Wikipedia talk:No original research/Archive 27

From Wikipedia, the free encyclopedia
Jump to: navigation, search

Page unprotected

It has been a month that the page has been protected. I have removed the protection. ≈ jossi ≈ (talk) 19:17, 25 October 2007 (UTC)

Thank you Jossi... I think that is the right move. Now... I strongly urge those of us who have concerns about the current state of the Policy to NOT rush off and make changes or try to impliment the ideas we have come up with over the last month. Discuss and build consensus first.
Now, Moving incrimentally, and hoping to build some consensus first... it has been proposed that we add the following language:
  • "Research that consists of collecting and organizing material from existing sources within the provisions of this and other content policies is encouraged: this is "source-based research," and it is fundamental to writing an encyclopedia. However, care should be taken to not "go beyond" what is expressed in the sources, use them in ways inconsistent with the intent of the source nor use the information out of context. In short, stick to the sources."
I know I liked this language, but had a question about where this might be added... and I don't think my question was ever answered. Are there any other questions or any outright objections to adding this? Blueboar 19:59, 25 October 2007 (UTC)
Sounds good to me. Even before that, though, we should replace the tag. Is anybody going to revert an inclusion of the template:disputedtag template? (Which would be really silly). COGDEN 20:26, 25 October 2007 (UTC)
Is there another tag that could be used instead of "disputed", maybe something along the lines of "under discussion" or something? wbfergus Talk 20:28, 25 October 2007 (UTC)
I like it as well, but perhaps some "wordsmiths", or others might find something in it that needs a bit of further refinement. I would support adding this to the beginning of the "Sources" section, before the PSTS sub-section. wbfergus Talk 20:27, 25 October 2007 (UTC)

Agreed that this is an acceptable paragraph as quoted just above. In my opinion, this brief paragraph could readily be positioned prior to PSTS or as a replacement for the last current paragraph of PSTS. ... Kenosis 21:05, 25 October 2007 (UTC)

It is essentially a rewording of the lede to WP:PSTS. Since the language seems well-within consensus and non-objectionable, I have made the change.[1] Vassyana 21:29, 25 October 2007 (UTC)
As a separate change, I have moved that material to the beginning of the "sources" section as a lede, as it seems like the most appropriate place.[2] If either of these changes are objectionable, I will not undo a revert, but please share your objections here. Vassyana 21:33, 25 October 2007 (UTC)
Vassyana's edits were in the right direction, as these edits sought to clarify existing and established understanding. What I do not understand is the flurry of reverts. ≈ jossi ≈ (talk) 23:44, 25 October 2007 (UTC)
I created a milder tag than disputed tag. Hopefully this will be sufficient. COGDEN 16:38, 26 October 2007 (UTC)

New edits... So far so good

I don't think any of the edits that have been made so far are at all controvercial. Nothing that hasn't been discussed and approved of has been added or subtracted. But please keep it slow... let's give everyone a chance to absorb these edits, and think about their impact. Then we can think about what (if anything) should be next. Blueboar 22:18, 25 October 2007 (UTC)

Fair enough. I was only trying to make a couple changes I believed were agreeable and within consensus. Some progress on the section, even if it does not address the principle points of disagreement is productive and healthy to a good working environment. I don't plan on making any more changes at the moment, but I may propose a few more changes to keep things moving in a positive direction. Vassyana 22:24, 25 October 2007 (UTC)
Exactly right in my book. Thanks. Blueboar 12:35, 26 October 2007 (UTC)

Wikipedia:No original research#Reliable sources

I would like to change the first paragraph of the section from:

To:

I believe this would provide more clarity, tighten the policy and differentiate it better from other policies. Objections? Agreement? Thoughts? Vassyana 21:46, 25 October 2007 (UTC)

I agree. That would be a good improvement. COGDEN 16:40, 26 October 2007 (UTC)
What would be good about the change? How would it improve policy? Vassyana 00:02, 27 October 2007 (UTC)

This seems fairly noncontroversial, so I was bold and made the change. Revert if there is significant objection, but please explicitly share your objection and its reasoning here (if you revert the change). Cheers! Vassyana 22:03, 27 October 2007 (UTC)

I support this language, and think it is an improved expression of the concept. ... Kenosis 22:20, 27 October 2007 (UTC)

Intro caveat in WP:PSTS

I've changed the introduction in PSTS from plain "here's the types" to a caveat that various fields have various definitions, to address the "not in my field" concern explicitly.[3] Acknowledging the existing distinctions may be helpful in clarifying the policy. If anyone objects strongly, please feel free to revert the change, but also let us know why the change was objectionable. Vassyana 22:22, 25 October 2007 (UTC)

I've removed it. (I'd intended to write "Removing wandering leading up to the basic Wikipedia assertion. Please put extras into a footnote or something" but while backspacing I hit the "Enter" key and it came out as "Re".) The opening paragraph should not be filled with explanations. Just cut to the chase as it was before, please. Thanks. ... Kenosis 23:27, 25 October 2007 (UTC)
Considering the number of people that relate the definitions to their field, it seemed appropriate. Why do you think it is out of place? Vassyana 09:30, 26 October 2007 (UTC)
WP needn't engage in apologetics to explain slight variances in interpretation of PSTS by several academic disciplines, especially not before making this assertion that is relevant to the readers/editors of WP, which can be stated in one very brief sentence. If they're judged to be needed at all, such apologetic caveats are just as easily put into a footnote, and I would presume that those diligent enough to complain that it doesn't match the PSTS distinction in their particular specialized field would also be diligent enough to respect that it's explained right there in the footnote. But if the extra explanation of what areas of endeavor are different from WP's use of the concept is included, please make clear at the outset of the caveat that WP uses PSTS in much the same way that professional library science uses it.
..... (I might also add that WP's function is in an important way very analogous to a library because both are not expected to synthesize original, novel interpretations of existing material, but to serve as conduits of existing published material --albeit libraries are mostly passive collections while WP is an active, dynamic encyclopedia. The fact that several disciplines use PSTS in somewhat differing ways is reflective of the fact that in those fields the goal is generally to facilitate original research, not to avoid it.) ... Kenosis 14:29, 26 October 2007 (UTC)
In information science, primary sources are nothing more or less than the original source of a claim, piece of data, et cetera. That is not how primary sources are defined or treated in Wikipedia. That being so, such a clarification would be inaccurate and misleading. Additionally, it's not apologetics, it's a caveat, which is not at all the same thing. All in all, your reply is just plain confusing. Vassyana 15:38, 26 October 2007 (UTC)
In short, please put extra explanations/caveats about who uses PSTS differently into a footnote.
..... As to the assertion that WP's usage is in some cases different from library science usage, it's a reasonable point, but only insofar as WP includes more in the primary source category than might sometimes be the case in a typical library (e.g., here). But the basic distinctions are identical. Virtually any experienced librarian would agree that if the topic is Aristotle, Aristotle's works are primary sources, books about Aristotle's works or ideas are secondary sources, and encyclopedic summaries are tertiary sources. From the fairly standard definitions offered by the University of Maryland library system: "Primary sources are original materials. They are from the time period involved and have not been filtered through interpretation. Secondary sources are accounts written after the fact with the benefit of hindsight. They are interpretations and evaluations of primary sources. Secondary sources are not evidence, but rather commentary on and discussion of evidence. Tertiary sources consist of information which is a distillation and collection of primary and secondary sources [giving as examples almanacs, encyclopedias and fact books]." In other words, it's very close to WP's usage. ... Kenosis 16:16, 26 October 2007 (UTC)
If we're going to keep the primary-secondary-tertiary model, we want that model to be the same as that given in primary source, secondary source, and tertiary source. Making them different leads to confusion. If we do make them different, there needs to be a big caveat. A footnote won't do. COGDEN 16:52, 26 October 2007 (UTC)
Maybe "we" do want a matchup with the articles on primary, secondary and tertiary sources. But who will maintain them? Remember, they are articles, and this is a policy, and thus it's essentially a cross-namespace citation if used in this way. One possible solution may be to disambig to a WP use of primary sources, secondary sources and tertiary sources. TBH, though, the basics appear compatible with existing article-namespace pages, which could of course be improved. ... Kenosis 17:02, 26 October 2007 (UTC)

I've added the caveat to a footnote, per Kenosis' strong objection. Vassyana 22:10, 27 October 2007 (UTC)

Another topic for discussion

On to something completely different. Another area of this policy that I have heartburn with is Original images. I have run across several articles here on WP, where there is clearly OR injected in the caption accompanying the image. The two images currently on the Korean Wall article are a couple examples. Originally, the first picture stated it was a view of the wall, when there is no wall visible in the picture (I subsequently changed it, but well before I ever ran across this policy). There is a fence in the picture, but no wall. Anyway, the entire article is unsupported OR, so I've nominated it for deletion, though some people so far have expressed a desire to merge it with another article. At any rate, it shows how OR can also be injected into article through another means, whether intentionally or accidentally, since the first picture was uploaded with no description, so another editor created an erroneous caption for the article to advance a point of view. wbfergus Talk 11:26, 27 October 2007 (UTC)

The Original images section deals with the images themselves, not the captions. I would classify captions as "text"... and as such, they would be subject to the same rules as any other text, such as that added in the body of the article. No? Blueboar 16:22, 27 October 2007 (UTC)
I kind of see what you're saying, and agree with it, but that would open up a whole new can of worms. If that was enforced, then the captions themselves would need to be referenced, and then that by itself would effectively kill any images that didn't come from a secondary source, and allow only a few from 'published' primary sources, but the remainder of 'primary source' images, i.e. peoples photographs from tours or whatever, would effectively be prohibited. Correct? wbfergus Talk 18:29, 27 October 2007 (UTC)
We allow original research in identifying what is in the image, so long as there is no serious dispute over it. This is not ideal, but it is a necessary practicality. The reputation of the person identifying what the image is of is important in helping to decide these things. We have a guy who goes to botanical gardens and takes excellent close up photos of specific species of plants for use on articles about that species. We trust him. On the other hand if some anon posts a pic and someone says wait a minute, is that really a so and so? Then we gotta remove it unless someone can find evidence (such as another pic of it in a reliable source that looks similar enough). WAS 4.250 19:05, 27 October 2007 (UTC)
Sounds like we have a sensible "rule of thumb", unofficial, determination here. It seems to follow the rules for WP:V... things that are challenged or likely to be challenged need verification. If it is not likely to be challenged, it can be placed on the back burner. After all, if no one doubts the caption, there is no need to worry about it being OR... but if someone does doubt it, some form of verification is needed and the rules for NOR could be applied. Common sense and editor consensus applies. The question is, do we need to explicitly state this in the policy? I think that depends on how common it is to have erronious or contested captions. Blueboar 19:47, 27 October 2007 (UTC)
Beans? Let's not add rules until they are needed. WAS 4.250 19:56, 27 October 2007 (UTC)

Another debate over the use of PS

I am currently involved in a debate at Wikipedia:Reliable sources/Noticeboard#Robert Spencer over when it is appropriate to quote and cite a primary source. Leaving aside the issues over whether Mr. Spencer is reliable or not on the subject of Islam, or under what circumstances he might or might not be reliable... the broader question is whether one can include a direct quote from an author's work, with full text attribution, and cite to that work. An editor has said that we can not do so, as that would be using a primary source and thus constitute OR. I disagree. I am contending that simply quoting a work (being very careful to keep what the source says in proper context) and citing to it is not OR. To my way of thinking, if one is adding a statement of opinion (with text attribution along the lines of: According to Robert Spencer, "blah blah blah") to an article, it is preferable to quote what the primary source says directly, and to cite it. Note that I am not talking about interpreting or analyzing what the primary source says in any way shape or form. I am simply talking about quoting the source. So which is right? Is quoting a primary source OR? Blueboar 16:25, 22 October 2007 (UTC)

Where is he quoted? If the subject of the article where he is quoted is either Spencer, or one of his books, then it would be primary source. If the subject of the article where he is quoted is Islam or something else, then it is a secondary source. If the subject is about him or his books, then I think a secondary source is needed to verify the a) notability of the quote and/or b) to what broader context the quote is interpreted. Especially with controversial subjects, the secondary sources are absolutely key to a) or b). If a) and b) are either incontrovertible (not often the case in situations where they're most often used at WP for controversial subjects), or they are used 'directly in conjunction with how they are used in secondary sources, (ie no alternative interpretation, no new argument made with it, no innuendos implicit from cherry picked new quotes, etc) then it is fine to quote primary source. The trouble with how quotes from primary sources are often used at WP is that editors are using them for their own arguments and emphasis, ie original research, with exactly those tricks--cherry picking quotes to make new arguments or "show evidence" of claims in alternative ways that aren't found in otherwise WP qualified secondary sources.Professor marginalia 17:12, 22 October 2007 (UTC)
Well, the issue essentially relates to using him as a source in articles on Islam (he is a notable, but definitely biased critic). The aguments against using him run the gammit from omitting him as "exstremist" to claiming that quoting what he says constitutes OR (as they are his own views, and thus primary). If you would be so kind as to pop over to the RS noticeboard, read the section and add your comments... it will help get us off the primary/secondary sidetrack and back onto the issue of under what circumstances he might and might not be reliable (which is what should be discussed there). Thanks. Blueboar 17:43, 22 October 2007 (UTC)
(edit conflict) As I said on that page, the definition of a PS is 'documents or people very close to the subject under discussion', and thus Spencer qualifies as a PS in subjects to do with his life, work, or the modern political criticism of Islam, in which he is a central figure. ('very close'.) I also said that naturally quotes from a man who writes a blog, a paper a week, and a book a year are problematic if they are not kept in context, and unless they can be said to represent the main thrust of his argument as determined by reliable sources that are surveying the subject in general. To repeat: Spencer is not a scholar of Spencer; he is not a scholar specializing in studying the criticism. Spencer is a student of Islam, and a critic of Islam; there's a difference. He is permissible as a secondary source on Islam in general (if perhaps a little unreliable); but is a primary source, by definition, on articles dealing with his books or his work (criticism of Islam.) I completely agree that quotes are cherry-picked in order to emphasize perhaps minor claims, or those that the author would not, in context, support, and thus we need to, especially in controversial subjects, use sources that can ensure that that does not happen. Relata refero 18:05, 22 October 2007 (UTC)
How can you say he is a secondary source on Islam in general, but a primary source when it comes to criticizing Islam?... he is a critic of Islam... that is what makes him notable. Your concept that he is a primary source on "his work (criticism of Islam)" is rediculous. By your reasoning you could not use a noted particle physisist as a source in an article on particle physics because that is his "work (particle physics)". You are completely misunderstanding what PSTS is trying to say.
Could someone misuse a quote from Spencer in a way that constitutes OR?... yes. You can misuse any source. But you can not simply say "It's a primary source, and thus you can not use it". WP:PSTS allows for careful and appropriate use of primary sources. If all you are doing is stating Spencer's view on something to do with Islam, without going further... as long as all you do is say: "According to Robert Spencer, 'Islam is blah blah blah'<citation to where Spencer says this>" you are not in the realm of OR. Blueboar 18:22, 22 October 2007 (UTC)
To take your particle physics example, excepting the rather small probability that the physicist is himself a fundamental particle and thus the subject of the article, the question does not arise. In Criticism of Islam, Spencer is himself the subject of the article. He does not write about the criticism of Islam! He writes about Islam itself, if critically. Do you now see the distinction? Relata refero 19:26, 22 October 2007 (UTC)
I would argue that placing this dispute in the realm of primary or secondary sources is incorrect. The use of that source is better assessed in the context of WP:NPOV#Undue weight and NPOV in general, as well as WP:SPS, notability, significance of the viewpoint, etc. ≈ jossi ≈ (talk) 19:16, 22 October 2007 (UTC)
Agree that this is primarily a NPOV issue, in that the intention seems to be to cite Spencer as a primary source of a notable minority view rather than as a reliable secondary source on the subject. Thus his statements or a summary of his statements can be cited as the fact that he's presented this view, and it should be set in the context of the majority view of specialists in the subject. Since it's a notable minority view, it should be possible to find secondary (or third party) reliable sources commenting on his view from the mainstream position to set the context. .. dave souza, talk 19:35, 22 October 2007 (UTC)
Jossi, could you please pop over to the WP:RS/noticeboard and explain the intent of PSTS as it would relate to this issue. They are now saying that Spencer is a secondary source on Islam, and he somehow is a primary source when it comes to criticising Islam... which to me makes no sense whatsoever (they seem to be saying that he secondary when making positive or neutral comments, but suddenly primary when making negative comments.) I whole heartedly agree that this is a NPOV issue more than anything else. This really started because there was a question over whether he was "extremist" source and thus unreliable (the answer came back mixed, with some people saying that he was, and others countering that he is surely biased, but not extremist). When I commented that, given his know POV, it would probably be best to attribute his views and phraise things as statements of opinions and not as statementa of fact, people jumped all over me and objected on PSTS grounds. This seems to me to be a classic case of misinterpreting PSTS in order to object to inclusion of controvercial material. As a respected editor on most of our policy pages, your views would be appreciated. Blueboar 21:39, 22 October 2007 (UTC)
Jossi's welcome to chime in, but your interpretation is quite off-base. I, who think he's extremist, sure, jumped all over your misapplication of PSTS which is not specific to this case. In terms of NPOV, as I said several times, he is a very notable critic of Islam, and his work is a major part of the subject of Criticism of Islam articles, so it would be ridiculous to leave a discussion of his work out. (Which is why I am puzzled as to how this means that the argument would lead to the exclusion of controversial material.) Furthermore, several users have pointed out that this argument is not limited to critics, but to other similar situations. Consider the fact that in an article on the decline and fall of the Roman Empire, we could quote Gibbon, but in an article on Writing About the Decline and Fall of the Roman Empire, we should quote those who study Gibbon.
Hence, while there's a WP:NPOV issue in Islam articles, sure, where I would prefer that large chunks not be devoted to Spencer (or Hitchens, or Praveen Togadia) any more than large chunks of any mainstream philosophical or religious articles are devoted to major popular critics, but it's not an NPOV issue in the related criticism articles. There it's a PSTS and NOR issue. Relata refero 05:17, 23 October 2007 (UTC)
Ralata, let me see if I understand what you are saying... are you saying that it is fine (within reason and not giving undue weight) to cite Spencer in our current articles: Islam and Criticism of Islam... but not in an article called, for example: Historical criticism of Islam? Blueboar 12:55, 23 October 2007 (UTC)
Ralata refero, you're working with an interpretation of PSTS that I have not seen any other editor hold. Spencer is a secondary source, except in articles about himself and (perhaps) the specific school of thought to which he belongs. I do not see how it is at all a concern of PSTS and NOR. Could you perhaps explain your rationale more clearly? Vassyana 22:35, 23 October 2007 (UTC)
Vassyana, you have indeed fairly precisely stated what I am saying. To continue the example, Spencer is a primary source in articles about himself, or articles about the criticism and critics of Islam. As he he does not discuss the criticism of Islam, instead he criticizes Islam, he is a secondary source for religion-related articles, but a primary source in the narrow subset of articles that are about him, his work or his views.(I see this as a fairly straightforward implication of our current policy. Perhaps I expressed myself poorly, becuase I think that this is how many people handle PSTS. If not, it needs to be rewritten.)
Blueboar, I think its fine to cite Spencer in an Islam article (leaving out concerns of undue weight.) THe Criticism of Islam article should have sections about Spencer others of his school of thought - Daniel Pipes is the only name I can think of off-hand - but should, rather than featuring wikipedians' summaries of what they believe are his main arguments, refer to and cite what reliable sources have said are his main arguments. (So, you see, I do not believe in 'excising' criticism.)Relata refero 16:32, 24 October 2007 (UTC)
OK, I get what you are saying now. I definitely disagree - In an article such as Criticism of Islam, I think we can and probably should refer to and cite Spencer directly. He is a notable critic of Islam and, as such, what he says is directly responsive to the topic of the article. We don't have to like or agree with what he says, but we should make note of it. It is not original research to do so. Blueboar 18:36, 24 October 2007 (UTC)
Could you respond to the specific points I have made? He is not "responsive to the topic of the article" (which is a little unclear). He has never, to my knowledge, written a book or a paper discussing individuals or methods of criticizing Islam. He has written books on Islam that are criticial. Do we agree on that much? If we do, all else seems to follow.
"We don't have to like or agree with what he says, but we should make note of it." Well yes, as I have already said twice, the article must make note of his views. But it isn't up to individual WP editors to pick and choose; we need reliable secondary sources to make that determination for us. Relata refero 20:24, 24 October 2007 (UTC)
I know that this page is paralysed by other concerns, but I'd just like people to check in and ensure that we are in agreement on this principle. We make note of people's opinions in the article in which they are primary sources, but NOR as it currently stands indicates we should use secondary sources on their work to choose what is important - even if we then cite chapter and verse in their own books as well. Relata refero 08:04, 29 October 2007 (UTC)

Sources - Disputed section?

If the section is disputed, please discuss and do not edit it. ≈ jossi ≈ (talk) —Preceding comment was added at 05:12, 29 October 2007 (UTC)

What have we been doing the last three months? Reverting the tag is silly. Of all things to start an edit war about, this is the stupidest. I'm going to wait a while (until you regain your sanity, Jossi), then re-insert the tag. COGDEN 05:36, 29 October 2007 (UTC)
I don't think it really matters whether we have a tag or not. The discussion is not going to end becuase the tag is added or removed. With Vassyana's proposal, we seem to be closing in on consensus language... let's focus on that instead of on whether there is a stupid tag or not. We have a proposal on the table (Vassyana's version). There is no point in making small edits to the extant policy section if we are potentially going to replace it with Vassyana's proposal. Let's settle the big issue first. Blueboar 13:32, 29 October 2007 (UTC)
This tag issue is becoming unhelpful, Codgen. Please consider dropping it and moving forward instead. ≈ jossi ≈ (talk) 15:46, 29 October 2007 (UTC)
It's important to have a tag, if for nothing else than to prevent the article from lying about itself. You'll notice at the top, there is a caption box stating that the article "has wide acceptance among editors and is considered a standard that all users should follow." Until that statement becomes true with respect to the PSTS section, we need a tag to alert editors that the section is not yet at consensus level. At the moment, the section is kind of experimental. If consensus emerges, then we can remove the tag. COGDEN 18:08, 29 October 2007 (UTC)
It's far better to have a tag on the specific section that says what's under discussion, rather than a protection tag on the whole policy saying it's in dispute. The new tag is a reasonable compromise to the disputed tag, and serves to direct people to the talk page to join the discussion. Dhaluza 23:28, 29 October 2007 (UTC)

Taking the dive (PSTS proposal)

User:Vassyana/NOR 002. This proposal is intended to replace the current PSTS section. I have reviewed the past (extensive) discussions about the PSTS section. I do not believe this draft meets all expectations expressed by all commenting editors. I do believe it addresses nearly all of the expressed concerns, beyond the existence of the categories themselves. Certainly, I feel it is a solid attempt at moving towards a compromise position. If you agree with the proposed change, please clearly share why for everyone's benefit. If you disagree, please similarly be explicit in your objection and its reasoning. Cheers! Vassyana 23:03, 25 October 2007 (UTC) The two proposals I've put forward and the unprotection, as well as the recent changes and the recent straw poll, have been announced on the policy village pump. 23:13, 25 October 2007 (UTC)

Changes

  • Added language to the end of the first paragraph of "secondary sources" in order to address concerns about the lack of a tertiary category and/or mention of the utility of broad references.[4] Suggestions for tightening up the language and further concerns about the proposal are welcome. 23:45, 26 October 2007 (UTC)
  • Added language to the second paragraph of the "primary sources" section, in order to address the legitimate concern about the clear consensus for the use of some primary sources (such as census data in city articles).[5] 21:59, 27 October 2007 (UTC)
  • Revised by Dhaluza and Vassyana.[6][7] (Total net changes since proposal.[8]) 15:24, 29 October 2007 (UTC)
* I would support that wording, as a good attempt to clarify existing application of this policy, without diluting its meaning or introducing new principles. It may need a bit of tightening, but it is actually pretty good as is. ≈ jossi ≈ (talk) 23:48, 25 October 2007 (UTC)
I agree with Jossi. WAS 4.250 18:38, 26 October 2007 (UTC)
Is the new proposal to remove tertiary sources from the basic approach? Tertiary sources turn out to be extremely useful in summarizing large bodies of secondary sources, such as we run into in religious articles, philosophy articles, even scientific articles and a host of other kinds of articles where such overarching summaries can be important in the context of NOR. Another thing I should mention is that if a whole lot of changing is about to be done to WP:NOR, it might be a good idea to let the wikiproject doing the audio recordings know what's afoot at present... Kenosis 00:40, 26 October 2007 (UTC)
Jossi, what can be tightened up about it? Anything in particular catch your attention or concern? Vassyana 09:42, 26 October 2007 (UTC)
I re-read it, and it is pretty good to go as is. ≈ jossi ≈ (talk) 15:42, 26 October 2007 (UTC)
It was one letter too long.  :) WAS 4.250 18:40, 26 October 2007 (UTC)
Good catch! :) Vassyana 23:57, 26 October 2007 (UTC)
Kenosis, I left out tertiary sources because they seemed superfluous. Many fields don't make a distinction between secondary and tertiary, and even those that do discuss the difficulty in determining the distinction in practice. Breaking it down into the two most common categories seemed liked a good move towards some clarification. However, it would be simple enough to note that review articles and subject references providing a broad treatment are useful in summarizing large bodies of literature and research. Since we work with summary style on Wikipedia, they are useful for avoiding original synthesis and undue weight. Would something expressing those principles help allay your concerns? Is there anything else that raises some issues for you? Vassyana 09:42, 26 October 2007 (UTC)
Without getting hung-up on definitions, I see a significant benefit to touching on the value of the "broad treatment" sources, such as general encyclopedias; they're good sources for estimating the appropriate weight to handle ancillary, dissenting, or trivial views in article subjects. Of course, disagreements over the question of weight given to particular views are one of the most common sources of edit warring.Professor marginalia 16:22, 26 October 2007 (UTC)
I've attempted to add language reflecting the concerns you and Kenosis have raised. See the "note" above. Vassyana 23:57, 26 October 2007 (UTC)
I really like the proposal. To me, one major benefit is that it keeps the distinction between primary and secondary sources in place, but directly relates the discussion of primary/secondary sources to the issue of Original Research... which I don't think the current wording does well enough. I especially like the fact that it discusses the limited situations when it is appropriate to use primary sources (perhaps an example will help?). Blueboar 12:55, 26 October 2007 (UTC)
Agree: I'd tweak it a little in addressing rare situations when articles may be justifiably based on primary sources alone. I'm not coming up with examples in my head where primary sources might suffice as only refs to the articles. For starters, how is notability demonstrated without secondary sources? Professor marginalia 16:36, 26 October 2007 (UTC)

I've no inherent objection to something along the lines proposed by Vassyana, though I think it needs cautious parsing while keeping the larger policy in perspective. Remember, please, that the way the currently discussed objections to PSTS came up in the first place was that in practice there were some situations that didn't appear to some users to easily fit into the basic PSTS distinction. Thus, for example, Vassyana's proposed statement that a single source can be considered primary in some applications and secondary in other applications. But remember also that the way tertiary sources came to be added to secondary sources about a year ago was in response to basic conceptual questions such as "Well then, what is Wikipedia?" (turns out that taken as a whole Wikipedia is in general a tertiary source that may sometimes draw on other tertiary sources). Overall I see some valuable points made in Vassyana's proposal that act to clarify some questions about PSTS but which may miss other points that have already been raised on occasion. ... Kenosis 16:56, 26 October 2007 (UTC)

The simple answer to "What is Wikipedia?" is "It doesn't matter." If someone needs to view in these terms, "Wikipedia is a (broad) secondary source." I've included language that attempts to address your earlier stated concerns about the lack of tertiary category. What other raised concerns do you think the draft might miss? Do you think this is movement in the correct direction? Vassyana 23:57, 26 October 2007 (UTC)
interim support with reservations: It's significantly better than the current language, and I'd support the edit in the interim toward a final draft, because it's a step in the right direction. In particular, I like the following:
  • It does away with the category of tertiary source, which is not needed and not helpful in defining the policy. The reason why a tertiary source like Wikipedia might not be an ideal source is because it is too "primary" (in the sense used here) because it is too raw and unsourced.
  • I like the statement that " A secondary source from one article may be treated as a primary source in another, because the focus has changed." I'd also like to see a statement that an entire work is not classified monolithically. A single reference may contain both raw ("primary") and interpretive ("secondary") portions.
However, it still has the following issues:
  • Using the terms primary and secondary in a non-standard way like this is bound to cause problems. If we use these terms, they should be defined the same way as in the articles primary source and secondary source. A solution would be to use made-up terms, like was suggested before. I'd suggest raw sources and interpretive sources. Raw sources include unprocessed data and other information not designed to directly communicate a conclusion or result to a public audience. Interpretive sources include sources designed to communicate a result or conclusion to an audience.
  • It classifies interviews in the "raw" (bad) category. They are indeed primary sources in the academic sense, but they should be included in the good category (defined here as "secondary", but probably best named "analytical"). Why? interviews are the most reliable and accurate sources of people's positions, opinions, and conclusions, and are very widely cited in Wikipedia. They should not be classed in the "raw" category.
  • It classifies artistic and fictional works in the "raw" (bad) category, but they can also be in the "interpretive" (good) category. Artistic works are frequently commentative (documentaries, for example). According to my definition above of "interpretive sources", these would fit in that category.
  • I don't think we need to say that the use of "primary" ("raw") sources should be unusual. For some topics, such as the thousands of articles on small cities in the United States, "raw" information (demographics, population, lattitude and longitude, etc.) is frequent. As long as editors follow the rules, and only make descriptive claims, etc., there's no reason "raw" information needs to be rare or unusual.
  • I don't think there should be a loophole loosening the rules for "raw" information contained within secondary sources. Just because you cite from a separate "interpretive" source doesn't mean that you are exempt from the rule of being descriptive about the raw data. Besides, citing a separate "interpretive" source does not add anything, and potentially may introduce errors. For example, the evangelical preacher misquotes and cherry-picks raw scientific data from a peer-reviewed journal. It's better to cite the peer-reviewed journal for that raw data. If you're going to use "raw" information, and you follow the rules, you should cite from the original source, if possible.
COGDEN 17:36, 26 October 2007 (UTC)
COGDEN, I don't see those flaws in User:Vassyana/NOR 002. Are we even reading the same thing? WAS 4.250 18:47, 26 October 2007 (UTC)
We're talking about the same draft. Why wouldn't it apply? Maybe there are minor tweaks I can make to address some of the concerns. COGDEN 05:03, 29 October 2007 (UTC)

(Outdent) I think changing the terminology may be a next step, but this draft in and of itself is a significant rewrite of the section. Let's take it one step at a time. You're probably aware that I support using other terms, but I'm wary of trying to do too much at once. Let's get a good formulation to work from and then address some of the further issues, such as the terms themselves. In the changes since posting the proposal, the consensus for the use of some primary sources has been noted, explicitly pointing out interviews and census data. I'd rather not make further changes in that regard, for now. I'm looking for an agreeable compromise and for things nearly all of us acknowledge are in practice and acceptable. We can address further changes after the first step of building an an agreeable basis for moving forward. Documentaries wouldn't be considered "artistic works" by most editors and would be a minority of such a category, even if included. Certainly, I've always considered them to be secondary sources, simply audiovisual instead of printed. That is the way I've generally seen them treated by Wikipedia editors, but if it were necessary a footnote could address documentaries. The key phrase with primary sources in secondary publications is: "as they are used". That is, if a reliable secondary source quotes or otherwise indicates a specific section of a primary source and provides an interpretation or analysis, then it is perfectly appropriate to say that primary source X says Y which means Z as long as it is cited to the secondary source saying that. Concerns about bias may be legitimate, but that is a concern for NPOV. However, the current revision of the proposal does include language cautioning about undue weight and the used of biased sources in relation to NPOV. I hope that address your concerns, at least for now. If you have further concerns or comments, please share them. Cheers! Vassyana 16:32, 31 October 2007 (UTC)

I haven't looked at Vassyana's proposal yet (maybe tomorrow). But, addressing COgden's statements above, I would have to clarify a couple of points on his statements. Raw sources (primary sources) can, and often do communicate a conclusion or result to a public audience. A perfect example from two fields I've worked in are blueprints (either mechanical, electrical, or architectural), and flow charts. They are raw sources, and they also communicate a conclusion (the finished product, or end result). The usually (but not always), show the steps required to progress from various 'building blocks' (components, materials, data sources, etc.) to the finished product, or desired end-result. wbfergus Talk 19:01, 26 October 2007 (UTC)
I've still to pull my thoughts together on Vassyana's proposal, but on a technical note blueprints were superseded by dyeline prints in the 1960s, so they're historic documents, and in any format they're technical drawings which need expertise to interpret properly. As primary sources they're likely to need considerable care. Perhaps you mean presentation drawings aimed at the general public? Beware of taking them at face value. .. dave souza, talk 20:16, 26 October 2007 (UTC)
Dave, I don't know what they technically were, but the place I worked at back in the early '80's still called them blueprints, and they were created the same way as I created them in school in the early '70's. In any case, since the technical wording doesn't really matter, many of them don't need any expertise to interpret. The ones we used (and some of the ones I created), were assembly drawings, showing how the various parts fit together. Those were designed so those with a minimal education (if even that), could look at them and see which piece went where, and in which orientation. Many architecural drawings are they same way, especially those that show the 'finished product', whether the outside elevations or just the floorplans. Those are primary, or raw sources, and they communicate a conclusion. They are rarely available in any definition of a secondary source, except possibly a couple hand-picked ones for a marketing brochure (since usually a set of blueprints can easily contain 40, 50 or more, detailing all 'pieces' or steps required to create the finished product). At any rate, whether they are historic documents or technical drawings, they are still primary or raw. Some may need technical expertise to properly interpret, but many do not, especially those created for use on an assembly line. wbfergus Talk 11:00, 27 October 2007 (UTC)

Vassyana: I would think that it would be a good idea to make your edit at this point. I think that there is broad support at least for a first baby next step. And that is a good thing ... I would want to go into the weekend with a show of some progress after the long and protracted discussion about this issue. ≈ jossi ≈ (talk) 00:22, 27 October 2007 (UTC)

I think this belongs in WP:RS since it is focusing on the reliability of the sources. I also disagree with the extreme depreciation of primary sources. Dhaluza 14:41, 27 October 2007 (UTC)
The principal focus is on original research. The closest it comes to defining reliability is in providing a rough definition of "reliable secondary source", which is nothing more than repetition of an uncontroversial broad definition based on existing policy. The draft certainly encourages reliable sources, but that is also noncontroversial and well-within community consensus. This sounds more like a nutshell repetition of objections to a PSTS/PSS section, rather than a response to this specific proposal. Is this proposal any better or worse than the current section? Is it a step in the right or wrong direction? As I politely request above, please be explicit in support or rejection and your reasoning for it. Vassyana 21:49, 27 October 2007 (UTC)
By my count the word reliable is used 10 times in the draft, so this is not a generic comment, it's specific to this draft which seems to be more about RS than NOR. Also if we are defining primary sources only to depreciate them as reliable sources, rather than to explain NOR, then I not only disagree with its applicability to NOR, I also disagree with its broad applicability to any WP policy. Dhaluza 18:35, 28 October 2007 (UTC)
Dhaluza, the difference is that in Vassyana's proposed language, we are talking about reliable primary or secondary sources in the context of what makes their use Original Research. We are not talking about what makes the source reliable in the first place (which would be discussed in RS). Blueboar 18:53, 28 October 2007 (UTC)
The recent expansions push it in the right direction. I've edited it to nudge it a little further. Dhaluza 09:28, 29 October 2007 (UTC)
Vassyana, what you have is very good. To me, it seems much clearer what the policy means and how it is 'supposed' to be applied. Some wiki-lawyers may have objections, or some wordsmiths may have a couple suggestions, but overall I see it as a major improvement over the current policy wording. wbfergus Talk 13:38, 29 October 2007 (UTC)

Personal testimony

Not adequately discussed in the article is testimony of an editor who is a direct witness to events. For example, there is a book that can be cited, Jonathan Karl, The Right to Bear Arms: The Rise of America's New Militias, HarperCollins, New York (1995) ISBN 0061010154, which contains an error, a quotation of a speech I made at a public event that was actually delivered as a printed handout. Short of gathering affidavits from other witnesses to the event, it should be possible for the editor to make a correcting comment based on him having been a witness, without that being rejected as "original research". This might usefully take a tag, "statement of personal witness" or some such.

Articles are almost entirely edited by nonwitnesses, but we need to allow for the participation of those who have been actual witnesses. Perhaps we could provide some system for filing images of verified affidavits for witness statenents. Jon Roland 14:57, 29 October 2007 (UTC)

Witness accounts are notoriously unreliable and have been known to contradict the body of other evidence. Beyond that, Wikipedia is predicated on the representing the body of published information. Between the two, there is little room for such a proposal. Vassyana 15:32, 29 October 2007 (UTC)
Wikipedia is not a first publisher of information. If there is a mistake in the historical record of an event, get the publisher to correct it on the next edition, or in an Erratum. ≈ jossi ≈ (talk) 15:45, 29 October 2007 (UTC)
Yep, unfortunately somebody can get something published, and then either intentionally or unintenionally inject "mistakes", and once it's published it carries much more weight than any eyewitness accounts. Then, trying to get the "mistakes" corrected afterwards is nigh onto impossible, and by then it seems at least one other "source" gets published using the same "mistake". Then, the "mistake" gets perpetuated and slowly becomes "fact" over the course of time. Publishers, especially of books, rarely are concerned with with being accurate, they merely want to get something out there so people can buy them. Then, the more controversy over the book, the better the sales. I've tried numerous times to get various 'things' corrected, jumping through all kinds of hoops presenting evidence and statements to no avail. Publishers aren't concerned once something is in print and being sold to people, unless it may turn out to be a financial "detriment" (lawsuit, etc.). wbfergus Talk 15:55, 29 October 2007 (UTC)
  • No way. Such a policy allowing this would be disasterous. I've already dealt with the situation where an editor embroiled in edit wars offered in all seriousness to do investigative type reporting IRL and take down new "witness" statements for the specific purpose of sourcing his otherwise completely unverified claims in his edits at WP. "On demand" generation of references would have terrible consequences in an open and anonymously edited encyclopedia.Professor marginalia 16:05, 29 October 2007 (UTC)
I recognize this suggestion has a history, but I did not find in that history my suggestion for the uploading of images of verified (sworn) affidavits, which are acceptable in a court of law. It seems to me that it would work to allow citing to such affidavits, or other online court documents, such as those one can access through Pacer, as a way to allow corrections of errors in published works by witnesses, without getting into the problems mentioned above. The suggestions mentioned below are all very well, but don't go to corrections in cited works, which errors might otherwise get propagated before someone comes along to correct them in a reputable medium, something that could take years or decades. Jon Roland 20:45, 29 October 2007 (UTC)
This is an anyone can edit and anyone can verify encyclopedia. It is not a place for original research. We are not a publishing house for information not available in other places. That is not what an encyclopedia does. No. We will not let anyone publish things here based on an something that is not publicly verify-able. We can however delete some things that are proved false to a trusted user such as a person claiming his birth date is wrong in the article on him. OTRS can be contacted for such purposes. We deleted that a covered bridge had no traffic when a trusted user saw it have traffic. But without a source that anyone can use to verify it we could not add that the bridge had traffic. Some things are judgement calls, but as a matter of policy there is no way Wikipedia is going to do as you suggest. WAS 4.250 21:31, 29 October 2007 (UTC)
"Online court documents" are published reliable primary sources that can already be used according to existing policy. They must be used with care, but they can be used. WAS 4.250 21:34, 29 October 2007 (UTC)
One important point in this is that they must actually be a court document... ie filed with a court of law. This is because we recognize court websites as being a reliable publishers of such documents. You can not simply write up an affidavit, scan it and put it on a website. We need the "impramator" of the court system, saying it is legit. Beyond that, WAS's comment is right on the money.Blueboar 22:14, 29 October 2007 (UTC)

Try adding it to Wikinews. WAS 4.250 16:40, 29 October 2007 (UTC)

The gist of the original question is whether personal knowledge of events is enough to trump a reliable source, not whether we want wholesale admission of personal accounts. The answer is still generally no, but it's a more limited question. If a person feels that he himself has been misquoted, slighted, or otherwise disserved by a Wikipedia article, then that person can bring it up on the talk page, and people will probably respect it. The discussions on the talk page about what actually happened, whether someone is portrayed unfairly or slandered, etc., follows a different discussion format that doesn't strictly adhere to NOR. In fact, OR and unverified claims are just fine on the talk page within some bounds. If it's a minor error of no consequence I would just fix it, and see if anyone challenges it. If they do, I'm afraid that a book by a reliable author trumps a statement of "that's not true" by a witness or someone personally involved to the events. If there's a place to correct errors in the official history, it starts somewhere other than Wikipedia. For example, if it were important enough and you got another reliable source somewhere else that the book is in error, you could say there is a conflict among sources, and people will choose the more believable one, all things considered. And in a practical sense, if you see something that's clearly wrong but sourced (e.g. tornado wind speeds are 300-500 mph, which they used to think in the 1950s and can thus be sourced), you can certainly use your personal expertise as the basis for finding better sources to set the article straight. Wikidemo 23:16, 29 October 2007 (UTC)
  • Personal testimony, observation, whatever is fully Original Research. There is absolutely no other way to describe it. --Rocksanddirt 00:27, 30 October 2007 (UTC)
I would say almost always Original Research... but there are a few exceptions... for example, if someone's personal testimony, observation, whatever, was included in a reliable source (say an autobiography). Assuming no one objected to the obvious conflict of interest, the person who was made the observation could quote that text in an article without violating NOR. It wouldn't be original to Wikipedia. Blueboar 03:13, 30 October 2007 (UTC)
maybe. Most of the time it would still be OR, and generally autobiographies are not reliable sources (I don't think). How closely are they fact checked? reality checked? --Rocksanddirt 04:06, 30 October 2007 (UTC)
we are getting way off topic... but Autobiographies can be reliable or not, depending on how and where they are used. context is everything. In an article about the subject of the autobiography, it should be reliable... at least for the subject's view of his/her life. (I would probably phrase any reference with text attribution to make this clear). But the point is, it would not be Original Research. In any case, I suspect my possible exemption would be very rare indeed. Ideally, the subject of an article shouldn't edit the article on him/her. They should simply comment on the talk page, and ask others to edit. Blueboar 13:54, 30 October 2007 (UTC)

To give another take - I essentially described a use of personal testimony related to Noddy Holder and I think it is relevant in the context of a sensible application of Ignore All Rules. If there is someone closely involved with a subject who attests that the article is incorrect and provides a plausible justification, then this should at least be the trigger for a close examination of the sourcing of the disputed comments. If the sources are not of a high standard of reliability, then we have just cause to treat the issue as disputed and remove or qualify it in some way, or use other sources to verify the testimony.

The issue is to make sure we do our best to be accurate, not do our best to follow the rules blindly. If someone can present evidence in good faith that an article is wrong, then we have a duty to reflect that. However, we must also be mindful that someone close to the source may not be objective and so we also have a duty to consider that. So as a rule, personal testimony is not sound for creating articles, but it may be a tool for challenging the veracity of an article. Spenny 14:30, 30 October 2007 (UTC)

Good wording of how the policies "should" be applied. No idea though how that concept can be written into existing policies without weakening the policies though. My word skills aren't up to that level, but I really like the concept. Eyewitnes-type testimony can be extremely unreliable, but then agin, the same can often be said (and documented), about secondary source type information, usually from someone seeking to establish a biased point of view. History articles (especially Military History) are notorious for this. As the saying goes, "History is written by the victors". It can also be para-phrased to "History is written by those who seek to make a name for themselves, by getting published". Accuracy or reliability can be easily ignored in history, often just by selective use of sources (just like this policy and the other policies seek to take aim at). Some historians are keen on trying to remain objective and nuetral and equally presenting all sides. Other historians, usually those trying to distinguish themselves from all of the other work already out there, instead selectively use every source they can find (whether reliable or not), to only advance a "non-mainstream" POV of the event. But, once their work is published, they can usually get enough 'supporters' to clamor that the work is the "authoritive definitions of the events leading up to, and through XYZ event". It then becomes a reliable and verifiable source, even if it's based on a highly POV bias and would otherwise be considered as a fringe theory. wbfergus Talk 18:26, 31 October 2007 (UTC)

Persecution

On many "Persecution of _____" articles, the following is common trend. Some wikipedian, looks up the definition of persecution and using that definition in combination with news reports to judge that a particular incident is an example of persecution.

For example (without pointing to any specific religion), a user concludes the following: "because persecution may be defined as the confiscation of property,reference to a dictionary and because the police of country A seized the property of a person who belonged to religion B,reference to a news report A is persecuting B".

Is it permissible to do this under WP:SYNTH?Bless sins 11:43, 31 October 2007 (UTC)

I'm not a great believer in the WP:SYNTH being that helpful as an explanation of policy. Rather this sounds like original research, inventing a premise and supporting it with a collection of partial citations. The real test is whether the concept as a whole has been presented in a similar way elsewhere. If not, then it contravenes NOR. If it has, then we still need to consider whether the use of such an argument falls within the scope of writing from a neutral point of view. With synthesis, you still get into the debate about whether what you are stating is an obvious and logical conclusion of those two facts, but in principle it sounds like you have a typical synthesis.
We've been working on the text of the policy: read the first couple of paragraphs under "Sources" on the policy page and tell me if that gives you the steer as to whether it is acceptable. Spenny 14:18, 31 October 2007 (UTC)
For Wikipedia to claim in a title or category or section heading or even in a sentence that so and so was persecuted, there must be a published reliable source that says that using that word. English has many words, each with subtle differences in meaning. If there is not even one published reliable source that uses a word, we should not either. Now once you have such a source, then the door opens up to including other published reliable sources that discuss elements or aspects of persecution such as confiscation of property so long as you can document that those are indeed aspects or elements and those elements are related to the subject of the article. But you can stretch this too far and it is within editorial discretion to insist a connection that can be documented should not for some reason such as context, relevance, importance, and all those other things that go into good neutral writing. WAS 4.250 18:51, 31 October 2007 (UTC)
Sourcing, verifiability, and synth don't really apply to titles and headings. We can't footnote them. I think it's just a style guideline issue of what things should be called. The title should be the subject of the article, and the heading should be the subject of the sentence. Inserting editorial comments into the title or headings is just incorrect. If something is inapt, unfounded, POV-ish, etc., just change it, talk about it, propose a move, etc. Overall, I would say that "Persecution" is too loaded a word to use, even if true, in most any article. But perhaps there are a few reasonable examples, like persecution of gays or Jews, where one could construct a valid article devoted to that subject specifically throughout history. But where persecution is just a way to characterize an event, it's too POV. Also, there's often a more precise way to say it. So, for example, countercultural people and communists were not persecuted (that might be true but that's not the precise issue) in the McCarthy era, they were blacklisted. Wikidemo 20:26, 31 October 2007 (UTC)
Actually "persecution" is an understatement for what the United States has done and continues to do with regard to organized efforts to promote communism. We literally and rightfully waged a "cold war" against them that included several hot wars. We warred against communism. Same as we are righteously warring against violent radical Islam right now in self-defense. WAS 4.250 22:23, 1 November 2007 (UTC)
If you're trying to defend McCarthyism as "righteous", you've lost me. Regarding the "evil empire" in Russia and the dangers of religious radicalism (most would not confine that to the Muslim faith), you and I may believe that but we still have to use neutral and precise language, no? A "war" is certainly more deadly than mere "persecution" (usually), but the other difference is that war is an objective term with a neutral meaning. Not so persecution. So we simply try to say what happened as best we can and let the reader make their own value judgments. Wikidemo 22:41, 1 November 2007 (UTC)
McCarthyism was not a legitimate part of the war on our communist enemies. Just as many of Bush's actions are not a legitimate part of war on those who organize with the purpose of destroying "the west". We do have to use neutral and precise language in words that we chose when summarizing but should repeat the exact language of our sources when making a specific sourced claim. If our source says "persecuted" then it is misrepresenting the source to tone it down. WAS 4.250 23:42, 1 November 2007 (UTC)
And that last bit is, of course, the key... as long as it is indeed the sources that use words like "persecuted" ... and not our editors who are using those words, then we are not dealing with NOR. There might (or might not) still be an issue for NPOV, of course, but it isn't OR. Blueboar 18:37, 2 November 2007 (UTC)

Catholicism and Freemasonry

I need a neutral party who understands the NOR policy to look at and comment on the above linked article, and to possibly act as a mediator. The article is contentious... it is also full of half substantiated POV "accusations" and "rebuttals" that I think too often cross the line into OR. Certain sections have serious SYNT problems (the typical situation being... A(Politician X did anti-clerical deed Y) + B (Politician X was a Freemason) = C (Freemasonry is responsible for anti-clerical deed Y) = C2 (Freemasonry is Anti-clerical). This problem is not just limited to the pro-church side of the coin... the Masonic rebuttals are sometimes just as OR. I have been calling for a complete re-write of the article for over a year now, attempting to find both POV balance and limiting the OR. I am not getting much support for this at the article. I could really use some back up on this. I am posting a similar request at WP:NPOV, since there are issues with that as well... if you know both policies, even better. Thanks in advance Blueboar 01:48, 1 November 2007 (UTC)

You're right about the SYNT to a degree, except that as far as I can see the reasoning is actually A+B=C is believed by the RCC,=C2 is believed by the RCC. Neverthless, C and C2 will have to be independently sourced from Catholic writing. Relata refero 08:55, 1 November 2007 (UTC)
The article definitely warrants a {{Original research}} tag. I didn't even get further than the first ref before running into a violation.
Text: The Church argues that Masonic philosophy discourages Christian dogmatism, and that it is anti-clerical in intent.[1][2]
Note #1: In Latin countries, the lodges have often attracted freethinkers and anticlerical types; in Anglo-Saxon nations, membership has mostly been drawn from white Protestants (citations follow)
Note #2: "French Masonry and above all the Grand Orient of France has displayed the most systematic activity as the dominating political element in the French 'Kulturkampf' since 1877" (cits follow)
This is not SYNthesis but outrageous OR nonetheless. Neither note backs up either of the two statements for which they are being ref'd.
A brief review of the other notes shows that majority of them belong inline (they might then make associated OR evident); an article with 144 references is almost certainly hiding OR with pseudo-refs, trusting that people will miss the relationship between note and text. Not NOR related, but a policy vio nonetheless is the preponderance of non-RSources. Quite remarkable for what is actually an academic subject.
-- Fullstop 18:52, 1 November 2007 (UTC)
I think there are quite a few other policy violations. This is a fork from Christianity and Freemasonry, and I would suggest it is giving a small observation (albeit of import to a specialist group) a large amount of coverage - without addressing the what the issues are in the main article. I think there is a hint that the article struggles to put together a good lead - if you can write a good lead, then you might have an idea of what the underlying detail might need to address.
My suggestion would be to pick out the core of the issue that is specifically Catholic and get that in the main article in summary style. Then decide if there is anything worth expanding out into detail. To me however, (and having a vaguely Methodist background and a grandparent who was a Mason and a Methodist in a different era) it smacks of point of view warring rather than an encyclopaedic neutral view of a minor debate. It's about as academic as a Dan Brown novel.
Anyway, I'd be bold and suggest a delete/merge for the page as a starting point, followed by a sort out of the main article which is not that good either. Then, having got the framework right, there might be a case for forking out the Catholic area, if the subject warrants it. Spenny 19:17, 1 November 2007 (UTC)
Much of the content seems to be perhaps not OR, but certainly unsourced. I actually am a Roman Catholic and much of the content reflects what I have previously heard regarding this matter from church bodies. The intelligent design article has 213 references because of the need to verify just about every statement made in the article, and I get the feeling that the same motivation may have been in place here. Having said that, "Sources of Catholic antagonism" could easily contain some OR, and definitely needs at least one citation. The structuring could use some definite work as well. It looks to me that the "Mexico" section is in particular dispute. Regarding that matter, and the talk page discussion, I note that there has recently been a statement from a Mexican Mason him/herself added to the talk page which seems to substantiate much of the current content. I wish that rather extensive quote were specifically referenced, but I imagine that the editor who posted it has one available. Otherwise, though, there was a serious degree of paranoia about "secret societies" in the western world in the 18th-early 20th centuries, and it seems to me, based on the evidence available, that much/most of the content which I myself might question as being completely objective and accurate is probably verifiable at least in so far as the people of the time thought about the matter. I would acknowledge that the article needs some serious work, and it might very well be a good idea to introduce additional references to counterindicate some of the existing content, but from what I know of the subject it seems to me to basically adhere to the verifiable sources I already know of regarding the subject. Whether that makes it objective is another matter entirely. John Carter 19:26, 1 November 2007 (UTC)
As I see it (as someone who fits both sides in the underlying issue, since I was raised Catholic and am a Freemason), we have three levels of "claims"... there is the fact that the Church is opposed to Freemasonry, and states that joining it is a serious sin (citable to numerous Papal documents)... that is clearly not OR. Nor is it POV to discuss this, so long as it is clear that this is the Church's view of the issue. Then there is the official "reasons why" the church holds these views... Most of this seems to be reliably sourced to various Catholic theologians and Papal documents... although some of it is referenced to the 1913 Catholic Encyclopedia, which is out dated even for the Church view on things. However, I do see some "selective" quotation going on. To me the article steps over the line in the latter half of the article, when it start into the historical stuff. This is where the OR seems to really be bad. These claims need attribution as well as stronger sourcing. I understand that many people believe these claims... but we need to make it clear exaclty who says what. Of course then there are the NPOV problems and weight issues. yikes. Blueboar 20:44, 1 November 2007 (UTC)
No real disagreements. I do note that I heard somewhere that the stance on Freemasonry in the US Catholic Church currently isn't quite as serious as it had been before, but I don't know if I can find a source to verify that. And the serious problem of the paranoia of the era mentioned by Voltaire, Disraeli, and others, could probably be specifically included as well, which as I remember extended into the early 20th century and Portugal. That would help verify that the claims were made. I can see a real quagmire about verifying the degree of involvement Freemasonry per se, as opposed to individual Freemasons, had in any of it, unfortunately. Certainly, a statement regarding the difficulties of proving "group" involvement there could be added, although I haven't a clue where to find something like that. With any luck, you as a member probably have more direct access to some information than I do, or at least knowledge of where to go looking for it. Maybe we could contact the Catholicism and Freemasonry projects directly, and possibly try to turn this into a collaboration of some group or other? John Carter 20:54, 1 November 2007 (UTC)

We now have a POV fork called Latin Freemasonry, which essentially copies much of the objectionable OR material from Catholicism and Freemasonry. The editor who created this (and was the major editor on the original page) does not seem to understand the magnitude of the OR situation here. I could use some help explaining it to him. He is a knit-picker who wants me to go through and list the OR point by point (presumably so he can argue why it isn't OR)... I just can't do that... the whole thing (even the title) is rife with OR. Blueboar 21:10, 2 November 2007 (UTC)

Primary, secondary, and tertiary

I'm unclear on where we stand with regard to the "Primary, secondary, and tertiary sources" section. Is there a consensus to replace its contents with User:Vassyana/NOR 002? I think everyone agress that it is better, even if some wish to tweak it further. So I'm gonna do that. Perhaps we can tweak it in place if people feel that additions such as bringing up tertiary sources are needed. WAS 4.250 18:38, 28 October 2007 (UTC

Are you talking about the proposed language at User:Vassyana/NOR 002? If so, I agree. If not, what are you talking about? Blueboar 18:41, 28 October 2007 (UTC)
Yes. I left out the "User:Vassyana/NOR 002" and we had an edit conflict that prevented me from fixing it before you replied. WAS 4.250 18:45, 28 October 2007 (UTC)

Ok, that's two of us for this version, and Jim62sch has decided to vote via edit warring to this version. WAS 4.250 21:39, 28 October 2007 (UTC)

Move slowly, WAS. We made some changes and that is a good think. Let it stay for a while and then we take the next step. Slow and steady wins the race. ≈ jossi ≈ (talk) 22:13, 28 October 2007 (UTC)
I agree with moving slowly, let others who haven't been participating in the discussions have a chance to see things in a piecemeal fashion, maybe a week or two between edits. This gives people who may be on vacation or otherwise engaged in non-Wikipedia work a chance to slowly absorb the changes, instead of coming back and finding what they would consider major changes. As far as those who decide to just engage in edit wars without participating in constructive discussions, I'd just say ignore them and any reverts. They know the rules and have been encouraged to participate accordingly several times, and they still decide to ignore active and constructive participation. Lets wait until next week before making any other changes to the policy, even if there is broad concensus for doing so. wbfergus Talk 13:34, 29 October 2007 (UTC)

I have tried to read all of this and get up to to speed. I think it fascinating how people keep pointing people to this article as if it was set in stone and not realizing how much debate is going on here. Several people have suggested that we use the same definitions of the terms, primary sources, secondary sources and tertiary sources, as are described in these articles. I would point out that the second and third of these articles seems to be almost entirely the usage in history and in particular has no mention of the usage in science. The first has a brief mention of the usage in science but I do not find it adequate. The use of the terms in science it not always the same as in history and there is quite a bit of confusion on wikipedia about this. It might be a good idea to resolve this first. --Bduke 00:21, 2 November 2007 (UTC)

I completely agree. We should get the articles (primary sources, secondary sources and tertiary sources) into good shape first, so that we have facts to work with.
Regarding sciences, it is my understanding that these terms are simply not used, certainly not even close to their sense in history/literature. In science, sources can be radioactive sources, with the primary source being the reference calibration source, and secondary sources being calibrated against the first, but I contend that this is a completely difference usage. There seems to be a lot of scientist wikipedians who are stirring the confusion. I think they need to accept that scientific language is not encyclopaedic language. The reporting of science is not science, and wikipedia is not science. Wikipedia is reference literature, and given that it is emphatically not news, why not treat it with the standards and terminology of historical literature. --SmokeyJoe 01:11, 2 November 2007 (UTC)
Because we have to fit our sources to these standards and terminology and I, for one, am not convinced that they do fit. I do not think any of us are stirring the confusion. There just is confusion. Historical language is not encyclopaedic language either. --Bduke 01:35, 2 November 2007 (UTC)
I don’t think that we “have to fit our sources” at all. WP:NOR and WP:N are not helpful policies/guidelines, except in their help at guiding WP:AfD. Secondary sources are the clearest way to demonstrate notability. WP:NOR and WP:N are just rules that get severely enforced on newbie editors who create low quality articles. As such, they are only applied to borderline articles. The secondary source test, aka WP:N based on WP:NOR, is just a test. An article passing the test can then use and organise its sources without reference to primary/secondary distinctions. Serious articles by serious editors need not bother with the primary/secondary source distinction. As has been thoroughly explored here in the last month, primary vs secondary distinctions has very little to do with reliability. If an article, such as the typical science article, is under no threat of being accused of being “nn”, then the sources are free to be used in the article without anyone attempting to distinguish secondary from primary. --SmokeyJoe 02:10, 2 November 2007 (UTC)
You are right for sound science articles, but I suspect the point I was trying to make is very important for dealing with articles on pseudoscience or just plain crap science where the editors try to use scientific sources. I need to think about it rather more, so lets leave it for now. BTW, I googled "Primary sources science" and got a bunch of links from libraries. They use the distinction but their approaches are not all exactly the same. --Bduke 07:46, 2 November 2007 (UTC)
Wikipedia is not required to explicitly follow the manner of usage of the concept of primary, secondary and tertiary sources as it may be applied in any other field of endeavor, nor is it necessary that the articles on primary sources, secondary sources and tertiary sources precisely track the application of the principle on this policy page. The WP usage already roughly tracks the usage in library science. Initially the primary/secondary source distinction was by directive of the founder, J. Wales, (cited at the bottom of the project page) and since then it has been developed by community consensus. It is true, as Smokey Joe just said, that the distinction often does not come into play at all, in much the same way that arguments over WP:NPOV and WP:V often do not come into play (such as in numerous cases where statements in an article simply aren't contested). Nonetheless the PSTS distinction has turned out to be quite valuable in articles on controversial topics where editors disagree about what material to cite. One of the more recent "poster children" for this principle has been in the article on homeopathy, where it is readily possible to cherrypick primary sources such as particular experiments, but where reliable secondary sources provide a more accurate interpretation of the overall body of research. ... Kenosis 15:39, 2 November 2007 (UTC)
That's a nice take on the issue, especially the observation that we shouldn't be worried about uncontested issues which seems to be forgotten at times. Spenny 16:25, 2 November 2007 (UTC)
As far as I know, Jimbo has never introduced any primary/secondary source distinction into Wikipedia, and he has never imposed any requirement on sources other than that they be reliable. However, I'm sure that Jimbo would agree that Wikipedia itself should never be a primary source, but he has never had any problem, as far as I am aware, with Wikipedia being a purely secondary source (i.e., citing primary sources without adding new ideas or research). We're not talking here about what Wikipedia is, though: we're talking about what additional requirements, if any, should be imposed on sources in addition to their being reliable. COGDEN 19:36, 2 November 2007 (UTC)
It's true that science itself does not really use a primary/secondary distinction. It is used, however, by researchers on the history of science. And that is what NOR is really about: what is the history of the original idea. According to science historians, the primary source of a scientific idea is the publication where that idea first appears. A secondary source is any publication that repeats or discusses the non-original idea. However, scientific secondary sources are almost always also primary sources, since to get published, a scientist always needs a new idea. The secondary source article is the primary source for the derived idea. That's pretty much a general principle: almost anything that is published is a primary source, because only original ideas get published. However, many if not most published sources are also secondary source, because they comment on prior sources. COGDEN 19:36, 2 November 2007 (UTC)
OK, that is a valuable insight. Let me add that with that prospective, review articles are then mostly secondary sources. Indeed they ought to be just secondary sources as they should not contain original work. However, in recent years review articles have contained original work that was not published elsewhere. We also need to realize that too many review articles in science are not independent sources as they mostly stick to work from the group of the main author. The days of balanced reviews of a whole area of science seems to have largely gone. There are also publications such as the specialist periodical reports of the Royal Society of Chemistry with really just mention original papers, summarizing them without analysis. These are useful to scientists but probably not useful to us. --Bduke 00:23, 3 November 2007 (UTC)
Cogden and Bduke, in my opinion this is trying to force PSTS into an overly rigid interpretation of the principle. It's not just review articles that are secondary sources, but the entire range of interpretation, which depends on numerous parameters including the size of the body of research, how controversial it is (e.g.homeopathy), whether it's controversial in the scientific community or just in a political or other applied context (e,g,, evolution and intelligent design), what practical applications may be involved (e.g., stem cell research), how widely misunderstood it may be in the public (e.g. quantum mechanics) and other issues that come into play. These and many other types of articles are not reducible to, say, "don't cherrypick the sources". or "use only reliable sources", etc.. We run into the same type of issue with philosophy articles, religious articles, history articles, and numerous other types of articles where editors disagree about the content of an article. The allocation of PSTS depends very much on the topic of the article (and also upon a reasoned participation among editors in discussion about the content of an article), not upon a predetermined, precise, legalistic interpretation that strictly defines in advance every possible permutation of the concept. ... Kenosis 15:27, 3 November 2007 (UTC)

Request for Comment - The Matrix Online

I have been involved in an edit war over certain parts of the article within The Matrix Online and the particular debatable subject is regarding complaints within the entire population of The Matrix Online. The Matrix Online, ever since the Sony Online Entertainment takeover in September 2005 has been under questionable management for some time, and a number of issues have arisen especially with subjects like the Live Events Special Interest Group, questioning an SOE employmee's competence after months of silence from this particular employee's silence over game developments, as well as questionable moderation conduct by player moderators, some which include deleting legitimate questions and complaints alike, as well as banning said players over such problems.

Article Dibol has written and constantly reverted by Oni Ookami Alfador.

Among other criticisms, the player community has consistently criticized the fact that unlike most of the SOE-controlled online games, the Matrix Online has been given the short end of the stick especially with little to no publicity, the handling of the Buddy Key program [9], questioning Joel "Raijinn" Sasaki's competence in carrying out his position as a Community Relations Manager shortly after Daniel "Walrus" Myers's disappearance as Myers was assigned to produce The Agency [10] [11] [12], slow responses to LESIG-related complaints with inactive liaison officers [13], [14], [15], as well as legitimate complaints and questions being frequently deleted by player forum moderators on Data Node 1 [16] [17] [18], as well as issuing perma-bans on said players for writing out such complaints under the charge of "excessive negativity" [19]. As a result, a player has emailed Courtney Simmons, the Director of Corporate Communications and Public Relations to bring the current problems to light [20]

Sources are cited and one could argue it's not original research, but there's one problem, "It's only given from message boards, and therefore unless a publication is given, it must be deleted." I am one of the players of The Matrix Online, and have joined the game since April 2005. I am one of those players fed up with the recent issues and could verify that these complaints are true, and the truth of the matter is that not one news source will be able to follow up on it because said reporter doesn't know the entire story, much less the state the game is in. The fact of the matter is that the game as a whole is generally obscure to the public, and that no one within Sony Online Entertainment, even the developers of the game would be willing or able to come forward and state their side of the story therefore little to no news coverage regarding the complaints the player community is having and sweeping it under the rug. On top of this, no gaming publishing site has ever seen this type of problem because said gaming publications have never covered the game as much as the entire playerbase. Dibol 06:13, 5 November 2007 (UTC)

My view is that this is the world of popular culture and different rules apply: you are not going to get learned journals or theses written on these subjects. There is an argument on notability, (should the article really be here?), but given that it is and it is notable for some people, we need to understand how we can sensibly work with Wiki policy in this style of article with the likely sources.
Are the sources as good as you are going to get? Have you attempted to be as neutral and fair as possible? I think you do have a problem in justifying it against NOR because you have examples of problems, but no one telling the story of all the problems.
So, is this a case of Ignore All Rules? I'm sympathetic to the view that Wikipedia has the space to be tolerant of the specialist articles and hasn't really got rules that work well with these popular culture issues. I think therefore, that as long as the article qualifies for inclusion, then you need to be flexible in the application of policy.
There is a specific issue you need to be careful of. If you are essentially publishing original claims that there is a problem with the product or the way it is handled, then you ought to have high quality evidence published elsewhere. Wikipedia should not be put in the position of having to handle claims against Sony for publishing defamatory comments, so the reverter does actually have a good point. It is unhelpful to express things simply in terms of policy if you do not give the reasoning behind why the policy is needed. Spenny 10:33, 5 November 2007 (UTC)

From my understanding, I thought Wikipedia is supposed to be an encyclopedia, but when it comes to articles on video games, the rules get in murky territory, as evidenced with what was written in that paragraph I was talking about, and I can't help but get the feeling that it's function is a double-standard in which there could be biases in "censorship for the sake of censorship" and the bureaucracy slowing it down. The facts remain the same, the player community is fed up with the way the game is handled, the game is heavily obscured due to aforementioned management which means no available news stories because said reporter isn't familiar with the game. I'm very doubtful that "a letter to a news site" even within the video game industry would work especially with the fact that the game has had very little publicity, and I'm very doubtful a reporter will do a serious article about it. Dibol 00:15, 6 November 2007 (UTC)

Dibol, I am concerned about your statement "The fact of the matter is that the game as a whole is generally obscure to the public"... this makes me question whether the game is actually all that notable, even in the gaming community. If we accept that it is notable enough for inclusion in Wikipedia, then we face the RS issue: there are all sorts of reliably published magazines and websites that cover gaming news. If these complaints have not been discussed in any of them, I have to wonder whether the complaints are really notable enough to discuss in the article. Ian has a point about it being OR without any media coverage. I think you are better served by drawing attention to the problems with Matrix Online in the industry media... then you would have reliable sources to cite in the article. Blueboar 14:19, 5 November 2007 (UTC)
If the subject is notable then some reliable sources will cover it. Gaming and entertainment are important industries and vetted publications cover important developments. DurovaCharge! 19:39, 5 November 2007 (UTC)
Unfortunately there are no reliable sources for these "criticisms." They are comprised almost entirely of forum posts of users. Every game has those who are not content with the way it is handled, made, etc. Unless these events are covered by a reliable source, I see no way to include it in the article without slapping a long-standing official policy in the face. I have already established in two seperate places in the article why this policy needs to be followed [[Talk:The Matrix Online#LESISG section removed|here], here, and later here in response to Dibol. The acknowledgment that this probably won't be covered by a secondary source is sign enough that this isn't notable or reliable enough to cover here. Wikipedia is supposed to be an encyclopedia. We collect information from reliable sources into one place; we don't conduct investigative reporting to determine how a certain group may feel about something.--Oni Ookami AlfadorTalk|@ 02:32, 6 November 2007 (UTC)
Oni, am I correct in assuming you intended to say: we don't "conduct investigative reporting to determine how a certain group may feel about someting" ? It may be frustrating that unnoted complaints such as this should not be discussed on Wikipedia, but given our rules... they shouldn't. Not without reliable sources to back up what we say. Blueboar 02:52, 6 November 2007 (UTC)
Agreed, and if you think the forum posts amount to something significant enough to be worth attention, you could propose a story to a regular publisher and see about getting it covered that way. Wikipedia isn't set up for that sort of thing. DurovaCharge! 03:01, 6 November 2007 (UTC)
Yes, it was a typo and I've corrected it, sorrry. So far every person who believes these entries should be in the article says something along the lines of "I know we have these rules and that its supposed to be an encyclopedia but..." Well, it is supposed to be an encyclopedia. I strongly encourage Dibol and anyone else who believes these issues need attention (and they may very well, its not my place as an editor to make that assessment) to make the effort to bring these issues to the attention of the video-game industry and its media. If it gets the coverage they believe it should, then it can be included in the article, but if we go ahead on this before then, its just going to set a precedent that the policies of Wikipedia, the guidelines with such wide acceptance among editors as to be a standard, can be ignored whenever they don't fit what a user wants to accomplish.--Oni Ookami AlfadorTalk|@
Not a happening thing. No precedents will be set by this incident, as we have specific policies that deal with these issues already, and these will not be violated to accommodate the inclusion of criticism posted in a discussion forum. ≈ jossi ≈ (talk) 22:23, 7 November 2007 (UTC)
People would also need to be reminded of WP:SOAP: Wikipedia is not the place to air grievances about a product or service. ≈ jossi ≈ (talk) 22:26, 7 November 2007 (UTC)

Press again

Happy to be edited, but I am not sure that this change has the intended effect. My original edit was just to raise the question of the reliability of newspapers without getting into depth about it - the depth is needed in the reliable sources guideline.

Periodicals does not, to me, have the general connotation of newspapers, so now we seem to have a policy that does not speak about newspapers at all. Although it is primarily an issue of reliable sourcing, it does have strong implications for NOR as the selective use of poor quality writing of dubious veracity does create the opportunity to write about things which are not commonly held views while giving the impression they have arisen from reliable sources (e.g. guest writing opinion pieces are not typical reliable sources). This becomes especially so when we consider the wide range of types of writing within a newspaper and the wide range of topics covered.

My wording attempted to address this without diving off at a tangent. The new wording seems to me to ignore the need to be circumspect about the press as a source.

Comments? Spenny 13:04, 30 October 2007 (UTC)

There is a difference between opinion pieces (Op-Ed) and the basic reporting of a newspaper. An Op-Ed piece is not a generally considered a reliable source...The exception, perhaps, being for a statement about the opinion of the author. And such a statement of opinion has all sorts of caveats and limitations in my opinion... I would say that the fact that the author holds this opinion should be notable in relation to the topic of the article (thus, their opinion is also notable)... It would say that it should be attributed in line so that it is clear that it is an opinion and not "fact". And there are probably other cautions to be given on "statements of opinion" that I am not thinking of at the moment.
But the basic reporting of news events in a mainstream newspaper is generally considered reliable. As it relates to NOR... what you are really talking about is selective use of the source. In this case, the source isn't the problem... the misuse of the source that is the problem. A piece from the source is being taken out of context to support OR. The policy already makes it clear that we should not do this. Blueboar 14:23, 30 October 2007 (UTC)
I'd take issue with the concept that mainstream newspapers are generally reliable. In the UK, every paper has a political standpoint and cannot be considered a reliable source for fact or opinion on political issues. With regard to academic articles, we have to bear in mind why articles are written; often it might well be that the journalistic sensation is the oddball opinion that is newsworthy just because it does not have the backing of mainstream science - medical cures would be a good example where the press are happy to publish articles on alternative medicine. With regard to other subjects, we need to be more and more mindful of WikiJournalism where we are the hidden source for the article.
Then there is the problematic issue of "What is a mainstream newspaper?". The Sunday Sport published stories of Hitler's Dornier found on the moon and other nonsense, so we might conclude it was not mainstream, but the likes of the Mail are likely to heavily distort reporting to suit their political or circulation agenda. Basically, the more controversial the subject, the more likely a newspaper is to be unreliable, which is the opposite of what we need. Spenny 14:42, 30 October 2007 (UTC)
These are excellent questions Spenny, and about time somebody asked them. The idea of "neutral" and "objective" journalism is becoming an oxymoron these days. How do we incorporate some of these concerns into this policy, is the $64,000 question... ≈ jossi ≈ (talk) 15:09, 30 October 2007 (UTC)
My view is that, in a sense, we don't. What we also should not do is assert that mainstream newspapers are reliable without question. I am happy that any newspaper article can allow people to pass the basic hurdle of NOR in the most simplistic sense but then we have to apply all those other tests. In some respects this is more an issue of NPOV. (Stops himself before generating another of his famous tangents). Spenny 16:34, 30 October 2007 (UTC)
Yes, I agree... its more an issue for NPOV (and also for WP:V and WP:RS). As far as NOR is concerned, the same rules apply for newspapers as for any other source. Don't misuse any source to back original research. Blueboar 17:01, 30 October 2007 (UTC)
(dindent)This "mainstream newspaper" thing is - like primary-secondary-tertiary stuff - entirely subjective. It is always material in context to which common sense must be applied, and one can't just say the medium as a whole would be subject to NOR or RS or whatever.
Example: Pravda could per-se be cited for a quotation of Brezhnev. A quotation of Brezhnev to tell us how wonderful communism is could be OR. A quotation of Brezhnev in an article of the Cuban missile crisis could be non-RS. But it is not necessarily OR or non-RS.
None of this would depend on Pravda's qualification as a "mainstream newspaper." Indeed, if Pravda cites Brezhnev on communism, that citation is a secondary source, not a primary one. If Pravda collates several different sources in an article on some aspect of Soviet communism, that article even becomes a reliable tertiary source.
As always, its a question of defining (and applying!) common sense; the use of arbitrary blanket definitions ("mainstream", "secondary sources" etc) can never work. Or to put it another way: Is sunshine better than rain? -- Fullstop 17:02, 31 October 2007 (UTC)
I think I have changed my mind from my first comment to the position that we do not take any position on newspapers. So when I look at the policy now and I see that it does not refer to newspapers at all, either for or against, I think it looks right. I think my edit to reliable sources, which has survived a whole two days unchallenged, (which I think is something of a record around here!), does try and reflect the points you raise which is a change from the previous assertion that mainstream papers were a reliable source. If the amendment survives the week, I'll look at bringing all the policy pages in line with a more cautious stance. Spenny 17:12, 31 October 2007 (UTC)
Good call, Spenny! ≈ jossi ≈ (talk) 18:29, 31 October 2007 (UTC)

Agreed. Each journalistic source needs to be assessed on its own merits, taken in context of the topic being presented, irrespective of whether the periodical is generally considered "mainstream". Even within a given newspaper, a story about a traffic accident (one of the examples presently used in PSTS) may be assessed as having a different level of reliability than a statement about national or international policy drawn from an "Op-ed" piece. I don't agree that it's entirely subjective as Fullstop said, but I certainly agree it cannot properly be simplified to whether the source is a "mainstream newspaper". ... Kenosis 18:49, 31 October 2007 (UTC)

sorry, poor choice of words. What I meant was "non-absolute." -- 17:27, 1 November 2007 (UTC) —Preceding unsigned comment added by Fullstop (talkcontribs)
Baby / bathwater. I can't speak to the UK or anywhere outside the US the mainstream press (e.g. NY times, WS Journal, Christian Science Monitor, Washington Post, LA Times - and online, MSNBC, Fox, CNN, etc), including the next couple tiers down (SF Chronicle, Sacramento Bee, Miami Herald, etc), are as solid as you are going to find, particularly for establishing notability but also for a wide range of material. You simply have to know how to use them. Don't trust the Miami Herald on the details of a scientific finding, or ancient history, but you can trust them as well as anyone on the biographical details of a local person, or business events in their scope of coverage. I don't think we have the equivalent of the Mail in the US. Anything that's more of a tabloid or partisan press would not be considered mainstream, however great its circulation. So, for example, the Washington Times (conservative POV despite full editorial process) or the New York Daily News (tabloid, despite pulitzer prize and huge circulation) are viewed as slightly less than mainstream, with a good dose of suspicion. Wikidemo 18:38, 31 October 2007 (UTC)
You simply have to know how to use them. which is the nub of the problem. You cannot convey that in a broad brush statement that mainstream newspapers are reliable sources which was how it stood a few days ago. (I don't think we are disagreeing, it is just a matter of emphasis or tone to resolve). Spenny 19:41, 31 October 2007 (UTC)
Let us not forget Freedom of the Press. The Pulitzer prize isn't given on the basis of neutrality, it's given on the basis of excellence (or something close enough to that) as determined by a panel of judges. It would be highly offensive if Pulitzer prizes went only to those who absolutely never rocked the boat.
Freedom of the Press means, among other things, that a lot of charlatans who have (in my opinion) bad ideas and motives are free to publish them. Freedom of the Press does not in any way limit or measure "reliability." Freedom of the Press, if actively used and applied, guarantees some unreliability: that's what it's for. Then the readers sift. [21] One would hope that most of the time garbage publications would fail to meet the criteria of the Pulitzer judges and in general they're probably fated to do so. But there's no guarantee.
My comments have little or nothing (I favor the latter) to do with the discussion. --Minasbeede 00:39, 13 November 2007 (UTC)
Right. And, the nub of the nub of You simply have to know how to use them is - as far as NOR is concerned - the word "use".
  • The "use" is what NOR policy covers (ala 'thou shalt not interpret or draw conclusions').
  • The "them" is what RS policy covers (as in your "... newspapers are reliable sources ...").
So, since this is NOR policy we're discussing, it's "use" that needs emphasis. The "them" (sources) is incidental to give the verb an object to bind to. ;-)
The nubs are thus: You simply have to know 1) how to identify appropriate sources; and 2) how to use those sources appropriately.
Newspapers - being sources - kick in at #1, while NOR - being evaluation - kicks in at #2.
-- Fullstop 17:45, 1 November 2007 (UTC)
  • If information taken from a newspaper (or any other source) is at all contentious, the inclusion of this information into the article can be done in proper journalistic format - i.e., instead of "In July 1987, Hitler was found on the moon", the sentence could be "In July 1987, the Times of London printed an article asserting that Hitler was found on the moon". Nothing in that latter sentence would be unfactual or actionable, and it wouldn't mislead a reader of the article. I've used this approach and have heard no complaints - I think it also makes the article look more encyclopedic. Thoughts? AllGloryToTheHypnotoad 16:53, 2 November 2007 (UTC)
I don't go for that approach - it opens the door for all sorts of nonsense. I don't think Wikipedia should abdicate its responsibility to produce high quality articles and you cannot produce good articles by delegating responsibility to the user to determine whether something is valid and useful. Your example is misleading in that in reproducing it without comment it gains implicit approval. Spenny 17:05, 2 November 2007 (UTC)
You say: "you cannot produce good articles by delegating responsibility to the user to determine whether something is valid and useful". I'm actually agreeing with you above: Wikipedia should be reproducing without comment, because adding our own comments violates WP:NOR - and also produces an online encyclopedia where 90% of all statements of fact are completely unsourced. My point above is, anything contentious A) should be sourced, and B) should be reported honestly - not by asserting the contentious fact, but only asserting that the source given asserted that contentious fact. (And even the article's assertion that the fact is "contentious" should be externally sourced and not just something asserted by an editor). That's what I meant by "journalistic", above. AllGloryToTheHypnotoad 14:18, 3 November 2007 (UTC)
My underlying principle is that when we edit we should be knowledgeable and it is not enough to copy sources but we should be using our knowledge and intelligence to use appropriate sources and we should not be seeking to absolve editors of using judgement. I think the issue is not about balance, it is not presenting a point of view, it is avoiding presenting wrong information. So my comment was meant to convey that if you use less reliable sources without comment, you are falling into an unnecessary application of "verifiability not truth" (a foolish principle in my book). That should be the last resort as we should be aiming for verifiability and truth. Spenny 17:12, 3 November 2007 (UTC)
We tell people "verifiability not truth" without getting into epistemology the same way teachers tell their students that the angles of a triangle add up to 180 degrees without getting into Non-Euclidean geometry. WAS 4.250 20:15, 3 November 2007 (UTC)
And - as far as NOR is concerned - we ought to be telling people "stick to the sources" without getting into what constitutes "sources." Let us not confuse - or undermine - one policy with another. -- Fullstop 17:02, 5 November 2007 (UTC)
I agree. The current version of NOR goes too far into territory covered by WP:RS and WP:V. The info on RS needs to be trimmed down or directly tied to NOR. Redundant repetition is not a good thing because it obfuscates the point we are trying to make here, and inevitably leads to contradictions between policies. The rule in technical specification writing is that you define a term only once, and then use it consistently thereafter. The same should apply to WP guidelines and policies. Dhaluza 11:16, 7 November 2007 (UTC)

Catholicism and Freemasonry again

A few weeks ago, I posted an RFC on this page (and at WP:NPOV) about the above article. Several of you responded, and expressed some dismay at the amount of OR and POV the article contained (Jossi, for example, called it a "POV and OR nightmare") and suggested either deleting the article or stubbifying it. Based upon your comments I did a major cut back, not quite stubbifying the article, but deleting the sections that I felt were the most objectionable (about half of the article). My stated reasoning behind this significant cut was so that we could rebuild the article in a way that does not violate NOR and NPOV. Unfortunately, my cut was viewed by some of the editors at the article as being POV in itself (since I am a Freemason, they view my concerns as being suspect). Rather than get into an edit war, I am once again calling for comment and help. I would like to ask that some of you return to the article and clarify your comments (or add new ones). Please help me explain why this article is such a "nightmare". Thanks. Blueboar 15:41, 13 November 2007 (UTC)

RfC: Proposal for WP:PSTS

Proposed: User:Vassyana/NOR 002 as a replacement for WP:PSTS. There may be more that needs to be addressed, but there seems to be broad agreement that this is at least a move in the right direction. We can always address more issues after taking this step, but if we can agree this is a good move from the current language, lets do that and then focus on refining even further. I am posting a notice of this proposal to the policy village pump (again) and listing this as a policy RfC. If someone knows of another venue I should post in to make sure the community is aware, please let me know (or by all means post a notice yourself). Vassyana 08:36, 4 November 2007 (UTC)

Request for comment

Discussion

This wording appears to have some virtues. I am just not in a position at this time to spend time on it. However, I do note that it says nothing about sources in science. I think you should ask the scientists on several WikiProjects what they think. --Bduke 10:33, 4 November 2007 (UTC)

The draft is an improvement over what we have now, but it unnecessarily uses primary and secondary sources as defined terms, rather than explaining the sourcing issues in plain language. As the draft fonow acknowledges, primary and secondary are defined differently in different fields, and editors will naturally revert to their familiar definitions because of the law of primacy. This has lead to counterproductive arguments over what is primary vs. secondary which distract from the real question of what is OR vs. NOR. Dhaluza 14:39, 4 November 2007 (UTC)
I do not disagree. I'd imagine the terminology and general language used would be one of the points covered after this step. There seems to be a lot of discussion over language and it's certainly worthwhile to explore. Vassyana 07:47, 5 November 2007 (UTC)
I like this proposal very much. While Dhaluza has a point, the simple fact is that there is substantial support (if not full consensus) for including discussion about primary vs. secondary sources in this policy. It has become clear that a version that does not include such a discussion will never be accepted. Compromise must be made on both sides of that issue. One clear benefit of this proposal is that it places the discussion of primary vs. secondary in terms of OR vs. NOR (something that the current PSTS section does not do well). Blueboar 15:22, 4 November 2007 (UTC)

I agree with Vassyana, Dhaluza, and Blueboar in that User:Vassyana/NOR 002 should replace for WP:PSTS. We should delay further proposed improvements for after that. As Blueboar pointed out, there is good evidence that this is the best compromise we are capable of reaching - and we have spent over six months on this. WAS 4.250 15:37, 4 November 2007 (UTC)

  • Agree This is far better than the current PSTS section, explaining things more clearly. Some tweaks will probably need to be added here and there as time progresses, but this proposed version is a major improvement. wbfergus Talk 17:43, 4 November 2007 (UTC)
  • No way Shall I "blue pencil" it? •Jim62sch• 21:25, 4 November 2007 (UTC)
This is not a vote. "No way" provides no useful information other than that you think you own the page. WAS 4.250 05:36, 5 November 2007 (UTC)
Really? I think I own the page? Damn that's news to me. Had you also noticed the second sentence, you'd know that I offered to edit the proposal in a separate place (that's what "Shall I "blue pencil" it?" means).
Let me put it this way: the propsal begins with an unnecessary digression, and the removal of tertiary sources is a non-starter.•Jim62sch• 20:13, 6 November 2007 (UTC)
I fail to see how an introduction nipping the "but I learned x sources are y" confusion in the bud is an "unnecessary digression". The fact that even single fields often have various (often incompatible) definitions of PSTS has been repeatedly covered on this talk page. The effect of the law of primacy should not be underestimated, but rather considered and countered (as needed). Could you perhaps better explain your objection? Could you also explain what is essential about tertiary sources that we need to include a distinction that isn't even clear in the fields that use it? Vassyana 23:16, 6 November 2007 (UTC)
Vassyana, I know you spent a lot of time on your rewrite, and you're to be commended for taking the time to do so. I'm afraid though, that I just don't see the flaws in the verbiage of the policy as it stands now. I'm not saying that your intro is poorly written, in fact far from it -- it's quite well-written, but I just don't see how it's necesary.
The conflation of secondary and tertiary sources troubles me greatly. One of the purposes of the section on tertiary sources is to sound a warning, or caveat, on their use -- by mixing the two, that caveat disappears and opens the door for much contentious sourcing issues in the future. •Jim62sch• 23:55, 7 November 2007 (UTC)
I don't necessarily agree that certain clarifications are needed or that certain aspects need to be emphasized (or that some of the approaches are correct), but I worked on incorporating many of them because of the long-standing dispute and confusion over the current language. I cannot emphasize enough that the draft is in large part an attempt to compromise between the very broad ranging positions that have been expressed here over the past several months. I've tried to retain the essence of PSTS that supporters of the status quo (or something similar) have identified. At the same time, I've attempted to revise with an eye towards addressing most of the reasonable objections to the current language. It's assuredly not perfect (far far far from it), but I think it does a good job of reconciling quite a diverse array of opinions. Regarding the caveat, given the large variety of definitions given even within single fields and the fact that the law of primacy perennially rears its head with the "here's what I learned" argument, I believe a caveat is highly appropriate and quite needed. For an example of the problem of varying definitions, see one of my replies to Kenosis.[22] Does that clarify the perceived need for such a caveat? Outside of the two objections you've listed (we'll continue to discuss those), do you think it is better, worse or about the same quality as the current language? Why?
The warning in tertiary sources just notes that some are more reliable than others. What is said about them there applies equally well to all sorts of sources. The language originated in addressing the use of encyclopedias, which are an extreme minority of tertiary sources. The vast majority of tertiary sources are covered by textbooks and review literature, both of which tend to be more reliable than average. I certainly could add a note that sources have varying reliability and that those authored by experts are generally more reliable, if that would help accommodate your concern. Please let me know your thoughts in return. Vassyana 14:33, 8 November 2007 (UTC)
(ri) OK, let me think on this a bit. •Jim62sch• 00:20, 10 November 2007 (UTC)
No problem. Take the time you need to collect your thoughts. Vassyana 07:16, 11 November 2007 (UTC)
WP:STEAM yet again. Spenny 21:29, 4 November 2007 (UTC)
Really, Spenny? He who asserts must prove. Prove. It is WP:STEAM how? Do tell. •Jim62sch• 22:18, 6 November 2007 (UTC)
My point is that the pair of you simply bowl in and say "no way" without being prepared to engage in discussion, which seems to be based on some view of your own omnipitence. Wikipedia does not belong to anyone, it belongs to the community. A substantial number of the community believe that there is something that needs fixing here. You simply assert that we are wrong, without really attempting to engage or understand, you assume that there is some evil doings going on here. Your superior attitude is offensive, it is the epitome of trolling, that is, the deliberate placing of antagonsitic comments to get a response. It is exactly the behaviour that brings Wikipedia admins into disrepute to the cost of the many worthy workers on our behalf.
Either be prepared to engage in civil and understanding debate on the issue, or if you believe there is malicious intent (which is clear from various talk page comments between yourselves, you do believe) then raise those issues formally, but do not disrupt the workings of Wikipedia based on your own prejudices. Spenny 00:42, 7 November 2007 (UTC)
Actually, there's an even more substantial segment of the community that feels that there's nothing to fix. They'll weigh in if this ever gets to that stage.
No one said there were "evil doings" here, merely that you want to fix what ain't broken. But seeing that you have inferred that we said that evil doings were afoot, please to 'splain.
I'm so sorry if my attitude offends thee. No really. •Jim62sch• 23:46, 7 November 2007 (UTC)
I'm also sorry for putting it so strongly, I just find that the statement that nothing is broken does not match my view. I think this is because you have not understood that my concerns are not about the academic world and high quality sources, but that policy needs to work in all areas of Wikipedia, for all editors. Without re-iterating the arguements too much, one reason I was motivated to weigh in here was that I found the policy pages were being used to produce POV articles and when I argued the case this very policy wording was used to defy common sense.
The specific example being using summary press articles to outweigh government Inquiry statements on the dubious grounds of the newspaper reports being secondary sources and the Government Inquiry being a primary source. The issues I had were: using the PSTS argument as a defacto means of excluding "unhelpful" quality sources without proper consideration of all policy (that's primary source, it is out, end of debate); the elevation of the press to an unsailable position (The press is secondary source, from that perspective, anything they report on is their primary source, therefore I can exclude all their sources.). This was being done by someone who exerted significant control over the wording of these policy pages - hence my cynicism that these pages are not broken.
I have no doubt that when applied to a scientific or historical subject with exemplary sources with experienced editors that the current wording works well. My premise has always been that the wording does not work for areas with poor or biased sources, for editors who are not skilled in the ways of academia. I believe therefore that it is important to get NOR back to a simple and obvious statement of what NOR is without the baggage of PSTS, and then allow PSTS to be used as a tool where it is appropriate, recognising that Wikipedia goes into areas which do not fit the academic model. Spenny 09:02, 8 November 2007 (UTC)
Agreed. There's one of the trolls, the other shouldn't be to far behind. wbfergus Talk 22:36, 4 November 2007 (UTC)
Troll, huh? Really? I don't think there are many trolls with three FA's and countless GA's to their credit. And by the way Fergus, the worst way to deal with me is to engage in ad hominems. Capisce? •Jim62sch• 20:13, 6 November 2007 (UTC)
Did Fergus just call you a troll? Wow, that is seriously news to me. Thanks for the information fergus. Orange</nt>Marlin Talk•

Contributions 20:18, 6 November 2007 (UTC)

Certainly seems that way. And, from what I can gather, so doing is a bad thing. Of course, I could be mistaken: maybe Fergus' fingers slipped on the keyboard, or maybe Fergus was just in a pissy mood. •Jim62sch• 22:22, 6 November 2007 (UTC)
Now that I'm back in town again I can reply. Yes, I most certainly did. Your past accomplishments with FA or GA articles does not prevent either of you from engaging in such behaviour, both here and on my own [off_my_page|talk page]. As long as you engage in reasonable discussions, presenting your side articulately, and refrain from "name-calling" or other abusive behaviour, then you obviously are not in the "Troll" category. But your previous behaviour here and on my talk page does clearly fall into the "Trolling" category. Much, if not most of your work, may clearly have been well spent doing constructive edits elsewhere here on Wikipedia, but your comments here (up until the last few immediately above this), clearly fell out of that category. Simply leaving comments here saying 'no way', 'non-starter', etc. without any further explanation does not help others to see why you may object. Your comments are then considered to be merely 'trolling'-type comments, as they do not attempt further discussion, but as an impediment to discussion. wbfergus Talk 16:06, 9 November 2007 (UTC)
(ri)Your link speaks for itself (although, likely not in the way you intended). Again I digress. Have you, perhaps, maybe, noticed a difference in the way I deal with you and the way I deal with Vassyana? Probably something there, eh? •Jim62sch• 00:24, 10 November 2007 (UTC)
Whatever twists your crank Jim. Your comments on my page came out of the blue. I never left you any comments anywhere, or made any statements about you that would entail any response, so either OM asked for your assistance in being a troll, or you followed him around to see what he was editing and where and then decided that he needed assistance as he was floundering. Here at least you are now constructively participating in the discussion, instead of exhibiting the previous behaviour of OM and yourself. Whatever caused the improvement, I really don't care. The point is that you are constructively participating now, instead of just popping in to leave a negative comment without any explanation or supporting statements that could be addressed. wbfergus Talk 19:16, 10 November 2007 (UTC)
Whatever twists my crank? ROFL. As for your strange assertions, let me just say that you're free to develop and believe delusions to your heart's content.
OK, so does anyone here really want to discuss the issue, or shall some of us continue to engage in a pissing match? •Jim62sch• 20:36, 10 November 2007 (UTC)
Well... I'm not sure if WP:STEAM actually applies... as I think the proposal is a majority opinion and STEAM only applies to squashing minority opinions. But that is a quibble, I do agree that Jim should at least explain what the objection is. Blueboar 03:30, 5 November 2007 (UTC)
Could you please provide something more constructive than a "no" without explanation? What is wrong with the proposal? Is it better or worse than the current language? Why? Vassyana 00:39, 5 November 2007 (UTC)
  • Disagree Neutral. Vassyana's draft completely omits the most common form of accepted primary sourcing used on Wikipedia: descriptions of artistic works (such as plot summaries). I think it could be sustantially shortened as well, and as Dhaluza points out, made plainer. I'll put my money where my mouth is and offer some revisions within the next two days to demonstrate the kinds of changes I have in mind.--Father Goose 00:05, 5 November 2007 (UTC)
It is certainly common, but also controversial. Rather than add more controversy to an already controversial section, that should be an issue tabled for further discussion. It is not a question of whether or not this proposal is perfectly correct. Rather, it is a question of whether or not this proposal is an improvement over the current language. Further revisions can always be addressed later. Is this proposal better, worse or equitable to the current language? Why? Vassyana 00:39, 5 November 2007 (UTC)
Exactly. Saying it could be better does not answer the question asked. This is a wiki. The process is incremental improvements - not freeze something a majority disagrees with until perfection is attained in a proposed replacement. More people like the replacement than like what we have now. Let's get back to normal non-ownership style editing. That is what made wikipedia great in the first place. WAS 4.250 05:40, 5 November 2007 (UTC)
It's worse than the current language by completely omitting this critical point. Whether or not some people oppose such sourcing, it is absolutely necessary to use it in certain limited circumstances. If this is not made explicit, people will purge such information as "unverifiable" (though it is verifiable), and huge holes will form in Wikipedia, particularly in our coverage of creative works. This is one point that cannot be glossed over or avoided.
As for "saying it could be better" -- like I said, I'll offer some revisions to make explicit what I feel needs to be changed.--Father Goose 06:37, 5 November 2007 (UTC)
Because the proposal omits a "critical point", that is not in the current language, the draft is worse than the current language? That makes no sense. If you want to include such language, we can always address that controversial topic separately. What makes this proposal worse, other than the absence of additional proposed language? Vassyana 07:47, 5 November 2007 (UTC)
The critical point that is in the current language is articles "should (1) only make descriptive claims about the information found in the primary source, the accuracy of which is easily verifiable". Your version restricts this to "only report what the source states". This omission makes it impossible to provide any form of basic description of a subject (such as a plot summary or something simple like the fact that KITT was housed in a black Trans Am) unless that description appears in a secondary source. This is vastly more restrictive than the current policy, and unnecessarily so.--Father Goose 06:43, 7 November 2007 (UTC)
Ah, my apologies for not understanding your concern. That was an error on my part in choice of language. I have changed that to "only report the content of the source", which should accomodate your concerns. It would plainly allow someone to saw KITT is a black Trans Am, for example. Is there anything else about the draft that raises alarm bells for you? Vassyana 15:56, 7 November 2007 (UTC)
That addresses my greatest concern well enough that I'll switch my position to "neutral". I still have some problem with "some occasions and exceptions" linking to WP:IAR and a single obscure exception. One should not invoke IAR to deal with known shortcomings of a policy, and one should not invoke it in a policy to handle "unanticipated cases" since that's one of IAR's jobs anyway. I'd be somewhat happier if it were rewritten as "some occasions and exceptions (such as WP:SELFPUB)", with the link to IAR discarded.
Then there's the problem that the whole thing is still rather muddled, and that we should possibly be considering moving the entire PSTS section to WP:RS. Fullstop and others are expressing this point well. But that problem exists with both the current version and the NOR002 version, so it's not specifically an objection to NOR002. I've been working on trying to generalize the principles behind PSTS, instead of just shuffle it around, and if I get anywhere with it, I'll share my ideas here.--Father Goose 20:23, 7 November 2007 (UTC)
The IAR language concern is a legitimate concern and your suggested replacement seems solid to me. I've altered that phrasing accordingly. If you think of anything else, please let us know. Your concerns and suggestions have been on the money, and the feedback for improvement is quite appreciated. Vassyana 21:07, 7 November 2007 (UTC)
I would additionally point out that the draft acknowledges that "there is a broad consensus for widespread use of some primary sources", with census data and interviews as examples. It uses two clear and noncontroversial examples. If someone would want to assert that a particular use of primary sources has broad consensus, it would be up to them to defend that assertion (like any other time someone declares consensus). Vassyana 07:55, 5 November 2007 (UTC)
  • Agree with the edit (though I don't fully agree with the draft). I think the edit would be a step in the right direction. I don't think we really need to listen to editors who say "I oppose draft X because it does not contain controversial language or theory Y". That's not a valid objection, and such statements do not annul consensus. I hear and agree with Dhaluza's statements above, but I'm willing to go with a draft that is less controversial than the current one, rather than insisting at this point that the expression of the policy have no controversy. (Eventually, all significant controversy must be eliminated, of course; otherwise, it's not a true consensus policy article.)COGDEN 00:53, 5 November 2007 (UTC)
You're free to disagree with others' views, but saying "we do not need to listen to them" does not lead discussion in a constructive direction.--Father Goose 07:33, 5 November 2007 (UTC)
  • Weakly agree. It is barely any different, only slightly better in some ways. I do prefer “raw facts” in the lead of the primary source definition. I don’t approve of the mention of “A secondary source” before its definition. There are many more than “some occasions” where primary sources prove useful. The biggest benefit of agreeing is that it encourages further efforts. --SmokeyJoe 05:27, 5 November 2007 (UTC)
Some improvement is better than none after the long period of disagreement. I agree that more improvement is necessary, but one step at a time. :) Vassyana 07:47, 5 November 2007 (UTC)
  • Comment. The first paragraph of the proposal is unnecessary and confuses the issues needlessly, IMO. The WP usage already closely tracks the usage of PSTS in library and information science. I adjusted the footnote so it reads accordingly here.
    ...... The proposed presentation also, IMO, gets bogged down in explanations that are perhaps best put into one or more additional footnotes to supplement existing basic material in PSTS rather than attempt to replace it.
    ...... I do see some merit in the passage explaining that a given source might be primary in one application but secondary in some other context. Perhaps this would be useful in a footnote as well. If it were concise enough (say, one sentence or two short sentences at most), I'd support its inclusion in the body text of the policy page. ... Kenosis 16:38, 5 November 2007 (UTC)
Thank you very much Kenosis. It is productive and constructive and positive to add parts of the proposal to footnotes in the current version. At this point I approve of any movement forward no matter how small. While one way of looking at it might be to see it as an attempt to see how little of the proposal can be added; I think it is better viewed as part of the beginning of a series of constructive incremental steps to improve the policy in ways corresponding to conclusions reached in a debate in which for over six months almost all discussion participants had a similar problem with the current PSTS section. WAS 4.250 17:36, 5 November 2007 (UTC)
This is something that has been covered multiple times in discussion. Even within library science, there are various definitions. Another common information science definition is that primary sources are simply materials that make original assertions or provide original material. Another definition in the same discipline, is that primary sources are purely source materials; even secondary sources drawn upon in other secondary references are considered primary sources. I will not revert, but rather ask you to reconsider your edit as it amounts to cherry-picking a definition and is misleading in that it implies there is a unified definition in that field (which is patently false). Vassyana 17:54, 5 November 2007 (UTC)
Misleading my foot. It is a general definition that is in fact fairly consistent across the field of library science, which is in turn very closely in keeping with WP's usage of PSTS. Specific interpretations of which sources belong precisely in which classification, with respect to a given topic, are no less specific or universally consistent than interpretations of WP:NPOV, WP:V or for that matter the rest of WP:NOR. They are all editorial policies that are defined case-by-case and article-by-article within a consensus process among those WP users participating in each particular article. ... Kenosis 19:02, 5 November 2007 (UTC)
There's a world of difference between primary sources as purely source material and primary sources as the original source. How NOR is interpreted in local consensus is an entirely separate matter from the various (and conflicting) definitions of sources used within a particular discipline. It is misleading to present (or imply) a false fact, such as that the definition provided is universal to information science. It's even worse when the second reference explicitly lists some of the various definitions of primary sources, rather than presenting the field specific definition (as implied by the edit). The page explicitly aims to present an overview of how they are used and defined, not present the library science point of view. According to this glossary, primary sources are original works in the creative and informational sense, as opposed to the historical sense used by your first reference.[23] Another presents primary sources as the original source of information.[24] Another relates primary status to closeness to an event, or rather first-hand accounts are primary.[25] This one uses the definition equating primary sources with source material.[26] This one supports the historical sense used by your first reference.[27] It's pretty clear your presentation of a universal definition just doesn't wash. Vassyana 21:33, 5 November 2007 (UTC)

I've taken the incentive to replace the prior version of "Secondary sources" with Vassyana's version of "Secondary sources". Personally I feel sure that's the strongest part of the proposal, and does not substantively change or diminish the present expression of the policy. The prior text in secondary sources was not well written in it's latest incarnation, and the examples of the traffic accident simply wasn't very useful. The edit is here. ... Kenosis 19:19, 5 November 2007 (UTC)

Before we make any changes, we really should allow time for people to discuss the proposal, for those active in the discussions to comment and for the request for comment to have a chance to draw outside opinions. I am going to revert the change for now, as the section is highly controversial and there's no rush. We can wait at least a few more days or a week to make sure there are no significant and well-founded objections. Vassyana 21:33, 5 November 2007 (UTC)
I may or may not be around in a few days. Since it's got my interest at the moment: If the proposal is to eliminate tertiary sources and give the long apologetic in the body text about how some fields may use it differently -- this belongs in a footnote at most. I must oppose the proposed replacement for PSTS. In addition, the removal of the many currently used examples of primary sources in favor of the proposed paragraph on primary sources doesn't represent an improvement, IMO. The proposed paragraph on "Secondary sources" does represent a significant improvement, mainly because that paragraph is presently in a severely deficient state that doesn't effectively convey the breadth of secondary sources across the wiki, while the proposed paragraph does much more effectively convey the typical range of secondary sources than does the current version. ... Kenosis 22:16, 5 November 2007 (UTC)
Why is the elimination of the tertiary source distinction problematic? Why do we need to distinguish them for the purposes of NOR? Why is the caveat so problematic? Why do we need an extensive list of primary sources? What is wrong with the primary sources section, beyond the shortened list of examples? Vassyana 10:07, 6 November 2007 (UTC)

(edit conflict) Among the benefits of User:Vassyana/NOR 002 over current wording is that it avoids this entire argument over what is the "real" meaning of "primary source" and makes clear that we are using some real world usages as a springboard to craft a new related meaning that is useful and relevant in the context of evaluating the useage of sources as references for claims in wikipedia articles. WAS 4.250 19:25, 5 November 2007 (UTC)

Sure, the expression of primary sources could use further discussion as well, so as to express it in a way that doesn't affect continuity of usage of the policy among those who've already taken it to heart across the wiki. For instance, the reason that original philosophical works and religious scripture are primary sources is that they are closest to the topic when the topic is the original philosophical work or the statements therein, or the original religious scripture or the statements therein, When the topic is whether the religious scripture accurately represents the historical information, as in the case of historical Jesus, then scripture might not be primary source and other artifacts may be primary sources instead. When the topic is Aristotle's ideas, Aristotle's works are primary sources. If the topic is Plato's or Socrates' ideas as talked about by Aristotle, then Aristotle's works are secondary. It depends on the topic and on the discussion among editors at a given article who are negotiating the content in the context of NOR. ... Kenosis 19:36, 5 November 2007 (UTC)
  • Comment: this new PSTS - though less foamy than its predecessors - still does not cut to the chase. And like its predecessors, this latest incarnation too has forgotten the context, which is how PST sources are related to 'No original research' policy.
-- back to basics begin --
The fundamental and general characteristics of PST sources are:
  1. A primary source is a piece of (typically novel) information that is subject(ed) to analysis.
  2. A secondary source is an analysis or interpretation of preceding material.
  3. A tertiary source is a non-analytical review/summary of existing analyses or interpretations. A tertiary source is "tertiary" by virtue of not being either a primary or secondary source: It is not primary because it neither introduces novel information nor lends itself to analysis, and it is not secondary because it does not itself analyze or interpret.
A Wikipedia article may not itself be either a primary or secondary source: it may not introduce novel ideas (original material), nor may it introduce novel analyses or interpretations (original research).
-- back to basics end --
Fullstop 22:10, 5 November 2007 (UTC)
This is not the view of PSTS presented by most editors. It would be an entirely different discussion. Though I doubt you'll receive much support for that approach (review the talk archives), I would recommend raising this as a separate proposal. The draft was intended to better frame the section in terms of NOR, and to compromise between the various concerns and views raised over the past several months here. How does it measure in those terms? Is this overall better, worse or just a sideways shift compared to the current PSTS section? Vassyana 10:07, 6 November 2007 (UTC)
  1. If your "draft was intended to better frame the section in terms of NOR" then the draft is not doing what you intended it to do. As NOR_002 concludes: "drawing conclusions not stated in the reference, or extrapolating a position from the claims in a source, is original research regardless of the type of source." If something can be original research "regardless of the type of source", then there is no need for a classification of "type of source" is there?
    ~ Note also that "drawing conclusions" or "extrapolating a position" can be summarized with "analysis" or "interpretation," which are the keywords of my "back to basics" summary of PSTS. As such, I'm rather surprised that you conclude that it is "not the view of PSTS presented by most editors."
    If this were really so, then NOR_002 is of course "not the view of PSTS presented by most editors" either.
    On the other hand, it appears that pretty much everything so far "is not the view of PSTS presented by most editors," otherwise we would have had a majority by now, wouldn't we?
  2. The fundamental "flaw" in all the drafts so far is that a) these definitions attempt to be comprehensive, b) the sheer length of the descriptions encourage glossing. Together, these factors discourage thinking about any description under review. You (and I) are just at fault as anyone else: your description can - in principle - apply to what I had written, but if it does not, then its either because you haven't thought about what I had written and/or I had misunderstood what you had written.
  3. To answer your "How does it measure in [terms of terms of compromise and in terms of applicability to NOR]":
    • Applicability to NOR: Given that
      a) the words "analysis" and "interpretation" are the key definitions of "original research" and that these words occur in every sentence in my "back-to-basics" definition, and that
      b) the ~125 words of "back-to-basics" actually lead up to the (last) two words "original research",
      I suspect that my "back to basics" definition of PSTS is slightly more related to NOR policy than anything so far.
    • In terms of compromise: I really think that if anyone were to earnestly try to find the kernel in what they think P/S/T sources are, they would find a match in my back-to-basics suggestion. The inverse - that my back-to-basics suggestion is the kernel of any notion of PSTS - would also be true.
      This is because "back-to-basics" represents the commonality of all definitions and objections so far. Your draft included.
-- Fullstop 17:37, 6 November 2007 (UTC)
Is this "back-to-basics" suggestion an explicit suggestion to be found somewhere? --SmokeyJoe 06:14, 7 November 2007 (UTC)
  1. The problem is not with the conclusion of your "back to basics", but rather the definition of primary sources you choose. It's so broad as to be useless. Additionally, plenty of secondary sources are subjected to analysis, but are not treated as primary sources in Wikipedia. It's most certainly not the way most editors treat or define primary sources.
  2. The current section is fairly comprehensive as well. While simplification is a fine goal, given the length of the controversial discussion and page protection of this policy, our first priority should be reconciling as many of the opposing views as possible. After an initial compromise to move forward from the dispute, discussing further refinement (such as using different terms or severe simplification) would be perfectly appropriate. We should be looking at ways to move forward, not demanding an extreme solution or nothing.
  3. Trumpeting your extreme simplification is not an answer to my questions. The implication that if people would just earnestly examine the issue (as if a large number of editors haven't over the past several months), they would "see the light" is degrading and insulting. Please actually answer the questions, explaining why or why not the proposal on the table is better or worse than the current PSTS section and explaining why or why not the proposal on the table is a reasonable compromise of the various opinions and concerns expressed here over the past several months.
Vassyana 23:04, 6 November 2007 (UTC)
    • Your "it's so broad as to be useless" is precisely the result of "reconciling as many of the opposing views as possible". A definition of PSTS with respect to NOR is indeed useless. You know that (as evident in the second last sentence of NOR_002) and I know that, and pretty much anyone who has thought about the permissibility of a source knows it as well.
    • Indeed, "plenty of secondary sources are subjected to analysis, but are not treated as primary sources in Wikipedia" is perfectly true. Back-to-basics does not say otherwise.
    • I (think I) am beginning to understand that you think of "sources" in terms of a book, a chapter, an article, a volume, a series. In reality though, a "source" can be as small as a phrase or factoid. A "source" has to considered in terms of "pieces of information" and in context of usage. A head count (raw or otherwise) is a primary source, the analysis of such a head count is a secondary source. And yet both appear in a census report. It doesn't matter what kind of source the census report is altogether, its the individual facts in them that are either primary or secondary.
    • "Raw facts" are primary sources because they do not analyze (themselves?). "Transcripts of the Nixon tapes" is a primary source because it is subject to analysis, not because it does not analyze (the "situation"). And historiographic accounts are not admissible for RS reasons, not for NOR, because when historiographers do analyze, they are usually flawed analyses. That is an issue for RS, not NOR.
    • Shakespeare is a primary source because he does not analyze Ovid (or whoever) he is using as a source. But a quotation from Shakespeare is not by-default a bad thing. If however, an editor were to conclude that Shakespeare was a racist because Othello is a negative character, then that would be OR. It is irrelevant which source were being cited: either Shakespeare directly or Bloom's observation that Othello is a negative character. "Do not interpret" is the crux of OR, not what kind of source is being used to interpret from.
  1. "We should be looking at ways to move forward, not demanding an extreme solution or nothing." After six months we are no further ("moving forward") then we were to begin with. As such, it might be a real step forward to reconsider whether the path being trod on has lost sight of where it is coming from or where it hopes to go. Its no surprise that NOR_002 is in the middle of nowhere, discussing the color of the grass by the wayside: The document is called NOR, but NOR is not the subject in that document. This is not moving forward; this is preserving the chaotic state of affairs.
  2. "is not an answer to my questions." If you think an answer does not suitably answer the questions you asked, then perhaps you need to consider that you had failed to phrase them in an appropriate fashion. Its not necessary to assume that the reader is an imbecile. Not that your latest phrasing accords with with you initially asked, but this new question is easy enough to answer:
    NOR_002 is no better or worse than the current PSTS section. Its still a load of dowah, undermining its own conclusion that "Drawing conclusions not stated in the reference, or extrapolating a position from the claims in a source, is original research regardless of the type of source."
    This is precisely the same idea (though slightly less convoluted) as that expressed in the conclusion of the current PSTS section: "Like primary sources, secondary and tertiary sources should be used in a way that does not give rise to new analyses, syntheses or original conclusions that are not verifiable from either the sources themselves or from other sources."
    The message is the same: both conclude that any kind of source can be misused for OR, and that - irrespective of kind of source - analysis and interpretation is a no-no. But both ramble on about kinds of sources anyway. Different nitty-gritty, but same ole blether. Not going anywhere, leave alone forward.
    The question you ought to have asked was: "Does NOR_002 make the the editorship-at-large actually understand NOR policy better?"
-- Fullstop 17:57, 7 November 2007 (UTC)
This is really starting to fork into an entirely separate discussion, so I will keep this relatively brief. It is not so broad as to be useless because of a broad compromise, but because the definition you offer is relativistic, providing no guidance or help in using sources. It simply says that primary sources are the subject of analysis. In the absence of a more comprehensive explanation the plain reading does indeed indicate any source subjected to analysis is a primary source, including other secondary sources. A "source" is commonly treated as a complete individual work, not individual claims within a work. If you wish to change that, I recommend starting by changing the common practice along with WP:V and WP:RS. It's noted that you disagree with the entire approach of the status quo. However, if you wish to participate constructively, you have to realize that is our starting point, that we need to work incrementally and that there must be compromise. Finally, you make the fundamental error that a conclusion about all sources indicates there is nothing else to say about differing types of sources in relation to NOR. That the conclusion limits us from making other statements, advice or restrictions regarding particular sources is a fallacious position on its face Vassyana 21:07, 7 November 2007 (UTC)
My definition of PST is not the point. This talk section is not about my definition, but the failures of yours, as indicated is the very first two sentences of my initial comment. My apologies for assuming that you were a thoughtful person who would assume that a comment tagged with *Comment* might actually be saying something that you could use, or even for assuming you would actually read or think about it. Apologies also for subsequently assuming that you were actually trying to understand where I was coming from.
That said,...
  1. Yes, the definition I provide is indeed relativistic. And this is what your definition should be too. This is because a) "secondary" would not be "secondary" without there being a "primary" (and vice-versa), and b) what is understood by "primary"/"secondary" is always relative to both the context of the source and well as the context of the material that the source is being cited for.
  2. And yes, the definition I provide "providing no guidance or help in using sources." And this is what your definition should be too. That is not NOR's business! WP:SOURCES is not WP:NOR.
    The conclusion might just as well read:
    A Wikipedia article may not itself be either a primary or secondary source: it may not introduce novel ideas (original material), nor may it introduce novel analyses or interpretations (original research).
    *poof* Sense maintained and no previous definition of PST (or even "sources"!) necessary. One could even keep the struck-out bit, and simply link to the relevant articles.
  3. And if I and others are supposedly making a "fundamental error" in concluding that an understanding of NOR is independent of an understanding of "differing types of sources," then perhaps - just perhaps - you need to consider that your NOR_002 is not doing anything to correct this (supposed) misconception. Instead, NOR_002 and current WP:PSTS actually tell us that the "kinds of sources" differentiation is irrelevant to NOR.
  4. I did not say the "conclusion [of NOR_002/current WP:PSTS] limits us from making other statements, advice or restrictions regarding particular sources." Thats your reading. What I did say was If something can be original research "regardless of the type of source", then there is no need for a classification of "type of source" is there?
    And if "kinds of sources" don't have to be described, then they shouldn't be. And if "kinds of sources" aren't described then all debate over what is/isn't a P/S/T source is moot. And so WP:KISS prevails.
-- Fullstop 17:36, 10 November 2007 (UTC)
I commented about how your definition was so broad as to be useless. The parallel you drew to the proposal is simply false. The draft is not so broad as to be useless. A completely relativistic definition that provides no context and guidance is however. Your definition is indeed salient, since you clearly have advocated it and framed your commentary on your "back to basic approach". Your continuing incivility is unwelcome, and I'd recommend you reconsider your choice of language and attitude. Engaging in gross simplification and insulting the intellectual capacity of others who disagree is distinctly unhelpful. You repeat the fallacy again while denying that you are making the assertion. Simply because original research can be engaged in regardless of the type of source doesn't mean that all sources are equally prone to abuse nor that there is an absence of anything else to say about sourcing in relation to NOR. In the absence of your interest in rationally and civilly discussing this matter, I have nothing further to say. Vassyana 00:14, 11 November 2007 (UTC)
  • Vassyana, it simply won't do to make backhanded allusions to irrationality, fallaciousness, "having an attitude," incivility, "engaging in gross simplification," "insulting the intellectual capacity of others" and heaven knows what else, just because you don't like (or don't understand) what I have to say. And don't dish out what you won't eat yourself.
  • You also continue to harp on the my definition although my definition is not the subject of this RFC. Its your definition that you put on the whipping post. So take the critique without whining and without getting defensive/offensive about it. In contrast, I have responded to every single one of your remarks, and in which I have repeatedly agreed with you, and in each case I have explained why they are - by design - the way they are. That you ignore that and simply reiterate your remarks is not fruitful.
  • But then again, your "definition was so broad as to be useless", "completely relativistic definition," "provides no context and guidance" etc are oblique and open-ended assertions that are actually meaningless without accompanying explanation.
    So,... if my response wasn't appropriate for what you had in mind, then its because you weren't specific enough. What, when, where, why, how.
  • Inversely, by cricitizing the alternate position, you are implicitely asserting that the opposite is true for your own draft. That is, you imply that it is necessary for your draft to be non-relativistic (about what?) and that it is necessary to provide context and guidance (for sources!) in NOR.
    But so far, you haven't provided a rationale as to why it is necessary to be non-relativistic (independent of context?), and why it is necessary to provide guidance for sources even though this is none of WP:NOR's business and is what WP:SOURCES is there for.
  • Indeed, "simply because original research can be engaged in regardless of the type of source doesn't mean that all sources are equally prone to abuse."
    But so what? Simply because not all sources are not equally prone to abuse does not mean that it is necessary to categorize these sources according to their susceptibility to abuse, or to suggest that there is a relationship between degree of susceptibility and some (perforce arbitrary) definition of PSTs.
  • "... nor that there is an absence of anything else to say about sourcing in relation to NOR."
    Indeed, but is it necessary?
  • NOR is simple. There is no need to make it complicated by discussing source types, even though such typing does not influence the understanding of OR.
    There is no such thing as "gross simplification." Simplification is never gross. A complex idea described in the simplest possible fashion is the ultimate sophistication.
  • But if sources must be explained on the NOR page (and thus partially undermining the purpose of the WP:RS and WP:SOURCES pages), then for heaven's sake explain why this is so, instead of simply insisting that it is so. What, when, where, why, how.
-- Fullstop 14:24, 11 November 2007 (UTC)

Support the inclusion of the paragraph on "Secondary sources", but only this paragraph. However, the words "offer a more independent view" in the proposed replacement for "Secondary sources" paragraph is not necessarily the case and should be further discussed IMO. Oppose the rest of the proposal as presently written. Additionally, I'd like to request discussion on the inclusion of the following sentence under "Tertiary sources": Such sources can be useful in providing a more general overview of topics where secondary sources are highly complex, in disagreement with other secondary sources, or otherwise difficult to summarize. IMO, this proposed sentence would help to put tertiary sources into better perspective. ... Kenosis 22:23, 5 November 2007 (UTC)

The newspaper report of an event is to an historian a primary source, not a secondary source. The Vassyanana proposal excludes newspapers from use as sources in Wikipedia.(except for their editorials). 03:37, 6 November 2007 (UTC)
Does it? Really? Where? ≈ jossi ≈ (talk) 03:43, 6 November 2007 (UTC)
  • Agree to Vassyana's proposed edit. The language lays out the potential NOR pitfalls of prim, sec sources. It eliminates the "tertiary" distinction which even experienced editors here aren't able to agree is necessary (there really isn't a significant distinction that need be made between sec and tert sources in terms of NOR anyway. I'd argue there is somewhat more value in separate "policy" points between sec and tert in describing pitfalls with NPOV.) This is not a departure in longstanding NOR policy, just a refinement. This proposal does not restrict in any way the use of primary sources for making uncontroverted claims. And we need to get on with it--we need "good enough" policies at WP which are stable, more than we need crystal perfect policies which are continually shifting. Professor marginalia 15:33, 6 November 2007 (UTC)
  • No way. Since webfergus, because he's the all knowing individual, will request reason, let me put it this way. I don't need a reason to stand on a principle to NOT fix what is NOT broken. I am unconvinced by the any of the arguments. In other words, the proposal is so minor as to be irrelevant, and what we have works so well. So there webfergus, am I a troll too? OrangeMarlin Talk• Contributions 20:21, 6 November 2007 (UTC)
If the proposal is "so minor as to be irrelevant", then why are you stating opposition instead of abstention or neutrality? It makes no sense to oppose what many support if there is little to no difference either way. Vassyana 23:04, 6 November 2007 (UTC)
While wbfergus has fallen for the irritated response, I'm afraid putting antagonistic comments up is the very definition of trolling, so, as an admin, you are not abiding by the spirit of what Wikipedia should be about.
Still, that aside, my main quibble with your response is that if, as you appear contend, the policy is expressed as the epitome of perfection, why are there so many diverse opinions on what the policy means or should mean? Spenny 00:27, 7 November 2007 (UTC)
Well, I don't think I would even deserve the backhanded characterization of me as a troll, but I was definitely passive aggressive to wbfergus. His comments were unnecessary. That being said, let me respond to you.
I thank you for responding sensibly and I appreciate for the tone in which you respond. I agree that his comments were unnecessary and antagonistic, but in responding in the style that you did, when you know people are sensitive then I suspect you know you are going to elicit a response. I simply ask that next time around, (and there will be I'm afraid!), you display a little more tact. There are ways of being robust without being dismissive. Spenny 09:39, 8 November 2007 (UTC)
From my personal observation, which is definitely original research, everyone tries to use rules to benefit themselves. I edit mostly science and medicine articles, and I frankly see some of the worst possible citations to defend an edit. Alternative medicine and Creationist cruft, utilizing bogus research and claiming the Journal of "I think I observe this" as the source of their POV, is the worst possible way for this project to go. I think people want to move Original research to mean slightly different things is a canard. Most of us with strong SPOV's, think that the NOR here is perfectly fine. Small changes, over several months, tend to move the ideal in a direction that will make this project no better than Conservapedia. So, you might observe diverse opinions, but what I think you're observing is POV warriors whining about the fact that they get reverted in their edits. Not sure that constitutes a great groundswell of support for change. OrangeMarlin Talk• Contributions 01:04, 8 November 2007 (UTC)
I responded to Jim above, and I think the issue with your position is the the same. You work in a particular area with an academic bent, using an academic perspective and the rules work well for you. My contention is that Wikipedia is a wider community and covers a wider range, with varying qualities of editor. My motives are: to have a policy that speaks to the average person, not just to academics (I actually had someone use primary sources as a phrase for the first time in the real world a couple of weeks ago and it was the first time in nearly 50 years that has happened); to ensure that PSTS is not used to falsely elevate poor sources by giving it precedence over other policy (specifically being careful about setting the press, very accessible via Google, as a sound source over other sources of merit); being aware that policy needs to work for notable subjects without good quality sources where the community clearly wants articles. An example of the later, which is more my field, is computer software where often the highest quality, definitive, source is either the manual or manufacturer sourced papers (for example IBM red books or Microsoft Knowledgebase). Bear in mind that these sources (close to the subject) I would have interpreted with the wording as being primary sources and at times people have sought to have PSTS wording that bans them in entirety.
My point is that you see PSTS as a tool for avoiding undue manipulation of sources: my point is that the current wording is used to manipulate sources for POV warrioring. A clear statement of NOR can be made without source typing which therefore avoids this manipulation, but that does not stop PSTS being used as a tool to support the NOR where appropriate. You see it as ideal and therefore any changes must therefore be retrograde. I wish I knew how to engage with you constructively and concisely (I'm not good at that!) to persuade you that this is a flawed position. Spenny 09:39, 8 November 2007 (UTC)
  • Disagree, for reasons similar to what Fullstop set forth above. While these definitions are obviously put forward in good faith, I honestly worry that they will only serve to confuse unnecessarily those who want to be guided by policy, while providing a club to those who wish to advocate their own agendas -- or to simply cause trouble. The primary problem with the existing definition of "primary sources" has been that its meaning differs between disciplines: a "primary source" in history is far different from a "primary source" in the research sciences, for example. Any changes should be made to fix this disagreement in meanings, while attempting to simplify the language -- the "back-to-the-basics" approach that Fullstop advocates. We are not writing a law here; we don't need to be exacting in the details. -- llywrch 22:16, 6 November 2007 (UTC)
Tut-tut, OM, you said no way. Like OMG, how horrid. Of course, I say of course, attempting to "fix" something that isn't broken is pretty horrid too. •Jim62sch• 22:26, 6 November 2007 (UTC)
And your evidence for that position is? Spenny 00:27, 7 November 2007 (UTC)
Please see my response to Fullstop above. Vassyana 23:04, 6 November 2007 (UTC)
I also have a few direct questions. Could you please explain how the proposal is worse than the current PSTS section? Could you explain why it is not an appropriate compromise of the various (and often conflicting) concerns and opinions expressed here over the past several months? Vassyana 23:08, 6 November 2007 (UTC)
  • Object. The original phrasing was accurate, more clear, and enjoyed wide support for over a year or more. I don't see any reason why it needs to change and this is not an improvement. FeloniousMonk 05:09, 12 November 2007 (UTC)
I think it would be very difficult to show what the genuine support for policy was over the last year based on the strong ownership that was enforced on the policy pages at certain times.
It might also be useful to suggest which version you thought was better. There have been some consensus amendments over the last few months to remove inappropriate statements, and is there now is not what was there a year ago. Spenny 11:21, 12 November 2007 (UTC)
I think FeloniousMonk overstates the original accuracy, clarity and support to the point of disingenuity. PSTS is not a work of art. Its inertia/stagnancy should not be mistaken for stability implying quality. --SmokeyJoe 14:31, 12 November 2007 (UTC)
The original phrasing was far from clear, contained loopholes, and hardly had wide support for over a year. Just in the last year from today, the policy itself has had over 500 edits. That is hardly a stable policy and clearly shows that numerous people had problems with the wording and attempted to clarify various parts of it. Further examination shows that the majority of those edits related to PSTS, which is also what most of the discussion on this talk page has centered around for the last few months. Saying otherwise, as stated above is either an inadvertant wording due to lack of researching (or more probably just going by memory without looking at the facts), or intentionally misleading. I would have expected better from an Administrator. wbfergus Talk 15:41, 12 November 2007 (UTC)
  • Agree I definitely am eager to see that there continue to be a distinction among types of sources. If it's necessary to give more latitude to primary sources, then I think Vassyana's draft is a good compromise. I really appreciate the work he's done on this. TimidGuy 16:14, 15 November 2007 (UTC)
  • Oppose. The proposed replacement is too wordy and it's hard to follow the train of thought. The current version is shorter and easier to read. Also, I disagree with requiring explicit in-text attribution or with stating that secondary sources are preferred. Thanks, however, to those who put a lot of work into it. Balancing the various concerns on this page is not easy. --Coppertwig 00:21, 16 November 2007 (UTC)

Arbitrary section break

First, I want to acknowledge that it's clear a lot of work has gone into this definition. And creating standards can be a useful thing: instead of putting effort into choosing between two or more alternatives, I can simply look to see what the Wikipedia policy is on a given situation is, & act accordingly.
However, my point in my objection above is that, as someone who struggles with issues of original research whenever I write articles, all of this effort fails to help me one bit. Maybe it's the subject I concentrate on: I write a lot of articles on history, & I find that, if I try to follow what I think is the intent of "no original research" (& I've been contributing to Wikipedia for over 5 years now, so I should have some idea about what I'm doing), I honestly can't think of any reason to treat primary & secondary sources differently. If Tacitus or Suetonius state A, I report (in effect) "according to Tacitus, A." If a secondary source like Edward Gibbon, A.H.M. Jones, or Ramsay McMullen say "A is supposed to mean B", I write (in effect) "according to Edward Gibbon, A is supposed to mean B". Citing a statement from Gibbon, another from Jones & a third form McMullen & saying that it proves some novel concept is, IMHO, just as much original research as if someone were to the same by citing from Tactius & Suetonius. Lastly, if a tertiary source like an encyclopedia says "C", I should take a step back, scratch my head & wonder why I am relying on an encyclopedia for my research. (Admittedly, there are reasons.) In short, for someone writing history (& related subjects), this detailed definition of the three types of sources is not a positive aid.
This detained definition does help two sorts of people: the extreme literalist & the motivated troublemaker. Wikipedia encourages the use of secondary sources over primary ones -- for reasons we need not go into here. Yet sometimes secondary sources get the facts wrong: respected authors confuse dates, botch translations, get sloppy in their research. Or sometimes they state fact A, & themselves reference a primary source; why not then cut out the middleman & cite the primary source alone? Despite all of this, there are people who will argue that Wikipedia articles should always avoid using primary sources, no matter the reason. (Note: this is not a "strawman argument"; I have seen many such arguments put forth by established Wikipedians in good faith. It is amazing -- & a little sad -- to see the many ways a simple statement of policy has been twisted.)
Now I came to all of this as an outsider: I saw Vassyana's notice at the Village Pump, & followed the link. I don't know what has been argued here before I popped in, & to be blunt, if this is a compromise of what has gone before then this effort is in a lot of trouble. I've stayed away from policy for a long while because I've learned that either I can spend my time arguing policy or writing articles; I decided I feel more productive writing articles. So I've mostly ignored whatever the policy about original research said; as I wrote above, I think I know what the policy is supposed to mean, & unless I encounter some odd case I have no reason to look at the policy; I suspect this is the case for most Wikipedians who are contributing content. So a group of people can write the most comprehensive, detailed explanation of these classes of sources -- only to have no one make good use of them, because we don't want to take the time to digest such a large slab of text -- & if in doubt, there's always the safety valve of ignore all rules. Or on the other hand, this group can write a less detailed definition that people will read & can apply to their own experiences, & it will be used & actually influence how things are done on Wikipedia.
A last note: I've tried to present the above in a dispassionate, & non-accusatory manner. If I've offended anyone, I apologize. I also apologize for the length of this response. However, I honestly feel that this rewrite is heading in the wrong direction, & that the direction Fullstop writes -- whether or not he did so in good faith -- a simpler, back-to-the-basics definition, is the right one. And if my comments are not deemed as useful, I'll be happy to leave & return to working on content. -- llywrch 05:43, 7 November 2007 (UTC)
Thanks for that "outside view". As someone who has long argued for a major simplification, in which I see the classification of sources as an unnecessary obfustication of the basic policy of NOR I strongly agree with your concern. The issue is how do we get from here to something more obvious and streamlined which does not go down a detailed rabbit hole. The long discussions were about saying that policy based on classification of sources was doomed to be unacceptable when what we were asking for could be summed up as stick to the sources. To attempt a compromise, those in favour of simplification have tried to move slowly, but the result is that we seem to have satisfied nobody - the simplication has not been achieved and the obstructionists for change claim any change is too much yet the proposed change is too small to be worthwhile. The key phrase of FullStop, which there is good consensus amongst the advocates of change is If something can be original research "regardless of the type of source", then there is no need for a classification of "type of source" is there? Spenny 10:01, 7 November 2007 (UTC)
Llywrch, I appreciate that you provide some well-considered comments. However, I am troubled by comments which undermine my good faith in your participation here. For example, you digress about tertiary sources, but the proposal quite clearly leaves out that distinction. That leaves in serious question whether you even looked over the proposal, considering the absence of tertiary sources is plainly obvious. Your comments (as a whole) are a tangent raging against the status quo and anything like it. If you choose to actually participate, you are quite welcome to, but outright ignoring the actual proposal on the table while soapboxing your opinion is unhelpful and a serious disservice to the editors here trying to work out a compromise. Vassyana 15:56, 7 November 2007 (UTC)
First, my comment about tertiary sources was meant as a joke. Obviously you didn't see the humor, but my intent was simply to diffuse the tension evident here. However, as for your comment about "outright ignoring the actual proposal on the table while soapboxing", my only response is that I'm not posting to stir up trouble -- I'm honestly puzzled that so much effort has gone over some definitions of tangential value to Wikipedia policy. I have read the proposal, but I didn't see anything there that helps to explain why so much effort should be expended on defining them. (Maybe there's more than meets the eye to this whole discussion.) As I wrote above, if the best thing I can do at this point is to bow out -- I will. -- llywrch 22:48, 7 November 2007 (UTC)
Please don't let my response discourage you from participating in the discussion. I thought the comment about tertiary sources was a bit flippant (in the lighthearted sense), but at the same time the proposal doesn't use tertiary sources. In combination with the fact that your post was more a statement against the existance of anything but a very simplistic version of the section (and a few other aspects of the post), it seemed like you weren't addressing the proposal or discussion at hand. I apologize for my lack of good faith. I simply should have asked for clarification. On that note, is the proposal better or worse (or just about the same mess :-P) as the current section? (Also, you've obviously put a lot of thought into this topic and I'd encourage you to stick around to help us further refine the section, whether or not this proposal succeeds.) Vassyana 23:44, 7 November 2007 (UTC)
Responding to a portion of your post, a lot of people believe primary sources are more likely to be misused for original research. For example, in topics related to Christianity it is generally (perceived to be) more likely that Bible quotations will be used inappropriately in relation to NOR, than a theologian's book describing the meaning and correlation of various biblical verses. Another problem is the (perceived) inherent original research in using some primary sources directly. This is particularly true of historical sources lacking the editorial oversight and/or modern credentialing generally required of reliable sources. For example, Caesar's Gallic War is widely considered a golden example of ancient reporting, but is also considered a masterpiece of propaganda and known to contain inaccurate hearsay. Without relying upon a reputable scholar to determine which parts are reliable, and to what extant, direct use of that primary source is almost unavoidably original research. (And that's a reputable example of ancient sources.) Just a couple of examples to illustrate what the problem as many see it is with such sources. Vassyana 23:44, 7 November 2007 (UTC)

(outdent) Sorry for the delay; I had some off-wiki business that kept me from thinking & writing out a prompt response. First, I think it is an improvement in that (if I understand it correctly) it removes a discussion of tertiary sources. Other than that, as I said above, because it doesn't trim back the wording I'm not entirely convinced that it is an improvement.

To respond to your second post, V., the reason we worry more about people misusing primary sources to create novel interpretations than about misusing secondary (or, excuse me, tertiary) is that in the later case, this misuse is far easier to identify & deal with: as important as a storehouse of facts a book like (for example) A.H.M. Jones' The Later Roman Empire is, obviously Jones presents only a selection of the material he has found, & with a patina of his interpretation. For that reason, I become concerned when I encounter Wikipedians who seriously argue that we should never quote primary sources; sometimes the secondary sources make their own mistakes.

From my (admittedly incomplete) reading of Wikipedia, the only case I've encountered where it is important to existing policy to distinguish between primary & secondary sources is in the formulation of notability: there are instances where it is very appropriate to use the existence of a secondary literature to prove notability. (However in some cases it's not as clear-cut, such as Ethiopian history: the secondary literature is not as rich as in the case of, say, Early Medieval European History, so some very important people & events are not covered. But I'm rambling here.) It might be best, for this reason, if this section were moved to WP:NOTE.

The problem with primary texts is the Wikiepdian who insists on reading into the passage his own interpretation, & cannot be shaken lose from that interpretation. Passages in religious texts are the best example of this, & they will just as eagerly read their own meanings into the secondary literature as well (e.g. "The book of Mark says this, & it is proven by Tertullian"). I don't know what to do about these people, except to let their own lack of ability to interact with other people defeat themselves.

As for your example of Caesar's Gallic Wars, IMNSHO the proper solution would be to quote what Caesar wrote but also quote later scholars who interpret him, as other points of view. This follows the principle of neutral point of view. Opinions in the secondary literature often mutate, & the predominant opinion of one generation may be reversed by the next; the text most often remains the same.

Sorry to have written so much again; I'm honestly not trying to soapbox here -- I have a blog to do that. -- llywrch 21:39, 9 November 2007 (UTC)

No problem! We all have lives. :) I don't think anyone is advocating, or has advocated, that we should never draw upon primary sources. The proposal itself moves to some degree in the opposite direction, identifying more clearly instances where the use of primary sources is accepted. (For example, noting the broad consensus for the use of census data and interviews. As another example, making it clear that "analytic, synthetic, interpretive, explanatory, or evaluative claims" contained within primary sources are acceptable when explicitly attributed in-text to the source.) As far as notability, I believe you are conflating third-party/independent with secondary. There is some overlap, but they are distinct from each other. For example, the commentary and interpretive analysis of the Tanakh in the Christian New Testament, or of the New Testament in the Koran, are certainly third-party, but are still treated as primary sources. (There would be additional concerns of reliability, but it's just an example to address the conflation of independent and secondary.) Just a side response, Tertullian is unlikely to be accepted as a secondary source by most editors, and in most of the relevant fields (anthropology, history, etc) he is a primary source (and treated as such under the proposed draft). On your response about the Gallic Wars, that approach would be perfectly acceptable under the proposal. (If you believe otherwise, please let me know the what language gives the opposite impression, so it can be repaired.) No need to apologize for the length. You're responding directly to the discussion and your post is plainly well-thought. Vassyana 23:35, 9 November 2007 (UTC)

Consensus and compromise

There are a number of editors on both extremes of almost literally all or nothing. On the "all" side, there are a few editors who object to any significant change from the current language. On the "nothing" side, there are a few editors who object to any change short of an extreme simplification or complete removal of source discussion from NOR. I believe both extremes honestly have the best interests of Wikipedia at heart.

Those trying to preserve the section are not trying to wikilawyer the place to death and those opposed to the section are not out to dismantle core policy. However, an all or nothing approach is not helpful, and I would even say it's contrary to the fundamentals of the wiki.

Consensus isn't built with a "my way or the highway" approach. Most often, it's built with compromise and incremental change. This policy has been highly controversial and debate has raged for months. This policy has even been repeatedly protected over the dispute, for periods that would be considered troubling for mainspace articles, let alone core policy.

If you're here to point out flaws and work with us to address them and improve, by all means, you are most welcome. If you're here to tell us it's your way or the highway, I'd venture to say you need to reexamine your approach and decide whether or not Wikipedia as a whole is a suitable place for you. Consensus is fundamental to the wiki and it does not arise among entrenched parties unwilling to compromise or work together in an intelligent and civil fashion.

A broad swath of editors have joined together in this discussion and even in supporting the draft as a positive change to one degree or another. Those editors range from those who have vigorously defended the current policy to those who have vigorously opposed the disputed section. The individual editors here and the community as a whole are under no obligation to entertain those holding extreme positions in discussions, even as we do our best to form some compromise between them. Similarly, we are only obligated to engage and consider those participating in productive discussion and consensus building. </end rant> Vassyana 08:00, 15 November 2007 (UTC)

Origins of this policy

I've wondered for a while why this section is needed. It's bloated and it's not actually dispensing policy. Well, in part it dispenses policy through the bullet lists. But this makes the page positively confusing: was this policy or is this policy? Can we take out the bullets that we need and scrap the remainder (or perhaps move it to a sub essay page)? Marskell 15:06, 15 November 2007 (UTC)

Programming Code

Why isn't all the code on the various computer science pages a violation of "no original research?" You have all this demonstrative code that wiki dudes have written just for the wikipedia with assertions like "this code does X and says Y about this language," etc. etc. There is example code all over the internet, in dozens of books, and in other sources that could be solidly referenced.

Basically, I can go and write a bunch of code and post it, and say what it does, and that isn't original research? You may say "Well other wiki dudes can go and verify the code," but I ask you this, is that not ALSO original, unverifiable research? —Preceding unsigned comment added by 141.154.109.226 (talk) 15:37, 15 November 2007 (UTC)

It may be helpful if you give us some example articles to look at, so we can see what you are referring to. Blueboar 18:09, 15 November 2007 (UTC)
If multiple editors can look at a statement, such as one that says "this code does thus-and-such", and see that it's correct, then it's considered verifiable. One can say "'radar' is a palindrome", or something similarly uncontroversial, without needing to provide a source.--Father Goose 18:48, 15 November 2007 (UTC)
Admittedly, I know very little about code, but I think the IP editor who raises the issue does have a point about OR... if the code was created by a Wikipedia editor, and not discussed by some form of reliable source prior to having an article written about it, then Wikipedia has become the primary source for presentation of that code ... and isn't the whole point behind WP:NOR to say don't do that?
As to WP:V... I think this is really more of an issue for WP:Notability than WP:Verifiability. If a bit of code has not been discussed by some form of reliable source (tech oriented media or websites, etc.) I have to question whether it is really notable enough for inclusion in Wikipedia... is it really encyclopedic? Do we have a set of notability guidelines that discuss these articles? Blueboar 19:38, 15 November 2007 (UTC)
I think that if a piece of code is able to meet WP:V, chances are that it's notable. According to the verifiability policy "...a reader should be able to check that material added to Wikipedia has already been published by a reliable source." Only a tiny percentage of code has actually been discussed (as opposed to merely being included) in a reliable source. I'd have to see an actual example to give an informed opinion, but in general I'd agree that unsourced code is probably OR. Chaz Beckett 19:53, 15 November 2007 (UTC)
I think this is both a concern of NOR and V. Generally, we expect information to be verifiable by any reasonable educated person, with a notable exception for foreign language sources. If the examples are original, it is then also a concern of original research, in addition to verifiability. Also, chances are the point could be explained (if in simplified fashion) in layman's terms without resorting to code examples and highly technical claims. Vassyana 21:18, 15 November 2007 (UTC)

Just as a short math equation needs no source, as long as it is supported by the surrounding sourced English, similarly a short piece of computer code does not need to be directly quoted in order to be used. An article that had nothing but code would be no good, but an article like Join (SQL) is improved by giving some example code. — Carl (CBM · talk) 21:06, 15 November 2007 (UTC)

Join (SQL) is a good example about what exactly is wrong with many computing and mathematics articles on Wikipedia. It is nearly incomprehensible to anyone without specialist knowledge. It also is a great example of the overcoverage computer and technical topics receive on Wikipedia (which is a natural consequence of its primary talent pool). Vassyana 21:24, 15 November 2007 (UTC)
Explaining things in a simplified fashion is also a goal, but there's no way to demonstrate the precise syntax better than just exhibiting it. The goal of our articles is to give both an adequately general and adequately detailed presentation. The lede of that article seems perfectly comprehensible if you accept that you might have to use the blue links to find the definitions of terms you don't know.
In any case, those code snippets are no more original research than the examples in this article on conjugation in latin, and are used for similar purposes. — Carl (CBM · talk) 21:30, 15 November 2007 (UTC)
That's just it, I don't think demonstrating the precise syntax of individual database commands and functions is appropriate for Wikipedia. This isn't Computapedia. Such excessive detail is no less cruft than fancruft, if a bit more respectable because "it's science" (so to speak). But, I think this is digressing into an entirely separate conversion, so I'll just agree to disagree. Vassyana 22:05, 15 November 2007 (UTC)
I think the Latin article is better, but still similarly poor. In either case, there are more than enough sources available to cite examples and their context. There are a ridiculous number of introductory Latin books available, that would probably add some better context and explanatory text. For the SQL commands, there are plenty of reliable sources that explain SQL and its commands. There is simply no need at all for original examples in either instance, as solid references are abundant and easily available. Vassyana 22:14, 15 November 2007 (UTC)
So you're saying you think that in addition to having examples of verb conjugations, the article on latin conjugation should explicitly quote those examples from some other book? I don't see how would that benefit the reader, or increase the accuracy of the article, given that the example is just a list of conjugated forms of some common verb. Similarly, I don't see the need to source the scale displayed in whole tone scale, even though a naive reader doesn't know how to read or write musical notation. — Carl (CBM · talk) 22:30, 15 November 2007 (UTC)
General comment: I've been away from this discussion for some time, and I congratulate those working on the PSTS section: the current version appears to me to satisfy the various concerns that have been raised and balance the various viewpoints. I like it. Re computer code: perhaps math proofs and computer code need to be specifically mentioned in NOR. I would like the policy to allow relatively simple computer code and math proofs that can clearly be seen as correct by people checking them. There's a problem with math stuff: if you follow exactly an example in a book it's plagiarism, but if you use your own slightly different example you can be caught by NOR. It's a bit different from the case with history or something where you can paraphrase a few words and it's still correct. With math, if you change one part of an example, you have to change a bunch of other parts to to keep the whole example accurate. It's also different from history in that history requires reliable sources to back things up, while math can be verified by reading the proof. The thing being proved should still be attributable to reliable sources. --Coppertwig 00:12, 16 November 2007 (UTC)
Please don't try to change the wording to "help" with math. Simple computer code and examples are already allowed by longstanding interpretation of this document, and the people who edit math articles find the current wording perfectly adequate for dealing with cranks without limiting our ability to write an encyclopedia. Trying to explicitly deal with math (and CS and the physical sciences) will just result in a lot of discussion in which nobody seems to agree. If it isn't broke... — Carl (CBM · talk) 00:45, 16 November 2007 (UTC)
I do have a question of the notability of a simple computer language command such as JOIN clauses... does it really rate an article in a general encyclopedia? Do we have articles on every other code command? If so, WHY? I can understand an article on the Language itself, but is a general encyclopedia really the right place for information on specific code commands? I would think that should be relegated to programing manuals. What is so notable about JOIN clauses and similar code that they rate an article? (OK, this isn't really the place to discuss this... I just had to say it.) Blueboar 02:42, 16 November 2007 (UTC)
In reply to several of the above concerns, I'd argue that programming code, even in short snippets, is not like short math equations. Math equations exist in a very specific context. If I say 2 + 2 = 4, it is understood by all involved that we are in the context of good, old fashioned base 10 integer arithmetic. With code, there are so many questions about the context that get left unanswered. What OS is this running under, what compiler, what version of the language, etc. etc. etc. And most of the code I see written here is stuff that could easily be taken from and cited from the language manual or other authority. Regarding the "if it isn't broken, don't fix it" argument, what exactly is NOR about? The code on many pages goes way beyond "short snippet" and into fairly long multi-line blocks. That's a lot of information content to take on good faith. —Preceding unsigned comment added by 141.154.109.226 (talk) 2007-11-16T16:41:24
Long extracts of code should be discussed and removed on a case-by-case basis. They probably aren't appropriate. Short (four to five line) blocks of code are usually fine, and they are not what NOR is about. Based on your contribution history, it doesn't look like you are actually active in editing computer science articles. That makes me think you might just be trying to fan flames with this discussion. — Carl (CBM · talk) 16:45, 16 November 2007 (UTC)
Carl, please assume good faith. I think the question is worth asking. It does not matter if the person who asks it is active in editing computer science articles... sometimes an "outsider", someone who is not intimately involved in a given subject area, can see problems that "insiders" don't see. That said, I agree with your conclusion that this really needs to be addressed on a case by case basis. Like anything else, Code can be OR... but it is not by definition OR. -- Blueboar (talk) 17:28, 16 November 2007 (UTC)

Do interpretive primary reliable sources need to be verified by a second interpretive reliable source?

User:Dave souza has introduced an edit and comments suggesting that even if a primary source contains its own interpretation of facts, an editor still must cite to yet another source (a secondary source this time) for a second interpretation of those facts. Do we really need to require two reliable sources on any given point in an article? If so, what's the rationale, and what does it have to do with original research? COGDEN 17:24, 16 November 2007 (UTC)

To clarify, my edit[28] "Undid revision 171795309 by COGDEN" which had the summary "Okay. Instead of "another," though, what about "a", since the primary source could contain its own interpretation."[29] This introduced a conflict with the next sentence which stated, "To the extent that an article or particular part of an article relies on a primary source, that part of the article should... make no analytic, synthetic, interpretive, explanatory, or evaluative claims, unless such claims are verifiable from another source." That complies with the requirement to base sections on third party sources, and to avoid SYN any interpretation by the primary source should be stated as a fact, as in "X interprets the data in this way". However, I note that an anon has now reverted[30] back to the Revision as of 17:45, 15 November 2007 by Philip Baird Shearer (rv changes by COGDEN as they are more than minor changes. Please discuss it further on the talk page) .. dave souza, talk 21:16, 16 November 2007 (UTC)
"Third party sources" does not mean what I think you think it means. It means non-self-published sources. That issue came up at WP:RS and that was our resolution. It does not mean that you can't use an interpretive primary source without backing it up with a separate secondary source. It's not original research to only use one reliable source instead of two. That's not the policy, and if that's how it's going to be interpreted, we need to revert the the language to that effect. COGDEN 23:23, 16 November 2007 (UTC)
Somehow I think the point is being missed here. •Jim62sch• 18:51, 17 November 2007 (UTC)
Re this edit: [31] This could be misinterpreted, since some people consider most scientific articles to be primary sources, and they usually contain interpretations of their own results. I suggest the following wording: rather than "must be verifiable" or "requires another source for that interpretation", I suggest "must be attributable to a reliable source". Rather than "unless such claims are verifiable" or "unless such claims are verifiable from another source", I suggest "unless such claims are attributable to a reliable source." Rather than "conclusions that are not verifiable" or "conclusions that are not verifiable from either the sources themselves or from other sources," I suggest "conclusions that are not found in either the sources themselves or other sources". Or it could be even stronger with "stated" rather than "found". The trouble with "verifiable" in this context is that it could be taken to mean that a person looking at the primary material would draw this conclusion, whereas what I think we want to say in these sentences is that the source should actually state the conclusion or interpretation. (elsewhere, it's clarified that in certain circumstances, conclusions based on very simple logic or straightforward arithmetic are allowed. Or at least it used to be. Where is that?) --Coppertwig (talk) 03:58, 18 November 2007 (UTC)

The interminable ramble

I've been shorter than is ideal in a few of my comments, so now I'm going to overcompensate and be longer than is ideal. What I'm offering here is a concept of what "research" means in the context of a college instructor who has taught courses on writing the research paper, and who is generally interested in questions of interpretation, production of meaning, and the like.

First, let me state unambiguously what is the default position of pretty much every credible instruction on research is: use primary sources. There are a multitude of reasons for this - one very standard text, The Craft of Research, says of primary sources "these are the materials you directly write about." Meanwhile, of tertiary sources, it is said that "they usually oversimplify, are seldom up-to-date, and are consequently mistrusted by most experts."

Second, then, some more general reflections on what we do here. I have seen, on several instances, language thrown around like "a primary source requires another source for its interpretation." It is difficult to describe just how ridiculous this idea is from a serious academic perspective, at least in the humanities. The idea that any source or text, primary, secondary, or otherwise, is going to somehow hold an absolute and clear interpretation of anything - itself or another text - is absurd. And the absurdity of this can quickly be seen in practice - consider the following hypothetical.

John Milton is, obviously, a major poet. But to interpret his writing - whether poetic or philosophical/political, in the current view, we need a secondary source. One like Stanley Fish - the most important Milton critic of the 20th century. Now, never mind that one of Fish's most famous essays on Milton takes explicit issue with the idea that "there is a sense, that it is embedded or encoded in the text, and that it can be taken in a single glance" (the exact idea that we are buying into when we believe that secondary sources can somehow be used purely and straightforwardlly.) What strikes me as so very strange is that the same exact essay, "Interpreting the Variorum" magically becomes impenetrable the moment we try to write about it, despite the fact that it was perfectly clear when we were using it to write about Milton.

This is obviously a very silly situation. And it gets to the heart of what is wrong with trying to use the primary/secondary/tertiary distinction on Wikipedia in a programmatic way - exactly what Fish says. The act of reading is itself interpretive. There is not some meaning that springs, fully formed, from a text - all reading involves interpretation, synthesis, analysis, etc.

This, obviously, gets at the heart of this policy. Because, from any mainstream humanities perspective, when talking about the act of textual interpretation, it is not possible to engage in any textual interpretation whatsoever without indulging in original research. Period. To read is to interpret. There is no independent, Platonic interpretation out there that we can grab and transmit through Wikipedia. Words do not work that way.

It should be noted, this is manifestly not a claim of postmodernist relativism (though it may be postmodernist). I am not saying that, because all reading is interpretive and there is no absolute and perfect interpretation that we can just glom onto and transmit as is it therefore follows that all readings are equivalent. "Colorless green ideas sleep furiously" does not, in any remotely useful sense, mean "I like yogurt." Value judgments about the quality of an interpretation do exist, and are valid. What it does mean, though, is that we have to acknowledge that "no original research" does not mean "no interpretive acts."

So how do we make these value judgments? This is where it gets tricky. In 99% of the intellectual world, we do this via expert peer review. But, of course, we've rejected that idea for Wikipedia. So what are we left with? Well, if we were the peer reviewing experts, we'd do it on a number of grounds - our own knowledge, our evaluation of the sources cited by the interpretation we're judging, other sources, both primary and secondary, etc. And while we reject credentialism, none of these approaches are lost to us here. Indeed, this seems to me the very basis of consensus and of our semi-anarchic model of article content. An article is never finished, its just not being objected to at a given moment.

So what becomes relevant is less an attempt to define absolutely what "original" research is (a difficult task at best, given the above observations about interpretation always being necessary and never being trivial) and more an attempt to figure out criteria that constitute good and bad interpretation. The social norms of consensus go a long way here - if an article is stable, we generally trust it somewhat. A secondary issue here is that we generally assume that if you're editing in an area, you know something about the area. I don't go removing OR from articles I know little about. I might add Template:fact to them, but I won't proclaim OR. On the other hand, if I'm reading an article I know something about and see a line that I can tell is a reader's interpretation of something that's vague in the original, out it goes. Remember - our anti-credentialism does not mean that we are an encyclopedia written by people who know nothing about the topic they're writing about.

A final reflection - this page does not describe a class of edit that is bad in and of itself. The problem with original research is not that it is necessarily bad - it's that we can't easily tell if its good or bad. NOR and V are, in the end, very similar policies. But in the end, content that nobody actually thinks is inaccurate should be given the benefit of the doubt, and articles should not be approached with Cartesian skepticism. This is the biggest problem with the primary/secondary/tertiary section - it's written, ultimately, with the intention of being applied by people who do not actually know anything about the subject they're writing about. This is not a sound way to write policy or an encyclopedia. This policy needs to be reapproached from the assumption that people who write about a given topic and edit a given topic presumably know something about it. That doesn't mean we don't verify what they say. But it does mean that we can temper our requirements with a deference to the idea that contributors are not idiots. The easiest way to identify whether a synthesis is novel is not to demand that it be cited to a secondary source - that just recreates the initial dilemma on a secondary source which, as stated above, is already a step removed from what we're trying to describe and adding a level of simplification and interpretation. The easiest way to identify whether a synthesis is novel is if somebody goes "Wait a minute, I know a bit about this, and I've never heard that before." Which sets off a complex process of demanding sources, finding sources, interpreting sources, coming to consensus about what a source says, coming to consensus about what in a source is open to interpretation. In other words, it sets off writing and editing.

Sorry that this is so long, but as I hope you see, the issues here actually are complex and nuanced, and don't lend themselves at all to bright line policy making. This is a very fuzzy, tricky thing to judge - what perspectives are important enough to go into a general summary and what perspectives are silly, novel, minor, etc. We cannot be programmatic here, and we have to accept that judgments on this aren't going to be able to be made quickly, simply, or effortlessly.

In any case, if you're still reading this, A), you're a lunatic, and B) thanks. Phil Sandifer (talk) 03:48, 17 November 2007 (UTC)

Since I do not want to be a lunative I have taken Phil's point and did not read this to the end. But I have one point to make concerning this: "it is not possible to engage in any textual interpretation whatsoever without indulging in original research. Period. To read is to interpret." I agree that "to read is to interpret" but I do not agree that from this follows the point that all acts of reading are ipso facto original research. This would only be true if every act of reading involved an original interpretation. I do not think this is the case. I think that most readers have available to them a limited, often culturally sanctioned, repertoire of interpretive strategies and, in the case of some texts, actual interpretations that they learned and further at some point learned not to question, so that when they read a text - Milton or the Bible - they do "interpret" but the interpretation is not really their own and not original. This I suggest is precisely why people can continue to write new interpretations (readings) of the Bible, Shakespeare, Milton - they are not just new in some "platonic ideal" sense ... they are new in the sense that they challenge conventional readings, the un-original interpretations most people have when they read "the classics." Slrubenstein | Talk 15:36, 17 November 2007 (UTC)
Well, here we have to be precise about what we mean by "original." Surely every interpretation is original in that it was created in the head of the reader. Now, sometimes that interpretation mirrors one that has been created elsewhere, but the act of interpretation is always original as it is being made. Even if I read Fish on Milton and then go read Milton to see exactly where Fish got his interpretation, I am not having Fish's interpretation again - I am reading and myself assembling Fish's interpretation, if not from scratch, at least from the ground up. Fish pushes this question quite far - further than I'm comfortable doing so here - but I think his basic observation that all interpretation is a creative act is true. And, furthermore, is important, because there are things that come up sufficiently routinely in interpretations that nobody has really found it worth writing them down. Plot summaries are the most basic example of this, but it's also true for extremely complex topics. Secondary sources on Derrida are rarely much easier than the primary source, and certainly rarely take up the task of explaining Derrida as such. People learn Derrida not by reading all the secondary sources until they understand but, among other things, by coursework with a substantial oral component, by reading Derrida, through conference presentations and conversations, etc. All of this is a vital part of how academic interpretations are formulated, and cannot be easily replaced with "printed secondary sources." To attempt to do so demands poor articles. (Which is why I have routinely said that you cannot write a decent article on Derrida where every line is sourced). This policy needs to distinguish between original research and "common knowledge among specialists that it was never worth writing down in a published source." Because the difference is huge. Phil Sandifer (talk) 16:16, 17 November 2007 (UTC)
This may be obvious, but as I see it the aim here is not to do research, but to produce a concise and accurate summary of published reputable independent research about the subject of the article in question. As you say, it's a judgement call whenever we read and summarise something, and equally judgement is needed in establishing the best sources. Looking at the original primary source is valuable as a way of seeing if there are obvious discrepancies between that source and independent commentary on the source, but if our expertise gives us a novel non-obvious interpretation of the source, Wikipedia is not the place to publish it first. The complex process you describe is indeed the motor behind improving articles, and in my opinion these constraints help the process. As a non-expert, .... dave souza, talk 11:36, 17 November 2007 (UTC)
I think it's inaccurate to say that the aim here is not to do research - given that we demand that people cite sources, we demand that people do research. It's not even fair to say that the aim here is not to construct an original argument - as one of the standard composition textbooks says, everything is an argument, and a Wikipedia article is no different - the lead is essentially a thesis statement declaring "these are the major things to understand about this topic," then the remainder of the article explains what those things are and supports its claims with evidence in the form of sources. (The argument is original because, well, we are not simply copying or summarizing other tertiary sources - we are engaging in this process of overview ourselves.) The line between what we do and what academic research does is narrow. To some extent, this policy is necessarily mistitled - we do not really mean "original" as such, as we expect people to synthesize secondary sources into a summary overview, nor do we mean "research" as such, since many of the articles we have we are the first people to write tertiary sources about. What we really mean is "no novel results" or "no new theories." And you seem to understand that - we do not want "novel non-obvious" content derived from any source - primary, secondary, tertiary, etc. But the controls we are putting in place - rigid distinction among primary, secondary, and tertiary sources - both do not produce the desired ban and needlessly ban things that are absolutely what we do want. Phil Sandifer (talk) 14:58, 17 November 2007 (UTC)
I agree 100% with Dave souza, and I think his comment gets right to the point of this ongoing discussion. Wikipedia is not a venue for frustrated academics who cannot get published in peer reviewed journals (and lest anyone think I am casting stones, I am referring to myself). I very much want to do the kind of research to which Phil refers and when I do I submit it to peer-reviewed publications or academic presses. There is a process by which my research based on primary sources will - or will not - achieve notability. But I will not use Wikipedia as a venue to publish what I ought to be submitting elsewhere. As Dave states, clearly and convincingly, this is an encyclopedia in which we provide an account of established research on various topics. This has been a part of NOR from the start: if someone has new research they want published, they should seek to have it published by a journal or publishing house. Only then can it be considered for citation in an encycloipedia article. By the way, I agree with Phil's point that all reading involves an interpretive act. But perhaps there is room for Phil and Dave to find common ground. The issue is indeed whether an editor uses Wikipedia to forward an original interpretation (analysis, synthesis, explanation, etc.) this is why NOR has never banned the use of primary sources - it has only restricted the use of primary sources so that they are not used to support original arguments or controversial representations of the source. The bias towards secondary sources, as I understand it, is that the person whose "reading" is being provided is not a Wikipedia editor, and is clearly identified - it helps us also comply with NPOV by ensuring that something added to an article is presented not as "the truth" but as a view, in many cases the view of the author of the (secondary) source. In other words, a secondary source is (or at least, often is) a text that admits that its account of a primary source is a "reading," an interpretive act, and thus, the view of the interpreter (author of the secondary source), in short, a particular and identifiable point of view. Slrubenstein | Talk 15:26, 17 November 2007 (UTC)
Yes, but your research on primary sources is not academic because it is based on primary sources - it's academic because of how it uses those sources, how it engages with existing secondary sources, etc. And one of the ways your research will live or die is on whether it's original - that is, sometimes you'll send it to a journal and they'll reject it as insufficiently original despite being based on primary sources. (Other times they'll reject it as excessively so because it doesn't use secondary sources enough.) The point being that the division between that kind of research and Wikipedia is not a division based on sources as such.
I think you come very close to a good rephrasing of the source distinction section. The issue is not whether primary sources are used, but rather whether significant secondary sources are written about in a NPOV manner. Perhaps this is the form the policy should take: "While primary sources are considered more reliable and complete for describing something, WP:NPOV requires that we give a thorough summary to all major perspectives offered in secondary sources on the topic." Reword away, obviously. Phil Sandifer (talk) 16:01, 17 November 2007 (UTC)
What I'm offering here is a concept of what "research" means in the context of a college instructor. Thanks, but not thanks. That is exactly what this policy is trying to tell us not to do. ≈ jossi ≈ (talk) 15:52, 17 November 2007 (UTC)
You would prefer shitty research that gives incomplete accounts of things? Because that's what you get working entirely from secondary sources. The policy is not telling us to abandon all training as researchers, scholars, and writers - it's telling us to summarize existing texts instead of creating new primary or secondary sources. Trust me - as a college English instructor, we spent a significant portion of class time on exactly this task. And still, rule #2 (#1 being "don't plagiarize") is "primary sources are more complete." And that's the issue - every time you move to a secondary or tertiary source you remove information and add other information, and get further and further from the object you're trying to describe. The policy that stops creation of new primary and secondary sources should not moonlight as a policy mandating bad writing. Phil Sandifer (talk) 16:01, 17 November 2007 (UTC)
I think there is a miscommunication here. Phil, when you invoke being a college instructure I think you can be "read" two ways. first, you are saying that you have and try to communicate to your students high standards. I doubt that Jossi would reject this point. Surely, you don't think Jossi wants a shitty encyclopedia? But I think you can be read another way: you are teaching your students how to do a particular kind of research. When I took college English classes one of the things I recall being taught was how to make an original argument (clearly expressed, well-organized, supported by appropriate evidence). This is my goal when I teach. Even an undergraduate research paper has something in common with an MA dissertation and a PhD. thesis - forwarding a well-argued and appropriately-supported argument in a clear and well-organized way. That is the kind of research I try to teach students. But I am not trying to teach them how to write encyclopedia articles. I think this is Jossi's point - the criteria for a well-researched research paper and a well-researched encyclopedia article are different. We want to encourage the latter, not the former. Are we misunderstanding you? Are you teaching your students how to write encyclopedia articles? If so, to be frank, I would be surprised that you ask them to rely on primary sources. Slrubenstein | Talk 16:20, 17 November 2007 (UTC)
The criteria are different, but I don't think the techniques are - and I would not direct students entirely to primary sources to teach them to write an encyclopedia article, but I would also not direct them away from them. Of course, all of this is made harder by the requirement that we not be credentialist. That is, after all, how most encyclopedias function - they get experts to write articles. And those experts do not, in my experience, ignore primary sources in favor of secondary ones. They give an overview of what the secondary sources say, but they also use the primary sources heavily, as, in the end, it is still the primary sources they are trying to summarize. And that's the point I'm trying to make - one cannot write well about X, whether on the level of summary or on the level of academic research, without engaging X directly. And, in the end, academic research still starts with the lit review and with summary of major texts. The difference between it and encyclopedia writing is not method - it's stopping point. Phil Sandifer (talk) 16:27, 17 November 2007 (UTC)
They have credentials because they are trained to use primary sources. I see NOR and V as being important precisely because one need not have credentials i.e. be trained in the kind of research you are talking about, the kind of training that is indicated by an academic credential, to edit at Wikipedia. EB hires people with credentials because those credentials presumably indicate that they are well-qualified to do original research. It is because they have credentials that EB does not have an NOR policy! Wikipedia makes the opposite presumption: that our editors are not necessarily qualified to do original research. Nor is it our role to provide people the training in how to do research that they might receive in a University. Our NOR and V policies should ensure that people who are not qualified to do original research or research based on primary sources can nevertheless contribute to this encyclopedia. Slrubenstein | Talk 16:37, 17 November 2007 (UTC)
Except EB does have a NOR policy. Or, at least, it is not a publisher of original research. I take Wikipedia's position on its editors to be "trust their competence but verify their work." But that doesn't mean a line-by-line audit either. Phil Sandifer (talk) 16:44, 17 November 2007 (UTC)
Phil, are you editing the same Wikipedia that I? We have thousands of contributors, and thousands more join each week, and the vast majority of us are not experts on anything we edit ... thus this policy is necessary so that we can continue to contribute. ≈ jossi ≈ (talk) 19:30, 17 November 2007 (UTC)
When you say that, do you mean that in the sense of "are not credentialed experts on anything we edit" or in the sense of "don't know a lot about the topics we edit." The former is true. The latter, I should hope, is not. Phil Sandifer (talk) 23:12, 17 November 2007 (UTC)

Phil, what on earth do you mean by "line-by-line audit?" I am flumoxed. Did you just write the above in response to my comment? Did I say anything about lin-by line audit? Or are you ignoring my comment and making a general comment on this policy? Then I am still confused - nothing in this policy suggests a line-by-line audit. Has anyone mentioned it on this talk page? is this just another of your red-herrings? Where did it come from?

On the other hand, don't you know that no wikipedia article has an "author," that all articles are collaborative, that anyone can edit at any time? Doesn't this (NOT this policy, but the wiki technology itself) mean that ay and every line I write in an article can be changed by any other editor? Isn't this just business as usual at Wikipedia?

I really wonder what you are talking about. It makes no sense to me. I'm genuinely confused. Slrubenstein | Talk 19:53, 17 November 2007 (UTC)

Sorry - I'll try to be clearer. Yes, Wikipedia demands a level of verification of its contents - that's obviously what WP:V means. But, as I've said, that does not amount to a Cartesian doubt where we must check and source every statement - which is a viewpoint that comes up in policy debates related to these issues. There is some situation in which we don't particularly worry how a claim is backed up. And I think that situation is more than just an incidental error on our part - there are things we more or less deliberately overlook. What I'm arguing is that figuring out what these are is important to figuring out how to appropriately calibrate this policy. Phil Sandifer (talk) 23:12, 17 November 2007 (UTC) Phil Sandifer (talk) 23:12, 17 November 2007 (UTC)

Thanks for clarifying. My own view - which I believe is not shared by many people, people who I bet irritate you as much as they irritate me - is that this matter is best dealt with on a case-by-case basis, by editors of a given article. I see V and NOR as guidelines and ideals and points of reference in disputes. If there is an article and a diverse group of editors are satisfied with it, great, leave well-enough alone. Busybodies who go around adding a "cite" template to every sentence do irritate me! But I do not think that the response to these irritants is to weaken policy. I think instead we need to educate people as to the value of policy, which as i said is primarily to help educate newbies, and as a point of reference in disputes. No dispute, no problem. My princple is this: the greater the dispute at a page, the greater the need to enforce a policy strictly. The more consensus among (ideally a truly diverse group of ) editors, the less need for wiki-lawyering. In short, I say let's not overcentralize and over-plan. Rely on the judgement of editors active on an article. If they all think the article is good, great! But when editors start warring then obviously claims in the article ae by definition controversial and it becomes very important to make sure that no onre is violating NOR, that claims are verifiable, etc. I am tired, i hope this makes sense. Policies and sub-groups of the wiki community need to work hand in hand; when one is weak the other must be strong, and when one is strong we can tolerate more weaknesses in the other. Slrubenstein | Talk 00:12, 18 November 2007 (UTC)

Clarification of "Synthesis of published material serving to advance a position"

I am seeking clarification on the section entitled "Synthesis of published material serving to advance a position". Specifically, I've seen some people arguing that this prohibits a "synthesis" of published information even if it doesn't advance a position, or advances a wholly uncontroversial one. In my opinion, collecting information from a variety of sources in ways that may have never been published before is one of the things that makes Wikipedia a great resource, and as long as doing so isn't used to promote a particular point-of-view or pet theory, I see nothing wrong with it. Any comments? Can anyone think of a way to make this section clearer to address these concerns? DHowell 23:32, 15 November 2007 (UTC)

Synthesizing sources is how we write an encyclopedia article. anything other than that, is just assembling an indiscriminate list. Synthesis should be interpreted as "synthesis that amounts to original research" as the development of a new position on contested matter, or a new correlation of previously uncorrelated disparate items. DGG (talk) 23:40, 15 November 2007 (UTC)
I agree with this, up to "a new correlation of previously uncorrelated disparate items." If such a correlation isn't serving to advance a potentially controversial position, should this be prohibited? Why? DHowell 00:11, 16 November 2007 (UTC)
Agreed, if the correlation was disputed, you'd be obliged to provide sources that attest to the correlation itself. But bringing together any form of material into a single article is just the basics of encyclopedia writing.--Father Goose 02:30, 16 November 2007 (UTC)
I have a problem with this line of reasoning. The whole point to WP:NOR is that wikipedia is not the original publisher for the ideas and thoughts of our editors. Wikipedia articles are not the place to correlate previously uncorreleted disperate items, even when doing so is not controvercial. That should be done elsewhere, published in an appropriate reliable source... once it is, then we can discuss what that source has to say. Blueboar 15:14, 16 November 2007 (UTC)
Why would you seek to ban things which by definition no one finds problematic? Don't write a rule where there isn't a problem.---- Father Goose (talk) 18:18, 16 November 2007 (UTC)
Well put. It is a fallacy to assume that knowledge exists in a pre-existing form that we can simply report on. There is no description without interpretation and synthesis. The question is distinguishing between acceptable and unacceptable synthesis. -- Phil Sandifer (talk) 18:31, 16 November 2007 (UTC)
Incidentally, the current synthesis example looks to me as if the main problem it has is that it is opinion. An opinion that might be reasonable given the two sources, or any on academic plagiarism, but the opinion is the only relevant problem. John Nevard (talk) 04:53, 19 November 2007 (UTC)

Primary source distinction

This really is not a good section, and never has been a good section. For one thing, the secondary/tertiary distinction is tenuous at best (at this point almost all secondary sources also use other secondary sources, at least in academic writing). For another, it flies in the face of the usual primary/secondary distinction, which is to point out that primary sources are more accurate because they don't introduce error. For our purposes, a primary source, assuming its relevance to the situation is clear, will always provide more detail and more information than secondary sources - this is very important. Thirdly, the idea that secondary sources are somehow inherently easier to interpret is just silly. Fourthly, this seems to me to serve mostly to confuse the real issue, which is original synthesis, which has very little to do with primary/secondary sources, and is actually a pretty subtle issue that our policy formation has basically steered around in favor of the broken primary/secondary/tertiary distinction.

So here's the question I think is really important here - what is the difference between a novel synthesis and an acceptable one? Thoughts? -- Phil Sandifer (talk) 17:46, 16 November 2007 (UTC)

Films are primary sources. Film reviews are secondary sources. An encyclopedia that cites a film review is a tertiary source. The Magna Carta is a primary source. A book about the Magna Carta is a secondary source. An encyclopedia that cites that book is a tertiary source. Whenever someone interprets a primary source, they become a secondary source. Editors on Wikipedia are not considered reliable sources for interpretations. Editors on Wikipedia do not have to edit under their real name (although editors on Citizendium do). Wikipedia articles do not cite who wrote every sentence in an article. Encyclopedia editors collate previously published information. The term "original research" refers to "unpublished facts, arguments, concepts, statements, or theories." If certain information hasn't already been published, Wikipedia should not be the first publisher. We should quote secondary sources that make syntheses, not put forth our own. Adding your own previously unpublished ideas into a Wikipedia article can be considered self-publishing. Primary sources are accurate when it comes to direct quotes — we do have WikiSource for source texts. But for any interpretation of a primary source, we must cite published secondary sources. I think the availability of a primary source is also an issue. We can certainly cite the Magna Carta in the article about the Magna Carta, but readers should not have to travel to where a copy of the Magna Carta is on permanent display to verify an article is correct. To answer your last question, a synthesis is acceptable if it's been previously published, outside of Wikipedia. --Pixelface (talk) 11:24, 18 November 2007 (UTC)
Films are are not necessarily primary sources. Primary/Seconday source distinction depends on the usage. A film about a subject can be a secondary source for the subject. It might demonstrate notability of the subject. It might offer an alternative perspective, or criticism, or analysis.
What’s a novel synthesis? I don’t think that a simple, generic formula can be used. I’d suggest that that the first test for a novel synthesis is that it gets challenged, and then the question should be resolved on the talk page. --SmokeyJoe (talk) 11:42, 18 November 2007 (UTC)


The distinction is not between a novel synthesis and an acceptable one. Rather, it is between a novel synthesis and a published one. As for the primary source/secondary source distinction, the issue is that in the context of NOR, we have experienced situations in which primary sources have been "quote mined",for effect or to push a certain POV, or to create a novel interpretation of that source. That is why it is better to rely on secondary sources that describe these primary sources rather that quote these sources directly. ≈ jossi ≈ (talk) 20:20, 16 November 2007 (UTC)
Except that logic flies in the face of the standard lessons of responsible research, which send you to primary sources specifically to avoid quote-mining and the like in secondary sources. And, of course, secondary sources can be just as easily quote-mined and selectively interpreted, creating portraits of the truth that grow ever more distant from accuracy. Yes, primary sources have been abused. But they are indisputably superior sources for almost all purposes than secondary ones. The responsible thing to do is to figure out how they have been abused and to caution against that - not to so actively discourage them in favor of inferior sources. -- Phil Sandifer (talk) 20:43, 16 November 2007 (UTC)

Oh dear God. Is the standard wording of this page really that interpretation of a primary source requires a secondary source? What does that even mean? Somehow words become dense and impenetrable to Wikipedians when they are employed in primary sources, but the language clarifies magically in secondary sources? WTF? -- Phil Sandifer (talk) 21:00, 16 November 2007 (UTC)

Most of us recognize that the section as it now stands is not a true Wikipedia consensus, and we've been trying to arrive at a better formulation that meets Jossi's concerns. Unfortunately, there is a great resistance to any kind of clarification here whatsoever, and we've had to proceed slowly. Eventually, we'll get to consensus and this section will reflect Wikipedia policy. COGDEN 23:30, 16 November 2007 (UTC)
If I'm reading correctly, jossi's concern is that primary sources can be used to push a particular POV. But as Phil Sandifer points out, secondary sources can also be used to push a particular POV. If so, why then isn't the NPOV policy sufficient to address these concerns? DHowell (talk) 02:20, 17 November 2007 (UTC)

No, that is not the main concern. This article is about OR and states that we do not edit into articles novel syntheses. We can describe syntheses that have been published by others. When we discuss primary sources, this is most pertinent, as any analysis of primary sources is indeed original research. Can people make novel syntheses of secondary sources? Sure they can. But that is not the point; the point is that any analysis of primary sources performed by a Wikipedia editor will be novel by default. ≈ jossi ≈ (talk) 02:52, 17 November 2007 (UTC)

That does not preclude the use of a primary source to state a fact, such as provide a number from census data. What this policy forces us it to avoid using census data to make a novel interpretation of that data. It is quite simple, really. ≈ jossi ≈ (talk) 02:56, 17 November 2007 (UTC)
Phil: I think I see what is the problem. You say: standard lessons of responsible research, which send you to primary sources. That is the way it works in high-school, college, or in academic research. But not for Wikipedia, and that is the reason we have this policy in place. ≈ jossi ≈ (talk) 02:54, 17 November 2007 (UTC)
Yes, but that amounts, in practice, to Wikipedia's research being irresponsible and bad. Writing entirely from secondary sources leads to bad research. Or, at least, to research that is not recognized by any reputable authorities as good. Wikipedia should not be rejecting academic standards and substituting its own. As for the statement that "any analysis of primary sources performed by a Wikipedia editor will be novel by default," sure. But the same is true of any analysis of any sources. Interpretation does not become magically clarified just because the source is secondary. It is always a complex act. The idea that there is, in secondary sources, some fully formed interpretation that can be excavated and presented straightforwardly is utter fantasy. Phil Sandifer (talk) 03:21, 17 November 2007 (UTC)
It's untrue that any analysis of a primary source is original research. Actually, any original analysis of a primary source is original research. If it's not original, it's not original research. That's why we have the word "original" in the name, otherwise it would just be "research". So if a Wikipedia editor is just citing an analysis contained in a prior source (be it primary or secondary) that can never be original research. Also, if an interpretation of a primary source is verifiable, there's no problem, because by definition, anything verifiable cannot be original research. COGDEN 03:12, 17 November 2007 (UTC)
Novel synthesis is certainly not forbidden. The writing of Wikipedia articles is all novel synthesis. The collection of ideas from reliable sources and the forming of them into never-before-published sentences, never-before-published paragraphs and never-before-published articles is totally allowed, provided these new paragraphs etc. do not present new ideas. Simple, straightforward logic and arithmetic is also allowed. Wikipedia is not limited to direct quotes of sources only. (If it were, there would be problems with copyright, as well as with cohesion and readability of articles.) Primary sources often contain interpretations. I agree with Paul Sandifer that secondary sources are not necessarily easier to interpret than primary ones.
The current wording is a good compromise between the position that primary sources have to be treated specially and the position that the same rules apply to all sources. The current wording spends time telling people how to be careful with primary sources, and then says "Like primary sources, secondary and tertiary sources should be used in a way that does not give rise to new analyses, syntheses or original conclusions that are not verifiable from either the sources themselves or from other sources." I don't see how people from either camp can complain about the current wording. --Coppertwig (talk) 14:32, 18 November 2007 (UTC)
That sentence is true, but it contradicts a couple of earlier sentences in the section, which state that while you can cite to interpretations found in secondary sources, you cannot cite to interpretations found in primary sources alone. (Which is stupid, because maybe the secondary source simply repeats the interpretation by the primary source without additional comment. How does that transmute the argument into something acceptable that wasn't acceptable before?) Also, the section still gives the implication that somehow primary sources are inferior to secondary sources for Wikipedia purposes, which is not true. COGDEN 04:57, 19 November 2007 (UTC)