Wikipedia

Wikipedia talk:No original research (draft rewrite 5th December 2004 to 5th February 2005)/Archive 4

< Wikipedia talk:No original research (draft rewrite 5th December 2004 to 5th February 2005)
(Redirected from Wikipedia talk:No original research (draft rewrite -- archive 4))
Archive 1Archive 2Archive 3Archive 4

sSee also:


Same problems

I entered a long critique above on this pages, to which there have been no responses. This might be because:

1.) It was felt to be so far off what people are concerned with, it was not worth discussing.
2.) It was not noticed.

If the former, where have I gone wrong? If the latter, I do think I have something to say, that the people working here are taking too much for granted, too much in group-think mode, a common problem in Wikipedia. A group of like-thinking people attempt to solve a problem that they see, spend weeks working up their ideas, and they find them solidly rejected when it comes to attempting to make them into Wikipedia policy, often for good reason. The solution in the opinion of most, fits no problem that they see, and introduces new problesm.

For topics of scholarly concern, Wikipedia is a tertiary source. In other words, Wikipedia generalizes and explains existing research on a specific subject, and is based on secondary sources, such as books published by a legitimate publisher and journal articles, that analyze, interpret, synthesize, or evaluate primary and other secondary sources. This does not mean that primary sources cannot be used to report fact. But at least concerning any subject that is a matter of academic research, we should exclude unattributed original interpretive and synthetic statements.

This simply isn't so.

First, attempting to distnguish between topics of scholarly concern and other topics makes no sense me. That this draft attempts to do this, that it seems compelled to do it, indicates something is wrong with the underlying philosophy of the draft . Everything in Wikipedia is a topic of investigation and scholarly concern, though we are mostly not supposed to be doing the primary investigation. We are trying to present valid facts and explanations to a reader and present facts. Where there is debate, we are attempting to present and explain the debate factually. We can only do this by approaching material in a scholarly fashion. So what are not topics of scholarly concern here? Things like Pokémon?

I know almost nothing about Pokémon, but only a few minutes on the web showed me that Pokémon has been the object of well-known academic study by Joseph Tobin ([1], [2], [3]), Anne Allison ([4], [5], and others. Academic conferences are held on the Pokémon phenomenon. [6], [7], [8]. The earliest paper I found is dated to 1997 (http://mentalhealth.about.com/cs/familyresources/a/pokemon.htm).

Second, basing material on secondary sources does not make Wikipedia a tertiary source as here defined. Wikipedia should not be doing original research at either the primary level, secondary level or the tertiary level, in the sense that original research is defined here, that is should not be presenting unique and unusual theories and explanations in its own voice. At the primary level, we don't talk about the Yacqui Indian who appeared to us and taught us the secrets of the universe. At the secondary level, we don't explain the real meaning of Oliver Twist by explaining hidden codes in the work that show that Charles Dickens was an illegitimate son of Queen Victoria. At the tertiary level we don't attempt explain the disappearance of the British Empire by notches in the Great Pyramid rather than from the normal economics, politics, and accidents of history.

But all those levels do appear in encyclopedias.

Wikipedia and any encylopedia reproduces and summarizes primary and secondary and tertiary material from other sources and even producing some of its own on occasion, as long as that does not involve presenting unique or original theories and explanations of its own.

Primary material is minimal, partly because it is idiosyncratic, and partly because it is mostly too long to present, other than a few pithy saying, some short verses, selected quotations, and so forth. Mostly, when primary sources are the point, an encyclopedia summarizes. It can present pictures. It can give us a very good representation of the Mona Lisa and other works of art, can show us Stonehenge and the Eifel Tower both through photographs and plans and maps.

We expect an encyclopedia to present secondary material, explaining, for example, the solar system and the laws of motion that govern it, to indicate the economics that drove the Age of Exploration.

We expect it to present tertiary material about the research behind the secondary material pesented and coverage of disputes where there are debates over differing secondary explanations.

But no encyclopedia has ever only been a tertiary source, unless there is something like an Encyclopedia of the History of Science. But even that has its primary sources, the writings of scientists themselves and writing of their biographers and other primary writings which would be embedded in the work, secondary material, explaining how the concepts developed, and tertiary material, indicating how the valuation of different concepts and different ways of looking at things changed. Wikipedia has never been intended to be just an "Encyclopedia of Scholarship". This new claim that it should be such, seems to be why suddenly there are two standards for sourcing.

The many lists that dot Wikipedia are an invitation to a kind of original research which does not involve original theories and explanations and there is no attempt to limit those lists to material not covered by academic journals.

Nor is tertiary research necessarily based on secondary sources in the sense of incorporating them.

Outside Wikipedia, a believer in modified Velikovsky chronology and cosmology, can present both secondary and tertiary material about the development of thought that arises from Velikovsky's original theories. And someone who disablieves all of Velkovsky's theores and derived theroes, may still complete reject secondary material, the theories themselves old and new, but fully accept the tertiary research by the same writer, that is the account of how Velikovksy's idea have been taken and worked over and argued by those who tend to believe in that kind of thing and a description of the different schools of research. Tertiary research is not necessarily directly based on secondary material at all, even though it is about secondary material. I don't believe in astrology. But I used to have what appeared to be a very good book by purported believing astrologers on attempts to prove that astrology actualy worked. The conclusion they came to is that none of the attempts were convincing. I mislaid it at some time, and cannot recall its name.

The periodic table does not magically become tertiary material simply because it is a copied from one source to another. It is still secondary material wherever it appears, an explanation of the relationships between the elements which are the primary material. And a dicussion of theories about it and how it came to created would be tertiary material.

The proposed draft, rather than strengthening the demand to rely on trustworthy sources, weakens that demand, by removing this demand from non-academic subjects which are confused with wrongly identifies with non-scholarly subjects. But there are no non-scholarly subjects. There are only non-academic subjects, and that only by virtue of them currently not being greatly covered by standard academic institutions, not be cause any subject is innately no academic. But that a subject is not academic by historical accident, does not prevent scholarship being carried out in publications and journals respected by those interested in such areas, perhaps better respected in those areas on the average than are many academic journals in their areas. There are poor publications and good publications.

And here still is that horrible phrase legitimate publisher which also occurs in the current version. Explain that phrase. Is the publisher of the National Enquirer a legitimate publisher? If not, is the Institute of Creation Research a legitimate publisher? Is Doubleday, publisher of The Da Vinci Code and books about Edgar Cacye a legitimate publisher? Is Duke University Press a legitimate publisher, the people who publish Social Text, the journal that accepted Alan Sokal's purposely and obviously fraudulent "Transgressing the Boundaries: Toward a Transformative Hermeneutics of Quantum Gravity"? Is the UCLA Department of Anthropology a legitimate publisher, the people who gave a doctorate degree to Carlos Castaneda for his interviews with Don Juan. I've seen too much crankdom in academia to think that insisting on academic acceptance of some kind will hold back the non-standard theories. I see what is being attempted here, but the problem with POV pushers of odd theories is not that the current policies aren't enough to stop them, it is that those policies aren't enforced, except after weeks or months of evidence and arbitration. Yes, we should rely on good scholarship, and yes, on the whole, academia is where you find the best scholarship and areas that are greatly covered in academia, and the crank theories that do flourish in academic don't gain a majority following, on the whole. (Though I gather that today Signmund Freud is generally looked on a crank by psyhcologists.)

But beyond that, one cannot go.

I've changed the draft to say 'probably exclude'. I accept that there are exceptions to the rule; but generally we should treat such publications as questionable materials. This is a good rule of thumb; because we are not in a good position to judge the quality of such materials, not least because of their limited avaliability.

We will find, however, that for certain subjects like local history that such publications will be the only secondary material avaliable; and so our articles will have to use them. But where we have alternative materials avaliable I think it is a good rule.

Of course, cite material that is more easily available when all else is equal, especially when the more easily available material often itself cites the less easily available material. But applying this as a rule would mean taking the word of a light and frothy treatment of some subject over carefully researched articles, just because those articles are not generally available. Or the more readibly available material may be very POV or very inaccurate. I know nothing about period furniture of any period. But I would expect those who do know such things would have access to books with limited print runs and journals with small print runs and hobbyist magazines that contain valuable information and I would expect them to use such material in appropriate articles. That "We", meaning most of us, cannot judge on the validity of such material is irrelevant. None of us are experts in all areas and cannot judge a large number of articles, cannot know which journals and which scholars are trustworthy, which are sometimes brilliant but have a particular bias that must always be taken into account, and which ones are flakey and produce shoddy or biased work.

That the average person cannot find and validate material printed in the The Comics Journal 5 or Witzend 4 is not a reason to avoid citing them as sources. It is all the more reason why obscure sources should be cited and perhaps quoted in full if relevent to an article. The more obscure the topic, the more obscure the references are likely to be, whether to obscure academic journals or to other publications. But Wikipedia is not paper, and we already have coverage of topics more obscure than anything in a limited-size encyclopedia. Let people who can cite primary sources that can be verified and obscure secondary sources do so. Express honest and reasonable doubt where sources seem dubious. Where there are few secondary sources, even where these academic sources, they are often very poor. (For example, most Tolkien academic literary commentary is notoriously poor, in part because Tolkien wrote in ways that go against current popular academic literary theories.)

Those who know obscure material in a certain area are those who should judge articles in that area and validate them or improve them. I cannot judge the validity of many articles in Wikipedia and would not try to do so. By the same token, I would not expect most Wikipedians to be able to judge the validity of many articles that I might and do write, to know which particular secondary sources are good and which are untrustworthy. But it still is the individual editors who must make such choices.

Removing a requirement for scholarship in supposed non-academic areas is not the way to force out supposedly crank theories in academic areas. And original research, as defined here, is irrelevant to an article that happens to be based on material not readily available to everyone, especially in areas where there are a number of people who know a particular field and where there is generally agreement that no-one is pushing special partizan POV. Again and again something will be on VfD because it is not verifiable, and someone who knows that kind of material, will easily verify the information. The POV-pushers are pushed aside eventually. The problem is not in the rules, but in lack of enforcement.

I someone is being continually harrassed, illegally, it does no good to make the laws stronger, if the harrassment is already illegal, especially if the new laws stop bad action by also stopping good action.

Jallan 23:54, 12 Dec 2004 (UTC)

Response

Jallan, these are very thoughtful coments but you were right above -- too long! I say this with respect and constructively. Can you break up your critique into smaller points? For now, I have the energy and attention-level to comment only on a few of your points. First, I think in some cases the problem can be both enforcement and the rule. Here I think it is a case of both. The rule, remember, is not just a rule, it has pedagogic value, explaining our policies to newbies. I think this rule needs to be clearer. Second, you are right that the distinction between scholarly and non-scholarly topics (a distinction I made) is false. Many things are of interest to different groups, scholars being only one. I am too tired now to figure out how to phrase it, but the point is an article shouldn't make scholarly claims without backing it up with scholarly research. When talking about pokeman, the views of fans and the company that produces the produce should also be represented but we still need to have valid sources. there has to be some way to make this point clearly. Finally, many encyclopedias are secondary sources (as I definied it; some people objected to my version of the intro, though I think the current intro has more problems) -- but remember that such encyclopedias are written and edited by scholars who ensure verifiability and rigor. Wikipedia rejects this system, but we don't reject all of its goals, e.g. verifiability and scholarship. If we reject their method, we need some other method. One is the darwinian process by which all articles are in progress and edited. Another, as you pointed out, is VfD. But I think this policy is a third leg of our method or system -- and an important one. I think that discouraging secondary research and encouraging tertiary is important when so many editors have such varied backgrounds. Slrubenstein
Jallan, I do respect what you say and value your judgment, but I was overwhelmed by the length.
Part of what I had in mind, but obviously didn't say, was that I planned to review your comments when we got further along, to see whether your concerns had been addressed.
The part you noted at the top of your first post has been reworded, although not exactly as you suggested.
Do you think the current Wikipedia:No original research needs no changes? Maurreen 00:44, 13 Dec 2004 (UTC)
Jallan, I'm generally in agreement with what Steven and Maureen say; but I would like to add why write an essay on what we are doing wrong; when you can contribute to the rewrite. We aren't appointed by anyone, we are just interested parties, who think the existing version of the policy isn't comprehensive enough. The more people that contribute to this rewrite the better. On your main point I agree that the phrase 'scholarly concern' is incorrect; but you haven't suggested an alternative to replace it.
An attempted fix:
"Where possible Wikipedia aims to be a tertiary source. In other words, a Wikipedia article generalizes and explains existing research on a specific subject, and is based on secondary sources of an academic nature, such as books published by a legitimate publisher and journal articles, that analyze, interpret, synthesize, or evaluate primary and other secondary sources. This does not mean that primary sources cannot be used to report fact. But at least concerning any subject that is a matter of academic research, we should exclude unattributed original interpretive and synthetic statements. On topics which have not currently attracted sufficient academic research other sources such as newspapers, magazines and government documentation may be used in addition."  :ChrisG 19:50, 13 Dec 2004 (UTC)

Jallan, I didn't understand what you wrote, which is why I didn't respond. I've just read it again and still don't understand. Two things would help me: (1) Could you give us your definition of primary, secondary and tertiary? (2) Could you give some examples of the issues you're raising? I should say I think I do agree with you that the distinction between scholarly and non-scholarly articles is a false one. I also agree that Wikipedia shouldn't be revealing facts, advancing theories, offering interpretations "in its own voice," as you put it, and that's a good way of putting it, but how do we express that so that other editors get it, because it goes to the heart of what we're trying to say?

But it still leaves problems: We don't want Wikipedia doing things in other people's voices either, if those other people are nuts, or too self-interested, or not edited well enough.

The legitimate publisher issue aside, here's an example of the "own voice" problem. Kenneth Bigley was a British hostage beheaded in Iraq. His murder was filmed and put on the Web. A Wikipedia editor watched this film and wrote in the article that there were continuty gaps in the film; that the actual death scene was partly obscured by the killer's legs; and that there appeared to be a bullet wound in Bigley's head, the implication being that the group didn't kill him by beheading but had staged a post-mortem scene. The fact that his body was never found might lend that idea some credence.

I checked all the news reports I'd read about Bigley and couldn't find references to this. I checked on Usenet and I found a very small number of posts about it; not enough to convince me it was an issue. So I deleted it. The editor came back to me, annoyed, and said he had watched the tape, and I hadn't, and possibly most of the journalists writing about it also hadn't watched it, so why should we -- who had not watched the video -- be allowed to write about it, but he -- who had watched it -- was not? Fair point.

I was not prepared to watch it, and he did not want to leave it out, so we compromised with the horrible "some of those who watched the video felt there appeared to be . . ."

What was the right thing to do here, given (1) the Bigley tape was primary source material, (2) Wikipedia is not allowed to analyse "in its own voice", and (3) no reputable (or even disreputable) news organization that I could find has referred to continuty gaps or bullet wounds, but (4) the other editor was right: he had watched the tape and I had not, so how come I was getting to comment? Slim 02:48, Dec 13, 2004 (UTC)

I think you were still right to delete it (and you should probably go back and delete it again). The problem that we continually have to wrestle with is that, just because we don't include some material in Wikipedia does not mean that material is bad/wrong. It simply means that Wikipedia is not the proper venue for such material (i.e. novel analysis/synthesis). If someone wanted to publish Pulitzer-level journalism, or Nobel-level science, on Wikipedia and Wikipedia only, then we would still have to turn that person down. (This probably needs to be emphasized in the no original research article because it comes up again and again and again...) —Steven G. Johnson 18:01, Dec 13, 2004 (UTC)
Yes, good point Steven. And I agree that it comes up again and again, both in the political/historical articles and in the science ones. Slim 23:50, Dec 13, 2004 (UTC)

Consolidation

I think this is a little confused because discussion of the draft rewrite is going on in two different places. I am going to move a bunch of comments from the main talk page to here. Then I'm going to archive a bunch. Maurreen 04:38, 13 Dec 2004 (UTC)

Redundancy with Wikipedia:Cite sources

I don't agree that we should essentially duplicate the material of the Cite sources page here. Not only does it make this page longer, less focused, and less readable, but it introduces the other usual problems involved with duplication — things will get out of sync, and people interested in citation policy will have to monitor both pages. I would suggest something short, like:

Since Wikipedia strives to be a tertiary source that merely summarizes well-established, published materials, it is especially important to cite sources that the reader can consult to verify an article and to find more information. Even if you are writing from your own knowledge, you should actively search for authoritative references to cite. See Wikipedia:Cite sources for more details and rationales, as well as an example citation style (although formatting is of secondary importance).

I don't agree, by the way, that citations are "especially important for controversial articles", as the current (modified) citation section states. Citations will come naturally in such articles, as people are forced to cite sources in response to challenges. The real problem is with non-controversial and obscure articles, which almost never get challenged, where someone will come along in 10 years and find all sorts of statements that are difficult to verify. In any case, this is the sort of thing that is better debated in the Wikipedia talk:Cite sources page. —Steven G. Johnson 21:12, Dec 13, 2004 (UTC)

problems with the current intro

I don't agree that the current version is better than the one I suggested (which doesn't mean the one I suggested can't be improved upon). I have three problems. First, if the problem with my proposal was that it didn't define original research, this version is no better -- like mine, it just provides examples. But I have a problem with the examples: they don't include other forms of original research (e.g. interviews, field research, experimental results) that are also inappropriate for encyclpopedias. Second, I object to the quote of Jimbo in the second paragraph -- it risks making him too much of an authority firgure. Third, it is strange for the intro to introduce the term "tertiary" source before it has explained what primary and secondary sources are. Slrubenstein 19:40, 14 Dec 2004 (UTC)

Your three points: (1) Interviews, field research, experimental results would all be covered by "unpublished concepts and ideas" and "assertions not published in a reputable publication," (2) I think Jimbo Wales should be right up there, because he came up with NPOV and "no original research" and they are the two concepts he often says are "non-negotiable." I feel that, by having him in there right at the top, we signal to people that this policy should be taken very seriously, (3) I agree that primary, secondary and tertiary sources need to be defined before being used here. Slim 22:16, Dec 14, 2004 (UTC)

I'm afriad I still disagree with you on the first two points. On the first point perhaps you misunderstand me. I don;t think it is covered by "unpubliched concepts and ideas" because concepts and ideas are generally associated with analysis and interpretation of data, whereas I am referring specifically to data itself. Interview data, field observations, assays, produce raw information which must then be conceptualized and thought about by the researcher. In most sciences, including social sciences, data collection and analysis are two separate stages of research, and produce different kinds of knowledge. As for the second point concerning Jimbo we may just have a difference of opinion but I hope others will chime in. I think very highly of Jimbo myself. But I think it would be a grave, grave error to imply or suggest that things said by Jimbo should be taken more seriously than things not said by Jimbo. I understand that ultimately Jimbo has a lot of power here. But it seems to me that his own vision of Wikipedia is more egalitarian. I think to use him as an authority would actually be to undermine Jimbo's vision. Slrubenstein

Re: (1) You could add "data" to unpublished concepts and ideas. In fact, I'll do it when I've finished writing this. Re: (2) I still disagree. I take your point about the egalitarian issue. But he has expressed it very well, I think, and his "novel narratives and historical interpretations" quote covers a few angles not explicitly covered elsewhere, or at least not so well. Finally, he has the only power around here, not just more power. Therefore, what he says does have to be taken very seriously; also in part, power aside, because the original vision was his. What do others think about this? Slim 02:22, Dec 15, 2004 (UTC)
I don't feel strongly either way. But the equivalent NPOV page quotes Jimbo. And I think his wording is good. About the various sourcing levels, I think the intro and overview should be fairly short and simple. Maurreen 06:14, 15 Dec 2004 (UTC)

Counter-response to responses to my objections

Primary, secondary, tertiary

I thought I was being clear. *Sigh!* Obviously not. There is a discussion of the terms at [9] which is probably as good or bad as any and states that dictionaries and encyclopedias are both secondary and tertiary. That I believe fits general perception. To introduce a perception in contradiction to the current statements without providing sources breaks the very rules that this new draft is attempting to further.

There is another discussion at [10] which states:

Secondary sources are used to write tertiary sources, which have a larger scope and ignore the details.

Do we really want Wikipedia to "ignore the details" in articles? That depends on the article. This account also states:

When a primary and a tertiary source contradict each other, it is better to rely upon the primary source and to explain why the tertiary source has another point of view, than the other way round.

I do not agree with that as a universal principle. But it is often true. That something is tertiary does not indicate it is better than primary or secondary. It still comes down to individual editors and groups of editors doing the final evalutation, based on sources available, whether primary, secondary, or tertiary.

Not allowed to analyze?

Wikipedia is not allowed to analyze in its voice? It does this all the time in article after article in what in includes, what it excludes, how it plays particular statements. Editors are surely supposed to do this, to recheck what sources say, to re-analyize, to re-evaluate, on a personal level, in order to properly evaluate primary, secondary, and tertiary sources available. The results of such evaluation on a personal level may sometimes by original or very non-standard. If original, Wikipedia is not the place for publication. If only non-standard, but agreeing with beliefs of others, even if those beliefs are generally labelled crank, then Wikipedia is a place where such beliefs can be published, but must be labelled as minority belief. Some articles here have behind them a great deal of analysis, indeed original analysis, about what should be included, what should be excluded, how exactly a particular statement should be phrased, and so forth. An article often states clearly that certain theories are commonly believed, that others have a minority following, or are supported by particular groups (religious or idealogical) that others are out of date or even fully disproved and so forth. Articles state clearly, hopefully giving sources, that particular pieces of information are factual and others are theoretical.

The Bigley case

In the case of the Bigley problem, as far as I can see, one editor is providing a viewpoint unsupported by any web links reference to hard publication. It is a claimed that the matter is discussed in a forum, but no link is provided to the discussion. If there were any evidence of this viewpoint, I would expect to find at least one sensationalistic website or one tabloid newspaper discussing the theory. An opinion without even that minimum support is basically unsupported. The editor states: "Bigley's death is not in doubt, but there's reasonable doubt as to whether the video is a true record of it." There is equally reasonable doubt as to whether any video is a true record to those who wish to doubt it, since all video can be doctored. And why not doubt Bigley's death also? There is nothing that some people will not doubt.

In short, the editor's only evidence is personal belief and unsourced claim that others perceive the same thing.

So how does this new draft, which specifically exempts non-academic subjects from validation, indicate more strongly than the current policies that this statement should not appear in the article? The proposed draft arguably does not oppose such statements at all, since this is not an "academic" topic.

But suppose a single academic source mentioned those doubts as worth taking seriously, did urge that the tape was doctored or that there were gaps in the continuity or both? Would that lend this viewpoint any more credibility than a similar article published in a tabloid newspaper? Is the academic journal Social Text, for example, more credible than the National Enquirer?

Scholarship

... the point is an article shouldn't make scholarly claims without backing it up with scholarly research.

More generally, an article should not make claims which most readers cannot easily validate without providing specific validation of its own. Scholarly may mean "learned and knowledgeable" or may mean specifically "academic", and I feel this draft too much inclines towards the second meaning, confusing academic with scholarly. For example, The Rolling Stone Dictionary of Rock and Roll is not a product of academia, but is very much a product of scholarship. Compilatons of sports statistics are similarly scholarly (in their field) but not academic. There are people like Jerry Bails, who is an academic, but has devoted his life to research primarily in what most think to be a non-academic subject [11]. Databases like this [12] are not created by the unscholarly. (There are fields where unscholarly research abounds, genealogy for example.)

Pedagogic value

This rule, remember, is not just a rule, it has pedagogic value, explaining our policies to newbies.

If the new policy is better, it should be justified on other grounds than being easier to explain. I don't feel it is better or easier to explain. I think this draft, more than the current draft, confuses original research with minority opinion, not necessarily the same thing at all. Is it really easier to explain that Wikipedia is only a tertiary source if in fact it isn't and, in my opinion, shouldn't be and can't be only a tertiary source? Encyclopedias echo secondary sources just as they echo tertiary sources. There are both generally authoritative secondary sources and generally authoritative tertiary sources and sources of both kinds with little authority.

I believe sourcing is the real issue. That controversial and obscure material should be sourced is easer to explain pedagogically than are dubious differences between academic and non-academic material and secondary and tertiary material. The current statements are probably strong enough, if they were pushed and were enforced. But this is difficult to do, when so many articles are entirely unsourced. Also, attempting to force sourcing at time of creation runs against the de facto Wikipedia tradition of semi-plagiarizing anything one can find on the web, a practice largely responsible for Wikipedia's fast growth.

Why not join in writing this new draft?

Because I believe the approach is wrong. It would be like a believer in evolution supporting the tweaking of a set of guidelines to moral behavior in the direction of creationism on that grounds that this would give the guidelines more pedagogic value and make them more enforceable. I think the increased reference to legitimate publishers is dangerous. I don't agree with the unsourced viewpoint that an encyclopedia should only be a tertiary source. I don't agree that there should be a distinction made between academic and non-academic subjects in respect to providing validation of statements made. The most that might reasonably be said, I think, is that information about a subject should be backed by academic sources in proportion to it being a subject of academic research, but without drawing a distinct line between academic subjects and non-academic subjects. Obscure and controversial information should be backed up for all subjects (and it doesn't hurt to provide references to more easily available and straightforward information also).

Jallan 05:09, 15 Dec 2004 (UTC)

Response

Jallan, you make some good points. But do you think the current Wikipedia:No original research needs no improvement? Maurreen 06:23, 15 Dec 2004 (UTC)

I agree that the distinction between academic and non-academic subjects is a false distinction. We need good sourcing for all articles, not just academic ones.
I didn't agree with the definition of primary, secondary and tertiary sources provided in the link Jallan gave (which I think was from James Cook University). I propose we barely mention these terms, because they're causing us a lot of confusion. Basically, claims in Wikipedia must have been published by a reputable publisher. People will argue about the meaning of "reputable," but most of the time, we know it when we see it. End of story. It's going to get too complicated otherwise. Slim 06:53, Dec 15, 2004 (UTC)
I think Slim's revisions have taken care of Jallan's objection's. If not, let us know. Maurreen 16:04, 16 Dec 2004 (UTC)


Yes. That perhaps fixes everything. For many of the articles I have been writing, a great deal of the article is pure secondary material, summaries and quotations from primary textual material, often replacing unsourced and inaccurate information taken from tertiary material, either unacademic or simply out-of-date. One could go to good sourced academic tertiary material to correct, but why, when one can go straight to the horse's mouth, to the primary textual sources? This is especially the proper route to take when that same data is also readily available for checking by readers of the article, on the web or in books currently available. And though, of course, I do check out good tertiary scholarly material as a guidepost to primary material that I might have missed and for possible differences of interpretation, I am not especially using that tertiary material as a source itself, only as a pointer to sources. That my findings agree with the tertiary sources and that other tertiary sources mostly agree with one another is because that is what is there to find and is exactly what almost everyone finds. So the article can be mostly written as a secondary source, giving those findings, providing, as much as possible, the basic information which may be and sometimes is used in theories.
"Reputable publisher" is reasonable, especially as the discussion makes clear this is not entirely straightforward. I have added to the discussion that divisions of general publishers may be very reputable. For example, Penguin Books is certainly reputable.
Jallan 02:45, 17 Dec 2004 (UTC)

Deleted primary etc. sources

I've just done another edit, and I got rid of the definitions that have been causing us problems. Let's talk about "reputable publications" instead of primary, secondary, tertiary because we're tying ourselves in knots. As usual, anyone should feel free to edit what I've written; re-insert what I've deleted. Slim 07:38, Dec 15, 2004 (UTC)

Linking to primary source material

Is the following an issue we should discuss on the No original research page, or does it belong elsewhere? An editor is trying to insert on the Kenneth Bigley page links to the video of his murder. I feel this is a snuff movie i.e. pornography and should not be linked to. In addition, the website carrying it carries other very extreme forms of pornography including bestiality. However, although the video is primary source material and cannot be used in the article (only published articles about it can be used as Wikipedia sources), can the video itself be linked to? Normally I'd say yes, of course, but what kind of policy could be formulated to exclude this kind of thing? Do others agree it should be excluded? Sorry if this is a tangent. Slim 05:50, Dec 16, 2004 (UTC)

I'm ambivalent. But I do think we should work toward at least guidelines for these types of things. I created Wikipedia:Ethics and law. So far, it is just links, mainly internal. Maurreen 06:45, 16 Dec 2004 (UTC)

"Normally" published elsewhere

Maurreen, regarding your latest edit i.e. that facts/ideas in Wikipedia should "normally" have been published elsewhere already. I worry that this opens a loophole for unscrupulous editors to squeeze through. Are you thinking of, say, a trial transcript that only a couple of people might have? I would argue that should not be used as a Wikipedia source, because the only way I could check it is if I were to purchase my own copy of the transcripts, at a cost of many thousands of dollars perhaps. I can't offhand think of a situation where Wikipedia could publish something not already published somewhere else, either on the Web or on paper, because the whole point of the verification principle is that readers should be able to check that what's in Wikipedia is correctly reported, and they can only do that if we rely on published material. I'm not saying there isn't an exception, just that I can't offhand think of one. Slim 06:32, Dec 16, 2004 (UTC)

I understand your concern and have changed it back, at least for now. I'm unsure about it myself, but I do think it deserves discussion, and that such a discussion would be more appropriate at Wikipedia talk:Verifiability. Maurreen 06:57, 16 Dec 2004 (UTC)

Examples

On second thought, let me give you a couple quick examples, and then we can decide whether a larger discussion is warranted. I don't know whether these statements are published elsewhere, but I doubt they would be disputed by anyone knowledgeable about the subjects. (I wrote a couple of these, and I might be biased.)

  1. List of common newspaper names
  2. At United States Marine Corps#Appearance: "Marines are often confused with soldiers, who are in the United States Army."
  3. At Copy editing#Tests and tryouts: "Most U.S. newspapers and many other publishers give candidates for copy-editing jobs a test or a try-out. These vary widely. They may be timed or not; they may last an hour to a week; and some are take-home tests. They may include questions, and usually include copy to edit. The length of the copy may be anywhere from single sentences to many articles." Maurreen 06:57, 17 Dec 2004 (UTC)

Point taken. It's just that I can see situations where editors will steamroll their way past that word "normally." And if you were determined to track down published sources for any of the above, you'd probably find them. Perhaps we could leave in the word "normally," but elsewhere make it clear that, if your edit is challenged, you must provide a reputable reference or delete the edit. Slim 06:36, Dec 18, 2004 (UTC)

Right. Will do here, perhaps such notice belongs elsewhere also. Maurreen 06:43, 18 Dec 2004 (UTC)

Footnoting

There is an interesting discussion going on just now on WikiEN-L about the mechanics of amending MediaWiki to allow the placing of footnoted references in an article. If such a feature could be added to the software it would probably encourage editors to be more focussed in citing their sources, to the betterment of the scholarly credibility of what we do. I will probably be corrected, but there seems to be no page devoted to discussing this particular idea. If there is, can somebody please add a link to it, since many of the comments in the thread impinge on what we are doing in these pages. Apwoolrich 07:44, 16 Dec 2004 (UTC)

Yes, it's an interesting and useful discussion. I'm not aware of any page dealing with it. Anything that encourages editor to provide references is most welcome. It's not just the verification issue; it's also that, if editors know they're going to be challenged for reputable references, they'll be more careful with their edits in the first place and the number of disputes should decrease.
I've got an interesting situation going myself regarding the no-original-research rule. I wrote an article called Rat Park, which is the name of a largely forgotten experiment that challenges the notion of drug addiction. I was aware when I was writing it, and other editors have since called me on this, that it was not NPOV because I wasn't able to find any published criticism of the experiment, so my article doesn't contain any, and it really ought to. So I've been searching around, actually telephoning researchers on the other side of the drug addiction issue to ask if they know of published criticism, and they've told me that there isn't any. They've said the study was ignored because it challenged the whole concept of drug addiction and therefore no one wanted to debate it. But they say it was a good study. The people telling me this are highly reputable; no one would doubt their qualifications. And yet I can't put a single word of this in the article because it hasn't been published anywhere. Not that I'm questioning the no-original-research rule, because I agree with it completely, but this kind of experience underlines that we're in the business of verification, not truth, and therefore providing reputable references is essential. Slim 06:59, Dec 17, 2004 (UTC)
Would the footnoting be part of Wikipedia:Cite sources? Maurreen 07:18, 17 Dec 2004 (UTC)
Yes, Maurreen, there are pages about this at http://en.wikipedia.org/wiki/Wikipedia:Footnotes. this is linked from the citations page but all the contributions in it are from earlier in the year. In this link is an experiment into how to make side notes, dated December, but none of the discussions on the WikiEN-L thread appear in there, which is a pity, since I think its an important feture and needs building upon. Apwoolrich 08:20, 17 Dec 2004 (UTC)
Tangent: I diagree that Rat Park is POV. Maurreen 07:27, 17 Dec 2004 (UTC)
I think the footnote discussion they're having is about the technical issue of adding a feature to the software that would allow footnotes to be created easily within articles. Thanks for your comment about Rat Park. I didn't intend it to be POV but I was aware when I was writing it that others would think it so, simply because of the lack of the other side; and indeed, I was dragged over hot coals for it. :-) Slim 07:32, Dec 17, 2004 (UTC)

"References" vs. "Sources"

I'm wondering whether "References" sections would be better titled "Sources". That way, it'd be less likely to be confused with general "Further reading" sections. Maurreen 07:18, 17 Dec 2004 (UTC)

I'd have no objection to calling it "Sources". Some academics might prefer "References", as that's the word used in scholarly books but, as you say, "Sources" might be clearer. Slim 07:34, Dec 17, 2004 (UTC)
"References" is more standard, I think, and standards in language trump logic. :-) Anyway, "Further Reading" should generally go into "References". In reality, there is no hard-and-fast distinction between "things that were used to write this article" (a view of citation that smacks of school assignments, to my mind) and "things that are useful to read with this article." (In scholarly publications, you often add many of your references after writing the article. This will be even more common with Wikipedia...if someone comes along, checks an article, and adds a good textbook reference, there should be no artificial distinction between references that were used by the initial editor and by later editors.) —Steven G. Johnson 21:59, Dec 17, 2004 (UTC)
When I was doing my postgraduate work, we had to stick to the standard definition of "references," meaning works used by the author(s) in the creation of the thesis. In Wikipedia, there are multiple authors so they'll be adding new bits of information all the time, in which case it's legitimate to add the new references too. The point of the references section is so the reader can quickly and easily see what your sources were, and what your train of thought was, which is very important in some academic subjects. But if "references" is used to refer to any further reading, there's no clear way for the reader to retrace the authors' steps. Even if that's regarded as a "school assignment" definition, I think it'd be good to stick to it, because some way needs to be found to encourage editors to concentrate on referencing/sourcing their claims. Over 40 per cent of Featured Articles apparently have no references, which is pretty sad if they're meant to be the best of Wikipedia. Slim 22:32, Dec 17, 2004 (UTC)

Move discussion?

Wikipedia:Guide_to_Layout#References essentially uses "references," "sources," "further reading" and "bibliography" interchangeably. Should we continue this discussion over there? Maurreen 08:11, 18 Dec 2004 (UTC)

The links to the mailing list posts don't go to the proper topics. I don't use the mailing list and don't yet know how to find the proper topics. Maurreen 06:38, 18 Dec 2004 (UTC)

Maureen, here is the link for the archives, which you can view by subject matter. You can also join the list if you want to.
I like what you've done with the paragraph after the "normally" sentence, and I like the "threshold for inclusion . . . is verifiability." Slim 07:58, Dec 18, 2004 (UTC)
To join the mailing list, go here and fill in your details, or if that doesn't work (and it sometimes doesn't), e-mail user:David Gerard at wikien-l-owner@Wikipedia.org Slim 08:14, Dec 18, 2004 (UTC)
Thank you for both. I don't think I'll join the mailing list; I don't like e-mail, but it's good to know where I can find info.
I like what you've done also. I think we're close to finishing this stage; the guidance is good, but a few spots could maybe be a little smoother. Maurreen 08:20, 18 Dec 2004 (UTC)
Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia_talk:No_original_research_(draft_rewrite_5th_December_2004_to_5th_February_2005)/Archive_4&oldid=182253006"