Wikipedia talk:No original research/Archive 2

Archive 1

Archive 2

Archive 3

Archive 4

Archive 5

A netural opinion?

This article even is not neutral. I guess that most of the other pages are though. Maby we should present another point of view here so that we dont break own rules of neutrality on our rule page of neutrality. (I'm just joking)

Template for violations

It would be nice to have a template {{original research}}, similar to {{disputed}} and {{NPOV}}, that could be used to mark articles that include material that is arguably original research and needs to be fixed (but which may be subject to dispute and can't be deleted out of hand without igniting an edit war). Currently, none of the dispute tags really fits this problem, which is rampant. —Steven G. Johnson 18:27, Dec 14, 2004 (UTC)

Currently, the closest tag is {{Cite sources}}, which may be adequate. —Steven G. Johnson 02:13, Jan 4, 2005 (UTC)

We have a tag Template:Original research. SlimVirgin 02:49, Mar 9, 2005 (UTC)

And they deleted Template:Cite sources. SlimVirgin 02:50, Mar 9, 2005 (UTC)

Bibliographies and etcographies

I have tried to find an answer to this, but can't: in compiling a bibliography, are the originals the primary source, or is the bibliography original research? Is there another category, "observation" perhaps?

For example, suppose I looked at J. G. Ballard and saw a reference to "The Wind That Came From Nowhere". I look on my shelf and see "The Wind From Nowhere". So I correct it. Or, is it only valid if I find a bibliography that someone else has prepared and use that? It is further complicated because several web bibliographies do list this (presumably) erroneous title, perhaps from a single bad source.

The same argument applies to, for example discographies, episode guides, etc. Can I produce a track listing by picking up a CD, or do I have to find someone else who did it?

Off topic, but perhaps someone can point me to where it is ON topic: what are the limits on bibliographies? Should they be complete e.g. is there any barrier to moving on from J.G. Ballard's books to a complete list of his short work, and each collection it appears in? A complete list of the contents of each issue of Astounding Science Fiction? Is the level of detail found in http://www.snpp.com/episodes.html and each page directly linked from its episode guide too much for Wikipedia? etc. Notinasnaid 10:16, 17 Feb 2005 (UTC)

What you raise is covered in the new draft of the policy (still in progress). You are not making original analytic, synthetic, interpretive, or explanatory claims -- so it isn't original research. Slrubenstein | Talk 16:59, 17 Feb 2005 (UTC)

Recent changes

This is really very good. Two queries:

(1)"Original research refers to original research by editors of Wikipedia. It usually does not refer to original research that is published or available elsewhere, though it might in some circumstances."

I added "usually" and "though it might in some circumstances," though arguably that makes the second sentence contradict the first, so it's not ideal. The reason I added it is that the arbcom, in their August/Sept decision, ruled that the LaRouche editors were engaged in original research when they quoted from LaRouche's original research. The way our sentence above stood originally, they would have been allowed to do that because the research was not theirs, though they would have fallen at the reputable-publication hurdle. Do we want to rely exclusively on that second hurdle to stop editing like that, or do we also want to be able to use the term "original research" to describe Wikipedia editors lifting badly done original research from outside sources?

(2) "For example, Wikipedia would not rely only on an article in a Socialist Workers' Party magazine to publish a statement about President Bush being gay. However, if that same claim was in The New York Times, then Wikipedia could refer to it (and probably also to its claimed sources). The political magazine could, however, be used as a source of information about the party itself."

I'm not sure what's meant by "and probably also to its claimed source" in the second sentence. Do we mean if the New York Times published a story that Bush was gay, and had obtained that story from the Socialist Workers' Party magazine, we could then refer to the SWP? We could, but I wonder if that's worth pointing out, because we can refer to anything that's in the New York Times. But perhaps I've misunderstood. SlimVirgin 02:46, Mar 9, 2005 (UTC)

Examples too broad; major change

For the record if nothing else, I'll repeat (but for the first time on this page) my disagreement with the following: "The phrase "original research" in this context refers to untested theories; data, statements, concepts and ideas that have not been published in a reputable publication; or any new interpretation, analysis, or synthesis of published data, statements, concepts or ideas that, in the words of Wikipedia's founder Jimbo Wales, would amount to a "novel narrative or historical interpretation"."

It's too broad, in my view. I know SlimVirgin disagrees. I don't want to make a big deal out of this particular point.

But I do think that these changes are major. And whether I agree with them or not, I think they should be brought to the attention of the wider community. Maurreen 04:21, 10 Mar 2005 (UTC)

Hi Maurreen, I've put the word out to a few people to stop by and read it, and there's also been a lot of talk on the list recently about the issue of original research and what's understood by it. Let me know if you'd like me to find the link for you so you can read through the archives (I'll get the subject headers and dates for you too so you don't have to pore through everything). SlimVirgin 06:47, Mar 10, 2005 (UTC)

Thanks, Slim. You don't need to look it up in the archives. I was thinking more like at some point, it could be pointed out at RFC or the Village Pump or both.

I haven't Wikipeding much lately or kept up with the changes. Other things going on right now. Ciao. Maurreen 07:36, 10 Mar 2005 (UTC)

New theories

I think the problem with how it's worded now is that what's not wanted are not new theories as such, but just new theories invented by Wikipedians! And you already covered that in the previous par. It's slightly misleading as it stands. I definitely think you could tweak it to ensure that there's a clear understanding that Jimbo didn't want to see a novel narrative on our part not a novel narrative per se! If some prof prints the quantum theory of elephants next week, we want to cover it, no?Dr Zen 07:43, 10 Mar 2005 (UTC)

Actually, reading further, I think you should tone down the whole approach to theories. You could arguably ban any deep discussion of intelligent design because it has not been peer reviewed. This is rather contradicted later when you say that theories with few adherents can be included!

Of course, this is one of those concepts where the thrust is very clear, and we mostly know it when we see it, but putting it into precise, legalistic terms is a lot more difficult. Dr Zen 07:46, 10 Mar 2005 (UTC)

As you say, the emphasis is on "not previously published" rather than new, and by and large that does mean we're talking about original research carried out by Wikipedians, which is not allowed. But it will also cover LaRouche-type situations, where an external group is basically inventing material (theories, purported facts) and publishing them: but because the publications aren't reputable/credible/authoritative (take your pick), we discount them, and treat those claims as though they are unpublished. So perhaps we should replace "new" with "unpublished." You're right: the thrust is clear, but just try writing it down! ;-) SlimVirgin 22:11, Mar 10, 2005 (UTC)

"Unpublished" is better. I think the key thing with the larouchian stuff is it is disreputable, so maybe we need to say more about what is disreputable. One thing we might want to add is that propaganda does not constitute a reputable secondary source, but can be used as a primary source describing the views of its author. ? Slrubenstein | Talk

Yes, good idea: the National Front's website can be used as a source to describe the National Front's views where the story or section is about the National Front. It could not be used as a source in a story about Jews or, indeed, anything else. SlimVirgin 22:35, Mar 10, 2005 (UTC)

Apple Pie

I can see the reference to the apple pie article was introduced after a lot of discussion but, to someone looking at this page for the first time, the reference does look rather bizarre! I expect it is an echo of the great apple pie controversy. Anyway, is it true that the apple pie article is based entirely on primary sources? For example, it tells us what the Random House Dictionary of Popular Proverbs and Sayings has to say on the subject. Even if the primary source claim is right, to me it really doesn't shed any light on the matter. Thincat 16:45, 11 Mar 2005 (UTC)

I am not sure I follow you. The policy claims that articles need to consult secondary sources. This is true for most articles, but there are some exceptions, articles where only primary sources suffice and don't violate the policy. Is it that you do not think this point is clear, or that you do not think "Apple Pie" is the best example? If the former, can you suggest clearer wording? If the latter, can you suggest better examples? Thanks, Slrubenstein | Talk 20:48, 11 Mar 2005 (UTC)

Thank you for your remarks and I'm sorry I wasn't clearer. Taking your questions in turn. The "no analytic ... claims" if an article is "based entirely on primary sources" is clear but puzzling. Surely if in an article I insert something based on a secondary source it does not affect what claims I may make eleswhere in the same article based on primary sources? I suppose "Apple Pie" is a possible example (but unfortunate in that I think it is at least partially based on secondary sources though these parts may have been added after "No research" was redrafted). As it is it seems rather arbitrary. I actually read through "Apple Pie" trying to see what was especially acceptable about it despite its reputed lack of secondary sources. I was being naïve! For something like a recipe for apple pie it may be entirely unknowable whether one is referring to a primary or a secondary source. However, the "current events" example is a good one and I think is helpful. I'll see if I can suggest a redraft (I'd post it in "talk" first) but I may not manage! Thincat 13:28, 14 Mar 2005 (UTC)

"Appropriate to the subject"

I do not see the words "appropriate to the subject" anywhere in the drafts or any discourse. Without this overarching guide, these rules about "original research" are a recipe for needless interference and groundless petty tyrannies. Not all Wikipedia entries need to be footnoted to a single standard.

Examples: Woody Woodpecker is a sensible report, currently without any documentation, apparently based on a good deal of perceptive understanding that would pass as "original research" if one were inclined to fault it. The original research embodied in Necktie#Ties as signs of membership is a contribution that should not be disallowed on petty grounds.

In the studies of mythology and art history, many sensible mainstream remarks are not susceptible to "proof." Look at Baroque and Rococo for examples. Aspects of Rococo that are pointed out in Francois Boucher's painting are either useful to the reader, or they are pretentious and confusing. The image is the equivalent of a cited source. Under rules that offer no flexibility,and which may not be "appropriate to the subject," many articles might be reduced to a mere tissue of cited quotations.

Take a tip from the way the expression "NPOV" has so consistently been reduced to a club to justify users' own POV, and retain some explicit flexibility in this area.

Two excellent rules for us self-important Wikipedians: Avoid unnecessary interference and Keep the reader firmly in mind. --Wetman 05:03, 12 Mar 2005 (UTC)

Wetman, I agree with the need for flexibility, but I disagree that it ought to be explicit, because some editors will take advantage. I also agree with your other two principles, and very much with the last one. Regarding your point that some subjects are not susceptible to proof, we don't require proof. The criterion for entry into Wikipedia is verifiability, not truth. We require that the material, if challenged or likely to be challenged, be referenced to a credible publication. That doesn't mean that every sentence needs a reference. It means that any sentence for which no reference can be found may be removed by any editor. If this were not the rule, we could all just be making things up. It also means that editors should use their commonsense and act like professional researchers, and not try to "get away with" unreferenced editing when no-one's watching what they're doing. They should anticipate future challenges and provide references for edits most likely to be challenged. Most sentences in most articles are not likely to be challenged, so no article will end up as a "tissue of cited quotations." As you say, keep the reader firmly in mind. The reader may want to verify what we've written. We should make it easy for them to do that, without going to extremes, of course. This is how we will build up a reputation for being reliable. SlimVirgin 05:41, Mar 12, 2005 (UTC)

"Appropriate to the subject" makes sense to me, but I'm not sure where it would go. Maurreen 07:39, 12 Mar 2005 (UTC)

I have no objection to adding something on "appropriateness" but the very first paragraph states that research might be excluded if it is not appropriate, so doesn't that make it implicitly clear that research must be "appropriate?" Also, I don't see how Wetman's objections have anything to do with appropriateness. If I understand Wetman's concerns -- well, I think the policy as written does address them explicitly:

In some cases, where Wikipedia articles make no analytic, synthetic, interpretive, or evaluative claims, Wikipedia articles may be based entirely on primary sources (examples would include apple pie or current events).

Doesn't this make it clear that not all articles need to follow this policy strictly? Slrubenstein | Talk 17:14, 12 Mar 2005 (UTC)

I think most readers will understand that documented sources appropriate to some articles would be preposterously pretentious in other. I contend that wit is even appropriate in some articles: but see Talk:Appositive! All too often, challenges oppugning "original research" are based on the challenger's ignorance in the particular area. See a good recently-handled case in point at Talk:Theophoric names. An unnecessarily owlish former page heading Theophory may have been part of the problem, doubtless. The idea that "any sentence for which no reference can be found may be removed by any editor" offers a grim future for all of us. And still there is to be no explicit call for flexibility and common sense because "some editors will take advantage." There are many ways some editors will take advantage: look at the current use of labels. --Wetman 03:35, 13 Mar 2005 (UTC)

Wetman, how would you want to express the flexibility issue? SlimVirgin 07:21, Mar 13, 2005 (UTC)

I think most editors understand that all policies must be applied flexibly (see the introduction to policies and guidelines). I just do not understand what Wetman is objecting to or asking for. He writes:

All too often, challenges oppugning "original research" are based on the challenger's ignorance in the particular area.

And yes, I know that this happens a lot and sometimes an editor might find it irritating. But so what? I firmly believe that this is a good thing in most cases -- an ignorant editor poses what I myself think is a stupid question -- but it makes me realize that many readers will probably ask the same question, and so it needs to be addressed, by adding more information about the sources. The end result is, the article is improved. This is what our process is all about. And in the case of trolls, whose demands and challenges about the sources soon reveal themselves to be based not on a good-faith desire to improve the article, but rather a POV warrior or simply someone who wants to throw a wrench into our process, well, this is easy too: ignore this persons challenges, revert his or her changes, and if they continue seek redress from the community. My point is, in neither case do I see anything that warrents a change to this policy. Wetman also states:

The idea that "any sentence for which no reference can be found may be removed by any editor"

Now, Wetman, you put this in quotation marks as if you are quoting this policy. But where on earth does it actually say this? Nowhere! The policy simply does not say this. So how do we respond when you insist that any sentence for which no reference can be found may be removed by any editor? We tell you this: please read the policy. Your interpretation of the policy is wrong. And Wetman, if you delete a sentence just because that sentence has no reference -- without following the policy and making sure all the conditions that call for a citation apply -- well, we will revert your change. If I am working on an article and someone deletes a sentence just because there is no citation, I will revert that editor. We have policies for a reason, they provide a common point of reference for all editors. You ae not allowed to create your own policy, "any sentence for which no reference can be found may be removed by any editor." That isn't what our policies call for, and you can't get away with it. Slrubenstein | Talk 18:30, 13 Mar 2005 (UTC)

P.S. Wetman, please, please do not invent misleading quotes that misrepresent our policy, it can only confuse newbies. Slrubenstein | Talk 18:41, 13 Mar 2005 (UTC)

I think Wetman may have been quoting me from this talk page. I would urge critics of this policy to read it carefully, because it seems to me that it's very clear. No one is saying that "the sun will rise tomorrow" needs a reference, and that without one it may be removed. But "Doctors have confirmed that Yasser Arafat died of AIDS" does need one, and if the editor who makes that edit doesn't supply a reference, it can and should be removed immediately. It's mostly a question of commonsense. If you're challenged, you must supply a reference. If you make an edit that is likely to be challenged, supply a reference. Don't add guesswork to articles, your own original research, things you think you've heard somewhere, or things you can't provide a reference for. Don't write personal essays. When you do provide a reference, don't use dodgy websites or other dubious publications. That's really all the policy says. SlimVirgin 14:24, Mar 14, 2005 (UTC)

Well, sure Slim, but you yourself were not quoting policy, and Wetman shouldn't have taken it that way. Moreover, you made it very clear that ediotrs need to use common sense and that this approach should be reserved for highly controversial claim More than that, you should actively search for authoritative references to cite.s -- two conditions that Wetman now seems to think s/he is proposing. It's just my impression that Wetman is not reading anything carefully. Slrubenstein | Talk 19:40, 14 Mar 2005 (UTC)

In fairness to Wetman, I think his concerns about appropriateness are well founded if you read the guidance as it stands. The guidance on citing sources is positively draconian. e.g: "More than that, you should actively search for authoritative references to cite." Authorative references for what I wonder? The best practise in Wikipedia and academia is to cite appropriately, e.g. one needs to provide citations for contestable claims. In view of this, I've added the following to the intro. (I've italicised the changes).

"It is very important to cite sources appropriately, so that readers can verify any contestable claims made in the article." :ChrisG 21:33, 14 Mar 2005 (UTC)

Good edit, Chris. SlimVirgin 00:00, Mar 15, 2005 (UTC)

Further comment to explain revert to myself. Citing sources must be appropriate, otherwise a person would have to put a reference for every claim in an article, we wouldn't be able to move for the footnotes. In practise in Wikipedia we ask people to cite appropriately any contestable claim. What constitutes a contestable claim is not something you can always know in advance, so often you would have cite a source for a specific claim in response to someone choosing to contest something. If you can't find an appropriate source, then the other contributor is within their rights to remove the claim. :ChrisG 00:13, 16 Mar 2005 (UTC)

Right. The point of this part of the policy is simply to make clear that the burden of evidence lies with the editor who inserted the claim that's being questioned. We don't want readers and editors to be left wondering what the truth is regarding such-and-such, and having to search around for a source themselves. Any editor (and any reader) should be able to request a reference of the person who inserted the material and that person must supply one, or risk having the material deleted. This applies to any material, but in reality most edits will not, in fact, be contested. SlimVirgin 00:26, Mar 16, 2005 (UTC)

Agreed. And this is also why I suggest 'contestable' claim rather than 'contested' claims. Some claims in certain subject areas will obviously require referencing from the get go. But for large parts of Wikipedia referencing isn't really necessary; because the debate is not so 'political' and/or 'academic'. Though again if you want to make a pop culture article featured you need to reference far more stringently, which again brings us round to appropriateness.  :ChrisG 01:00, 16 Mar 2005 (UTC)

The way I judge appropriateness when I'm editing is by asking myself: if I were the reader, would I want to know where this information had come from, either because I might be suspicious of it, or very interested in it. And that question is going through my mind the whole time I'm writing. (That's over and above citing sources for reasons of intellectual honesty, which is a given.) SlimVirgin 01:09, Mar 16, 2005 (UTC)

An example of officious interference

At Talk:Capitalism today, the following notice appeared:

"I removed this sentence:

The first use of the word "capitalism" in English is by Thackeray in 1854, by which he meant having ownership of capital.

because it is unverifiable. Let's have a proper citation, and a fuller paraphrase, if not an actual quote."

We are going to be seeing a rash of text removed from Wikipedia as undocumented "original research" or "unverifiable" in the near future. Let us remember that "first citations" of English usage are reported in OED. Genuine concern might be better expressed by providing the OED citation, or simply asking for it. Removing text in this fashion is officious: it benefits neither Wikipedia nor the reader—a person not often considered in discussions on this page. Let us be sure that our censorship is not based on our ignorance.--Wetman 23:55, 14 Mar 2005 (UTC)

That's exactly the kind of sentence that needs a reference. However you found out about it, it's surely worth supplying that reference, so the readers and other editors (who are readers too, don't forget) can verify it easily. SlimVirgin 00:04, Mar 15, 2005 (UTC)

Bear in mind, not everything can be linked to on the Internet. Sometimes you need to go to a library.--Ben 06:07, 15 Mar 2005 (UTC)

My point, perfectly clearly made, is that genuine concern might have been better expressed by providing the OED citation, rather than by officiously removing the offending sentence. As SlimVirgin stated previously, "any sentence for which no reference can be found may be removed by any editor." It may not be policy, yet, but it is an increasing practice among some Wikipedians, as my example demonstrates. Common sense is as rare as good manners. --Wetman 04:15, 15 Mar 2005 (UTC)

Wetman mentions good manners, and I suspect he is pissed off at me because I removed this passage in question (and perhaps because of what I wrote above, to which he has not responded). What Wetman fails to provide is the context: the last month or so there has been some heated debate over "the definitions of capitalism." My concern is two-fold. First, I am not quite sure Thackeray means "having ownership of capital." Wetman writes "Genuine concern might be better expressed by providing the OED citation" and I am not sure what he means -- if I knew the citation obviously, I would have put it in myself. As a matter of fact, I suspected that the cite comes from The Newcomes except Thackeray doesn't actually define the word there. So I thought, perhaps it comes from some other essay Thackaray wrote. Bette for whoever put this in to provide the citation. But this is my whole point. Wetman thinks there is something wrong with my having taken the passage out. The problem is not my having taken it out, the problem is someone put it in without any source at all. By the way, OED is the source only for the claim "first use" which I am not challenging. The real source for this is The Newcomes. It is editors who put such information in without any sources or citations who are doing the real damage, not I. When people put stuff like this in without any citation, no one can verify it, no one can look for the passage to see if this is indeed what Thackeray meant. All of us are aware of the way urban legends and myths gain a hold in the popular imagination (an analogous example: there are still many people who thought that in Columbus's day people thought the world is flat. Try to imagine how widespread this claim would be today if whenever it was mentioned the source was provided!). I did not delete the passage, I removed it to the talk section until someone can fix it. This has been common practice at Wikipedia since I have been here. This is not "interferance" and there is nothing "officious" about it. Slrubenstein | Talk 17:10, 15 Mar 2005 (UTC)

People who have been active in reworking this policy are attempting to effect a shift in the culture of Wikipedia. Up until now NOR has been enforced in a sloppy and weak fashion, if at all. People just need to learn to provide sources when they add to an article. I really do not see why whoever wrote this passage could not minimally have written: The word "capitalism" first appears in Thackeray's novel The Newcomes, published in 1954. How hard is this? Slrubenstein | Talk 17:19, 15 Mar 2005 (UTC)

Wetman, I really don't think good editors like yourself have anything to worry about regarding this policy. No one wants to turn articles into lists of citations or quotations, or keep interesting edits out of articles. One of the advantages of Wikipedia is that it can be a little bit quirky, which is what makes it interesting. It's a question of ensuring accuracy and verifiability too, and I believe we can have them all. SlimVirgin 03:58, Mar 16, 2005 (UTC)

Attempts to refute a position

The existing No original research guidelines are fairly clear that original attempts to refute a given position in an article are "original research", yet I don't see that as clearly here. In my experience one of the more significant problems of original research are editors who see a position in an article they don't like, and therefore come up with an argument to refute it. I think we need to make it clear that refutations must also come from published sources, and not just arguments people think up on their own. Jayjg ^(talk) 20:43, 8 Mar 2005 (UTC)

Quite recently - four months ago - someone seems to have added, with little if any previous discussion, that something is original research if "it purports to refute another idea". This is far too vaguely phrased, and - sure enough - is being used exactly contrary to the No original research policy: to prevent criticism of claims too bizarre to have attracted any significant quotable criticism. By this dictum as currently phrased, I could find a quote of someone saying any conspiracy theory too extreme or weird to easily find a quote dismissing - eg, Burton's concept of a "Sotadic zone" - and remove any efforts to point out facts that cast light on its implausibilities or flaws on the grounds that they were banned by this policy. This needs to be removed. - Mustafaa 20:41, 22 Mar 2005 (UTC)

Hi Mustafaa, every part of the draft of this policy was discussed for months, so nothing is there that was just edited in by one person and not talked about. What that phrase means is that editors are not allowed to make cases of their own in order to refute a position. They are only allowed to report the refutations or criticisms of others. For example, an editor recently put forward a refutation of special relativity, and when it was checked by an editor with a PhD in physics, it was found to be quite an interesting refutation; but because it had not been published anywhere, it was not allowed to be edited into the article. So in theory you are right: you could find a crank theory that no one has bothered to dismiss, and you could edit it into Wikipedia without criticism. However, in practice, there would be very few, if any, examples of this, because anything so crankish that no published criticism of it exists, will almost certainly not have been put forward in a credible publication, and therefore would be ineligible for entry for that reason alone. If you did find a crankish, uncriticized theory in a credible publication, then it would still be the case that, in the interests of NPOV, other opposing theories about the same subject could be put forward. NOR and NPOV need to be read together as they're the two core policies and they're complementary. What editors are not allowed to do is to write personal essays, searching out facts and constructing arguments of their own. When you wrote above that this is being used exactly contrary to NOR, did you have a specific example in mind? SlimVirgin 21:13, Mar 22, 2005 (UTC)

Discussed where? Can you provide a link? The specific crank theory I was thinking of is Yehuda Bauer's claim that opposition to the existence of the state of Malaysia is anti-Malay - an idea so ridiculous and counterfactual that no one, so far as I know, has ever attempted to refute it or even noticed it, yet one that has been inserted as a quote into Arab-Israeli conflict (an article, incidentally, whose length would be about halved if this ridiculously excessive policy were followed strictly.) - Mustafaa 22:41, 22 Mar 2005 (UTC)

You mean this: "Yehuda Bauer, for example, says that: "If you advocate the abolition of Israel ... that means in fact that you're against the people who live there. If you are, for example, against the existence of Malaysia, you are anti-Malay. If you are against the existence of Israel, you are anti-Jewish." (As this is quote, it needs a citation by the way.) This is Bauer's argument. If you were to mount an argument against this — an argument of your own — you would be carrying out original research. If you report Bauer's argument, you have to report someone else's argument against him. Of course, it need not be a published argument specifically addressing what Bauer wrote. It can be a published argument that addresses an argument like Bauer's that was made by someone else. You would have to stick very closely to that type of argument though, and not extend it to bring in other arguments you felt you could use. The key is this: as soon as you find yourself looking around for material with which to build a case, you're probably doing original research. SlimVirgin 22:52, Mar 22, 2005 (UTC)

---

I agree completely with Mustafaa — published arguments are sufficient, all my articles are written in this way, and I am confident they meet scholarly standards. But as for his mistaken impression of Yehuda Bauer (a product of the incomplete and out of context quote above), see my correction at Talk:Arab-Israeli conflict#Revisited. El_C 11:23, 3 Apr 2005 (UTC)

---

Arguing speculatively against it might be original research. Pointing out known and citable examples that disprove a claim, and noting the fact that they do disprove a claim, is not "original research" and should not be classed as such; yet the current wording could be, and sometimes is, interpreted to allow such an absurd classification. I can find no authoritative source anywhere to deny that President Bush visited Mars last weekend; if I find a conspiracy theorist who believes that, is it "original research" to point out that this is physically impossible? - Mustafaa 23:08, 22 Mar 2005 (UTC)

Yes, it would be original research for you to point out that Bush visiting Mars is physically impossible. But note (a) you would not find that claim in a credible publication and therefore it wouldn't be in the article anyway; and (b) if you did find it in a credible publication (e.g. the New York Times), then they would have pointed out that it's impossible, so then you could attribute it. I can't think what a "known and citable example that disproves" Bauer's argument would be, because it's an argument, and so you can't "disprove" it as such. Can you say what it is that you want to say about Bauer's argument? If I had an example to work with, I could perhaps more clearly explain this. SlimVirgin 23:15, Mar 22, 2005 (UTC)

I disagree; it would not be original research to point out that Bush visiting Mars is physically impossible, and more to the point it would be crippling to Wikipedia to impose a taboo on saying such a thing. I think we may need more people's views on this... I note that you still haven't given a link to wherever this discussion took place (mailing list, perhaps?) (But since you ask, the immediate example is that of course one can be against the existence of Malaysia without being anti-Malay. One can think that it should be broken up (say, into its separate emirates), that it should be incorporated into a larger state ("Pan-Malayo-Polynesia", maybe?), or even that it should be incorporated into China or Thailand without having anything against the Malay population. It's honestly harder to imagine why he would think this than why it's wrong.) - Mustafaa 23:36, 22 Mar 2005 (UTC)

The Bush example is not a real one, so take my answer with a pinch of salt. It's better to stick to actual examples so we don't go off on tangents. The discussions about this page took place on the draft talk page, which I'll have to look for. But you see you're saying "of course," Bauer's just wrong. That's you arguing, and that's exactly what we're not allowed to do. We can't use our own deduction, or our own opinions. I take your point: that there are times when things are so obviously right or wrong that it's absurd to call for sources. But I don't agree that the Bauer example is one of those. I'd also guess that if you looked hard enough, some journalist or commentator somewhere will have published against that kind of argument, and it's just a question of finding it. SlimVirgin 23:46, Mar 22, 2005 (UTC)

Sorry, it was at the top of this page, but I didn't see it. You can find the archives here Wikipedia talk:No original research (draft rewrite). Best, SlimVirgin 23:49, Mar 22, 2005 (UTC)

Thanks for the link. Judging by this, it looks to me like this particular part of the policy has scarcely been discussed at all, let alone a consensus established. There's a post introducing it (Eclecticology, 3 Dec.) followed soon after by a post arguing against it (Wetman 6 Dec.) and much later by Jayjg's post arguing for it. I'm inclined to suggest that this should be removed pending further discussion. - Mustafaa 00:31, 23 Mar 2005 (UTC)

It's also been discussed by e-mail, and in various forms on the mailing list. It would be better not to remove it, though by all means start up the discussion here. User:Jayjg should be invited to join in as he has a very strong grasp of what counts as original research; as does User:Slrubenstein: he was also involved throughout the draft discussions, but I believe he's away for a few days. SlimVirgin 00:54, Mar 23, 2005 (UTC)

Well, I wasn't notified, but I finally discovered it. Jayjg ^(talk) 03:04, 3 Apr 2005 (UTC)

I don't see it on the mailing list - do you mean Wikipedia or Wiki-EN? - Mustafaa 01:04, 23 Mar 2005 (UTC)

In general reply to SlimVirgin and Mustafaa, perhaps part of the difficulty is that, of the six types of item to be excluded, all are “new” or “introduces” except the refute item. Also, “purports” is rather pejorative. I suggest changing “it purports to refute another idea” to “it introduces an original argument refuting another idea”. Also, although material cannot be maintained unless it is from a reputable source, I think it would be helpful to change the second point of the next paragraph “reported in newspapers or news stories” to “reported in a reputable newspaper”. I also suggest deleting “independently”. Does this mean independently of the editor? . This requirement does not seem to apply elsewhere. Is it permissible to cite one’s own peer-reviewed publications? Thincat 12:00, 23 Mar 2005 (UTC)

"It introduces an original argument refuting another idea..." would certainly be superior to the current wording, insofar as it explicitly does not cover non-original arguments, which the current wording, strictly speaking, does. My worry is that this wording might still be (mis?)interpreted as banning stuff like writing that (to take a hypothetical example) "Baron Von Munchausen claims to have seen a herd of wild orangutans on an iceberg in Greenland, although orangutans are unable to survive at temperatures below zero." (Imagine the potential rewrite: "Baron Von Munchausen claims to have seen a herd of wild orangutans on an iceberg in Greenland. Biologists have not commented on his claim. The Home Manual of Orangutan Care claims that orangutans are unable to survive at temperatures below zero."!) - Mustafaa 03:39, 25 Mar 2005 (UTC)

If an argument is truly "non-original" then it will be cited somewhere, by definition. And "Baron Von Munchausen claims to have seen a herd of wild orangutans on an iceberg in Greenland; the Home Manual of Orangutan Care notes that orangutans are unable to survive at temperatures below zero." is far superior to the first version. BTW, biologists will undoubtedly have commented on this somewhere as well. Jayjg ^(talk) 03:04, 3 Apr 2005 (UTC)

It is not true that "if an argument is truly "non-original" then it will be cited somewhere." If you tell any biologist that you saw "a herd of wild orangutans on an iceberg in Greenland", they will tell you exactly why this is impossible, which makes that argument distinctly non-original (and certainly worthy of inclusion here) - but I guarantee you can find no citation to literally say "it is impossible to find a herd of wild orangutans on an iceberg in Greenland". "Baron Von Munchausen claims to have seen a herd of wild orangutans on an iceberg in Greenland; the Home Manual of Orangutan Care notes that orangutans are unable to survive at temperatures below zero." is far inferior to the first version, stylistically and by massively understating the objection's correctness by failing to make it clear that any sensible observer, not just the HMOC, knows this is impossible. - Mustafaa 05:34, 3 Apr 2005 (UTC)

Biologists will undoubtedly have commented on the temperatures at which an orangutan can exist; a simple mention of that fact, with a footnote, or any other reasonable source, takes care of the problem, and is again vastly superior to the uncited argument in the first version. Jayjg ^(talk) 07:23, 3 Apr 2005 (UTC)

To the contrary. When an argument's correctness is self-evident, uncited is preferable to cited. If someone cites the claim that ~~p => p, the only possible conclusion would be that they know no symbolic logic whatsoever. If someone cites the claim that orangutans can't survive in Greenland, the only possible conclusion would be that they know no biology whatsoever. There is a limit past which citation makes an article look uninformed, rather than improving it. - Mustafaa 08:03, 3 Apr 2005 (UTC)

Uncontroversial facts generally do not require citation; as soon as reasonable controversy is raised, NPOV requires citation. The Baron von Munchausen example is a strawman argument, since extreme minority views need not be cited (and thus no refutation is required). However, if the view that orangutans can live in Greenland does indeed become more widely held, then it must be cited, and countering views must be presented. And if it does indeed become more widely held, rest assured there will be plenty of biologists who respond to the belief and who can be cited to counter it. Jayjg ^(talk) 09:18, 3 Apr 2005 (UTC)

NPOV and No original research

As soon as reasonable controversy about the argument's accuracy is raised, NPOV requires citation. When the only controversy is over whether the argument constitutes original research or not, NPOV does not require citation. As to the allegation that this is a straw man argument, I selected Baron Munchhausen deliberately, although I mislinked him - he is certainly notable, but do you know of any scientists who have bothered to refute his claims? - Mustafaa 10:11, 3 Apr 2005 (UTC)

Mustafaa, I don't know where you get the idea that NPOV requires citation when accuracy is challenged, but NPOV does not require citation when OR is alleged. Any edit that is challenged, on any ground whatsoever, requires a credible citation. If no credible citation can be found, there's probably something wrong with the edit: that is, it's false, and/or original research, and/or a tiny-minority position that should not be mentioned. SlimVirgin ^(talk) 18:47, Apr 3, 2005 (UTC)

If the controversy is about an argument's accuracy, the NPOV demandes citation. If the controversy is about an argument's originality, then NOR demands citation. And if the argument itself is an extreme minority opinion, then it need not be mentioned at all, except perhaps in an article about the author of the argument. Munchhausen is notable (as a liar), but his extreme minority opinions on all sorts of topics are not. They can be cited in the article about him, but they need (and in fact should) not be cited in article about Orangutans or Greenland, as the NPOV policy makes clear. As SlimVirgin has pointed out before, NOR works in conjunction with NPOV and Cite your sources. Jayjg ^(talk) 18:03, 3 Apr 2005 (UTC)

Taking the Baron Munchhausen hypothetical above, I am assuming that the Baron's alleged statement about what he saw would be removed from the Greenland and Orangutan articles without the need to add a sourced statement in the article itself, correct? But in the Baron Munchhausen article, his statement and the statement from the Home Manual of Orangutan Care would both remain in the article, because the Baron's statement, although factually untrue, says something about the Baron's honesty/sanity/etc. However, adding a further, unsourced sentence saying "The Baron was insane when he made the claim of seeing orangutans in Greenland" could be prohibited as original research, correct? Jim^{Collaborator «talk»} 10:57, Apr 3, 2005 (UTC)

Absolutely, because all sorts of other explanations for his claim (lying, really bad eyesight, etc.) could be envisaged. - Mustafaa 11:10, 3 Apr 2005 (UTC)

Yes, that's right. A more typical case of original research in the Munchausen article would be one created by his defenders "e.g. The Baron did not actually make this claim, but his detractors later attributed it to him"; in my experience original research typically comes about when someone sees an argument they don't like, and feels a need to "defend" against it. Jayjg ^(talk) 18:03, 3 Apr 2005 (UTC)

I think in general assumptions about users knowledge should be avoided, and that this is supported by Wikipedia:Explain jargon. Many articles already contain basic information supported by one unobtrusive citation at the end of the paragraph. Many contributors feel Wikipedia's reputation rest on citation, and hope for a day when all facts are supported by multiple references. Hyacinth 08:29, 3 Apr 2005 (UTC)

Explaining what you're saying is certainly important, but is a totally separate issue from citing it. You wouldn't footnote a claim as tautologous as ~~p => p, or cite it as "according to Hofstadter 1980"; rather, you put a good introduction to symbolic logic in the bibliography and leave tracking down the enormous list of people who have said that ~~p => p to the reader, should he or she be bored enough. Conversely, you wouldn't footnote or cite the claim that ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~p => p (an utterly trivial corollary of the former); rather, again, you'd put a good introduction to symbolic logic in the bibliography and leave it as an exercise for the reader. - Mustafaa 09:16, 3 Apr 2005 (UTC)

It ought to be cited as Hofstadter 1980, if the claim was first made by Hofstadter; otherwise not. It's not clear what you're saying here, except that citation gone wild is a bad thing, and everyone agrees with that. (The symbol you're using by the way is for implication; for a tautology, it would be better to use equivalence <=> .) SlimVirgin ^(talk) 18:55, Apr 3, 2005 (UTC)

Thanks to Jayjg for bringing this new discussion to my attention. The whole point of the purported refutation clause had to do with introducing arguments that don't need to be made. Thus if Bauer claims that a united Malaysia is anti-Malay, or if the National Enquirer claims that George W. Bush was abducted by Martians during an airforce training flight, or if Baron Munchausen claims that he saw Orang-Utans in Greenland these are all merely claimns that do not require refutation. The only point that can be challenged is whether the sources if fact made those claims. National Enquirer has been published for many years, and it is not up to the person challenging the claim to go through every issue of the publication to not find something that is not there. It would be the duty of the person introducing the statement to specify the page in a particular issue where the claim is made.

Any attempt to disprove these things only appears or "purports" to disprove them. More often, it is only excess verbiage that further confuses the issue. One needs to distinguish between "Baron Munchausen claims that he saw Orang-Utans in Greenland," and "There are Orang-Utans in Greenland." Only the latter needs to be addressed on a substantive level. The kind of argument under discussion is common in areas relating to what some call "pseudoscience", where a great deal of zeal is applied by opponents refuting claims that were never made. If a seer claims to have communicated with your grandmother's spirit you have no way to prove or disprove that claim by the application of logic. Attempting to do either is futile. Eclecticology 19:20, 2005 Apr 3 (UTC)

Scientific Data Presentation vs. No Original Research

Scientific articles on Wikipedia benefit from being able to show the actual data. So for a while now, I have been donating figures to Wikipedia that replot scientific data from published sources. See: User:Dragons flight/Images.

However, data selection and presentation is something of an art, and data by itself can call for conclusions. In making this post, I am looking for some community feedback regarding how the policy of no original research relates the preparation of scientific figures.

I am assuming that merely replotting and compositing someone else's data (e.g. Image:Instrumental Temperature Record.png) would never qualify as original research, however there are a number of cases that are more ambiguous than that.

In order to offer the best possible presentation, I have sometimes been creating comparisons among data that may not exist in quite the same form in the research literature (e.g. Image:1000 Year Temperature Comparison.png) In some cases, I have been scaling or smoothing the data to make comparisons possible. I believe that all of these manipulations are reasonable and follow well established principles, however, if I am the first one to smooth the data in a particular way or make a particular comparison, then one might argue that it is original research.

Of potentially greater concern is the presentation of scientific data when the data itself calls for a conclusion (e.g. Image:Carbon Dioxide 400kyr.png). I've been fairly careful to not draw any conclusions from any of the plots I have made that can't be supported by other published research (though in some cases better documentation of supporting material is probably appropriate), but would it be a problem in Wikipedia, to say that X data supports the conclusions of Y even if neither X or Y actually chose to point out the connection?

Of the plots I have prepared probably the most problematic (on several levels) is Image:Holocene Temperature Variations.png. First I am combining data in a way that is strightforward, but has never been done before for this particular set of data, and then I am saying that the average thereof supports the conclusions others have made about temperature changes during the Holocene. Even if you agree that it is okay to say that X data supports the conclusions of Y, I can understand someone who might still have a problem with allowing that the average of X1, X2, X3, etc supports the conclusions of Y is an appropriate statement.

Since I am asking for community feedback, there is a related, but somewhat off topic issue on which I would also appreciate feedback. All of the images I have produced are used in articles on Wikipedia, but I have also chosen to use the Image description pages themselves to not only document the data used and describe the construction of the figure, but also to provide some summary of what it means. One user has objected to me, perhaps paradoxically, that the image description page should not be used for extended description and interpretation. In essence saying that those details should only be given in the main namespace. Personally, I feel that since the Image namespace is dedicated to the images that this is the appropriate place for any extended description or explanation of images that might be useful.

Thanks in advance for any guidance.

Dragons flight 21:43, Apr 9, 2005 (UTC)

The crucial test is this: can your replotting of data be used to suggest or confirm a new synthetic or analytic claim, interpretation, or explanation? If the answer is no, I think you have strong grounds for including the data. If the answer is yes, then you should represent the data as it is plotted in whatever source you are drawing on. In the case of Holocene temperature, I don't think it is enough to say that your presentation of the data supports existing conclusions. If you are the first person to use this data in this way to support the conclusions, well, this is exactly the kind of argument that can be published in a peer-reviewed journal mdash; but not here. If others have used the same average to support the conclusions mdash; in other words, if all you are doing is making aesthetic changes to the way the data is being presented – then I think you can add this data to Wikipedia. Slrubenstein | Talk 16:57, 15 Apr 2005 (UTC)

I'm not sure whether the policy prohibits most presentation of published scientific data, but if it does it seems to me to be pretty blatantly a bad policy that would effectively preclude Wikipedia from any usefulness as a science reference.--dsws 19:28, 14 July 2005 (UTC)

Are we offending someone?

"If you have a great idea that you think should become part of the corpus of knowledge that is Wikipedia, the best approach is to publish your results in a peer-reviewed journal or reputable news outlet, and then document your work in an appropriately non-partisan manner." I'm not sure...this doesn't offend. --VKokielov 07:39, 15 Apr 2005 (UTC)

I don't think it is offensive. It is a constructive suggestion for those who might feel frustrated by this policy. Slrubenstein | Talk 14:28, 15 Apr 2005 (UTC)

Steve is absolutely correct. Indeed, this is the same advice that one is given in any field, for any purpose. People who intend to "contribute" to the corpus of mankind's knowledge solely by publishing their own hypotheses on websites are trolls or cranks. No serious and sincere academic, scientist or scholar has a problem with being told that they should publish in a journal. RK 14:32, Apr 15, 2005 (UTC)

I see no reason why it would offend, or why it would be relevant even if it did. Jayjg ^(talk) 16:02, 15 Apr 2005 (UTC)

Perhaps VKokielov could say more about why it might be offensive. SlimVirgin ^(talk) 16:48, Apr 15, 2005 (UTC)

I think I'm letting my emotion get the best of me again. This "you" is about the average Wikipedia reader, no? And the one whom the entire expose is to teach? It looks like poking fun, no less poking fun than would be to tell whomever grumbles about the prices of stamps that he should become president. --VKokielov 22:48, 15 Apr 2005 (UTC)

This was written, I believe, with editors in mind who in the past have attempted to introduce new scientific theories into Wikipedia, theories they have developed, but which haven't gone through peer review. I see what you mean, though: that someone might interpret it as sarcasm. Perhaps it should be reworded slightly. SlimVirgin ^(talk) 01:37, Apr 16, 2005 (UTC)

This interpretation is an unfortunate consequence of the de facto monopoly of research institutions on academic journals and conferences. But don't forget, anyone can publish, no matter what their affiliation or background, with just a little effort making the right contacts. And web pages aren't a bad place to start - a number of articles published as web pages have later drawn serious attention, been published in journals, and been cited. But perhaps I digress. Deco 28 June 2005 18:13 (UTC)

In fact, I think it is written very clearly to address itself not to the "average" Wikipedia editor but to those specifically who have developed scientific theories (and we certainly want these people contributing to Wikipedia too, right?). Slrubenstein | Talk 17:35, 16 Apr 2005 (UTC)

Believe me, they're the last ones to need us to tell them what to do. --VKokielov 02:01, 17 Apr 2005 (UTC)

The great could be seen as a bit patronizing — what about those ideas which are WP-worthy in notability but are far from great and are, in fact, lame(!)? Otherwise, it's a constructive advice that I offer routinely. El_C 21:34, 16 Apr 2005 (UTC)

'Openly' Partisan Sources

The policy says that references to openly partisan sources should be avoided. Unfortunately there are many institutions, particularly Washington think tanks that provide research to order just as there are companies that will perform 'astroturf' (fake grass roots) campaigns for a fee.

I think that it is important that partisan sources be treated as partisan even if they are pretending to be neutral. If an institute has received a donation of $100,000 from Exon it should not be cited as an impartial authority on global warming. If it does not cite its sources of funding at all it should not be considered impartial.

The same can also be said of industry analysts. In the computing industry it is well known that if you want a favorable report from certain 'analysts' the we that you assure that is to either commission the report directly, or if the group does not do that type of research you pay for 'consultancy' or a 'presentation', the going rate being $30,000 to $50,000 for a day of this 'work' or less.

Further, how do we know sources are not "partisan", i.e. presenting a version of the truth convenient for a special interest, e.g. the company publishing the material and its shareholders, a political party, or a funder of the research?

There is a well documented history of omission of facts and re-emphases from US news reports, for instance. Should we exclude the NYT? Michael Moore? Sean Hannity? What about framing the assertions these made and stating the omissions-- is this original research, a "non-standard synthesis" if the particular comparison has not been made before? If so, should we prohibit all original research? I agree that conclusions should not be drawn from the comparisons, but shouldn't it be allowed to make them in the first place? Mr. Jones 07:01, 6 August 2005 (UTC)

Like pornography it is often easier to spot partisan sources than to define them. For example a 'private research institute' with an address in Washington DC should probably be automatically considered partisan. Michael Moore and Sean Hanity are both obviously partisan sources, the NYT is not perfectly objective but considerably better. The big problem in this area comes from the research for hire think tanks.