenotalk 23:32, 15 July 2009 (UTC)[reply]
205.205.172.20 You seem confused. You are lashing out at the wrong person. In fact, lashing out in general is not how we do things. Please be very sure of your facts before making any sort of accusation. Chillum 01:23, 16 July 2009 (UTC)[reply]

Proposed policy by Danglingdiagnosis

Danglingdiagnosis has proposed a policy regarding possible concerns of health consequences as a result of Wikipedia articles. As it relates dirrectly to this article I have posted it here. See¨: here. --Doc James (talk · contribs · email) 18:03, 15 July 2009 (UTC)[reply]

Completely unrelated, but...

...this is one of the most extensive discussions I've seen on a single topic, and I'm happy to see almost all of it was conducted civilly with little edit warring. That's how things should be done here, and regardless of the consensus you guys come up with, well done! :) Regards, --—Cyclonenim | Chat  19:23, 15 July 2009 (UTC)[reply]

I agree. While parties on both sides have been stubborn(for better or worse) there have been only occasional assumptions of bad faith, even less instances of incivility, and even less still edit warring. Compared to some debates I have been involved in, this has been very productive. Chillum 01:12, 16 July 2009 (UTC)[reply]

A new look at old arguments

Some of the arguments that others have made center on the harm that might come from a posting of the images. From my perspective, the harm has clearly been established. The norms for the test in question were collected at great expense, time, and effort. These norms are based on individuals who have not previously seen these images. By widely disseminating these images, the Rorschach Test page on Wikipedia is contributing to the invalidation of those norms. This is, in effect, harming the legitimate financial interests of corporations who invested in those norms, but more importantly, the scientific community who uses them to perform basic research on issues related to the greater societal good such as the assessment of suicidality and thought disorders (i.e., psychosis). I will point out that relatively few tests exist that have been validated for these uses--the Rorschach has. Now, I believe that this is strong evidence for the harm that may be caused by continuing to post the images on Wikipedia. Because of this, it seems to me that the onus is now on those who would see them posted to prove that there is in fact no harm. The logic for why the potential for harm exists is enough for them to be removed in my opinion. It is now an empirical question as to whether they do, but until there is conclusive evidence that posting them will not cause harm, I would vote for caution and go with removing them.

The argument made by some that they are posted elsewhere on the internet is not really worthwhile. Most mothers, in raising their children, addressed this point by referencing the hypothetical situation of having all of your friends jump off a bridge. In case that is too cryptic, the point was, "Just because someone else is doing it doesn't make it a good idea (or correct)."

Finally, and I think this has to do with ethical issues as well, I would analogize this issue to what I imagine the issues are for the Wikipedia page for Pornography. Why does that page not include graphic images of pornography? Why no links to videos of pornography? An argument might be made that images of pornography would enrich the page and the coverage of the content. As it stands now, a user may not get an adequate notion for what they are talking about when they discuss pornography? Don't you think that a large number of people would like to see explicit images posted on that page? My assumption is that the some members in the community are offended by the posting of these images, right? I don't think that quoting some Wikipedia policy on nudity here would suffice to rebut this point, in fact it would merely augment it. The Rorschach images are exactly the same. A portion of the community, psychologists and allied professionals, are offended by Wikipedia's refusal to curtail the posting of what we see as protected test material. It is as if there has been a gross exposure of images that should remain outside of the public view, for ethical reasons I might add.

For these reasons, it is clear to me that the debate should end, consensus or not, and the Rorschach images should be replaced with non-official exemplars of what an inkblot looks like. Dolphinfin (talk) 01:43, 16 July 2009 (UTC)[reply]

Someone needs to develope norms in people who have already seen the images. That would be the most responsible and ethical thing to do. Hopefully this controversy will promot this.--Doc James (talk · contribs · email) 02:53, 16 July 2009 (UTC)[reply]
There is a serious misconception among those who do not understand psychological testing that a set of norms solves most of the problems associated with improper exposure to the Rorschach images. It has been argued (unbelievably) in another section of this talk page that a new set of inkblots could easily be normed and all the problems will be resolved. Now it is being posited that a set of norms for those who have prior exposure to the image will effectively eliminate the problems with test validity and prior exposure. The reality is, to those who have had a modicum of training in psychological testing, norms only scratch the surface in the development of a useful psychological test. Exner developed his set of norms on the Rorschach, but those norms had little meaning if there was no research base for interpreting the meaning of the test scores. That only came about after many years of research that used those norms and applied them to hundreds of behavioral and personality characteristics. The Rorschach (nor any good psychological test, such as the Wechsler intelligence scales) did not suddenly become a valuable psychometric instrument simply because norms were collected. If that was the case, we could easily produce hundreds of new psychological tests at very little cost. Norms only start the process. The real value of the test only develops after lots and lots of research that comes after the norms. So no, it's not true that a set of norms on those who have prior exposure to Rorschach images would do very much to improve the Rorschach. That would only come after several hundred research studies based on those norms. As for the statement that this controversy on this talk page will prompt collection of another set of norms ... dream on. A number of major researchers involved with the Rorschach have been laboring intensively for more than a decade trying to develop a new set of international norms, as well as a new set of norms for children. It has been an excruciatingly difficult process because collecting norms is very challenging and very expensive and time consuming. So this discussion's impact on whether any new norms will be developed will, at the very most, be a drop in the bucket. Ward3001 (talk) 03:36, 16 July 2009 (UTC)[reply]
Dolphinfin, you make a few points I would like to quickly address...
  • That harm has been clearly established: The fact that reasonable people disagree would seem to disagree that it is clear. While, yes, the presentation of the images on the related article page may "[harm] the legitimate financial interests of corporations who invested in those norms" the financial interests of corporations are not a realistic benchmark. If it were Wikipedia itself could be considered doing harm to encyclopedia publishers and many articles about numerous subjects could be considered harmful to one entity or another.
  • The onus should be on those to prove there is no harm: It is a logical fallacy to prove a negative. More so, in discussions of censorship the norm is to place the burden of proof and harm on the side of those that wish to remove or censor the material.
  • Rorschach images vs. pornographic images: That is, for all intents and purposes a closed debate (and also a classic example of the censorship debate I mentioned above). Those that believe pornography should be censored have proved their arguments in the public sphere and Wikipedia operates under that consensus. The debate we are having here is (hopefully) moving toward a consensus about these images, but until that consensus is reached, comparing what Wikipedia does with pornography to what is being done with the Rorschach test images is apples and oranges.
  • Psychologists and allied professionals, are offended by Wikipedia's refusal to curtail the posting...: While psychologists and allied professionals may be offended about this issue, I'm sure I could find an article here on Wikipedia that would offend me. If all "offensive" articles were edited to make everyone happy, then this would be a very small Wikipedia, indeed. That said, the offense is not a slur or attack against psychological professions, but a presentation of fact, the dissemination of which causes upset. No where in Wikipedia's rules is there a clause for removing factual information because someone or some party wants to keep it secret.
  • ...of what we see as protected test material: While you may see it as protected test material, the copyright on the images has, in fact, lapsed. The APA, nor any other professional organization, has any right to use these images any more than a member of the public. They are in the public domain. While psychological professionals may not like this fact, it is the crux of most of this argument, that a portion of the test materials (the images, but not any other related materials that may still be under copyright) is freely and rightfully accessible to the public, and, as such, can be used responsibly in academic and informational settings, or they can even be used irresponsibly in any number of fashions.
In the end, from everything I have ready here over the last few days, I believe that this argument at Wikipedia likely will rest on the grounds that: because the images are in the public domain, and because there is not a consensus, or even vigorous public debate, in the society at large (as one does find in the discussion of pornography), these images will end up remaining on the related article page. --Raukodraug (talk) 17:44, 16 July 2009 (UTC)[reply]

Social Irresponsibility

Is Wikipedia a socially responsible entity? Here is the intro to the Wikipedia page on Social Responsibility: Social responsibility is an ethical or ideological theory that an entity whether it is a government, corporation, organization or individual has a responsibility to society. This responsibility can be "negative", meaning there is a responsibility to refrain from acting (resistance stance) or it can be "positive," meaning there is a responsibility to act (proactive stance).

I believe that the display of these images, at the insistence of relatively few individuals operating on this page, constitutes a socially irresponsible stance towards this test and the images. I believe that it is socially irresponsible because it is at the expense of a marginalized and often persecuted group of people (i.e., mental health patients).

Whether policy or Wikipedia rules exist to support the presentation of the full set of images on this page is irrelevant, the choice to do so is socially irresponsible, and we as members of society, often agree that certain behavior can be prohibited based on this criterion (e.g., yelling fire in a crowded theater, it harms no one directly, but there is a potential for harm).

I would ask those who are in favor of maintaining the images on the page, is your curiosity (or whatever it is that is motivating you) worth more than a parent's ability to have their adolescent child validly administered this test as part of a difficult differential diagnosis assessment? If so, please let me hear your arguments below. I would be interested to know what you see as more important than that. Dolphinfin (talk) 04:58, 16 July 2009 (UTC)[reply]

WP:ISNOT your socio-culturally localised idea of appropriate behaviour. Fifelfoo (talk) 05:12, 16 July 2009 (UTC)[reply]
Those who actually administer difficult differential diagnosis assessments are no doubt aware that the images can be obtained at libraries. I'd assume that most of us agree that teenagers should be chased there wherever possible? Although many local libraries are suffering from budget cuts, thankfully almost all now at least have means to obtain online access to provide some resources that would otherwise be unavailable. Then again, I suppose some libraries (and perhaps some who try to administer difficult differential diagnosis assessments) may differ in their policies. Steveozone (talk) 05:48, 16 July 2009 (UTC)[reply]
If someone is on the internet then the images are already well available to them. If they typed Rorschach test into google then would have found the images. You can't put an egg back into the shell. Chillum 13:24, 16 July 2009 (UTC)[reply]
Yes, by all means, let's reduce Wikipedia to the lowest common denominator on the internet. If something is "already well available" then it certainly belongs in Wikipedia. The previous gold standard (producing a quality encyclopedia, such as Britannica) should be replaced by the new gold standard: let's emulate the vast number of useless, incompetent, and/or dangerous websites that are now abundant on the internet. Ward3001 (talk) 15:49, 16 July 2009 (UTC)[reply]
Actually, the point could be argued that presenting the images in context with documented, verifiable, and vetted information about the test is, in fact, a more responsible way to present them than the way the may be found elsewhere, be it on the internet or in the library.
In other words, the images are out there and, if someone is looking, they will be found. I would much rather have the interested party learn about the test as a whole than simply find the images elsewhere on the web and draw incorrect or ill-informed conclusions about their use and importance.
As for the general points about social responsibility, I would look at the points in context of the Wikipedia project. When someone is searching Wikipedia they are looking for information about a topic. Removal of the images would remove valid information about the topic (while some argue the value of the images as information, their validity in connection to the Rorschach test isn't really in question). If the Wikipedia project is tasked with, and expected to, present valid and verifiable information on a topic, then it could be argued that it would be irresponsible to suppress the inclusion of the valid and verifiable images.
To directly address Dolphinfin's point about having the test validly administered to a adolescent, let me ask another question... as a parent of two children, if a psychologist were to approach me about administering this test to one of my kids, shouldn't I, as a parent, be able to research and find good information, including the images, so that I may make a responsible decision about the proper care for my child? And wouldn't removing the images be socially irresponsible in the context of Wikipedia being a source of information for many casual researchers and interested parties. --Raukodraug (talk) 17:02, 16 July 2009 (UTC)[reply]
Raukodraug, if your adolescent (or for that matter, an adult member of your family) was suicidal but not telling anyone, and if the Rorschach could detect that (which in some cases can occur), and if your adolescent was hesitant to take the Rorschach, would it be a good idea to show the images in the Wikipedia article to your adolescent so that he could see that it's only harmless inkblots? Ward3001 (talk) 17:15, 16 July 2009 (UTC)[reply]
Would I show my kids? No. Would I want to see for myself before it was administered to one of my kids? Hell yes I would. It is my job to protect my children and, to do so, there is information that I learn and do research on that I never show them, but that still effects the decisions I make to protect them.
More so, the fallacy here is that if they are suicidal and not telling anyone... then who's administering the test??? Really, if the Rorschach test were the only tool psychologists had to diagnose mental illness then this whole argument might be turning a different direction, but it's not. In fact, it's not even close. If Wikipedia were printing the contents of the DSM-IV then I might question the idea, but these images are near 80 years old and the scoring system is nearly 50 years old.
In fact, if a psychologist wanted to use the Rorschach test on one of my girls what would I say? I would probably ask the psychologist why he/she is using a test with such questionable validity on my child? And I'd expect a damn good answer. --Raukodraug (talk) 18:00, 16 July 2009 (UTC)[reply]
"Would I show my kids? No. Would I want to see for myself before it was administered to one of my kids?": OK, what if an adult who is in treatment but not acknowledging his suicidal thoughts? If that person is curious about the test, is it a good idea for him to look up the images in the Wikipedia article?
"More so, the fallacy here is that if they are suicidal and not telling anyone... then who's administering the test??? ": Sorry, but this comment makes little sense. People with suicidal ideation (and sometimes intent, acknowledged retrospectively) often are administered psychological tests when they show up in a psychologist's office because of depressive or other serious mental health symptoms, even if they do not acknowledge suicidality. I see no "fallacy".
"if the Rorschach test were the only tool psychologists had to diagnose mental illness then this whole argument might be turning a different direction, but it's not.": Does that mean that if we have one diagnostic tool that might do a fair job of identifying serious mental illness, then we don't need to worry about the value of another tool that could add to the diagnostic precision or effectiveness? Does that mean if a simple interview can sometimes lead to an accurate diagnosis, then we don't need psychological tests at all?
"If Wikipedia were printing the contents of the DSM-IV then I might question the idea": Do you believe the DMS-IV does a better job of identifying all important aspects of mental illness than the Rorschach? Do you believe that DSM-IV has never had any serious challenges to its accuracy and effectiveness in diagnosis?
"the scoring system is nearly 50 years old": A gross inaccuracy. Take a quick look at the Rorschach literature over the past 30 years. The coding and interpretive system have been under constant revision, including within the past four years. Ward3001 (talk) 19:25, 16 July 2009 (UTC)[reply]


The images themselves are not necessary to answer client questions about the purpose, adminstration, use, and interpretation of the test. Professionals talk with clients all the time about the tests they plan to use. But we do not show them the tests themselves ahead of time. We explain what the items are like, or we use sample items that are not on the test. To show them actual items would be unnecessary and could potentially interfere with the usefulness of the data the tests would yield. You don't demand that the College Board let you see all the items on the SAT before you sign up, do you?
Raukodrag, the current system is very current, with the most recent edition of the scoring system being published in 2005. Oh, and the diagnostic criteria from the DSM-IV-TR are available (by permission from the publishers) on the Internet already.
What does a professional say? Personally, I say that I find that the Rorschach is a very useful tool for understanding people in a holistic way, and that my experience and that of my colleagues, as well as that of a very large body of scientific research, has been that, although it has rotten face validity and seems almost silly, it turns out to be extremely useful for actually answering the questions that brought that person in for assessment in the first place. No one forces anyone to take any test -- if the kid refuses to participate in the process, then fine, we'll work with the information we can get through other means, and we'll do our best. But we don't generally ask our auto mechanics to explain what wrenches they use -- if you don't trust the professional, find one you do trust.
Hm. I've been restraining myself from explaining more fully about why psychologists want to keep tests secure, but it doesn't seem to be getting through to many folks in this discussion what the nature of the harm is that we're talking about and why we think that WP should conform to the usual encyclopedia practice of not publishing protected tests. Psychologists use tests, not just the Rorschach, but many protected tests, to make high-stakes decisions about individuals. Some of those individuals are seriously mentally ill, and some of them are dangerous to themselves or others. They're very often the folks behind scary or tragic headlines. Test security is protected in order to increase the likelihood that the psychologists can make good decisions, preferably before the headline stage.
Some common examples... 'This individual has a history of violence against his children. He is now petitioning the court to regain custody of those children. The court would like an expert opinion as to whether that is a safe bet.' 'This individual has been hospitalized for suicidality. The insurance company wants her released soon, but we need to have a clue as to how likely it is that she is still a danger to herself.' 'This individual is on trial for a crime, and his attorney contends that he does not understand the charges against him and cannot participate effectively in his own defense. The court would like an expert opinion as to whether he's competent to stand trial or whether he's just faking it.' 'This individual has been convicted of repeated counts of sexual violence. However, he has served his term and is due to be released to the community. The court would like an opinion as to whether he remains dangerous to others and what level of supervision would be appropriate.' Often, what the psychologist finds helps create a more compassionate plan for the individual. While it's never possible to make these decisions with absolute confidence, having more secure instruments helps the process a lot. I'm not trying to raise scare tactics here -- previous attempts to make this point about the nature of the potential harm have not been heard. Mirafra (talk) 18:32, 16 July 2009 (UTC)[reply]

Raukodraug, If you are actually concerned about a test, I would suggest discussing it with the practitioner who is proposing it. This is what you might do with an oncologist or gastroenteroligist who is suggesting a colonoscopy. This test is a useful clinical tool, and has shown effects sizes for some things that are higher than tests you trust to work quite well, like the Pap smear (here is the link [36] you can compare the Rorschach and other psychological tests to common medical tests by their effects sizes in that PDF. As others have pointed out above, you do not need to see the images to evaluate reliability or validity of the test. I would again admonish you to discuss any test you are taking with a professional first. I would agree that they should give you a good rationale and discuss any dangers or concerns. You do not need to see the blots for that purpose. The analogy to the SAT above is the correct one. We agree that these test times can remain secret because they serve a society purpose. Others may think that the SAT can be used to keep worthwhile kids out of college, but that isn't the point. As for the DSM (I through IV-TR) these never have been, nor ever will be secret. These are diagnostic criteria and have no strong reason to be protected.Dolphinfin (talk) 19:32, 16 July 2009 (UTC)[reply]

A Consideration on Harm, Protection of Test Materials, and the Use of the Images in Question

Thanks for all the users who have given their thoughtful comments in the above sections in response to my previous posts.

Up above (in Arguments Pro) I questioned who was harmed and in what manner. I now realized that the two groups I recognized as potentially being harmed (those in need of psychological evaluation and those performing the said evaluation) were both slightly off the mark. In the posts above I see that the harm being done is to the test itself and it's usability and validity in respect to the public availability of the images. With the harm that is done to the test's usability and validity by the images public display, the desire to protect the test materials becomes a very real concern for professional organizations such as the APA and members of those organizations. As such, the concerns presented here are very real and very appropriate.

Here's the grand problem. It is not the posting or presentation of the images that is causing the harm. The harm was inflicted upon the test when the images lost their copyright protected status and entered the public domain. At that moment, the core of the test became freely, and appropriately, available to the public. I'm not sure who the copyright holder was, but in failing to secure the continued protection of the images, the APA and psychological professionals have been place in an unenviable position attempting to censor the appearance of the image in public.

Many of the arguements presented in favor of removing the images make perfect sense in respect to the question of whether the test is useful and whether the test materials should be protected. When approaching the question from that direction I find no real debate. Sadly, a significant portion of the test materials were not protected and now are in the public domain, and none of the arguements I've seen presented have really answered to the copyright/public domain question. And that is the foundation of the usage question.

While it really concerns me that these images were allowed to fall into the public domain (and it worries me that such a useful test has been harmed by neglect/oversite on part of the copyright holder or his designary), the images are directly tied to the article and have value in presenting accurate illustrative information. Therefore, unless an immensely convincing arguement can be presented that challenges the use of the images under the public domain... an argument that, in all reality, would need to be a legal argument challenging the images' current public domain status... then I don't see a point at which the images will be permanently removed. In fact, without legally changing the public domain status of the images, the battle to keep them out of the public eye is one that cannot be won, because the images will always be freely disseminated and those wishing to keep them out of sight will always be a step behind.

Psychological professionals and organizations have a right to be upset and angry about this issue... but the anger and action needs to be in the direction of those that failed to protect the copyright status of the images and not those that use the images to legitimately illustrate and present useful information to the public. --Raukodraug (talk) 20:33, 16 July 2009 (UTC)[reply]

Raukodraug, some of your points are accurate, but you yourself are off the mark in some of your comments, perhaps because previous comments from other editors did not adequately explain some of the details. First, regardless of the issue of copyright, the harm isn't just to the test. The harm is to the patients who could be served by the benefits of information acquired from the test. If you have a contaminated medical test, the harm isn't just to the test; you also are harmed. Secondly, the publisher of the Rorschach test has no culpability regarding copyright. The images were created by Hermann Rorschach, and the copyright to the images expired a certain period after his death. The publisher of the Rorschach test has never had a copyright on the images; it only had trademark on the test itself. In other words, I cannot legally sell the Rorschach test without violating the rights of the publisher, but I can legally reproduce the images because they are not copyrighted. There is a difference. (Incidentally, no responsible psychologist would administer the test from images not acquired from the publisher because of the possibility that the psychometric soundness of the test could be compromised in doing so.) Thirdly, it isn't just the professionals and organizations who have a right to be angry because of any harm; it is also the patients who have been deprived of better care that can result from valid test results. And finally, the argument that the damage was done when the copyright expired has some merit, but it does not exonerate Wikipedia (or anyone else) in whatever reponsibility it has for being the vehicle for the harm. There have been legitimate disagreements in this discussion (including in the archives) regarding Wikipedia's ethical responsibilities in that regard, but whatever responsibility Wikipedia has, it is not reduced by the fact that the images are in the public domain. The old saying "two wrongs don't make a right" applies; the fact that other sources may have inflicted harm does not reduce Wikipedia's additional responsibility for adding to that harm (whatever that responsibility might be). Ward3001 (talk) 21:20, 16 July 2009 (UTC)[reply]
1. You place the harm to the patient, but that harm comes from using a compromised test. The test was compromised when the images entered into the public domain. That brings me back to the point that the harm is to the test in the form of it being compromised. A patient is only harmed when a practitioner chooses to use a test that has been compromised.
2. I do understand that the copyright to the images were held by the creator, but that copyright can be transferred to a rightsholder (often an heir or a foundation) which can retain the legal right to renew copyright. Because this wasn't done harm came to the test.
3. True, patients can also be angry... but as with the professionals and organizations, their anger should be directed at the entity that allowed the test to be compromised... the original copyright holder.
I understand that some may wish to argue that Wikipedia is doing harm by displaying these images. Those who wish to do so will continue to do so. But I want to point out one very important thing... These images belong to the public. They belong to everyone. No one person, organization or entity has any right to censor the images. Even for the public's own good. (And to head off a discussion of pornography... those images are censored by what amounts to social consensus not any single person, organization, or entity.)
Argue all you want about whether Wikipedia furthers the harm to the test, because it's a moot point. Wikipedia did not and compromise the test, nor did any other site on the internet. This compromise occurred when the images entered the public domain. Some may not like the fact this occurred and may want to blame sites like Wikipedia and accuse them of harming the test and patients, but that is little more that blaming the messenger. Blame needs be laid where blame is due.
Finally, I am truly sorry that the test has been compromised. It was a useful test. But from here forward I will consider the Rorschach test compromised. Any harm to a patient from the use of a compromised test falls on the shoulders of a practitioner choosing to use the compromised instrument. --Raukodraug (talk) 22:16, 16 July 2009 (UTC)[reply]
  • "You place the harm to the patient, but that harm comes from using a compromised test": Nonetheless, Wikipedia can cause harm (regardless of whether other sources cause harm) when it exposes the images.
  • "that copyright can be transferred to a rightsholder (often an heir or a foundation) which can retain the legal right to renew copyright. Because this wasn't done harm came to the test."": Only partially correct, and it misses my most important point in that regard. The copyright was held by the Rorschach family until a specific period of time after H. Rorschach's death. There is a limit to renewal of copyright. That's why books go into the public domain at a certain point, even when all the possible copyrights have been upheld as long as possible. The copyright to the images existed until it was no longer legally possible to maintain the copyright. This has been argued repeatedly here and is a simple matter of copyright law that has been confirmed by legal experts who edit here. Copyrights cannot be held indefinitely. Read Copyright (and please keep in mind that Wikipedia must primarily follow the laws of the United States since it's main servers and its corporate structure reside primarily in the United States). The fact that the copyright to the images expired was not due to anyone's negligence, including the Rorschah family or the pubishers of the Rorschach test. To try to shift blame to either of those clouds the issue.
  • "their anger should be directed at the entity that allowed the test to be compromised": That would be the creators of copyright laws that do not allow endless renewals of copyright. That's not exactly an easy target for a patient's anger.
  • "Argue all you want about whether Wikipedia furthers the harm to the test, because it's a moot point": It's a moot point from the point of view that Wikipedia does not violate the law. It is not a moot point regarding whatever responsibility Wikipedia might have not to further any harm. As I said previously, there have been legitimate disagreements about what extent Wikipedia has such a responsibility, but you again have clouded the issue by making a blanket statement that "it's a moot point." It depends on whether you're talking about the law or talking about ethics. If you want to argue that Wikipedia has no ethical responsiblity here, you're entitled to your opinion. But don't try to present your opinion as irrefutable fact. It is a fact that Wikipedia does not violate the law by displaying the images. It is an opinion that Wikipedia has no ethical obligation to exercise some control over display of the images.
  • "Any harm to a patient from the use of a compromised test falls on the shoulders of a practitioner choosing to use the compromised instrument": And prior to that, it falls on the shoulders of those who decide to display the images, including those who do so on Wikipedia. Ward3001 (talk) 22:35, 16 July 2009 (UTC)[reply]
I present to thee a hypothetical organized religion centered around H. P. Lovecraft's Cthulhu mythos. One of the central tenets of the religion is that those who aren't properly, slowly introduced to the ways of the Great Old Ones by a licensed priest become corrupted - that those who have read central parts of the stories in their native tongue cannot have their eternal soul saved by the church's ministrations. All information offered is restricted in the organization until one has attained the proper level of committment and training. Now, all of Lovecraft's works have entered into the public domain recently, in (death+70 years) countries. Does Wikipedia publish detailed (but forbidden) incantations that are necessary for an ignorant but interested reader to get a grasp of the religion? Even if some consider Wikipedia to do great harm by releasing this information, doesn't Wikipedia have a duty not to restrict access to publicly available, public domain content based on such opinions? The above example could just as easily (given different copyright circumstances) be a set of objects used to divine the next Dalai Llama, the secrets of Scientology, the number of days that marijuana shows up in a drug test, or publicly accessible details of a crime, knowledge of which the police would prefer to use to identify the perpetrator. There are great numbers of things certain people would rather not have on wikipedia, and proving harm to some interested party is frankly immaterial to the mission of wikipedia. Only in the grossest cases of direct danger (such as a kidnapped journalist), and only then in the worst of references (local Afghan sources) is deliberate censorship ever approved of by the userbase.173.66.253.109 (talk) 23:17, 16 July 2009 (UTC)[reply]
Hmmm... a hypothetical argument. Perhaps it might be more constructive if we applied ourselves to the issue at hand. I like agile minds, though. Welcome to our discussion.
As for censorship, how do you feel about a proposed compromise to add a hide/show button that surrounds images? Is that still censorship? Or is that something else for which we have no word? You may be better equipped than I am to explain it. Danglingdiagnosis (talk) 03:26, 17 July 2009 (UTC)[reply]

Picture of Hermann Rorschach

I have found a better copy of the same picture of Hermann Rorschach than the one we had. I have replaced it. I hope we can all agree this is an improvement. Chillum 23:00, 16 July 2009 (UTC)[reply]

Definitely an improvement. Thanks. Ward3001 (talk) 23:05, 16 July 2009 (UTC)[reply]
Very nice. I always cringed at the grainy one. –xenotalk 23:08, 16 July 2009 (UTC)[reply]