Wikipedia talk:No original research

From Wikipedia, the free encyclopedia
  (Redirected from Wikipedia talk:NOR)
Jump to: navigation, search
Peacedove.svg The project page associated with this talk page is an official policy on Wikipedia. Policies have wide acceptance among editors and are considered a standard for all users to follow. Please review policy editing recommendations before making any substantive change to this page. Always remember to keep cool when editing. Changes to this page do not immediately change policy anyway, so don't panic.
Frequently Asked Questions (FAQ)
I disagree with the definition of secondary source.
Wikipedia mostly follows the definition in use by historians, which requires more than simply repeating information from some other source or rearranging information from the author's notes. The earliest definition of a secondary source in this policy was in February 2004 "one that analyzes, assimilates, evaluates, interprets, and/or synthesizes primary sources".
WikiProject Spoken Wikipedia
WikiProject icon This page is within the scope of WikiProject Spoken Wikipedia, a collaborative effort to improve the coverage of articles that are spoken on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
 
Shortcuts:

OR/SYNTH contravention?[edit]

Statistics Canada publishes demographic census data on visible minorities. The order it presents the minority groups are as follows, based on largest to smallest populations: South Asian, Chinese, Black, Filipino, Latin American, Arab, Southeast Asian, West Asian, Korean and Japanese.

My inquiry is two-fold:

  1. Is it contrary to WP:OR/WP:SYNTH to present this data within tables in an order of groups inconsistent with the order published in the StatCan source? If no, would it more likely be a contravention of WP:NPOV?
  2. Is it contrary to WP:OR/WP:SYNTH to use different terms for the groups than those published in the StatCan source?

The concern about #1 is that some editors may reorder the groups to match their preferences for whatever reason, or it can be perceived that they have been intentionally ordered to align with someone's preferences.

An example of #2 is using "African" rather than "Black", or using "White" rather than "Not a visible minority", or in the case of Demographics of Canada#Visible minority and Aboriginal population, aggregating "like" groups into greater groups (like rolling Chinese, Korean and Japanese into "East Asian").

For your background, consider this discussion that triggered this inquiry.

Cheers, Hwy43 (talk) 04:05, 25 January 2014 (UTC)

Note the other editor involved just made a change. My inquiry above was made on behalf of both of us. Hwy43 (talk) 08:00, 25 January 2014 (UTC)
I concur with you and Skookum1. UrbanNerd's concerns should be disregarded.—S Marshall T/C 10:09, 25 January 2014 (UTC)
Thanks S Marshall. His concerns aside, can you confirm the above are technical violations of SYNTH/OR? Hwy43 (talk) 17:55, 25 January 2014 (UTC)
They're certainly in the topic area that SYNTH/OR touch upon, and I definitely wouldn't want people changing the terms for ethnic minorities willy-nilly. However, I feel that in this case they're not violations. I feel that a key aspect of SYNTH/OR is when you go beyond the sources to suggest conclusions that the sources don't imply. I don't feel you're doing that, so I don't feel that a technical violation is taking place.—S Marshall T/C 19:59, 25 January 2014 (UTC)
I'd be concerned about declaring people who pass for being white as actually having European origin, since they might not. You could misrepresent the source that way. But in this context, you could probably use Black and African interchangeably, since white Africans (e.g., Afrikaaners) aren't actually "visible minorities" in the sense that they're talking about in the source. This sort of change falls under the category of "explaining the source" rather than "materially misrepresenting the source". WhatamIdoing (talk) 01:38, 27 January 2014 (UTC)
I think you have just made a classic common American centric blunder, the peoples north of the Sarah are just as African as as those from sub-Sahara, yet I suspect that people from such an ethic background are more likely to be classified by the Canadians as Arab rather than Black as they say it is based on visibility. Whether or not that is true using African for Black distorts the survey. In the same way are not Arabs "West Asian", if not what does "West Asian", encompass? For reasons like this, if the authors of the Canadian survey have not defined precisely what the terms mean, is is probably best to leave the wording as is. -- PBS (talk) 13:20, 27 January 2014 (UTC)
The fact that it is a Canadian survey on what people look like is exactly why I believe that their category of "African" excludes non-Black Africans (from Afrikaaners to Berbers to Arabs and more), and includes all dark-skinned people, even if their ancestors immigrated two centuries ago. Also, they probably include immigrants from Jamaica and other Carribbean countries in this category, even though many of them don't see themselves as being from Africa. WhatamIdoing (talk) 15:32, 19 February 2014 (UTC)

We frequently enable users to reorder data tables, so I see no problem with orders. Terms is a more difficult question, though. Aggregation becomes more problematic, it seems like synthesis to me. SamBC(talk) 00:04, 27 January 2014 (UTC)

Thank you all for your insights. Cheers, Hwy43 (talk) 08:59, 28 January 2014 (UTC)

Third party sources and tertiary sources , Hitler and Mussolini (OR rules)[edit]

Simple question - 1. Are a third party source and tertiary source the same ? And a second longer question - A :DVD-documentary about Mussolini includes a speech that this Italian dictator held in Bari in 1934, after the murder of :Austrian dictator Dolfuss. Mussolini says (it is subtiteled sound film, and a part of the documentary)
"With a serene contempt can we reguard some doctrines from the other side of the Alpes, doctrines which has been created by a people that even didn't knew the art of wrighting, and couldn't write their own history at the same time as Rome had Caesar, Vergilius and Augustus ! It is indeed at the shores of the Mediterranian Sea that they all have been born, the great religions, the great philosophers, the great writers and an Empire which has made an indelible impact on the history of all civilized peoples !"
Can this be used as a source for instance in a historical (and relevant) article to support a statement like "Mussolini :wasn't very fond of eighter the Nazists or Germany from the beginning of the Third Reich" ? And finally - 3. May a statement (at a relevent location) like "Hitler hated Jews" and use some phrase from "Mein Kampf" as source be OR or wrong in any other sence ? Appriciate serious answers. Boeing720 (talk) 19:11, 6 February 2014 (UTC)
Question 1: As far as I know not. A primary source is someone giving their opinion or eyewitness report. Examples are the speech by Mussolini, and blogs. A secondary source is an analysis of a finding (or a primary source). Most academic papers are secondary analyses of an observation A tertiary source is a source where the body of knowledge is aggregated and made accessible to a larger audience. Wikipedia is a tertiary source, as are many textbooks use for classes.
Re question 2: The actual speech by Mussolini is a primary report. The statement you suppose is an interpretation of that primary speech by Mussolini and would likely count as original research as it is not what Mussolini actually said. The same goes for your Mein Kampf statement. If you find a quote which state "I hate jews" in Mein Kampf it would be a primary source, if you interpret phrases it would probably be original research indeed.
In any case both examples would build heavily on primary sources, which is not to be preferred, especially for a topic where there are literally thousands of secondary and tertiary sources that provide the claims you want to make. So basically you don't need any of this in this example (so why bother). Arnoutf (talk) 22:17, 6 February 2014 (UTC)
Thanx! So if the narrator in the documentary (or an author of an analyzing historical book) uses this speech as an example, then the statement would be OK (the statement by the narrator or author, not the Wikipedia-editor, to be absolutley clear) ? Boeing720 (talk) 03:30, 8 February 2014 (UTC)
The narrator would be a secondary source; and using that is not Original Research. Whether the narrator should be listed a RELIABLE secondary source, is another issue. Arnoutf (talk) 11:07, 8 February 2014 (UTC)
I sooner ment the narrated text. The reliable person(s) must be the editor and/or the (executive) producer, I assume. Boeing720 (talk) 01:35, 9 February 2014 (UTC)
Boeing720, I think you'll want to read WP:Party and person and WP:Identifying and using primary and secondary sources. WhatamIdoing (talk) 16:43, 19 February 2014 (UTC)

Counting -- possible addition to WP:CALC[edit]

Has there been any discussion about the possibility of adding "counting" to WP:CALC as not being "original research" ? Just looking to be more clear. In other words, do I need to have a source that says Tom and Carol Brady have 6 kids combined, or can I just look at Greg, Peter, Bobby, Marcia, Jan, Cindy ... and count up to 6 without sourcing it? I think so.--Paul McDonald (talk) 21:13, 7 February 2014 (UTC)

Please do not add that to WP:CALC. This is simply a question of understanding/reading a particular source and about different ways to summarize it correctly.--Kmhkmh (talk) 14:24, 8 February 2014 (UTC)
I don't think that needs adding – it comes under 'routine calculations' to me, so is already covered, and is only one of many examples of such and we don't need to go adding every variant. If 'routine calculations' seems vague then that's by design: what is routine is determined on a case by case basis by consensus among editors.--JohnBlackburnewordsdeeds 14:55, 8 February 2014 (UTC)
I once asked if calculating population density for an area is OK, based upon area and population. Even adding some areas their population, and calculating the overall population density. This was OK. I also recall that if the article´in question is of mathematical type, a higher degree of math may be alouded as NOR. If this was to any help Boeing720 (talk) 01:50, 9 February 2014 (UTC)

I think it is more one example where (or it is possible to imagine) a new or an isolated wikipedist lost good contribution because some one other wikipedist, use self interpretation of WP:NOR or WP:CALC rules, for delete good contributions at Wikipedia. I started Wikipedia:About Valid Routine Calculations (today an abandoned essay but we can rewrite and restart it) for this cases: an essay is a "consensus reference" for any one, without the support of an "expert", show or check what is wrong with your own text, or what kind of abuse the another wikipedist is doing. --Krauss (talk) 16:20, 9 February 2014 (UTC)

Please keep in mind that even simple counting can be abused. For example, we all agree blueberries are fruits, and so are watermelons. But counting 1 blueberry and 1 watermelon up to two pieces of fruit in the context of daily intake of food is nonsensical. The current vague routine calculations would allow consensus to block such cases while it creates no problem for regular counting. Arnoutf (talk) 16:50, 9 February 2014 (UTC)
Also note that you cannot formulate 100% foolproof guidelines, that is guidelines protecting against any conceivable form of abuse. Any guideline can be abused and any guideline is subject to interpretation (to a degree). In doubt it is up to involved editors to decide in an individual case (usually via consensus) whether a source is used according to guideline (in doubt in its spirit rather than its literal reading if every single sentence) or not. If such such consensus is not possible then there project pages where you can request the assessment of univolved editors/rd opinions.
The longer and more detailed a guideline gets, the lower are the chances of editors actually reading and heeding it.--Kmhkmh (talk) 17:34, 9 February 2014 (UTC)
  • As far as the specific example is concerned:- if I have a source that says Tom and Carol Brady's children are called Greg, Peter, Bobby, Marcia, Jan and Cindy, then it would be reasonable to say "Tom and Carol Brady have six children" and cite it to that source. That's not original research. It's simply summarising the source, which is the encyclopaedist's most basic task.—S Marshall T/C 14:26, 10 February 2014 (UTC)
My point exactly!--Kmhkmh (talk) 14:33, 10 February 2014 (UTC)
I felt the need to put together an essay. Wikipedia:Counting and sorting are not original research. Please take a look and comment/change/modify: it's a community essay of course!--Paul McDonald (talk) 15:58, 13 February 2014 (UTC)
The important point is that counting is not OR unless it is reasonably disputed. You should feel free to simply count up Barack Obamas children. However, there is nothing "simple" about counting up the number of Clint Eastwood's children (for example). WhatamIdoing (talk) 16:46, 19 February 2014 (UTC)

Minor clarification add "primary"[edit]

Any interpretation of primary source material requires a reliable secondary source for that interpretation. A primary source may only be used on Wikipedia to make straightforward, descriptive statements of facts that can be verified by any educated person with access to the primary source but without further, specialized knowledge.
It's straightforward from the examples (i.e. summarizing a book requires access to the book (primary source), interpreting the book requires a secondary source that interprets the book). The policy is referring to the primary source, not a source of special knowledge . --DHeyward (talk) 03:07, 26 February 2014 (UTC)

Interviews clarification[edit]

WP:NOR#defs defines "(in some contexts) interviews" as primary sources. In what contexts would an interview not be a primary source? 88.75.174.246 (talk) 12:58, 28 February 2014 (UTC)

Oh, if you interview the local history professor on the radio and say, "So tomorrow's a big anniversary in history, so why don't you tell our listeners about it?", and he replies, "Sure. Tomorrow is the 400th anniversary of our town being spared destruction during a religious war, because the mayor could really chug wine. Mayor Nusch slept for three days afterwards."
An interview that tells someone's own ideas is primary; one that is basically a talking textbook lesson is not. WhatamIdoing (talk) 16:52, 28 February 2014 (UTC)


To qualify as a secondary source, it must contain secondary analysis. From WP:SECONDARY, "A secondary source provides an author's own thinking based on primary sources, generally at least one step removed from an event. It contains an author's interpretation, analysis, or evaluation of the facts, evidence, concepts, and ideas taken from primary sources." For example, a 60 Minutes investigative interview piece would be secondary to the extent that it mixes their own fact-checking, analysis and other content along with subject's own words.
But a 60 Minutes-style interview with some genuine secondary content is pretty rare. When all the questions are just variations on "tell us more about yourself / what you think / what you've done / what you plan to do / something technical / ..." and the only real content is the subject's own words, that is always WP:PRIMARY as regards the interviewee and would not, e.g., be suitable for establishing notability of that individual. (Fundamentally, you can't make yourself notable just by writing more or talking more. Other people have to talk about you.)
But even if an interview is a primary source as regards the interviewee, it might still be an appropriate secondary source for a different topic. For example, an interview with an author of a how-to book on programming would be primary as regards the author but could be secondary regarding the software topics discussed. Msnicki (talk) 17:10, 10 April 2014 (UTC)

WP:OI[edit]

(At least to me) WP:OI isn't as clear as it could be.

Currently the central sentence reads as follows:

Original images created by a Wikipedian are not considered original research, so long as they do not illustrate or introduce unpublished ideas or arguments, the core reason behind the NOR policy.

To my judgement, that comes more or less down to the tautology:

Content created by a Wikipedian is not considered original research, so long as it is not original research.

As far as I can see, the same rules apply to adding images (or graphics) to an article, that as well apply to adding text to an article: Created content is most welcome (as in self-made photos/graphics or self-written text), but it has to be verifiable by a published source.

How about changing it to something along the following line, and making it the first in the WP:OI paragraph:

Adding self created content to further the coverage is most welcome at Wikipedia. Content like photos or graphics ("Original Images") created by a Wikipedian are not considered original research, so long as the content is verifiable by a published source.

If there are any exceptions with regards to NOR and images, then they should be spelled out more clearly – at the moment I can not see any exceptions, and in my view the same NOR-rules that apply to text apply to images as well.

My mastery of the English language is not the best, and in addition I am not very good in navigating all the WP rules, so I would ask if someone more knowledgeable could think about this, and make some changes to that paragraph if deemed useful.

I see at least one caveat: There are potentially quite a few original images of places used in Wikipedia (e.g. images from small towns or publicly not accessible places) that are possibly not verifiable by a published source at the moment (e.g. no StreetView coverage). Tony Mach (talk) 12:56, 8 March 2014 (UTC)

I agree for graphics; the problem with the many current graphs is that when challenged in relation to them being original research, creators of graphics frequently refuse to provide sources; so in my view some stricter verifiability criteria for graphs would be very welcome. For photos it is a bit trickier. E.g. a photo of the Statue of liberty is not easily verifiable, nevertheless such a photo would be easily "recreatable"; so perhaps a slightly different phrasing there. Arnoutf (talk) 14:08, 8 March 2014 (UTC)
This text would pretty much ban all photographs. It's not just places. What's your published source for saying that this selfie is actually a picture of you? That this thing on a microscope slide is actually the thing that it looks like? That this picture of something that looks like an apple really is an apple?
Our goal with photos isn't to get something that's verifiably "the thing". It's to get something that shows the reader what the thing looks like. We don't want disputes from AIDS denialists about whether we can find a published reliable source that says this exact picture is a picture of HIV. We just want something that looks like what we're claiming it is. WhatamIdoing (talk) 20:07, 20 March 2014 (UTC)

List of American beach volleyball players[edit]

What's your opinion on List of American beach volleyball players... I know very little about beach volleyball, so who decides that this is a list of well known players? Says who? Should it be sent to AfD? Thanks, JMHamo (talk) 16:15, 17 March 2014 (UTC)

Looks fine as a typical list. While I didn't look at all the entries, it appears that they are all notable, American, beach volleyball players. The sourcing is in the respective articles on the individuals. Any entry that doesn't meet the inclusion criteria should be removed.
However, there's a category that it is duplicating. I'm not sure what the consensus is on categories vs lists covering the same topic. --Ronz (talk) 17:06, 17 March 2014 (UTC)

Explaining rejections?[edit]

WP policies like WP:OR are intended to facilitate the presentation of good content and reject mere speculation, or OR. A militant rejection of 'crank' content is necessary because there are many attempting to promulgate their own views, which is unacceptable unless these views also are those of reputable sources.

It is my observation that WP rejections are tending more and more toward overly brief in-line editor comments and a resistance to more complete explanations. I wonder if there is a place in WP policy for a recommended procedure to be followed; a sequence of steps that could direct discussion toward sources and prevent descent into put-downs and arbitration? For example, in the case of rejection of a submission it could be recommended that:

  1. Rejection should be polite;
  2. Rejection should cite the policy that has led to the rejection and supply some indication of what is in violation;
  3. Rejections with an in-line edit should assume that further discussion could go to the talk page, and suggest that possibility;
  4. If a talk-page discussion takes place, the rejecting editor should identify specifically what caused the rejection, and exactly how it contradicts the cited policy; not just blanket-reference the entire contribution, and admonish the contributor to "read the policy".

Possibly these steps are too obvious to require elaboration in WP policies, or perhaps it is felt that these steps are already present in existing policy? Often in-line edit statements strike me as peremptory, and when a contributor requests more information a confrontational atmosphere shows up right at the beginning. Brews ohare (talk) 19:51, 17 March 2014 (UTC)

This may conflict with WP:BOLD, possibly even WP:DR.
It would be nice to see more emphasis given to focusing on content and focusing on the sources (which should be detailed somewhere in our policies/guidelines. --Ronz (talk) 20:51, 17 March 2014 (UTC)
I think that your post covers about 6 much-needed different areas of good behavior, and I don't see any that concern the specifics of this policy; I think that they actually are about a missing policy. North8000 (talk) 21:55, 17 March 2014 (UTC)
PS when you noted that this policy exists for a reason, I thought you were going there (i.e. to note the reason when deleting and not just the policy) but I don't think that you did. Sincer3ely, North8000 (talk) 21:57, 17 March 2014 (UTC)
When an editor has attempted to insert material over multiple articles and not gained consensus for any of them the patience of others wears thin. When policy has been explained to that editor many times but they simply don't listen it wears thinner. When they go to policy discussion groups (this is the third time) to try and win the argument another way it finally goes. Any review of Brew's history on Philosophy articles (and his permanent ban from editing all physics articles) will show the pattern of behaviour. Put simply if you have explained policy and content issues once, but an editor refuses to accept that then short edit summaries are reasonable. When the editor calls an RfC and gets no support but carries on anyway the edit summaries just get shorter and shorter ----Snowded TALK 06:09, 18 March 2014 (UTC)

Snowded doesn't like the process of content development, which is too demanding of his time and intellect, requiring reading and assimilating sources and actually formulating summaries of them. So he has taken the route of claiming my attempts to fix the uncountable gaffes in philosophy articles as a kind of plague rather than a necessary clean-up operation. Brews ohare (talk) 16:11, 18 March 2014 (UTC)

This is a WP:CANVAS violation, and inappropriate use of this forum.
There are some good points in the initial comment [1], but they have nothing to do with WP:OR. --Ronz (talk) 16:26, 18 March 2014 (UTC)
Ronz: I agree about the misuse of this forum to engage in extraneous debate. To get back to work, the proposal is indeed more general than WP:OR although WP:OR is a policy very often invoked in rejections. Can you suggest a better venue for this discussion? Brews ohare (talk) 16:39, 18 March 2014 (UTC).
I already pointed out WP:BOLD and WP:DR, especially noting WP:FOC, which is a part of WP:DR. WP:FOC begins with a link to WP:EP. --Ronz (talk) 18:14, 18 March 2014 (UTC)
Ronz: Thanks for these links. I wonder if you agree with me that these links basically cover the territory, although perhaps an itemized list would add some emphasis? Unfortunately, these policies are not followed, so one has to wonder if it is clarity that is the issue. Brews ohare (talk) 18:24, 18 March 2014 (UTC)
I suggest you choose one of the talk pages to the policies/guidelines listed, and discuss the matters there. --Ronz (talk) 20:32, 18 March 2014 (UTC)
Ronz: I wasn't aware of all the attention to these issues already in the links you provided. I am going to drop the subject, as I'm now persuaded that policy is not the problem on WP; it is the failure to follow policy. Brews ohare (talk) 16:48, 19 March 2014 (UTC)

What do cautions about primary sources mean?[edit]

The policy WP:OR states:

"The prohibition against OR means that all material added to articles must be attributable to a reliable published source, even if not actually attributed."

Later, regarding sources,

"The only way you can show your edit is not original research is to cite a reliable published source that contains the same material"
"In general, the most reliable sources are:
  • peer-reviewed journals
  • books published by university presses
  • university-level textbooks
  • magazines, journals, and books published by respected publishing houses
  • mainstream newspapers"
"Policy: Unless restricted by another policy, primary sources that have been reliably published may be used in Wikipedia; but only with care, because it is easy to misuse them."

Clarification is requested here about the meaning of "used with care" and "misuse". In a conversation with editor Snowded,12 I suggested this phrase "with care" is a caution that it is easy for a WP editor to slip into synthesis not found in the sources, or into statement of the WP editor's personal interpretation. However, any rejection of a presentation of this material is not to be based upon the controversial nature of the subject, but upon any misrepresentation of the sources, for example, by failure to note there is controversy surrounding the subject, or by ignoring some aspects of the subject that should come up."

In short, material that is controversial or rapidly changing is not excluded as OR simply on that basis, so long as accurate presentation of what is contained in reliable published sources is adhered to, and care is taken to present all sides of any controversy.

Does that seem to be what the policy requires? Brews ohare (talk) 22:07, 19 March 2014 (UTC)

In the policy on primary sources the sentence after the one you quote is this: "Any interpretation of primary source material requires a reliable secondary source for that interpretation."
So using primary sources, in particular devising an 'accurate presentation' of them, requires a reliable secondary source or sources. Without a secondary source your presentation might be quite different from another editor's selection and presentation. So any presentation should be based on secondary sources.--JohnBlackburnewordsdeeds 00:02, 20 March 2014 (UTC)
That isn't how I read it Blackburne - I'd write this line as follows:
"Any interpretation of primary source material requires a reliable secondary source for that interpretation."
The 'interpretation' that cannot be made is a WP editor's personal interpretation of what the primary source said, in contrast to what the source itself has to say. That is, I'd take it that only a secondary source is allowed to interpret what a primary source said.
Thanks for bringing up this point for further clarification. Brews ohare (talk) 05:19, 20 March 2014 (UTC)
Selection is interpretation ----Snowded TALK 05:26, 20 March 2014 (UTC)
I'd understand that simply reporting what a primary source says is not 'interpretation' unless it reports the source inaccurately. Of course, an improper selection, like taking a line out of context, can be removed on that basis. Brews ohare (talk) 05:37, 20 March 2014 (UTC)
Brews, if you take (as you have many times) strings of quotes with some minor commentary between them then, aside from the stylistic issues, you have determined which quotes and which authors. Recently you suggested that half a loaf was better than one and that rather deleting your quotes more material should have been added. The role of an encyclopaedia is not as a recording device for notes for people studying the subject, it is to fairly and accurately summarise the field ----Snowded TALK 06:40, 20 March 2014 (UTC)

There is another way of looking at it. Just quoting primary sources is unencyclopaedic. I.e. an article that consists just of a quote, or a list of quotes, is not really an article. Any section similarly consisting just of quotes is not encyclopaedic. Adding text giving the source, e.g. "this is a quote from A" doesn't change that and is redundant as that's what references are for – readers can follow the links to them so there's no need to clutter the article with such attributions. Anything else, such as saying "this source is important", "this source is controversial", "this source is contradicted by this other source" is commentary, and as such must be based on reliable secondary sources or it's original research.

So using just primary sources is either unencyclopaedic or original research. Articles should be based on secondary sources, and once these are used and the article is properly written based on them then there are relatively few uses for primary sources. Quotes when the quote is being discussed. Examples of language when the use of language is being discussed. Plot/chapter summaries for creative works that are the main subject of the article. But actual encyclopaedic content should be based on secondary sources.--JohnBlackburnewordsdeeds 12:30, 20 March 2014 (UTC)

Selection of sources, and deciding which information from sources to include in an article, is not original research, it is source-based research. Source-based research is the method used to write Wikipedia; without it, Wikipedia cannot exist.
The full sentence about misinterpreting primary sources is "All interpretive claims, analyses, or synthetic claims about primary sources must be referenced to a secondary source, rather than to an original analysis of the primary-source material by Wikipedia editors." From this we see that the caution against interpreting primary sources does not apply to the ordinary process of reading: scanning with the words, understanding the meaning of the words with the brain, deciding what is relevant, and restating the meaning in the editor's own words. The caution against interpretation applies to more advanced forms of interpretation, which is described as "original analysis", that is, reaching a conclusion by combining material from various parts of the source, or by comparison of material in the source with material in other sources. Jc3s5h (talk) 13:42, 20 March 2014 (UTC)
The issues raised by Snowded and Blackburne about the use of quotations are a side track: the use of quotes is not the subject here. Brews ohare (talk) 14:54, 20 March 2014 (UTC)
I disagree with you Brews. Quoting is in practice about the only way to fairly use a primary source (as anything else would be interpretation). But providing a long list of quotes is distinctly unencyclopedic; so the quoting discussion is fairly central to the use of primary sources. Arnoutf (talk) 19:15, 20 March 2014 (UTC)
There is no difference in how to properly use a secondary source or primary source. It's just that since primary sources are more apt to be one-sided or promotional, a greater effort must be made to maintain a neutral point of view in the article. The adoption of Arnoutf's understanding would destroy Wikipedia. Jc3s5h (talk) 19:42, 20 March 2014 (UTC)
That's not the policy. As you state it there is almost no difference between them, as secondary sources can be biased and only give one side of an argument. But the policy is clear: "Do not base an entire article on primary sources, and be cautious about basing large passages on them." (emphasis as given). Primary sources should not be used the same way as secondary sources.--JohnBlackburnewordsdeeds 19:51, 20 March 2014 (UTC)
(ec)Well it hasn't destroyed it yet, we normally expect secondary sourcing to validate what is chosen or used as material. Primary sources, with care, can be used to expand that. There is also a major difference, a secondary source is peer reviewed, a wikipedia editor even being careful is not ----Snowded TALK 19:54, 20 March 2014 (UTC)
For purposes of this policy, the no original research policy, there is no difference in how to properly use an individual primary or secondary source. For purposes of the notability policy, there certainly is; that is the most direct reason for not basing an entire article on primary sources. The NPOV policy is apt to make it inadvisable to base long passages on primary sources. But there is no difference in how to properly paraphrase or summarize a source, whether it is primary or secondary.
The statement that secondary sources are peer-reviewed is simply not true as a general statement; some of them are, and some aren't. Likewise, some primary sources are peer-reviewed, and some aren't. Jc3s5h (talk) 20:10, 20 March 2014 (UTC)
Having been through the process from both sides and in both cases there is a very big difference between a peer review for a secondary source and a primary one and it relates to purpose. The purpose of a secondary source is to summarise a field (in the main) which is why we use them. A primary source generally advocates a position within that field. ----Snowded TALK 05:18, 21 March 2014 (UTC)
Quick plug: WP:Secondary does not mean independent and WP:USEPRIMARY cover a lot of important background information on figuring out what a primary source is. (And Jc3s5h understates the case; most peer-reviewed articles are not secondary sources (which are mostly review articles—the names are confusingly similar). WhatamIdoing (talk) 20:20, 20 March 2014 (UTC)
Arnoutf's claim here is wrong. It's also not representative of Arnoutf's practice, because here we see him adding a primary source to an article, without quoting anything. That's not cherry-picked; that's just the last time he edited any article.
We have two main problems with primary sources. One is the person who uses a perfectly decent primary source as a jumping-off point for his "special" theory (e.g., "this source says that pouring oil directly on cancer cells will kill them, so I wrote that petroleum products cure cancer patients"). That's pure OR, and we really want to discourage that. The other is the person who uses a biased/unrepresentative primary source. That's actually a WP:DUE problem, but it's a DUE problem that is far more likely to occur with primary sources than with secondary sources. We really want to discourage that, too.
The net result of these two separate problems is that we really, really want to discourage people from using primary sources, and that's why this section has been phrased so strongly. There have been discussions in the past about splitting this off into its own policy. If it were on a separate page, the fact that "primary = caution" might make more sense. The need for caution is not OR-specific. WhatamIdoing (talk) 20:20, 20 March 2014 (UTC)
Two things to this: (1) I was quoting an article in a quality newspaper and not the press release by NS (which would be the primary source here) (2) I never stated quoting was the only way, but it often is the only way to use a primary source when there is potentially disputed information. For undisputed facts there is no difference between type of source (and I have not yet encountered a single editor complaining). I may have phrased my earlier comment somewhat strongly on that account though. Arnoutf (talk) 18:12, 21 March 2014 (UTC)

Regarding editor discretion in selecting primary sources, what really happens (and a common unspoken practice in Wikipedia) is that where there is no objection, it is editor discretion. If I'm writing about a ship and 2 sources give it's length at 827' and the other gives it as 8,270' (a ridiculous number) I'm going to pick one of the two "827'" sources, and nobody is going to drag out the rule saying that that editor discretion was improper. If I start trying to POV an article by saying "polls say" and then inserting and summarizing my cherry picked poll results, someone is going to pull out the primary source rule and remove my material, and rightly so. North8000 (talk) 21:53, 20 March 2014 (UTC)

Editor discretion in using primary sources is limited by WP:OR in a number of ways. The preamble in this policy is:
"Research that consists of collecting and organizing material from existing sources within the provisions of this and other content policies is fundamental to writing an encyclopedia. Best practice is to research the most reliable sources on the topic and summarize what they say in your own words, with each statement in the article attributable to a source that makes that statement explicitly. Source material should be carefully summarized or rephrased without changing its meaning or implication. Take care not to go beyond what is expressed in the sources, or to use them in ways inconsistent with the intention of the source, such as using material out of context. In short, stick to the sources."
As pointed out by Jc3s5h policy says:
"All interpretive claims, analyses, or synthetic claims about primary sources must be referenced to a secondary source, rather than to an original analysis of the primary-source material by Wikipedia editors." (Underline added)
The policy apparently supports the use of primary sources as long as the WP editor doesn't tinker with the gist of the source or add the WP editor's original input. That seems to be contrary to the view that the policy takes the stand that "we really, really want to discourage people from using primary sources" or that "Quoting is in practice about the only way to fairly use a primary source (as anything else would be interpretation)". As Jc3s5h has put it, paraphrasing the policy regarding source-based research:
"the caution against interpreting primary sources does not apply to the ordinary process of reading: scanning with the words, understanding the meaning of the words with the brain, deciding what is relevant, and restating the meaning in the editor's own words."
He adds the remark, which seems quite correct to me, "Selection of sources, and deciding which information from sources to include in an article, is not original research, it is source-based research. Source-based research is the method used to write Wikipedia; without it, Wikipedia cannot exist."
A good deal of the resistance to this policy stems from a reluctance to get involved in the work of writing WP. Clearly, if a contribution is a WP editor's best effort to summarize the sources they are aware of, a criticism of the contribution will engage the critic in discussion of the way sources are presented, and possibly in the introduction of other sources known to the critic. That is going to be a talk-page engagement that editors trained in the brief "one-line edit comment" reversion of crank contributions are unprepared for. A much briefer engagement is feasible if the critic can simply say that nothing but (preferably verbatim) descriptions of primary sources taken from secondary sources are permissible. Brews ohare (talk) 14:04, 21 March 2014 (UTC)
However, as has been observed, WP articles are not intended to be a brief version of those summaries already to be found in secondary sources. A restriction like that would chop most WP articles down to a superfluous and largely out-of-date treatment of most of its topics. Brews ohare (talk) 16:06, 21 March 2014 (UTC)
The Wikipedia system is fuzzier than most imagine, including the rules and definitions of terms. Sometimes that's a good thing, sometimes that's a bad thing. In this case I think that the rules and definitions are a bit fuzzy but that they mostly work reasonable well. North8000 (talk) 16:27, 21 March 2014 (UTC)
"a reluctance to get involved in the work of writing WP" please stop imputing motives to other editors simply because they oppose you writing personal essays Brews. It doesn't help ----Snowded TALK 17:33, 21 March 2014 (UTC)
Following the above discussion, I don't see much fuzziness in the portion of this policy under discussion. That is why I wonder about interpretations that conflict with its wording, and straying away from the topic at hand instead of staying on point. I don't see these divergences as a concerted effort, but more as an instinctive reluctance against becoming involved in talk-page development of content that can become extended and tedious, however necessary it is. Brews ohare (talk) 22:44, 21 March 2014 (UTC)

Neutral notification of ongoing discussion at Talk:Bradbury Building[edit]

In lieu of starting a full-blown RfC I would like to invite interested editors to comment on an ongoing discussion between myself and Beyond My Ken at Talk:Bradbury_Building#Popcult_removals_by_alf_laylah_wa_laylah. It's been open for a while and no one else has weighed in. It involves issues related to this policy, so I am dropping this note here.— alf laylah wa laylah (talk) 03:40, 9 April 2014 (UTC)

A further exploration of WP:OR policy[edit]

In an earlier thread What do cautions about primary sources mean?, some discussion of a particular form of article structure arose, namely, an article consisting of a narrative supported by citations or quotations. A question needing further examination is whether such an article is OR simply by the very nature of its structure .

Now if a WP editor actually makes some extrapolation or interpretation not actually contained in the cited sources, but the editor's own original contribution, everybody agrees WP rules that out as WP:OR. There is no quarrel about that.

But by virtue of the very structure of such an 'essay' article, some selection of sources has been made, and only a selection of the ideas in these sources have been presented. Thus, by its very structure the presentation involves an assessment by the WP editor, and the issue is whether it is therefore a violation of WP:OR because no WP editor can be allowed to exercise such an independent judgment. Some editors use this argument on Talk pages to say there is no point or purpose in discussing contributions of this kind. That is, because they are OR on the face of it, discussion of any such contribution is a waste of time.

So the issue comes down to whether it is placing too much discretion in the hands of the WP editor to let them (i) select which published works like monographs, anthologies or even journal papers, will be reported upon, and (ii) select which ideas within these works will be reported upon? Do existing policies like WP:Undue and WP:RS provide sufficient safeguards without refusing any article with this structure?

In considering this issue, one might ask two questions (i) Is this strict interpretation of WP:OR the way WP articles are in fact constructed? and (ii) If this standard is applied to existing WP articles, how does that limit what WP can do? Must WP contain only condensed summaries of already published reviews?

Without a clarification of WP policy in this regard, every time a fight arises it will have to be settled by the parties themselves or be taken for adjudication. Of course, adjudication of a what is seen as a content dispute will not happen, and so adjudication will end the dispute by banning one or all parties in one way or another. A clarification of WP:OR is preferable.

A basis for clarification could be the observation by Jc3s5h in the earlier thread. In part the comment is:

"Selection of sources, and deciding which information from sources to include in an article, is not original research, it is source-based research. Source-based research is the method used to write Wikipedia; without it, Wikipedia cannot exist."

Can some such modification to WP:OR be made so it cannot be used to justify deletion of useful material, and to avoid even the discussion of such deletion? Brews ohare (talk) 16:41, 12 April 2014 (UTC)

Comments[edit]

  • No, not required as this allow for inclusion of inappropriate material that would fail OR or other policies. For example, the inclusion of a minor/insignificant viewpoint connected to the topic that is only covered by a single but otherwise reliable source that would affect the neutral POV of the article or give too much undue weight to that view point; such a inclusion should be removed and further discussion needed to discuss inclusion. (arguable, this was a point of discussion about six months ago here and the answer was still no there). --MASEM (t) 16:50, 12 April 2014 (UTC)
→ Masem: Is your comment that articles of the described structure are not allowed on WP, or is your objection to the wording by Jc3s5h? This wording is intended as a basis for further discussion, not as the exact wording of a modification. Of course, policies like WP:Undue, WP:NPOV, and WP:RS would be observed in a permitted article. Brews ohare (talk) 16:58, 12 April 2014 (UTC)
What Jc3s5h is correct. Your jump of logic from that statement to "this means as long as its sourced it must be kept" is not. --MASEM (t) 17:06, 12 April 2014 (UTC)
No inference that anything sourced should be kept is meant. Any acceptable wording would have to avoid the possibility of that interpretation. But I'm sure that articles structured as described can be limited to avoid endorsing such excesses, aren't you? Quite likely, existing policies like WP:Undue, WP:NPOV, already insure that error won't arise? Brews ohare (talk) 17:21, 12 April 2014 (UTC)
What you are arguing and have argued before is that you feel you should be justified to be able to add material from a reliable source as such to avoid the claim that it is OR and prevent that from being removed even with discussion just because it meets OR. That would prevent other policies like NPOV from being used to remove it. Additionally, there can be additions that would be violations of OR (along the lines of SYNTH) even if they come from reliable sources. You're free to add material to an article but it would be completely improper to say that it cannot be removed if someone feels it violates policy or negatively impacts the article. --MASEM (t) 17:33, 12 April 2014 (UTC)
→ Masem: You seem to want to make this discussion of WP policy a discussion about me. It isn't.
→ The part of your comment that is pertinent to a discussion about policy is your statement that WP:OR should not supersede other policies like WP:NPOV. That is absolutely the case, and such a view is not suggested. Likewise, WP:SYNTH is completely applicable. There is no intention that material should be forced to be retained if it satisfies WP:OR but violates other policies.
→ What is proposed for discussion, is that it should be impossible to use an appeal to WP:OR to deny an article of a specific structure simply because of that particular structure and regardless of its content . That does not suggest any other policies should be ignored. Brews ohare (talk) 18:03, 12 April 2014 (UTC)
  • The underlying issue here is the degree of confidence WP has in the process of contributions by non-experts. Those with little confidence suggest that articles can contain only reports upon sources that come from reviews that in turn report about those sources. Those with more confidence suggest that non-experts can report upon the contents of all kinds of reliable sources so long as they observe certain restraints like WP:NPOV, WP:Undue and WP:RS. This confidence rests in part upon the idea that the deficiencies of non-expert contributions will be corrected in due time by virtue of the many contributors involved, each bringing their particular knowledge and correcting and expanding articles.
How much confidence in the process of correction and expansion involving many non-expert editors do we have? Does it extend to allowing articles structured as a presentation of sourced views so long as the presentation sticks to what the sources say and observes WP policies and guidelines like WP:NPOV, WP:Undue and WP:RS? Or do we have to deny anything that is not a condensed version of an elsewhere published summary account? Brews ohare (talk) 18:25, 12 April 2014 (UTC)

This is yet another reflection on a core point. By a rigorous interpretation of wp:nor, about 80% of Wikipedia violates it. Most of the "80%" "slips through" by two methods:

  1. Most of it is such that it takes a really rigorous interpretation of wp:nor to consider it a violation, and for various reasons that usually doesn't happen.
  2. The rest of it gets by because nobody challenges it. Challenges happen for good reasons (= use of the policy for its intended purpose) and bad reasons (=is-use of the policy, not for bad purposes which are its intended purposes)

The system works better than my post implies, but where it breaks down is in the important minority of cases where wp:nor is being mis-used. I think that some more work to do a little more delineation between extraction/summarization and WP:OR would be a worthy quest. North8000 (talk) 22:55, 12 April 2014 (UTC)

North8000: You may be accurate in saying 80% of WP violates WP:OR when strictly interpreted, which I take as the interpretation disallowing the presentation of sourced views even if the presentation sticks to what the sources say and observes WP policies and guidelines like WP:NPOV, WP:Undue and WP:RS. If that is so, it would seem WP in reality functions as though WP:OR allows the presentation of sourced views so long as the presentation sticks to what the sources say and observes WP policies and guidelines like WP:NPOV, WP:Undue and WP:RS. It would then seem likely that WP:OR should reflect reality and be worded unambiguously as supporting this activity that reflects what actually is going on. However, it seems you might adopt the view that even if WP:OR as it is now does generate heated and unacceptable results on occasion, things aren't so bad: the de facto system works and the problems with the de jure system are infrequent, so let this Rube Goldberg arrangement continue. Let lie the sleeping dogs of policy review, or something like that. What is your stance? Brews ohare (talk) 16:11, 13 April 2014 (UTC)
My view is that of Jc3s5h, which is that WP:OR actually is in support of the presentation of sourced views subject to various policies like WP:Undue and so forth, but unfortunately some editors don't think so. Consequently, the policy wording should make its intent clearer. Brews ohare (talk) 13:28, 14 April 2014 (UTC)