Wikipedia talk:No original research/Archive 59

This is an archive of past discussions about Wikipedia:No original research. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 55

←

Archive 57

→

Proposed modification of WP:OR regarding figures

Should this addition be made to Wikipedia:No_original_research#Original_images?

"In the case of figures intended to illustrate text, even if nothing like a proposed WP figure exists anywhere else at all, it is acceptable on WP if it faithfully depicts text that is not OR. Whether a figure of this type constitutes original research is contingent only upon its accuracy in representing the accepted text it illustrates."

I understand the present policy to be consistent with this proposal. This addition would rule out explicitly a view such as "A diagram is not exempt from normal sourcing rules". Brews ohare (talk) 15:22, 6 August 2013 (UTC)

Comments.

I believe there are a great many figures on WP that are completely original illustrations of accompanying text, and some of which are unrelated to any published figure. Nonetheless, there are misconceptions regarding WP policy in such matters, so an amendment is needed. The issue is not gray, but black and white - either a diagram on WP must approximate one in print (inviting copyright restrictions), or that is not necessary, and felicity to text is a sufficient criterion. Present policy says the latter, and suggesting policy is consistent with personal judgment about this requires a rewrite of policy. Brews ohare (talk) 15:37, 6 August 2013 (UTC)
I oppose this change. Editors should use their best judgment in every case, which may include a very strict interpretation of verifiability in contentious articles. There are misconceptions in both directions here. WhatamIdoing (talk) 15:52, 6 August 2013 (UTC)
As noted above this is a content dispute. You should not be trying to rewrite policy to help you win a content dispute; you've tried this many times before so should know this by now. Resolve the issue on the talk page, recruiting other editors perhaps on a project talk page if there are only two involved.--JohnBlackburne^words_deeds 16:03, 6 August 2013 (UTC)

Blackburne: You are wrong - you mistake a general issue with the instance leading to it. Brews ohare (talk) 16:08, 6 August 2013 (UTC)

If this really has nothing to do with the dispute, then feel free to come back here a couple of months after that dispute has been resolved. WhatamIdoing (talk) 20:23, 6 August 2013 (UTC)

Why on Earth should a general issue have to wait for a minor dispute between two editors to end? Your reason presumably is because you cannot see that this is a general issue, even when it is stated in general terms that depend in no way upon this other matter. Brews ohare (talk) 21:42, 6 August 2013 (UTC)

Because it is very difficult for editors to completely separate a proposed change from a particular instance, and it is impossible to avoid some people believing that it is an attempt at WP:GAMING. Additionally, the resolution of the particular, concrete dispute can provide some insight into how the community sees the matter in practice. WhatamIdoing (talk) 03:16, 16 August 2013 (UTC)

Oppose. This issue can be dealt with on a case by case basis according to appropriate policy and guidelines that already exists. --- Steve Quinn (talk) 21:05, 6 August 2013 (UTC)

Steve: The appropriate policy is WP:OR#Original images. It encourages Editors to upload their own original figures. Now, this policy may be clear to some, but to others, any original figure is OR or SYN. Why do you not favor ending any perceived ambiguity by making policy explicit that an original figure illustrating some WP text is required to be an accurate depiction of the text, and the policy of original research applies to the text not to the figure illustrating it? Brews ohare (talk) 14:20, 7 August 2013 (UTC)

~~No way~~ I just posted at RSN where "is a map a reliable source?" was asked. The answer is no. For everyday "the sky is blue" questions, a map or diagram are fine as sources. For anything remotely contentious, a secondary source is required to interpret the document. Johnuniq (talk) 02:17, 7 August 2013 (UTC)

Johnuniq: I wonder how this process you envision takes place. Here is a possible way to do this that you might comment upon:, An explanation is made in text, in words, with sources, and this text is critiqued to be sure it doesn't exhibit OR or synthesis. Then a figure illuminating this text is created, and this figure is then criticized as to whether it accurately portrays the text. Thus OR and SYN apply to the text, but fidelity to the text applies to the figure. I am sure this process works; do you have an alternative that might work too? Brews ohare (talk) 03:00, 7 August 2013 (UTC)

Oh, sorry, but coming here straight after the RSN section I linked above, my mind was directed towards a different issue from what you have raised. Sometimes an editor will say that a diagram they found somewhere verifies fact X (the diagram is given as a reference for X). I now see that the issue under discussion concerns when an editor-created diagram is original research. That's much more tricky. In general, everyday matters which do not raise objections are not original research and are good ("the sky is blue", or the diagrams at parabola). I do not think there is anything useful that can be said as a generalization regarding when a diagram is a faithful depiction of verifiable facts—it's yet another issue which would need to be discussed, if objections are raised. I have not looked at the background that you linked above, but I have struck my "no way" as that was a misunderstanding, and I might just sit on the fence, although from what I've just said I don't think any general text would really be helpful. Johnuniq (talk) 06:28, 7 August 2013 (UTC)

Johnuniq: Thanks for a thought-out response. I feel that the application of OR and SYN to text is documented. On the other hand applying these policies to an original figure as allowed by WP:OR#Original images is nigh impossible, and is just a quagmire for unending dispute. You have said the same. My view is that for that subset of figures used to illustrate text, an approach based upon the figure's accuracy in depicting the text is a more straightforward way to proceed. Brews ohare (talk) 13:19, 7 August 2013 (UTC)

Oppose. I think insertion of the proposed text will be read to imply that original images that are not based on text, but which can be verified by reference to reliable sources in some non-text media, would be disallowed. For example, there may not be any pictures of a living person who is the subject of a biography that have copyright licenses compatible with Wikipedia, but there are many such photos in reliable sources that can be viewed for verification. A Commons editor takes a photo of the article subject and puts it in the article. That is entirely compatible with existing policy, but the proposal would imply that it is not acceptable. Jc3s5h (talk) 15:11, 7 August 2013 (UTC)

Jc3s5h: Thanks for the observation. You point out that a figure sometimes can be authenticated by comparison with published figures, for example an original photograph claiming to be that of a pigeon can be compared with published photos to ascertain that it is indeed a picture of a pigeon. The proposal doesn't intend to eliminate this kind of verification. But in cases where there is a completely original figure involved, and a similar figure is not available in print, an approach to verification is to compare the figure with the text it describes. This proposal is an alternative, not an exclusive, means for critiquing figures, and sometimes the only way to do it. Perhaps you could suggest some rephrasing of the proposal to make this point obvious? Brews ohare (talk) 15:32, 7 August 2013 (UTC)

Leading section

I think that it is more clear if in the leading section the two words must exist are in Italic, in stead of only the word must right now. Sander.v.Ginkel (talk) 07:45, 15 August 2013 (UTC)

It would be even clearer if we said must have been published and still be accessible to at least some members of the public. I've forgotten the details, but at some point we actually had a problem with someone claiming that his source "existed" (original copies of handwritten letters by his grandmother or something) and was therefore useable, even though it had never been published anywhere. WhatamIdoing (talk) 03:19, 16 August 2013 (UTC)

Revised proposal regarding OR and figures

An earlier proposal met with difficulties, so a revised version is presented below intended to meet those problems by carefully restricting the type of figure to which it applies.

Should this addition be made to Wikipedia:No_original_research#Original_images?

"In the very specific case of a figure in the form of a flowchart or graph illustrating a specific text in an article, such a figure is acceptable on WP provided it faithfully depicts the associated text, and that text itself meets WP policy on OR. Whether a figure of this type constitutes original research is contingent only upon its accuracy in representing the accepted text it illustrates. Of course, meeting OR is not the only possible criterion to apply in accepting a figure." Brews ohare (talk) 15:31, 15 August 2013 (UTC)

As things stand at the moment, objections have been raised that figures that are a combination of published figures, or are variations of published figures are declared to be original research simply because of original elements or organization, even though these diagrams faithfully depict text established to be acceptable. Argument over these matters is directed by mistake at the originality of the figures themselves instead of their accuracy in depicting the text they illustrate. Brews ohare (talk) 22:44, 15 August 2013 (UTC)

Comments

There are a great many figures on WP that are completely original illustrations of accompanying text, and some of which are unrelated to any published figure. Nonetheless, there are misconceptions regarding WP policy in such matters, so an amendment is needed. The amendment makes clear that the OR issue is not gray, but black and white - either a diagram on WP must approximate one in print (inviting copyright restrictions), or that is not necessary, and felicity to text is a sufficient criterion. Present policy is sometimes interpreted as saying OR policy on such figures is open to personal judgment, but that is is too vague. Policy is governed by felicity to text, and not upon extraneous considerations. Brews ohare (talk) 15:31, 15 August 2013 (UTC)
Oppose. Will be interpreted to require removal of perfectly acceptable, uncontentious figures. Jc3s5h (talk) 17:06, 15 August 2013 (UTC)

Jc3s5h: Would you please elaborate upon this scenario? Brews ohare (talk) 15:26, 17 August 2013 (UTC)

Oppose. The present policy is clear and fine. Whether any particular image matches the policy can be decided on a case by case basis, and disputes can be resolved the usual way. Frankly it comes up very rarely: images are far more likely to be removed for relevance or copyright/fair use reasons, or as better ones are found.--JohnBlackburne^words<sub style="margin-left:-2.0ex;"deeds 17:45, 15 August 2013 (UTC)

Brews ohare, Could you explain how the following excerpt from policy doesn't already cover what you are trying to add?

"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."

Thanks. --Bob K31416 (talk) 14:12, 25 August 2013 (UTC)

Bob: Thank you for the opportunity to explain the need for this change. As I said myself, present policy is consistent with this proposal. However, as is abundantly clear from others' remarks here, there is a faction that thinks present policy leaves the matter open to 'interpretation' as to when OR or SYN has occurred in a figure, and that judgment has nothing to do with the text the figure illustrates. In my opinion, and in yours, there is no such thing as OR and SYN in a figure if it sticks to faithfully illustrating acceptable text. If the amplification of policy suggested here is included, that erroneous attitude would be eliminated, leaving one less breeding ground for the rancor it generates, and cases ArbCom must deal with. Brews ohare (talk) 15:41, 25 August 2013 (UTC)

Please consider the possibility that the problem isn't with the way policy is currently worded when it comes to a particular content dispute. --Bob K31416 (talk) 17:10, 25 August 2013 (UTC)

Bob: Thanks for another opportunity to elaborate the reasons for this proposed clarification. The origin of my objection, as you point out, has nothing to do with the possible value of this clarification. If the policy clarification would plead only a single special case, as Blackburne suggests, or introduce collateral damage that was unacceptable as Blueboar suggests, then no change should be made. However, the scenarios of Blueboar where unintended consequences could occur actually are not possible under the clarification, and the argument of Blackburne that the only case of its sort ever to arise is the one subject to this RfC is improbable in view of the opinion expressed by several here that applying NOR and SYN to figures is just fine even when they accurately reflect acceptable WP text. So in answer to your suggestion, yes, the problem is indeed with the way policy is currently worded. Brews ohare (talk) 23:32, 25 August 2013 (UTC)

Again, if as you assert this is a widespread problem please provide links to the disputes. You may think it "improbable" that it's only come up once but as the editor wanting to change policy it is for you to provide evidence that there is a problem that needs addressing. If it is a serious problem it should be easy to find examples.--JohnBlackburne^words_deeds 15:45, 26 August 2013 (UTC)

Blackburne: I think clear arguments are presented that a minor clarification would make WP:OI more useful. Neither your objections nor Blueboar's discredit the proposal. Engaging in a dispute over whether problems are 'widespread' and just what criteria specify a 'real' example adds nothing to the discussion. Brews ohare (talk) 16:41, 26 August 2013 (UTC)

COMMENT - I think our current wording works best... specifically because it is flexible and doesn't try to spell out in detail what is and is not acceptable. In broad terms, I appreciate what Brews is attempting to say. If a user created chart/graph/data presenting image sticks to simply illustrating the text of the article in graphic form, it should be OK (as long as the text itself is well sourced and not OR). However, it is very easy to unintentionally slide over the line into OR (especially SYNTH) when presenting data in pictorial form... even when we try to "simply illustrate the text".

This potential for OR makes it very difficult to write definitive policy statements about user created data presenting images... So much depends on examining the specific data presented in the image. For example, 99% of the information presented in an image might be absolutely fine, with only one small bit of data slipping into OR. Take that one bit out, and the rest of the image is acceptable.

And because so much depends on the specifics of the individual image, I don't think we can (or should) say more than we currently do. Each image has to be judged on a case by case basis. As long as we focus on the principles that underlie WP:NOR when making those judgements, we generally get it right... attempting to create a more narrow definition of what is and is not acceptable, and in what situations, will just create more problems than it resolves. Blueboar (talk) 15:47, 25 August 2013 (UTC)

Blueboar: You express unease that a figure may 'slide over the line' into SYN or OR and WP needs a sufficiently 'flexible' policy to allow handling such cases. You example a figure 99% right and only 1% wrong, and thus slipping through the cracks. Does the suggested policy change facilitate such problems? No, it does not. If the data in the figure is represented in the text, then the error in the figure is also in the text. The proposal suggests that the erroneous text should be challenged and corrected. Of course, fixing the figure will follow the fixing of the text. On the other hand, if the figure presents material not in accompanying text, but presents information supplemental to the accompanying text, it is immediately clear that this figure does not fall into the category of the proposal, which is limited to figures faithfully illustrating accompanying text.

The problem with the present wording, Blueboar, as your comment illustrates, is that its 'flexibility' not only allows for incorrect interpretation, but invites it on a case-by-case basis. This case-by-case orientation suggest criticism that should focus on incorrect text should be directed instead at the accompanying figure, and this focus injects the notion that these figures can somehow be judged for OR and SYN independently of the text they faithfully illustrate. Brews ohare (talk) 16:13, 25 August 2013 (UTC)

I should expand on my comment as I think it's relevant here. I think the current image policy exists mostly for two reasons. To allow for photos of e.g. buildings, people, animals, artworks to illustrate articles. These photos are easily created by editors and can also be found online with suitable licenses. And for diagrams, often ones that exist in books, such as textbooks used as sources. The diagrams can't be used directly, much as the text can't; while the text can be paraphrased the diagrams can be recreated.

In almost all cases the diagrams on WP are similar to those in textbooks, academic papers and similar sources; there is no OR as the editors can point to those sources. They rarely need to as those working on the topic will be familiar with the sources and so know what is appropriate and relevant. Editors rarely need to create diagrams unlike those in sources. If none exist, in any of the sources, then it is likely the topic does not admit to a graphical representation. This is the case in many mathematics articles, where the topic is to abstract to be easily diagrammed.

So whether an image is OR very rarely arises; almost always it's a photo or a diagram like those in sources. As such is is best resolved on a case by case basis, and doesn't need a change in policy. This might change if the issue arose again and again, so e.g. if there were many disputes over image OR that editors find it hard to resolve within the current policy. But I can't think of any, apart from the one preceding this discussion at Talk:Conceptualization (information science).--JohnBlackburne^words_deeds 19:17, 25 August 2013 (UTC)

There are very many examples of original and acceptable figures that do not fit into your scenarios, Blackburne. Here are a few. The reason they have not been subject to ridiculous objections is just that there weren't ~~ridiculous~~ ill-informed editors to raise them. However, policy should be clarified to make sure the ridiculous does not happen. Brews ohare (talk) 19:35, 25 August 2013 (UTC)

I mean instances of disputes over images. As for the rest of your comment: WP:NPA.--JohnBlackburne^words_deeds 20:29, 25 August 2013 (UTC)

Brew's I suggest you redact (not strike) that remark fully, given your block history its enough to earn you another lengthy one. The problem at the article concerned is that (in my opion) you are using the diagram to create a synthesis of material in several sources. So you are not illustrating the material you are creating something original not fully contained in the test. The fact that you keep coming here to try and get policy changes to support your position on a content dispute on an article is disruptive of itself.----Snowded ^TALK 06:13, 26 August 2013 (UTC)

Snowded: It would be helpful if you read the discussion here. Besides other matters, it discusses the use of a figure to describe text. In such cases, many here feel the present policy is clear that 'original' figures are fine if they do not introduce controversial material without sourcing. In particular, if they simply illustrate text that is considered acceptable, the figures also are acceptable. Your approach is a clear indication that the policy is not clear to you, and for editors like you, a clarification would be helpful. Brews ohare (talk) 14:59, 26 August 2013 (UTC)

Please stop assuming that because people disagree with you that they can't have read the material. I have, including the multiple comments from editors who think there is no need to change policy. I have also read the personal attacks and if you don't redact them I am going to raise an ANI report, or ask an admin who has handled you before to look at the issue. Your call, but consider this fair warning. ----Snowded ^TALK 16:48, 26 August 2013 (UTC)

Snowded: Nothing you say above relates to the consideration of the proposed clarification. I reiterate that you do not understand the policy as written, and that this misunderstanding is very clear from your remarks about it. Therefore, a clarification would be helpful. Brews ohare (talk) 20:22, 26 August 2013 (UTC)

I don't see you redacting or apologising for the 'ridiculous ill-informed editors' comments. If you don't do that by tomorrow UK time I am going to raise it along with your various edit comment summaries. Your call ----Snowded ^TALK 20:54, 26 August 2013 (UTC)

Snowded: This discussion is hardly the place for threats. If you interpret my remarks personally, that is upon you. Stick to the discussion. Brews ohare (talk) 21:30, 26 August 2013 (UTC)

Difficult not to take "ridiculous ill-informed" other than personally Brews. Also its not a threat its a promise. You have a long track record now of snide comments and direct attacks in edit summaries and elsewhere. It matches behaviour for which you have previously received both blocks and topic bans. Your call where you take this. I am going to bed, but will put time into a report tomorrow if needed ----Snowded ^TALK 22:25, 26 August 2013 (UTC)

Another thought... consider the following situation... an article contains a complex chain of information in the text. It combines information from several sources, but in a way that does NOT violate NOR. Then someone comes along and says "you know, we could make all this much easier for the reader to understand if we present it in graphic form". He/she creates an image. At this point that image would meet Brews's proposal... faithfully illustrating acceptable non-OR text. Now, lets say another editor says "Wow...yeah... that is much easier to understand. In fact... it is so much easier, why do we even bother trying to explain it in written words? Why not just present the information in graphic form." So, that editor removes the complicated chain of text that the image was illustrating. Does the image suddenly become unacceptable OR because it no longer illustrates information presented in the text? If we accept Brews's proposed language, I could see someone arguing that it would. That bothers me. Blueboar (talk) 16:16, 25 August 2013 (UTC)

Blueboar: You raise the important point that history can enter into the use of policy, and the policy must think ahead. The example you suggest is that a figure that was once included under the proposed change can, through later editing, become a stand-alone figure if the text is removed as a result of the 'one figure is worth 1000 words' principle. However, once the text is stripped away, the remaining figure is no longer within the category of 'figures faithfully illustrating accompanying text', so the newly stand-alone figure must stand or fall under OR and SYN applied to itself. Any policy change must make this point clear. Brews ohare (talk) 16:24, 25 August 2013 (UTC)

It seems likely that a stand-alone figure challenged on the basis of OR or SYN will result in some supporting text being added to supply sources where the information underlying the figure can be found. That will bring the figure back into the category of 'figures faithfully illustrating accompanying text'. The discussion is then brought back to a consideration of text. Brews ohare (talk) 16:34, 25 August 2013 (UTC)

What should not happen is that a figure be labeled SYN because it illustrates points raised in several sources (for example, definitions) or OR because it uses a graphical organization somewhat different from published figures (for example, bars instead of pies). Such 'artistic license' is explicitly allowed by policy if it is clarified that all the figure must do is correctly illustrate an acceptable WP text. Brews ohare (talk) 16:43, 25 August 2013 (UTC)

Meta comments

Brews, have any of your proposals for NOR ever been accepted? It seems like there's a new one every time I turn around. WhatamIdoing (talk) 03:21, 16 August 2013 (UTC)

He only comes here when he has a content dispute elsewhere :-) You could also ask him if anyone has ever responded to his multiple RfCs etc. etc. ----Snowded ^TALK 05:30, 16 August 2013 (UTC)

Would you two care to take a little time off your food fight to comment regarding improving policy in this area? Brews ohare (talk) 14:46, 16 August 2013 (UTC)

There is nothing wrong with the policy. What is wrong is trying to change policy to win a content dispute. To do this repeatedly, despite being already warned about it, is disruptive.--JohnBlackburne^words_deeds 16:55, 16 August 2013 (UTC)

Blackburne: Unwarranted unsubstantiated pure bull. The discussion stands on its own as a clarification of policy, and your spite and bile have no place here. Brews ohare (talk) 02:05, 17 August 2013 (UTC)

Brews, this isn't spite and bile, at least not on my part. The question is about your effect on this page. You seem to make a lot of proposals here, and I can't remember any of them having any result other than a refusal. If my impression is accurate, then it would probably be best overall for you to stop making policy-related proposals. (By "best", I mean that you would quit wasting your time making these doomed proposals, and multiple other people would quit spending hours refusing to accept any proposal that you make.) I'd rather have this be voluntary, but we could look into an RFC/U or see whether ArbCom is willing to consider expanding your existing TBAN. WhatamIdoing (talk) 02:51, 17 August 2013 (UTC)

Whoa there! There's no earthquake here, no seismic activity, no Amazon butterfly's wing causing chaos in Kansas. No need for venting or vomit. It's just a proposal for a tweak of policy. No hidden conspiracy, no unholy contagion. WP:AGF, bro. Brews ohare (talk) 15:15, 17 August 2013 (UTC)

I am assuming good faith, Brews. (And I'm "ma'am", not "bro".) But I'm beginning to lose my ability to assume competence. Being a useful encyclopedia editor does not make you a capable policy writer. It's a different skill set. If you're not very good at writing policies (e.g., if you've noticed that your proposals are never accepted), then it would be efficient for you to give up on policy proposals. WhatamIdoing (talk) 05:40, 24 August 2013 (UTC)

Sorry for my mistake in gender. My assessment is that changes in WP policy hardly ever occur. I don't think this rigidity stems from a lack of good proposals for change, or a lack of need for change. But despite the situation, one can try to clear things up so unending Talk page disputes can be reduced by having very clearly stated policies that don't require ArbCom actions to settle them. If things on WP were different, those with more skill in drafting policy could help to address a problem when one surfaces. Instead we have this kind of flack. Brews ohare (talk) 11:31, 25 August 2013 (UTC)

I helped improve a guideline just the other day, based on a good talk-page suggestion from an unregistered editor. On the same talk page, there's another editor griping about how it's impossible to get consensus to make any changes to that particular guideline. I don't think it's impossible to change guidelines and policies. I do it fairly regularly. It depends more on the proposals you making. WhatamIdoing (talk) 23:57, 26 August 2013 (UTC)

Well... it does tend to be hard to make changes to our core policies (like this one). Even small changes can require a lot of discussion and debate before a consensus forms to either accept them or reject them. Of course, I happen to think that is a good thing. Our core policies shouldn't change easily. Blueboar (talk) 00:07, 27 August 2013 (UTC)

Please help to conclude discussion about numerical data

The discussions about WP:CALC and summaries of numerical data are on hold: the detailed rules and recommendations are "in consolidation and review" here.

Please collaborate with critiques, editions, etc. at the essay Wikipedia:About Valid Routine Calculations.

--Krauss (talk) 02:16, 18 August 2013 (UTC)

Wikipedia:About Valid Routine Calculations

Someone has started an essay Wikipedia:About Valid Routine Calculations intended to augment WP:CALC. Ostensibly it seems to me to be much more permissive than the WP:CALC, and to disregard the consensus of many past discussions on this talk page. Sławomir Biały (talk) 11:07, 19 August 2013 (UTC)

Yeah... it does seem a tad more permissive than was intended by CALC. I would suggest either working with the essay's author to amend that, or starting a "counter-viewpoint" essay of your own, to express your interpretation of the policy. Blueboar (talk) 15:05, 25 August 2013 (UTC)

Justifying "widely" by using numerous sources

Often an article says "X is widely regarded as Y", without providing a reference that actually says "widely", in whatever words. Instead, some number of references are supplied, each saying "X is Y" or "X is regarded as Y". It is left to the reader to judge whether this number of references equates to "widely".

IMHO, merely providing a bunch of references, then saying "widely" should be regarded as Original Research. (After all, an equal number of references saying "X is NOT Y" could possibly be found.) Also IMHO, the definition of Original Research should be expanded to explicitly include this situation. Lou Sander (talk) 18:35, 26 August 2013 (UTC)

Could you be more specific as to what you are referring to. Sometimes X really IS widely regarded as being Y. Blueboar (talk) 22:38, 26 August 2013 (UTC)

Yes, it's normally considered original research to determine that something is "widely" the case without good sourcing that says so. Formerip (talk) 22:47, 26 August 2013 (UTC)

I would go even further: it is not even OR; it is a typical WP:WEASEL: how widely? why says so? how widely the 'opposite-X' is regarded as 'Y'? etc. Sounds seriously but in fact close to useless as fact under close inspection. Of course, we may not be able to prevent wikipedia from entering such weaselisms from sources, but better not multiply them ourselves. Staszek Lem (talk) 23:36, 26 August 2013 (UTC)

Generally speaking, I find that this problem is best addressed by all the editors doing an open-minded review of the available sources, and then honestly asking themselves whether this description is accurate. If it is, then they should accept it in the spirit of accurately summarizing the entire corpus of sources (i.e., what editors are supposed to be doing here); if it's not, then they shouldn't permit it. You simply aren't going to find an explicit statement for some things, e.g., that the current US President is "widely regarded as being a US citizen" or "the Earth is widely regarded as being mostly round" because these facts are so widely accepted that nobody felt it necessary to expressly say so: they just say that he is a US citizen and that the Earth is round. WhatamIdoing (talk) 23:47, 26 August 2013 (UTC)

Well, I don't think anybody would want to put the two your explicit examples into wikipedia either. I stand to my point that "widely regarded" is "mostly widely weasel". Case in point: your first example: the phase "the current US President is widely regarded as being a US citizen" may sound rather ridiculous for one who is aware that POTUS is a CITUS by definition of the office, unless one have heard you know what. But in the latter case the "widely regarded" may sound as a snide remark, rather than encyclopedic information. And in fact, a complementary statement would be encyclopedic instead: "some attempted to argue that the current US P to illegally occupy the office". And most surely both your and mine statements about POTUS can be readily supplied with solid refs, if challenged. Staszek Lem (talk) 01:58, 27 August 2013 (UTC)

Therefore I'd rather reuse yours: "the editors doing an open-minded review of the available sources and then honestly asking themselves," why would anyone write "widely regarded"? Is there a significant dissent? Was there a small but vocal controversy? etc. If an answer on one of these provocative questions is "yes", then it is probably a topic of a paragraph or two, and while writing one will readily come with both good references and more sensible phrasing. Otherwise your version of "open-minded review" is plain post-factum OR (WP:SYNTH). Staszek Lem (talk) 01:58, 27 August 2013 (UTC)

What I'm saying is that saying "widely", etc. without a source that also says it, should be specifically identified as Original Research (though maybe it's enough to call it weaseling). The specific activity that I've often seen is to say "widely", then support it with a bunch of citations, none of which says "widely". The second paragraph of Rupert Sheldrake contains an example. Lou Sander (talk) 02:39, 27 August 2013 (UTC)

Actually (IMHO) the use of "widely" in the second paragraph of Rupert Sheldrake could be OK, because the secondary sources cited in support do make statements about the general view among scientists. The sources cited say, "Almost all scientists who have looked into Sheldrake's theory consider it balderdash." and "...most biologists considered Sheldrake's theory of morphic resonance hogwash...". As long as the sources used are reliable (I haven't checked), that use of "widely" is supported. Zad68 03:01, 27 August 2013 (UTC)

Actually in is not OK. I am sorry to rephrase myself for the third time, but please, please try to remember that the words "widely" and "general view" are way not the same. (I trust you may find 10 differences and a partridge in a pear tree yourself.) And there is no reason to substitute one for another, especially with these sources cited at hand. Of course, there are legitimate usages of the word "widely". There are even cases when 29% is wide enough to be "widely". At the same time English language is flexible enough in expression: "vast majority", "significant majority", "a large number of", "vocal minority"... -- all of them may be collapsed into "widely", but will it be informative? Staszek Lem (talk) 19:03, 27 August 2013 (UTC)

Staszek Lem, if I were doing a GA review of an article and saw such a use of the word "widely" like that I'd flag it as WP:WEASEL and ask for it to be corrected. WP:WEASEL gives "it is widely thought" as an explicit example. It can be a WP:OR problem as well, as laid out in the original question (who is doing the counting of the sources? how are we sure all the most authoritative sources are being reviewed?) but WP:WEASEL should stop it in its tracks before we even have to worry about that. Zad68 02:56, 27 August 2013 (UTC)

"Historical inaccuracies" sections

A recent thread on the OR noticboard ("The White Queen (TV series)#Historical Inaccuracies") has brought up an issue that has been bothering me for a long time. It is quite common for articles on historical novels, films TV shows etc to have sections about historical inaccuracies. It has long been asserted by some editors that it is WP:SYN to list such alleged inaccuracies unless they have been noted by WP:RS with specific reference to the film, novel etc in question. In other words, the argument is that it is not enough to cite a reliable source on the historical facts themselves. I want to question this, because I think it's one of those areas in which anti-OR dogmatism damages the project. The point of WP:OR is to prevent people presenting their pet ideas as though they are accepted fact, are more significant or have more justification than mainstream opinion asserts. In this case, the effect is actually to suppress mainstream opinion in order to leave deviations from historical fact unchallenged. This is because the content of the novel or film can be described without citation, but comment on it cannot. I have long felt that this is a kind of perversion of the spirit and intent of WP:OR and that we should stop repeating the mantra that reliable sources have to have directly commented on "inaccuracies" in films and TV shows with historical content before we can note these.

As far as I can see cases like this are similar to the issue of whether making arithmetical calculations (WP:CALC), or writing translations, are OR. If there is nothing to dispute then the concept of OR is inappropriate because there is in fact no "synthesis" occurring. If a film portrays, say, the Battle of Waterloo being fought in the same time as the Battle of Trafalgar, I cannot see how it is in any way OR to assert that this is inaccurate. If there is no ambiguity about what the film/novel shows and there is no uncertainty about the historical facts, then there is no original research, because no new idea is being "sythesised". It's not a new idea that the battles did not happen in the same year. Now, I accept that there are ambiguous cases, but uncertainites about the facts are a separate issue. By adopting the approach advocated by some editors we damage the encyclopedia by excluding information that is verifiable and which serves the purpose to inform. In addition, as I said above, this is actually one of those aspects of historical fiction that people are most interested in, for perfectly good reasons. They want to know how much of it is true. We do not serve readers by excluding information that is not in dispute. I think we need a clarification of OR policy similar to WP:CALC to address this issue. Paul B (talk) 18:03, 29 August 2013 (UTC)

You have a very real and good point here, although I see one possible problem, even if it is a minor one. There are a lot of "alternate history" fiction out there, like that of Harry Turtledove, which is more or less based on the "what if" scenario. Articles on such works of fiction could easily be bogged down with the specific details regarding how they deviate from reality. Also, the same could be said about a lot of contemporary fantasy, including a lot of comic books, like say Aquaman, which seem to be based on what is more or less the real world, with some variations, like, with Aquaman, the existence of his Atlantis. Personally, I wouldn't want anyone to think Braveheart is historically accurate, it isn't, but I don't know how much space in our content to give such material in general. Having said that, personally, I wouldn't find making it possible to add a little material along the lines of "a fictionalized account of (whatever)" in a lot of cases of this type, although, I suppose, in at least some cases, the movie or whatever might be based on a seriously "alternative" (or fringe) theory which isn't necessarily "fictionalized." And I don't know how many might qualify as that. I agree there is a problem, but be damned if I myself know how to rewrite things to address it. John Carter (talk) 19:01, 29 August 2013 (UTC)

Historical inaccuracies sections, like "differences from the book" or the like, fall into WP:TRIVIA and even if obvious (like a show stating an event happened in one year when there's universal agreement it happened in a different year), should only be included if other sources have pointed them out, unless, as John Carter points out, this is intentionally part of the show's fiction, a "what if" show if certain events happened in different ways. (I can't think of any immediate clear examples, but one close case that comes to mind is the alternate universe in Fringe (TV series) where certain events happened differently and shaped that universe in a different fashion that leads to the fundamental premise of the show. However, at the same time, these differences have been noted in sources so there's no guesswork here). So unless the historical inaccuracy is critical to the show's plot, such sections should be avoided unless called out by sources. --MASEM (t) 19:06, 29 August 2013 (UTC)

strongly support Masem. "historical inaccuracies" in a work of fiction are only important in an encyclopedia article if we can include a an answer to a "Why should we care?" or "What was the impact?"- and both of those require that third party sources have noted the inaccuracy and its impact. If inaccuracies are important and relevant to an encyclopedia article, then reliable sources would have covered them. We do not do the investigation and research, we merely aggregate what professionals have already published about the topic. CALC covers a very different situation from what is being suggested. I dont believe there is any way to write guidance about interpretations of history or cultural programs that can effectively identify clear cut boundaries.-- TRPoD aka The Red Pen of Doom 00:58, 30 August 2013 (UTC)

These responses are, I think, to some extent addressing a separate issue. Yes, articles can be bogged down in pointless detail when people want to list too many of the deviations-from-fact, and there is a tendency for such sections to grow out of control like those "in popular culture" ones. That's also true of the "differences from the book" sections, as you say. But that's a quite different question from the matter at issue: which is what constitutes WP:OR. I wouldn't want to add pointless material like "King Fred was 5 foot tall, but the actor portraying him in the film is 5 foot 6". Of course that's silly, but that's an issue of undue weight, not of OR. Where the historical context is significant and relevant I think we should be able to add content without being accused of WP:SYN. Re John's comment about rewriting, I don't think the SYN section needs to be rewritten, but we could have a clarification paragraph like WP:CALC. Paul B (talk) 19:18, 29 August 2013 (UTC)

(e-C) You're right, but the hard part there can be determining in some cases if there is someone out there somewhere who honestly thinks that the movie might be presenting "the truth" as they see it. So, for example, I could see something like "the movie presents a fictionalized account from the perspective of a person not known to history" for some movies about history told from the viewpoint of an unknown, like Forrest Gump or Apocalypto, or maybe "the story presents a version of the Battle of Trafalgar in which the Egyptians win" or "promotes the view that Richard the Lionheart was a bisexual" or some similar phrasings like those. I can certainly see maybe allowing some such constructions in instances when the context is significant to the work in question. But there might well be all sorts of problems in determining how significant they have to be.

Regarding the addition about a clarification paragraph, I could agree with that, if we had RS which indicated the "nonstandard" nature of the ideas being discussed. John Carter (talk) 19:31, 29 August 2013 (UTC)

If we can do it for these cases, what's different about doing it for fringe stuff such as Where Troy Once Stood? Dougweller (talk) 19:42, 29 August 2013 (UTC)

We really have three situations here:

1) Historical Fiction... ie purely fictional stories set in the past. We don't expect fiction to be historically accurate, so pointing out where some work of fiction gets history wrong really is TRIVIA. That said, in the case of Alternate History fiction (or "what if" fiction) to note the point of divergences (as these are important to the plot.)

2) Biographical fiction... ie novels and movies that claim to present the life of a historical person. We do expect a biography to be accurate... but we also understand that biographical fiction may take liberties. So pointing out where a work of biographical fiction gets it's history wrong is appropriate if the error is a doozy... but don't note minor trivial errors. To prevent OR, only note errors that have been highlighted in sources.

3) Pseudo-scholarship... ie books where the author is at least claiming to present fact... actual history. We expect scholarship to get facts right... and if the author does not, I don't think it wrong to highlight it. For example, if an author makes a big point of saying that King Athelstan traveled to York on a given date... and reliable sources say that Athelstan actually was in Winchester at the time, note the disagreement and cite the reliable sources. Blueboar (talk) 21:03, 29 August 2013 (UTC)

I have always thought to correct approach to historical inaccuracies in fiction was simply to link to articles on the actual historical figures/events. Readers who want to learn the actual history will follow the links. People do not read about a fictional TV series to learn about history, and if they do, we can't help them. The same goes for lists of "scientific inaccuracies" in the articles of some sci-fi movies. If the inaccuracies are so significant that journalists/historians/scientists have written about them, then they may be appropriate to mention. Someguy1221 (talk) 21:11, 29 August 2013 (UTC)

Not a bad approach, but it presents a problem regarding what Blueboar above calls pseudo-scholarship, including some of the whack-job pseudoscience. I know for a fact, for instance, that there have been a lot of recent popular books regarding the field of early Christianity which have put forward ideas which have little if any support in the fields of biblical studies/ancient near east studies/christian studies, but which have gotten generally neutral-to-favorable reviews along the lines of "interesting idea," but may not be submitted to any academic journals. So, the library reviews might say "recommended" because the conspiracy theory people like books like these, even though the academic journals, if the works were submitted to them, might call them unfounded, irrational, highly speculative, or, well, nuts. In such cases, yeah, the material based on the independent reliable sources under the current structure would be generally positive, even if, unfortunately, the ideas presented in the works are, maybe, held to be about as reliable as saying that the Greek or Egyptian equivalent of John F. Kennedy was the bastard descendant of their equivalent of Abe Lincoln by their equivalent of Sally Hemmings. In a lot of material of that type, the existing library journal reviews might basically say, "could be" (because no one knows enough to say it can't be) and "interesting" (because it is interesting and appealing to those who find such appealing), and might even recommend that libraries buy it. But that don't mean anyone in the relevant academic community thinks the writer said anything worthwhile. And, unfortunately, that sort of situation happens more than I think a lot of us would like. John Carter (talk) 21:23, 29 August 2013 (UTC)

Are we talking about articles on pseudoscholarship-related topics, or the use of pseudoscholarship sources? Someguy1221 (talk) 21:56, 29 August 2013 (UTC)

Probably, to some degree, both. Some editors probably know more about this than me, but at least in the broad fields where history and religion interact there have been claims in the past when people have attempted to use a recently released popular book as evidence of some new and groundbreaking theory, and in some cases, like Zoroastrinaism, let's say, which doesn't get a lot of coverage, it can be really hard, and sometimes impossible, to say that a popular book by an academic might not be in some way reflective of an idea gaining broader acceptance. John Carter (talk) 22:04, 29 August 2013 (UTC)

WP:NOR/WP:SYN and retroactive use of new names

FYI, I have started a discussion here about whether or not using a person's new name when discussing periods of their life during which they were known by another name violates wp:Verifiability, wp:No original research and/or wp:Synthesis (as some have claimed).

The note at the top of this page suggested I post this to WP:NOR/N, I but decided that because it pertained to a phenomenon (the retroactive use of a new name), not just to any specific instance of the phenomenon, it made the most sense to leave the notice here.

-sche (talk) 17:37, 30 August 2013 (UTC)

Take a look at how we deal with cities that have changed their names... we refer to St. Petersburg for most of the article, but use Petrograd and Leningrad where historically appropriate. I don't think there would be a problem doing this with people. Blueboar (talk) 23:22, 30 August 2013 (UTC)