Wikipedia talk:No original research/Archive 6

Archive 1

←

Archive 4

Archive 5

Archive 6

Archive 7

Archive 8

→

Archive 10

Test Case

As a bit of a test, I have listed 9-11:_The_Road_to_Tyranny for deletion. It is a self-published video promoting the author's crank theory about whacky 9/11 conspiracies. The Wikipedia article editors of the page have provided a long, detailed synopsis of the video, which (IMO) (may now be relegated to edit history) serves primarily to promote the video and its theories. It is not otherwise sourced. So, is a synopsis of a self-published (Internet-published, in this case) crank video encyclopedic? If it is, may I load up all my home videos on the Internet and write pages about them :-) -- Gnetwerker 17:25, 14 April 2006 (UTC)

I have not read the articles, but please note that Alex Jones (journalist), the author, has an article in WP. If he is notable enough to have an article, articles on his works can be created. BTW, this video generates 140,000 pages in Google. The problem is not with having an article, but what the article contains. An AfD will fail IMO. I will vote for keep, cleanup and sourcing. ≈ jossi ≈ t • @ 18:25, 14 April 2006 (UTC)

After reading the articles, and digging a bit deeper, I would argue that the video may not be notable enough to warrant a full article. Maybe a delete and merge onto te author's article would be more appropriate. ≈ jossi ≈ t • @ 18:38, 14 April 2006 (UTC)

I know the AfD has closed, but I took a look, and it seems to me that the example illustrates an NPOV problem with articles on theories that have not been discussed in the literature. One way to identify a crank is that no one has bothered to argue with him. Since there appears to be no independent sources about the theory, the only possible critique of the film would be OR. Doesn't one of our policy pages say, in effect, that if one cannot be both NPOV and NOR, then the article doesn't belong? Of course, if we can't form a consensus to delete it, then it matters little what the policy says. Robert A.West (Talk) 02:57, 22 April 2006 (UTC)

controversy over authority

This part of WP:NOR came up in an AFD: "In some cases, there may be controversy or debate over what constitutes a legitimate or reputable authority or source. Where no agreement can be reached about this, the article should provide an account of the controversy"

It's being claimed that it means that if wikipedians can't agree on the notability or reliability of a single source, that source should be included in the article anyway, and the article should note that wikipedians don't agree on its inclusion. I believe that is misreading the above, since such an interpretation would violate Wikipedia:Avoid self-reference. I think the above means that when there is a difference of opinion about a source that the information in that source should be compared to what other sources say and that the differences between those sources should be addressed in the article. Part of the policy was left out when it was quoted in the afd, which reads in full: "In some cases, there may be controversy or debate over what constitutes a legitimate or reputable authority or source. Where no agreement can be reached about this, the article should provide an account of the controversy and of the different authorities or sources. Such an account also helps ensure the article’s neutral point of view." Шизомби 02:25, 18 April 2006 (UTC)

I think you are right, but would like to hear from others who have a lot of experience here. Ultimately, what we editors think of a particular source is not as important as what other published accounts say about a source (e.g. if one source actually and explicitly calls into question the validity of another source). I would only add that often disputes among well-informed editors in fact are an index of disputes among authors of verifiable sources. Slrubenstein | Talk 03:10, 18 April 2006 (UTC)

I agree with Schizombie's interpretation. The sentence means that, where no one source stands out as authoritative, an account of the controversy regarding the subject should be given — not the controversy regarding the source, unless the latter is part of the subject. SlimVirgin ^(talk) 03:23, 18 April 2006 (UTC)

• Because that section uses passive wordage in stating, "there may be controversy..." the original policy-writer's intent is unclear-- and thus, my reading of the section is slightly contrary to what the editors have suggested above. Obviously, an article couldn't very well include a discussion of disagreements between WP editors; that would be Original Research itself. My initial impulse is that rather than referring to a controversy between Wikipedians about the legitimacy of a source, the section may instead be about a controversy among sources themselves as to whether a particular one is reputable. This goes towards what Slrubenstein suggests above about "disputes among well-informed editors". Providing an account "of the different authorities or sources" may mean, in effect, stating that some "authorities" disagree on the merits of others as sources. For example, the Holocaust revisionism article, has a discussion of the trial and controversy between David Irving and Deborah Lipstadt, and shortly afterwards there is this statement (which I suspect the folks above would agree with): "Finkelstein's work is rejected by much of the mainstream Jewish community as well as many scholars."

  However, an additional interpretation is the section does mean that in those rare cases of a strident disagreement between Wikipedians over the legitimacy/reliability of a source, the portion of an article that is based on such a source might include a disclaimer ("an account of the controversy"), such as, "The legitimacy of the source for the following section is disputed". This would be no more a violation of "Avoid self-reference" than putting an "Original Research" or "NPOV" tag on sections of articles in dispute. (Note that "Avoid self-reference" itself is a style guide and not a policy, and likely needs to be cleaned up, too.)

  Both of these views, IMHO, are appropriate, and not mutually exclusive; but in any event, the particular section being cited should be clarified with more direct language.—Leflyman^Talk 07:49, 19 April 2006 (UTC)

Clarification to Top-of-Page Warning

I've edited the warning at the top of the page to make clear that "changes" do not include grammatical corrections, links, clarifications, or examples. As I hope I've made clear, I do not agree with this policy, but I am deferring to those more experienced than I. In any event, my view is that bad policy clearly stated is better than ambiguous bad policy. When the main page is unblocked, the statement there should be changed too.Ragout 05:41, 19 April 2006 (UTC)

• I removed your modification of the notice because 1) it was unnecessary ; 2) it was grammatically horrid; 3) included an inaccurate claim that adding "clarifications or examples" to the policy does not require consensus -- on the contrary, it still does. In the case of the present disagreement, the consensus of those who actually understand the policy agree that SlimVirgin's changes were in accordance with it.—Leflyman^Talk 22:23, 19 April 2006 (UTC)

Of course it's necessary: witness all the controversy regarding "changes" vs. "clarifications." Your other points (and Askolnick's similar point) are well taken. Thanks. I tried to change as few words as possible, but the result seems to have been poor grammar and clarity. I'll try again. Ragout 03:38, 20 April 2006 (UTC)

Ragout, in your next proposal, the most important is to qualify "consensus" with "of those who actually understand the policy". In fact, in practice, consensus is almost always used with this implicit qualifier, not only for edits on the policy, but also for edits on ordinary articles. What is interesting is that most of the debates around the policy pages are between those "who actually do not understand the policy" and those "who actually understand the policy". Many amongst those "who actually do not understand the policy" try to clarify it. They are often supporters of views that are held by minorities in the Wikipedia community, but still perfectly verifiable and significant views. They often have less edits because they get discouraged with the policy before they can get more. The latter oppose. They even suppress the opinions of those "who actually do not understand the policy" using the argument that they do not understand the policy and do not have enough edits. Perhaps they are afraid that those "who actually do not understand the policy" are just trying to have a policy that they can understand and use with authority to present their view (and get more edits). -Lumière 04:37, 20 April 2006 (UTC)

Lumiere, don't you think this defense of your conduct would be better placed where it belongs, on the Requests for Comment-Lumiere page? [1] Once again, you appear to be hijacking a discussion to promote your personal agenda. In so doing, you are only providing your accusers more evidence of your disruptive conduct. Askolnick 13:15, 20 April 2006 (UTC)

In the course of editing the top-of-page notice, I eventually realized that it is not the same as the main policy page notice. So I'm going to edit the talk page notice to make it consistent. I think there needs to be a very strong argument for having two different notices. Ragout 03:55, 25 April 2006 (UTC)

Right after I wrote this, I noticed that SlimVirgin had edited the top-of-page notice that appears on WP:OR, WP:V, and WP:NPOV. The intention seems to be to discourage edits to the policy pages. Since this was done without discussion, I reverted it too. Ragout 04:11, 25 April 2006 (UTC)

To be clear, on April 22, SlimVirgin changed the 2nd sentence of the top-of-page notice from this :

Feel free to edit the page as needed, but please make sure that changes you make to this policy really do reflect consensus before you make them.

to this:

Please make sure that any changes you make to this policy really do reflect consensus before you make them.

And this was done while both WP:NOR and WP:V were protected! Ragout 04:20, 25 April 2006 (UTC)

Does size always equal quality of reputation or relevance of research?

This is my first day here so I can't be too terribly demanding, but I really must at least ask the question after reading this policy regarding what qualifies as "reputable."

Charles, what a breath of fresh air you are. We've had new editors turn up trying to change the policy no less. Demand away. :-) SlimVirgin ^(talk) 12:13, 20 April 2006 (UTC)

"...Does it have a large or very small readership?...does it have a large, permanent staff?...If you heard that the publication you are about to use as a source was considering publishing a very negative article about you, would you (a) be terrified because you suspect they are irresponsible and do not fact-check; or (b) feel somewhat reassured because the publication employs several layers of editing staff, fact-checkers, lawyers, an editor-in-chief, and a publisher, and will usually correct its mistakes? If it is (a), do not use it as a source. If it is (b), it is what Wikipedia calls 'reputable'."

While I certainly agree that fact-checking and responsible publication practices are essential to take any paper seriously, this doesn't mean there must be a "large, permanent staff" to do so. Ever heard of I.F. Stone? There are many publications that told the truth during the buildup to the U.S. invasion of Iraq as well, should they be ignored because the New York Times was beating the war drums at the time?

Reliable newspapers and publishing houses do tend to have a permanent staff, while not invariably huge, usually not tiny either. Can you think of an example of one you'd regard as reliable that doesn't? SlimVirgin ^(talk) 12:13, 20 April 2006 (UTC)

I thought Charles' example of I. F. Stone, publisher of I. F. Stone's Weekly, was an excellent example, perhaps the classic example. Doug Henwood's LBO and the Cook Political Report are reliable self-published newsletters. Josua Marshall's Talking Points Memo also springs to mind. Ragout 13:40, 20 April 2006 (UTC)

Such examples are particularly poinant regarding smaller-scale matters in local communities. With the national, state and regional press naturally ignoring what can often be bitter and brutal political and economic battles in rural areas, who are we supposed to turn to for research? Sometimes the one or two established daily newspapers in the area both get it wrong, are we to assume they are always right just because they have the cash flow and discount newer or smaller publications don't?

I hope the established folks here don't take this the wrong way, but as someone who is starting up a community paper, it'd be nice to see the open source and free-thinking community on the Internet be a little more open to research that isn't churned out by media conglomerates which more and more tend to shy away from the kind of real investigative journalism that makes history. Especially in small communities such as Humboldt County, the alternatives are few, but we are just as deserving to have our history recounted.

--Charles Douglas

Size is not the main consideration. Wikipedia insists on verifiability from reputable sources because it is a tertiary source, and is entirely dependent on the quality of the information in the primary and secondary sources it draws from. If information in Wikipedia is wrong, we need to be able to say, it came from this source. We therefore limit the sources that we may use in hopes that quality sources will get things right most of the time. We look to peer-viewed journals for scholarly subjects, to books from publishers that have an established record and professional editors, and to periodicals that fact-check material before publishing it. On the converse, we do not use self-published books, supermarket-tabloids, blogs and web-forums and other sources that have not established a reputation for fact-checking, except under carefully limited circumstances. I can't say whether a new community will be accepted as a reputalbe source by other editors, but I would look for attribution (the name of the author) and indentification of sources for any article that would be cited in Wikipedia. I will also comment that I have seen 'histories' in comunity papers that got things very badly wrong because the author didn't bother to check his or her facts. -- Donald Albury^(Talk) 12:04, 20 April 2006 (UTC)

Charles, I've inserted a couple of comments into your post, and also want to add that the point of the policy is that for a publication to be what we call "reliable," we expect it to have some kind of fact-checking process or editorial oversight. A very small publication that is, in effect, self-published can't be regarded as a reliable source, no matter how good it might be in reality, because we simply have no way of judging how good it is. SlimVirgin ^(talk) 12:13, 20 April 2006 (UTC)

Charles gave a convincing example of a reputable self-published newsletter, and I've added a few more. The paragraph in question says "there is no clear definition," of a reputable source (outside of academia). So it doesn't rule out self-published sources, just suggests that self-publication is an important factor, and a very big strike against the source. Ragout 13:55, 20 April 2006 (UTC)

The rules about self-published sources are described on WP:V. SlimVirgin ^(talk) 04:37, 21 April 2006 (UTC)

Thanks for the pointer. But getting back to Charles' question about I. F. Stone, it seems like Stone's newsletter would pretty clearly be one of the rare exceptions mentioned on WP:V (were Stone still alive & publishing). Ragout 05:16, 21 April 2006 (UTC)

That's what I meant by saying it was dealt with in another policy. Where the self-publisher is an acknowledged expert in the field, or a well-known journalist, the material may be used, though still with caution. SlimVirgin ^(talk) 02:59, 22 April 2006 (UTC)

Punctuation correction needed

"where an article (1) makes descriptive claims the accuracy of which is easily verifiable by any reasonable adult without specialist knowledge, and"

should read

"where an article (1) makes descriptive claims, the accuracy of which is easily verifiable by any reasonable adult without specialist knowledge, and"

-- SGBailey 13:17, 20 April 2006 (UTC)

Thank you. :-) SlimVirgin ^(talk) 04:35, 21 April 2006 (UTC)

No necessity to cite

The policy should explicitly state that just because someone put something in an article without a citation, it doesn't automatically mean it's "original research." You would be amazed at the number of people on Wikipedia that don't understand this. I had an arbitration case against me that accused me of original research, and the ruling was against me and I was put on probation. But, the arbitrators never even bothered to request sources for ANY of my edits. I was ready to supply sources for any edit that was questioned, but, they never bothered. They either just didn't care to take the time to check if my edits were indeed original research or they just didn't understand the policy. I'm always careful to enter in information that I can present a source for if someone requests it. This shouldn't happen to anyone else. Please mention in this article that not attaching a source to an edit does not automatically "original research." Do not claim or convict someone of "original research" without requesting sources for their edits. Thanks. RJII 02:53, 22 April 2006 (UTC)

I am curious, why didn't you volunteer the sources? It would seem the obvious reply to make. While WP:CITE is a guideline, not a policy, I can't see much reason not to cite a source, unless the point was really trivial and you were 99% sure it could also be found in one of the already-cited sources. Ideally, shouldn't the sources be in the article's bibliography, so that an interested reader can do further reading? Robert A.West (Talk) 03:13, 22 April 2006 (UTC)

WP:V is policy, Robert. WP:CITE is just a style guide: a how-to. SlimVirgin ^(talk) 03:31, 22 April 2006 (UTC)

I didn't volunteer any sources, because there was never any specific edits that were accused of being original research. I've contributed a huge amount of edits to Wikipedia, so, what was I supposed to cite? It was just a vague charge of a pattern of "original research" by a couple people trying to get rid of me. During the case I pleaded for any arbitrator to request a source for ANYTHING that I entered in Wikipedia, but they couldn't be bothered --they just didn't care or didn't understand the policy. I'm guessing they were thinking "let's just put him on probation so we don't have to deal with it --no harm done." And, I do very often attach a source to my edits, but not for everything. RJII 03:18, 22 April 2006 (UTC)

Again, it needs to be made explicit that there is no policy requirement that an editor attach a source to any of his edits. All that is required is that the information is sourceABLE upon request. RJII 03:25, 22 April 2006 (UTC)

I would consider policy statements that unsourced material may be deleted or moved to the talk pages, for example, to be a policy requirement. There may be more explicit statements I don't recall right now. But remember that this is an encyclopedia, with encyclopedic standards, and like any other encyclopedia, sources and references need to be given, not to prove the information is correct, but so that readers can use them to expand their own research. This is what an encyclopedia is for. Coyoty 02:14, 23 April 2006 (UTC)

It should at least be a guideline, if it isn't already. Imagine that you write some nice proza based on a book that you have but you fail to provide it in the references. A couple of years later when you are busy doing gold mining in Timbiktou, another editor who used another book with a differing opinion asks for the reference on the Talk page. No answer. Obvious outcome: deletion or text replacement. Harald88 10:17, 23 April 2006 (UTC)

I think what Coyoty was thinking of is in Wikipedia:Verifiability#Burden of evidence, which says, The burden of evidence lies with the editors who have made an edit or wish an edit to remain. Editors should therefore provide references. I think that is fairly explicit. If you want a edit to remain, provide a reference when you add the material. It may be harder to dig up the reference later when the material is challenged. -- Donald Albury^(Talk) 10:20, 23 April 2006 (UTC)

Sure, it's a good to provide sources. But, I still have a legitimate point. It's not a requirement that a source is cited, unless one has been requested. This needs to be made clear, because people are being persecuted for "original research" when their edits may, in fact, not be original research. The only way to determine if something is truly original research is to request a source. The mere lack of an attached source in an article does not prove that it is original research. (A policy that required every statement in an article to have an attached source would be absurd). RJII 17:11, 24 April 2006 (UTC)

Unprotecting

I think everyone has cooled down, so I am unprotecting. I urge anyone who wants to make a substantive edit to discuss it here first. If there is a violation of 3RR by more than one person in the next couple of days, I think it should be protected again. Slrubenstein | Talk 16:56, 24 April 2006 (UTC)

No necessity to cite (second)

Why was the mention that there is no policy requirement to cite a source deleted? This is a true statement. It needs to be clarified that it's wrong to assume something is original research without requesting a source. An unsourced statement may well not be original research. Most sentences in Wikipedia ARE unsourced, but are not necessarily original research. RJII 17:32, 24 April 2006 (UTC)

You should go to WP:V if you want to discuss the need to cite sources, RJ. If an editor requests a source, and you don't provide one, they are entitled to remove your text, whether they believe it's OR or merely unsourced. If you're annoyed about an arbcom ruling, you have to take it up with them, not try to change a policy to reflect your opinion about the ruling. SlimVirgin ^(talk) 17:43, 24 April 2006 (UTC)

I'm well aware that if a source is not cited upon request that the material should be deleted. I fully support doing that. I do it myself. If I find something dubious, I request a source. If none can be provided in reasonable time by anyone, I delete the information. I expect and hope others would hold my edits to the same standards. I'm not attempting to "change a policy." I'm attempting to point out that there is no policy requring that an editor cite a source. Just because there is no source attached to a statement in an article, it does not mean it is original research. The way to determine if it is original research is to REQUEST A SOURCE. This needs to be made clear. To just assume something is orignal research because there's no footnote attached to a statement, is improper. Do we really want people going around deleting all unsourced statements on Wikipedia without attempting to determine if it is original reasarch or not? What is important is that the information is sourceABLE. RJII 17:53, 24 April 2006 (UTC)

The policy does say that "citing sources and avoiding original research are inextricably linked: the only way to demonstrate that you are not doing original research is to cite reliable sources ..." I think your point is best brought up on WP:V, which is where sourcing is discussed. SlimVirgin ^(talk) 18:09, 24 April 2006 (UTC)

Yes, of course "the only way to demonstrate your are not doing original research is to cite reliable sources." However, there is no policy requiring one to demonstrate that he hasn't conducted original research. There is no policy requiring the citation of sources. It should be stated that "the way to attempt to verify if something is original research is to request a source." RJII 18:16, 24 April 2006 (UTC)

This policy describes what OR is. People can therefore spot it, or know they're doing it themselves, with or without a source. Please take discussion of sourcing to Wikipedia talk:Verifiability. SlimVirgin ^(talk) 18:22, 24 April 2006 (UTC)

Certainly this article should suggest HOW to determine if something is original research. It doesn't. It's negligent in that respect. The way to determine if something is original research is to request sources (or look for them yourself). This needs to be made clear. You can't just assume something is original research because a sentence doesn't have a footnote attached. RJII 18:24, 24 April 2006 (UTC)

WP:NOR is clear enough about what is required vis a vis sources, and what is or isn't OR; it doesn't require a lengthy discourse about methods for finding original research. Jayjg ^(talk) 19:02, 24 April 2006 (UTC)

I'm not requesting a "lengthy discourse." I'm requesting that it be stated that just because no source is attached to a statement it is not a foregone conclusion that the statement is original research, and "the way to attempt to verify if something is original research is to request, or look, for a source." RJII 19:27, 24 April 2006 (UTC)

There are a lot of things in life you aren't required to do, but if you don't do them, the results are mediocre and no one wants to use your work. Businesses don't succeed by doing only what's required of them, and artists don't get well known by putting in minimum effort. If you think including sources is a chore instead of part of making a good article, and you think making good articles is too much work, then maybe you should do something else more satisfyingly adequate. Coyoty 21:03, 24 April 2006 (UTC)

That's not my point at all. Do you realize how ridiculous the encyclopedia would be if we attached a source to EVERY SENTENCE? I attach sources to everything that I think is not common knowledge or may be controversial, but I don't attach a source for every single sentence I add to Wikipedia. That would be insane. It needs to be made clear that just because a source isn't attached to a statement that it is not a foregone conclusion that it is original research, and that to determine if it is, sources should be requested. It's very simple. RJII 04:01, 25 April 2006 (UTC)

And do you realize how useless the encyclopedia would be without sources cited? Citing a single sentence will always be the exception. Depending on many factors, citations can be for the entire article, a section, a paragraph or a sentence. And do multiple citations make articles in scientific journals or well-written histories look ridiculous? I will say that refusal to provide a source raises a strong suspicion of OR. The policy is that if anybody requests a citation for anything in an article, it should be supplied. Period. -- Donald Albury^(Talk) 13:14, 25 April 2006 (UTC)

Why am I not getting through here? Of course a source should be supplied upon request. My point is that UNTIL a source has been requested it should not be assumed a foregone conclusion that a statement is original research. There is no requirement to attach a source to a statement. Therefore, to help verify if something is original research one should request (or look for) a source. That should be made clear in this article. RJII 14:21, 25 April 2006 (UTC)

Point to where someone has been accussed of OR before anyone asked for citations. I want to see what you are complaining about. -- Donald Albury^(Talk) 17:04, 25 April 2006 (UTC)

Maybe you haven't been dealing in controversial articles. That's all I deal with. I can't even count the number of times I've seen myself or others being accused of original research without being asked for sources or things being outright deleted as "original research" without a source ever being requested. This is routine on Wikipedia. RJII 17:16, 25 April 2006 (UTC)

If this is how you respond to requests for citations, I don't need to spnd any more time on this. -- Donald Albury^(Talk) 19:39, 25 April 2006 (UTC)

The complaint is that unsourced information is summarily deleted. The solution is to source the information and not give any excuse to delete it, simple as that. Coyoty 01:46, 26 April 2006 (UTC)

So attach a footnoted source for every single sentence in Wikipedia? That's ludicrous. RJII 04:17, 26 April 2006 (UTC)

If the information in each sentence is from a different source, then it will require a citation for each sentence. But, in most circumstances that will not be necessary. There is nothing ludricous about that. -- Donald Albury^(Talk) 11:08, 26 April 2006 (UTC)

Original information exception

This is very similar to the "no need to cite" discussion above, but that seems more oriented to cases were a person has citations but they don't see it necessary to put it up. I want a clarification on when the information in the article falls far below the "research-bar" and is just information. For example, if you work at Microsoft and happen to know that the new version of Windows had been given the in-house nickname "skyway" and you wanted to put that information on the entry, would you have to cite it? You didn't do research, you are the person people would goto for research. Nor is this the sort of thing that would be published in a "peer-reviewed" journal. This is a far-cry than the crack-pot physics theories wikipedia is trying to rule out by this policy. What if you worked with a philosopher who had an entry on wikipedia (as Susan Haack has an entry, a woman who is currently my prof) and you happened to know her middle-name? Or nick-name in the department? That sort of information is likely never to be published anywhere to be cited. Furthermore, it doesn't seem to need such citation to be allowed. What I am arguing for is an Original information exception to the No original research policy. This is for information that is too "small" to be research and information that the poster of the information might be an adequate source. —Preceding unsigned comment added by Atfyfe (talk • contribs)

If the information is notable, it will almost certainly be published somewhere. To use your example, I'd be surprised if Windows in-house nicknames aren't published, or mentioned in interviews, etc. And it isn't at all clear to me how one would define "small" for such purposes. JoshuaZ 22:13, 24 April 2006 (UTC)

No, this goes back to the Emo (slang) article, where you're dealing with no verifiable sources and no unbiased common knowledge (just look at all the hateful comments left in the article's vandalism). There is a lot of notable, unbiased information, but the only place where it has been published is on Wikipedia. All other possible resources are personal sites or sites making biased attacks. I'm sure there are more examples than this topic, but this is the only one I've personally dealt with.--The ikiroid (talk parler hablar paroli 说 話し parlar) 23:28, 24 April 2006 (UTC)

Remember, verifiability is a policy. We only get around to considering 'notability' if the material is verifiable. One problem is, if you can't supply reliable sources for verification, how do you know it's notable? -- Donald Albury^(Talk) 23:36, 24 April 2006 (UTC)

Exactly right. SlimVirgin ^(talk) 04:39, 25 April 2006 (UTC)

See: "Talk:Unofficial combat badges of the United States Army" for the origins of this question. The kind of situation there is probably what the user is talking about. -Husnock 15:49, 25 April 2006 (UTC)

Tertiary Sources

SV, you are incorrect in your quick RV comment about tertiary srouces. Wikipedia contains a number of 1911 Britannica articles that are out of copyright. These are Tertiary sources (just like Wikipedia). Tertiary sources out of copyright can be included. In no stretch of the imagination does copying a tertiary source make it a secondary source. --Tbeatty 21:12, 26 April 2006 (UTC)

There is a level of abstraction between copying a (out of copyright) tertiary source and citing it for a fact. I would much prefer that we stick to the former and avoid the latter. --Mmx1 03:53, 27 April 2006 (UTC)

These are not copies. They are Wikipeida articles that are edited like any other article. The rules are for articles. We simply allow these "tertiary sourced" articles to exist. They don't violate the rules so the rules shouldn't be written to imply that they do.--Tbeatty 05:11, 27 April 2006 (UTC)

We cannot say that sources are not required, only to say later that they are. ≈ jossi ≈ t • @ 03:45, 27 April 2006 (UTC)

We have have a number of usnourced 1911 Brittanica artilces. In fact it's a Wiki project. We take them on faith from a tertiary source. --Tbeatty 05:11, 27 April 2006 (UTC)

Huh? The 1911 Brittanica is the source for those articles. When we put the {{1911}} tag on articles, we are citing the source. Whether we copy the PD text with no changes, or copyedit it, the 1911 Brittanica is still the source. -- Donald Albury^(Talk) 11:58, 27 April 2006 (UTC)

It is a source, just not a very good one. We all know there are things that are flat out wrong in 1911EB, and need to be edited out. What's really needed is higher quality sources for that material. - Taxman ^Talk 13:48, 27 April 2006 (UTC)

Agreed, but until more reliable sources are found, we go with verifiability, not truth. -- Donald Albury^(Talk) 14:43, 27 April 2006 (UTC)

Please read the context of this discussion. Brittanica is a tertiary source. Citing Brittanica is citing a tertiary source. If the edit was to make it policy that tertiary sources cannot be used, then the Brittanica project would be against policy. --Tbeatty 16:33, 27 April 2006 (UTC)

And in practice, it is not always easy to decide whether a source is secondary or tertiary. If an author refers to both primary and secondary sources in writing an article or a book, is that work then secondary (as it uses primary sources) or tertiary (as it also uses secondary sources)? -- Donald Albury^(Talk) 21:45, 27 April 2006 (UTC)

I think you ae mixing up two issues. First, I personally am against using encyclopedia artiles as sources, NOT because they violate NOR (they do not) but bcause I think it is just silly or one encyclopedia to use another encyclopedia as a reference. If the other encyclopedia is so good, why should people com here and not go there? Moreover, what makes that encyclopedia any good is that it is based on real esearch; so should we. Second, what is the specific status of 1911 Bitanica aticles here? I suggest that they are not sources in the sense of this policy. When Wikipedia started out it had few articles and great ambitions. One way to bridge the gap is to ceat stubs - otherwise crappy articles, really just placeholdes for future, real articles. Since 1911 EB is public, we in efect used it to create stubs. In other words, no one usede the 1911 EB as a source for an article, the original and entire 1911 EB article was used as a place-holder. As we work, I think most would agree we need to rewrite all these EB articles, and replace them with good up-to-date articles. Until we do that, someone using Wikipedia can at least read the 1911 article. Using 1911 means we have far fewer gaps than new and groing encyclopeida would normally have. That is all. It is not a real article, or a source for an article, just a stop-gap place-holder, a temporary thing. Slrubenstein | Talk 19:28, 29 April 2006 (UTC)

Help with possible original research

Hey, I need help at Earth Day. There might be some original research involves and the page needs a neutral third person to check it. __earth ^(Talk) 03:43, 29 April 2006 (UTC)

I will take a look. ≈ jossi ≈ t • @ 04:07, 29 April 2006 (UTC)

What counts as a reputable publication?

What does this section have to do with this policy? Hyacinth 22:09, 29 April 2006 (UTC)

1. It is an obligation of Wikipedia to its readers that the information they read here be reliable and reputable, and so we rely only on credible or reputable published sources. See " What counts as a reputable publication?" and " Reliable sources" for discussions on how to judge whether a source is reliable. john k 22:16, 29 April 2006 (UTC)

The point is (and I thought that this was clearly stated, but now I can't find it in the article,... then where did I see the explanation?), that it's still original research if you use your own homepage as source. Harald88 20:30, 30 April 2006 (UTC)

Sidebar?

Is this recently added sidebar {{Associations/Wikipedia Bad Things}} appropriate in a policy page? ≈ jossi ≈ t • @ 15:26, 3 May 2006 (UTC)

I kind of like it, Jossi, but I won't mind if you want to remove it. SlimVirgin ^(talk) 20:22, 4 May 2006 (UTC)

Copyright issues

Let me try to briefly summarize this surprising policy and its blatant contradiction with copyright policy, as I see it (please see Wikipedia:Copyrights):

One: No new, previously unpublished material is allowed. Fine. Strange, but fine.

Two: If a contributor writes a new article (or adds new text to an existing article), based on previously published materials (books, magazines, webpages, newspapers, etc.) then almost invariably a self-appointed vigilante will appear, will decide that such material is, must be, in violation of such "copyrighted" reputable publications, and will delete or slash it, and altogether cripple it miserably.

This is no invention, it's a fact that happens too often in Wikipedia. Wonder how much traffic is devoted just to battles consisting of removal and replenishment of materials in articles. I also wonder how, despite this seemingly perennial confrontation, has Wikipedia managed to reach the astounding number of a million-plus articles. Amazing.

Can someone please help clarify this confusion? AVM 20:39, 4 May 2006 (UTC)

I'm not sure what your question is. In order to avoid OR, we should cite our sources, and in order to avoid accusations of plagiarism or copyright violation, we should also cite our sources. So the two policies are mutually reinforcing in that sense, not contradictory. Also, please sign your posts on talk pages. Cheers, SlimVirgin ^(talk) 20:21, 4 May 2006 (UTC)

We're looking for an encyclopedic summary of published mainstream thought on our subjects. The only time any real conflict between WP:NOR and copyright concerns should ever emerge is in cases of there being only a single source on the subject of our article. If that happens, don't get too ambitious; we need to then restrain ourselves to brief paraphrase. If you have an example of your concern, feel free to link to it. Jkelly 20:24, 4 May 2006 (UTC)

Your edit of your original question didn't really clarify where your confusion is coming from. Jkelly 20:47, 4 May 2006 (UTC)

Hi AVM. There isn't really a contradiction. Wikipedia is not a place to add original research. In other words, editors shouldn't add articles that are something they made up. As for the ssecond point, copyright law applies to Wikipedia as it applies elsewhere. So editors should not just copy material wholesale from somewhere else because it is copyrighted. However, copyright covers the expression of an idea, not the idea itself. So if you write an article on MSNBC (say) that describes a new fad for chocolate-covered bats, then you can't just cut and paste the lietral text of the MSNBC article into a Wikipedia article, but you can add an article about chocolate-covered bats and reference the MSNBC article to provide a verifiable source showing its not original research.

This is how Wikipedia works. Of course, my characterization of copyright above is a simplification, but its generally accurate. Good luck, Gwernol 20:31, 4 May 2006 (UTC)

RfC: Interpreting WP:NOR in the context of examples of Mnemonics

Talk:Mnemonic#RfC:_How_should_WP:V.2C_WP:NOR.2C_and_WP:CITE_be_applied_to_unsourced_examples_of_first-letter_mnemonics.3F. The article contained about seventy-five unsourced examples of "first-letter mnemonics," probably representing a mix of well-known but uncited mnemonics, unpublished orally transmitted folk culture, and original creations. Should the WP:V, WP:CITE, and WP:NOR policies be interpreted as allowing such material, on the basis that it is self-verifying (i.e. anyone can see by inspection that the initial letters of "Kinky People Can Often Find Good Sex match those Kingdom, Phylum, Class, Order, Family, Genus, and Species, and the source of the mnemonic is of no practical concern)? Dpbsmith (talk) 22:27, 5 May 2006 (UTC)

I see no harm in this, Dpb, and as you say, people can see for themselves whether they're right or not. SlimVirgin ^(talk) 22:33, 5 May 2006 (UTC)

Me too. There is no major harm in this. --Siva1979^{Talk to me} 16:59, 7 May 2006 (UTC)

Original research in external links?

There's a discussion in Wikipedia talk:External links about whether or not original links can constitute original research. Wikipedia:External links lists, among links to avoid:

Any site that contains factually inaccurate material or unverified original research, unless it is the official site of the article's subject or it is a notable proponent of a point of view in an article with multiple points of view. (See WP:RS for further information on this guideline.)

However, I was under the impression that WP:NOR only applied to the actual content of articles. Having read it along with WP:RS a bit more carefully, I understand that exterior webpages about something that hasn't been talked about in reliable sources can also constitute original research. Is that right? How could we make things clearer on Wikipedia:External links?

(This is pretty much a copy of what I posted on Wikipedia talk:Reliable sources) Flammifer 16:39, 7 May 2006 (UTC)

This policy would seem to mean that links to mainstream journalism would constitute original resource. The key word would seem to be "unverified", but I'm not really sure what that means. If a Washington Post article says something about a subject based on interviews, leaked documents, or whatever, that would seem clearly to consist of "original research" by the Post reporter. I suppose one could say that this is "verified original research," but I'm not sure what that means. I think we really ought to get back to "reliable sources." Original research presented by "reliable sources" like mainstream journalism is acceptable to link to. But the policy would seem to currently exclude such things. john k 23:05, 8 May 2006 (UTC)

Original research by others is fine so long as it's published by a reliable source; it's unpublished OR i.e. OR by Wikipedians that we don't use. SlimVirgin ^(talk) 10:33, 9 May 2006 (UTC)

John k:as far as I can tell, the policy is ok with original research in reliable sources.

SlimVirgin: But how about the case in between - original research that was never published by a reliable source (which would cover most websites), but not written by a wikipedian either? We shouldn't use it as a source for the article, OK, but how about just listing it in the external links? Flammifer 11:46, 9 May 2006 (UTC)

It seems to me that Wikipedia:External links should be changed from "unverified original research" to "original research from non-reliable sources," on the basis of your responses. I think that you can link to sites with unpublished OR if they represent a significant POV. If it's just one guy's crank theories, we probably shouldn't link. I'd also suggest that uncited material is not necessarily original research. A website can contain material which is uncited, but which obviously derives from reliable sources. john k 16:09, 9 May 2006 (UTC)

Flammifer, the standard for external links in what many editors now call the "further reading" section (because the references section usually contains external links too), is much lower than for sources, so it's probably okay to include such links so long as they're not too strange or offensive. It's best not to include blogs or personal websites unless they belong to the person the article's about. SlimVirgin ^(talk) 16:24, 9 May 2006 (UTC)

New synthesis, etc.

The discussion of what constitutes new synthesis or analysis in this article seems to me to be far too broad, and doesn't contain any caveats. Let's repeat it here:

Editors often make the mistake of thinking that if A is published by a reliable source, and B is published by a reliable source, then A and B can be joined together in an article in order to advance position C. However, this would be an example of a new synthesis of published material serving to advance a position, and as such it would constitute original research. "A and B, therefore C" is acceptable only if a reliable source has published this argument in relation to the topic of the article.

Is "A and B, therefore C" really only acceptable if a reliable source has published this argument in relation to the topic of the article? This would seem to suggest that A=An undisputed fact about the subject of the article + B=a dictionary definition of a term defined as "Something possessing characteristic A" cannot be put together to form C="The subject of the article is B", unless somebody has specifically said "The subject of the article is B".

If applied strictly, this policy could be used to exclude virtually any material which is not direct quotations from a secondary source. Every time we paraphrase something, we can be accused of conducting original research "Sure you've shown sources that say that X was born in Finland and has an ethnically Finnish name and his first language was Finnish and served as President of Finland, but it's original research to say 'X is Finnish' because you can't find a source which uses those words." I think we'd all recognize that such an argument is ridiculous, and that saying "X is Finnish" is not original research. But the current wording of the policy seems to support such a view, or at least not to rule it out.

As I write, I realize that an earlier statement, apparently trying to say the same thing, is even worse:

That is, any facts, opinions, interpretations, definitions, and arguments published by Wikipedia must already have been published by a reliable publication in relation to the topic of the article.

This way leads to insanity and complete unworkability, I think. It has the same problems as the quote I discuss above, but is even broader, in that it apparently doesn't actually go to new synthesis or analysis, but to any presentation of facts in the article. For instance, if an article on the Byzantine Empire is offering a sourced statement about the Seljuk Turks - say, what area they originated from - this statement would have to be excluded unless we can find a source which discusses this fact in relation to the Byzantine Empire. This is ridiculous on its face. At the very least, it should say "Any facts, opinions, interpretations, definitions, and arguments published by Wikipedia must already have been published by a reliable publication in relation to the topic under discussion." An article may (briefly) deal with subjects that are only peripherally related to the topic of the article. It is completely ridiculous to demand that any statement about such peripherally connected subjects must be sourced with sources that are directly about the subject of the article.

I'll add, as I think others have in the past, that the prohibition on new synthesis is ridiculous. As I think others have previously pointed out, new synthesis is exactly what an encyclopedia does. The principal meaning of "synthesis" is, according to the American Heritage Dictionary of the English language "The combining of separate elements or substances to form a coherent whole.". Any article consists of synthesis, and the prohibition on new synthesis would mean that we couldn't have any material on wikipedia. I don't even find the philosophical definition of synthesis as "deductive reasoning" to be terribly problematic. Deductive reasoning is often misused by annoying people, but that doesn't change the fact that, given valid premises and valid operation of deductive reasoning, any conclusion of deductive reasoning is a priori true. The problem with people using deductive reasoning is that they generally use it wrong - they either operate based on invalid premises, or else make conclusions that aren't justified by the premises. Banning deductive reasoning would, again, prevent us from doing virtually anything on wikipedia, because any kind of paraphrase involves the use of deductive reasoning. We ought to be very careful about displays of "deductive reasoning," but the way to defeat the problems posed by POV-pushers claiming things are "simple facts" derived through "deductive reasoning" is not to ban the use of what is essentially a basic tool of human reasoning, without which it is completely impossible to write an encyclopedia article. An encyclopedia article is explicitly a work of synthesis, in that it is taking various sources and combining them into a whole. If we can only put together sources which some existing secondary source has already put into a coherent whole already, we are completely useless.

What we really intend to ban here is new analysis. The kind of "synthesis" which seems intended to be banned here seems actually to be analysis disguised as synthesis.

A final comment, which is that these particular parts of the article seem to have been at least partially derived out of disputes over articles relating to the Israeli-Palestinian Conflict. "Tough cases make bad law," says the old adage, and I think it applies here. When it comes to a subject that raises very strong emotions on both sides, there are bound to be issues where policy comes into question. I think it would be best not to base our general policies on issues arising out of such things, because the result is bound to turn out extreme, as I think it has in the case of the WP:NOR policy as it stands today. It is particularly troubling that the sole example of original research in the article is actually a direct quote from one such dispute, with the names changed. I think what is needed on this page is to kind of go back to basics and figure out what is not original research. The current definition would seem to define most of wikipedia as original research, in that it contains new synthesis. john k 22:59, 8 May 2006 (UTC)

Ah, someone who agrees with me, that it is impossible to create an encyclopedia without doing original work in the normally accepted sense. I also agree that the difficulty seems to be that a sensible policy has been steadily expanded to turn it into a dispute resolution tool, with the result that it is in danger of becoming absurd. Sandpiper 15:34, 26 May 2006 (UTC)

I am a new editor. I came to this policy page to understand what was expected of me. If I interpret this article very strictly, all I can do is post information from reputable public domain or free-license sources. If I interpret it with moderate strictness, I can post paraphrased and shortened versions of one reputable source. If I interpret neither strictly nor liberally, I can post paraphrases and abridgements of several sources, so long as the sources are in agreement, or they address different topics and the topics don't interact within my article.

To give a fiticious example, I could write that Peter Piper was born in 1496 (source 1) and opened his pickle factory in 1520 (source 2). But I couldn't write that Peter Piper's parents were born in 1497 (source 2) and Peter was born in 1496 (source 1) because that would imply that Peter was born out of wedlock.

As a new editor I found the entire policy off-putting and tempted me to surf elsewhere.

Also, even before I read the discussion page, I noticed that some parts of the policy tasted like the aftermath of editing wars. Gerry Ashton 20:22, 3 June 2006 (UTC)

Help with what constitutes original research

Recently there's been some strife regarding a group of pages that relate to the discussion on Age of consent. The sub pages (eg. Age of consent in Australia and Oceania for sex) list the laws pertenant to the subject with a brief discussion of what they mean.

My question is, does citing Law clasify as original research? At university we were always told under pain of death two things; First - don't plagiarise; Secondly - no original research. In order to satisfy these two points we had to always cite our sources. This meant pointing to both the Law itself (caselaw and legislation) and pointing to reputable discusions of said laws.

Has there been a position set on this at WP as yet? --Monotonehell 07:21, 9 May 2006 (UTC)

If you mean citing legislation, that wouldn't be original research, although you'd need to be careful that you cite current legislation and that you're not using it in any kind of a novel way. But if you mean citing case law, that probably would constitute OR, because you'd be picking and choosing which cases to cite and how to interpret them. It's safer to use a secondary source for that. SlimVirgin ^(talk) 10:30, 9 May 2006 (UTC)

Thank you, that was my take on the policy. --Monotonehell 11:43, 9 May 2006 (UTC)

Got a question

See two edits here and please clarify, should original research be allowed to make something as these two say, NPOV or accurate or both? Or should wikipedia have original research if an article has no reliable sources except those that editors say are innacurate and NPOV? Some claim original research applies to only "It introduces original ideas" and not to requiring reliable sources as citation. DyslexicEditor 10:56, 9 May 2006 (UTC)

Can I use Time as a secondary source?

Can I use this edition of Time magazine, featuring articles on Harry Dexler White as any other secondary source? Time Archive Stor stark7 19:19, 9 May 2006 (UTC)

I would say yes you can. Time is a reputable, independent source. Gwernol 19:30, 9 May 2006 (UTC)

Yes, definitely. SlimVirgin ^(talk) 19:33, 9 May 2006 (UTC)

Ok, thanks --Stor stark7 17:32, 14 May 2006 (UTC)

I would agree as well. - Patman2648 23:45, 16 May 2006 (UTC)

Use by cranks

One of the problems with No Original Research is that it is often used by cranks seeking to exclude any claims that might disprove the pet theory they are peddling.

So for example suppose that Prof A has a paper published by the journal of Economics describing some fancy new statistics technique. The journal publishes the paper because it is a journal of techniques not policy.

Suppose then that the paper contains a conclusion that black is white on the basis of this new statistical technique. Advocates for the B=W theory then peddle the paper in the blogosphere and it is peddled on Wikipedia.

Since the theory is clearly a crank theory no journal thinks it worth rebuttal. The cranks then refuse to allow any negative comment on the crank claim because there is none in scholarly journals - but the paper is torn to shreds in the blogs. --Gorgonzilla 20:40, 14 May 2006 (UTC)

That's correct. If the black=white theory is published by a reliable source, it may be written about in Wikipedia. If it's bad enough and important enough, someone will publish an article about it. We're not here to publish what Wikipedia editors think is true, false, or important, but what reliable published sources are writing about. SlimVirgin ^(talk) 17:34, 14 May 2006 (UTC)

I agree but I think that there needs to be some statement to the effect that a single statement in a single journal does not amount to irrefutable proof. Account has to be taken of the number of papers citing the original claim, the date of publication etc. If a paper has just been published in a journal and six professors tear it apart in their blogs I think it is reasonable to cite the scholarly rebuttals. --Gorgonzilla 20:40, 14 May 2006 (UTC)

Blogs may be cited so long as they're written by widely acknowledged professional expert in their field and there's no reasonable doubt about their identity, but caution should be used, because if the comment were interesting or important enough for us to publish, someone else will have published it, and if they haven't, that should tell us something. Also, there's nothing standing between a professor and the act of publication on his blog, so he's being less careful than he would otherwise be and there's no third-party editorial oversight. SlimVirgin ^(talk) 03:26, 17 May 2006 (UTC)

Question on synthesis of Maps

I have a question about a map showing language distributions. The author has apparently used several language map books to make an image of the distribution of the German language (and dialects), pre-, and post- ww2. Se these two discussions

• Talk:German_language#The removed Maps
• Wikipedia_talk:German-speaking_Wikipedians'_notice_board#Disputed map

Does this constitude synthesis forbidding the use in wikipedia? --Stor stark7 17:40, 14 May 2006 (UTC)

The author has apparently used several language map books I would like to add that the author didn't present to us only language map books as his source, but rigged voting results(ignoring such facts as moving Germans to the region of voting or the fact that different nationalities besides Germans have voted for Germany), allegations against Polish state, data that comes from war situation when German military was present in Poland. All of this contradicting scholary sources indicating much more smaller presence in Poland by German military(including Reich statistics which aren't likely to downplay it). There is no clear source for the image and inclusion of emotional attacks against Poland in the text given as "source" seems to indicate POV. --Molobo 17:50, 14 May 2006 (UTC)

Edits in April

Sorry to dredge up the past, but SlimVirgin seems to have been making a WP:Point when she (?) added the Chicago Manual of Style example to the article (which looking at the edit history was not an uncontroversial change). The example is taken directly from a dsipute in which this user was involved at Talk:Norman_Finkelstein#OR. (Note, I was also involved in this dispute and opposed SV's point of view). Regardless of who was right, it was surely inappropriate for this user to edit the policy while citing it in the dispute, so I ask that the inclusion of this example (and the strict interpretation of OR which it illustrates) should be reconsidered (and of course accepted if there is a consensus in its favour). Cadr 17:47, 14 May 2006 (UTC)

Wikipedia talk:No original research/archive5 Jkelly 00:25, 15 May 2006 (UTC

Did a consensus ever get reached? I couldn't find any evidence of it in the discussion (on archive 5 or archive 6) The repeated suggestion that it was only "new" editors who were opposing SlimVirigin's interpretation of OR on the Finkelstein article was patronising and false -- I've been here for just over 3 years. Anyway, I suppose it isn't feasible to have another discussion of that kind, but it does seem rather inadvisable for SlimVirgin to have made such extensive edits immediately after (and in fact during, so far as I can see) engaging in a debate over the interpretation of the policy. That stands regardless of the user's overall contribution to drafting Wikipedia policy. I apologise for not looking through the talk pages myself in the first place, and for bringing this up a month later -- I only just noticed the situation. Cadr 01:32, 15 May 2006 (UTC)

There was a very clear consensus in its favor, as I'm sure you know, Cadr. SlimVirgin ^(talk) 01:32, 15 May 2006 (UTC)

I looked through the discussion pages and I didn't find it. Maybe it's there -- I'm not suggesting it isn't, I looked through quite quickly -- but I'd like a pointer to it. Cadr 01:34, 15 May 2006 (UTC)

I think all of the changes made by SV in April were problematic, and clearly designed as ad hoc responses to then-current content disputes. As I noted above, I think the phrasing of the OR policy which resulted from this period is indefensible and unworkable. I'd prefer that we go back to the pre-existing version and discuss how it could better be elucidated. john k 01:18, 15 May 2006 (UTC)

As I recall, 17 editors who are familiar with the policy confirmed that no change had been made. I added an example of what the policy already said, that's all, and I'm not going to go over it again. John, I'm surprised at you, given that what you're calling the "pre-existing version" says the same thing as the current page does. You'd have to go back around 15 months, and possibly more, to get rid of the bit you don't like, and I think it may be implied even in those earlier versions (though I'm writing this from memory). And for the very last time, I did not cite the policy while editing it to support what I was citing. I cited what was already there, and then added the dispute (once I had stopped commenting on it) to the policy page as an excellent example of the type of OR some users just don't understand. SlimVirgin ^(talk) 01:31, 15 May 2006 (UTC)

SV, you were still engaging in discussion on the Finkelstein article after your edits to the policy, so far as I can see. Edits to the OR policy at around 08:00 on 11th April; comment on the Finkelstein talk page signed at 23:59 the same day, while there was still no clear consensus on that talk page. That's partly what I object to. However, I don't dispute that you only cited the older version of the policy during that discussion. Cadr 01:42, 15 May 2006 (UTC)

If I only cited what you're calling the "older" version (which was the same as the version after my edits), then there's no problem, so I don't see your point. This is the first edit I made, [2] which repeats what the page already says elsewhere, and a few minutes later, I added this example, [3] which was a disguised version of the OR I had seen someone try to add to Norman Finkelstein, and which is a very typical example of the type of issue some editors have difficulty with: they think if A is sourced and B is sourced, there's no problem adding A and B, even if no source ever mentioned A or B in relation to the topic at hand.

So far as I can see, I made one comment to Talk:Norman Finkelstein after that, [4] and it only repeated a point I'd made earlier, then I had nothing more to do with it (and in fact had very little to do with the dispute before that either: I was a briefly interested passerby, no more). SlimVirgin ^(talk) 02:51, 15 May 2006 (UTC)

My first point was just to establish that it isn't strictly true that you "...added the dispute (once I had stopped commenting on it) to the policy page as an excellent example of the type of OR some users just don't understand..." I also rather resent the implication that I "don't understand" OR -- had you thought that perhaps we might just disagree over what it is/ought to be? I would have thought it should be fairly self-evident why concerns might be raised by someone editing a major Wikipedia policy during (or shortly after, depending on how you look at it) engaging in a debate over the interpretation of that policy. It is, perhaps, the sort of thing that there ought to be a policy against ;)

they think if A is sourced and B is sourced, there's no problem adding A and B, even if no source ever mentioned A or B in relation to the topic at hand.

For the record, I don't find that an accurate characterisation of the discussion on the Finkelstein article itself, but we can't go into that here. In any case, I happen to think that's an overly strict interpretation of OR. If Joe Bloggs says "Black is white", I see no problem in citing a source which says "actually, black isn't white", even if no other source has made the connection -- so long as NPOV is not violated.

Black is not white

It's fine so long as the addition of "black is not white" does not "advance a position or opinion an editor may hold, or to support an argument or definition s/he may be trying to propose," as the policy says. On Finkelstein, one of the editors was trying to argue that someone was or was not guilty of plagiarism (I forget which way round it was), which was the problem. SlimVirgin ^(talk) 17:34, 15 May 2006 (UTC)

The thing is that I can't see the relavent distinction between my "black is white" example and your adaption of the Finkelstein discussion on WP:OR. It seems at best a very subtle one. Also, wouldn't your "A, B therefore C" criterion prohibit any reference to the source saying that black isn't white? The article would be saying, "Joe Bloggs says 'black is white', Joe academic says 'black is not white', therefore according to Joe Academic Joe Bloggs is wrong". And that is prohibited according to you, isn't it? Though it seems perfectly NPOV (assuming that there aren't hundreds of other academics who disagree with Joe Academic, etc. etc.) So I'm confused...Cadr 18:05, 15 May 2006 (UTC)

The Finkelstein section was about whether one of the authors had committed plagiarism. So the way to write that is to say "A says he did, B says he didn't, C says he did according to this definition, D says that definition is wrong and that the author did commit plagiarism according to other definitions" and so on, sticking only to what commentators about this situation have said. What the editors on the page did was to move outside that structure. They added their own view: "According to the Chicago Manual of Style, what D did was not plagiarism" (or whatever the example was: I forget it exactly). Three problems with that: first, the Wikipedians are putting their opinions on a par in terms of importance and relevance with the opinions of the main commentators; two, they are deciding that the Chicago Manual of Style is the best source to quote; three, they're attempting to decide the issue. In other words, the Wikipedians have become commentators on the issue. That is what is not allowed.

With your "black is not white" example, it's only fine to add "by the way, black is not white" so long as there's nothing at stake. A real example I can think of from a recent page was about a white supremacist who had said to a journalist: "Go don a yarmulke, dance a seder and drink some small child's blood. You belong with the Jews." An editor had added in parentheses that a seder is not a dance. It was OR (an unpublished synthesis of published material, with the synthesis being "X said go dance a seder, but a seder is not a dance"), but it's accurate and there's no harm in it, because it advances no position.

I still don't see any distinction here. There is plenty at stake in the "black is not white" example, i.e. whether or not Joe Bloggs is right. In the Finkelstein article, it was actually the case that various commentators had referenced the CMS (Dershowitz mentioned it originally, here's a source which challenges his interpretation [5]). Even if this was not the case, I don't see anything inappropriate in providing the reader with a definition of plagiarism from an authoritative source in the context of a debate over plagiarism. No position is advanced by doing so, unless it is stated or implied that the cited definition of plagiarism is correct. Of course, one has to be careful to cite a representative definition (or set of definitions), and so on and so forth, but all of that is obvious. There is no inherent problem with citing some definition(s) of plagiarism in that context. Cadr 22:36, 15 May 2006 (UTC)

In borderline cases, judging which examples sit on which side of the fence will boil down to experience. With experience (and perhaps without it), you'll develop a sense of when you've crossed the line into acting as an advocate of a certain position, rather than just reporting what people have said about it. It's the "acting as an advocate" role that the NOR policy tries to limit. SlimVirgin ^(talk) 18:55, 15 May 2006 (UTC)

Well, if that's a personal "you", I have more experience than you here, in terms of the time I've had my user account. Cadr 22:36, 15 May 2006 (UTC)

No, it was a "one," not a "you," and I meant number of non-minor edits to the main namespace. SlimVirgin ^(talk) 03:22, 17 May 2006 (UTC)

Anyway, as I said in the first post I made on this talk page, I just think the edits ought to be reviewed in light of your concurrent/immediately precedent edits to the Finkelstein discussion (and possibly some other pages according to the talk archive, but I haven't checked this out). I was not aware when I first posted of the earlier talk discussion, which did this to some extent; but as I said I didn't see that a consensus was really reached, which some of the discussion here would seem to bear out. Cadr 14:19, 15 May 2006 (UTC)

No, there was a very clear consensus on that point, and if you look properly you'll see that everything I cited on the Norman Finkelstein page was already in the policy and had been there for at least a year, perhaps longer. It's perfectly standard (and indeed is to be expected and hoped for), for experienced editors to add examples and clarifications to policy pages after coming across issues people have difficulty with, so long as the policy is being expanded and clarified, and not changed. That's what I did, and 17 (or so) other experienced editors who are very familiar with the policy agreed with me. SlimVirgin ^(talk) 17:28, 15 May 2006 (UTC)

Again, I looked at the talk archive, and there didn't seem to be any consensus. All the people who disagreed with you at the start still disagreed with you at the end. Also, re your "look properly" comment, I've already explicitly acknowledged that what you cited on the Finkelstein page was not new. I still think that editing a policy in the same period of time that you were engaged in a debate regarding that policy is a bad idea. Clearly, your edits were influenced by the debate, given that you used it as an example. Whether or not the policy was being changed or clarified is open to debate. Cadr 22:36, 15 May 2006 (UTC)

They were new editors, and at least two of them were only out to cause trouble, as witnessed by the vandalism spree that one of them went on around the same time. You'd do better to pay attention to the 17 or so long-term editors who agreed with the example, and who confirmed that it was just an example of what was already there. Finally, my edits weren't in the slightest bit influenced by a brief discussion about Finkelstein, except that I saw it as a good example of the "unpublished synthesis" mistake. I've been editing this page or its draft for around 18 months, and absolutely nothing happened on Finkelstein that was in any way new. SlimVirgin ^(talk) 23:14, 15 May 2006 (UTC)

Though I take your point about the relative levels of experience, I mostly agreed with what the inexperienced editors were saying. Ragout made some pretty good points which were generally just answered with "we're more experienced than you and we drafted the policy". That's no way to build a consensus. I'm a long-term editor, and I found the inclusion of a distorted version of the discussion on the Finkelstein page rather inflamatory, and rather transparently designed to nudge the policy towards a specific interpretation. There was as I saw it a degree of arrogance involved, since the discussion over the interpretation of OR on the Finkelstein article was still going strong when you decided that your interpretation was correct, and "clarification" of the policy was required. I don't want to put any words into John Kenney's mouth, since his views on this issue seem a good deal more nuanced than mine. But he seems to be an experienced editor who is not 100% convinced that your changes were just clarificatory. Cadr 00:00, 16 May 2006 (UTC)

With respect, I believe you don't understand the policy, and I feel that's true of John too. It actually boils down to "don't add your own opinions." But people's opinions can creep in in various ways, so we have to outline those ways. One of the ways it happens is when editors create syntheses of material they've read elsewhere in a way that builds their case for them, but they think it's okay because it's all true and sourced. So the policy prohibits "unpublished syntheses that appear to advance a position or opinion an editor may hold, or to support an argument or definition s/he may be trying to propose." One example of that was the Finkelstein case (where the editors' position was: X did not commit plagiarism). Another one, if you want to get away from that example, and I think we should, is the Holocaust example I give below (where the editors' position would have been: Hungarian Jews would have acted differently had they been told). Perhaps rather than arging back and forth, you could give me a concrete example, apart from these two, of something that has been, or would be, excluded by the wording of this policy, and which you think should not be excluded, and we can take it from there. SlimVirgin ^(talk) 00:18, 16 May 2006 (UTC)

The Finkelstein case wasn't an example of that at all (even accepting your definition of OR) because Dershowitz himself had cited the CMS, and several other sources had commented on this. This is partly why the use of that example in the policy grates (even with names changed). Anyway, venting aside, I think what's making this difficult is that I basically agree that we shouldn't create original syntheses in order to advance positions, etc., etc., but in the particular case of the Finkelstein article, I didn't see that happening. Even supposing no-one had referred to the CMS, I do not see that any position is advanced by referring the reader to a representative definition of plagiarism. That's just hepful, so long as it's carefully worded. So I feel that your example in WP:OR imposes a certain interpretation on phrases such as "advancing a position" which I find unreasonable. To get your main question, I would suggest the "black is white" example again. Maybe it would be better to use an example with a slightly less absurd proposition, so we can have clearer intuitions. Suppose X is an advocate of an ethical and political philosophy based on evolutionary psychology. X is not one of the foremost advocates of such a philosophy (though he has published a couple of articles in reputable journals), so there is no critique specifically addressing his work which can be cited. However, in the exposition of his ideas, it would not be NPOV to fail to mention that his philosophy is somewhat controversial. It would be appropriate to cite more general critiques of moral philosophies based on evolutionary psychology and briefly explain how they apply to his ideas, without taking a stance on whether these critiques were correct. This would appear to violate "[the]...precise argument, or combination of material, must have been published by a reliable source in the context of the topic the article is about." The word "precise", especially, is getting in the way of writing a good, informative article. WP:OR should not be a catch-all policy as it is at the moment -- we ought to exercise our judgement in deciding where source-based synthesis is appropriate. To my mind, it's appropriate when it doesn't violate NPOV, or introduce patently nonsensical arguments. Cadr 01:42, 16 May 2006 (UTC)

Answering below. We'll have to talk about the Finkelstein example some other time, because I can't even remember it, but it was a very clear example of OR. I'll re-read it when I have time. Response to your new example below. SlimVirgin ^(talk) 02:02, 16 May 2006 (UTC)

Just to note, I don't think I ever said that everything I dislike about the current page came in the recent edits. But I do think that the recent edits took an already misguided part of the NOR policy (the references to unpublished synthesis as forbidden) and made it completely indefensible by explicitly defining "unpublished synthesis" in an incredibly broad manner. YMMV. john k 03:05, 15 May 2006 (UTC)

Here's what I think is another good example, one that I'm currently wrestling with myself, because I'd like to add it to a page I'm working on, but it would be OR to do so. I'm writing about Rudolf Vrba, an Auschwitz escapee who brought news of the camp to the Allies in April 1944. His report was not distributed widely enough or fast enough, and as a result (he argues), the Jews in Hungary continued to board the trains taking them to Auschwitz, believing they were going to be resettled, not killed. Vrba, backed by some scholars, says if they had known the truth, they would have run, hidden, or fought back. Other scholars say this is false and that Vrba's information would have made little difference if it had been distributed earlier, because the Jews did already know about Auschwitz, and even if they didn't, they probably wouldn't have believed it and wouldn't have acted on the information. That's the basic debate.

Holocaust survivor Elie Wiesel has talked about his experience as a Hungarian Jew of being taken to Auschwitz, and has said: "The last transport left the station on a Sunday morning. ... It was less than three weeks before the Allies' invasion of Normandy. Why did we allow ourselves to be taken? We could have fled, hidden ourselves in the mountains or in the villages. The ghetto was not very well guarded: A mass escape would have had every chance of success. But we did not know." [6]

This is clearly relevant to Vrba's point and backs it up, and I would like to add it. But to do so would be original research, because Wiesel has never commented on the Vrba controversy, and no historian has ever mentioned Wiesel's statement in relation to that debate. Therefore, for me to decide to synthesize these two accounts would be my POV and my OR. It is this kind of research that the "no novel synthesis" rule seeks to avoid. SlimVirgin ^(talk) 03:06, 15 May 2006 (UTC)

This is a borderline case, and I think there's a fair amount that could be said on either side of the argument, but I think you're right that it should be excluded. But it seems to me that the potential problem with this material is not that it is an original synthesis. It is that it is an implicit original analysis - that is to say, the Wiesel quote is being brought in as part of an analysis of whether we should believe Vrba's version or his opponents'. As I said in my previous comments that everybody ignored, original analysis presented as synthesis needs to be excluded. But original synthesis as such should not be removed. I think that the older version (the one you say has been around for 15 months) made clear that only certain kinds of original synthesis were invalid - ones that were trying to advance a POV, iirc, was the main mention. But I stand by my point that any encyclopedia article is going to contain synthesis, and that this in no way constitutes OR. john k 03:29, 15 May 2006 (UTC)

It still says that, John: "any unpublished synthesis of published material, where such a synthesis appears to advance a position or opinion an editor may hold, or to support an argument or definition s/he may be trying to propose." SlimVirgin ^(talk) 03:34, 15 May 2006 (UTC)

Indeed it does. But it also says other, much broader things, like any facts, opinions, interpretations, definitions, and arguments published by Wikipedia must already have been published by a reliable publication in relation to the topic of the article and "A and B, therefore C" is acceptable only if a reliable source has published this argument in relation to the topic of the article. john k 03:39, 15 May 2006 (UTC)

Yes, therefore, meaning that if a conclusion is drawn from A + B, that conclusion must have been published by a reliable source. In other words, we may not publish "any unpublished synthesis of published material, where such a synthesis appears to advance a position or opinion an editor may hold ..." The two points are saying the same thing. SlimVirgin ^(talk) 04:08, 15 May 2006 (UTC)

But it depends on what "A" and "B" are. The way it is written now seems to be essentially banning all independent use of human reasoning capability. This is, I think entirely unworkable, and I agree with Cadr that the main issues with this are going to come down to the NPOV rule, not the NOR rule. john k 16:57, 15 May 2006 (UTC)

No, it doesn't depend on what A and B are, and it doesn't "ban all independent use of human reasoning". It prohibits unpublished snyntheses that appear to "advance a position or opinion an editor may hold, or to support an argument or definition s/he may be trying to propose." It's actually very clear. SlimVirgin ^(talk) 17:23, 15 May 2006 (UTC)

But the parts I quoted don't say that it's only when they're trying to "advance a position or opinion an editor may hold, or to support an argument or definition s/he may be trying to propose." It says that this is true in all situations. Which was my point. john k 20:34, 15 May 2006 (UTC)

We can't keep repeating the same words throughout again and again. It always means that unpublished syntheses are not allowed "if they advance a position or opinion ..." etc. It's a question of common sense. This sentence is an unpublished synthesis. SlimVirgin ^(talk) 20:36, 15 May 2006 (UTC)

Well, I think we should try to be clear on issues like that, or we get into trouble with people misinterpreting things. It seems fairly clear to me that the new language added seems to ban all unpublished synthesis. I'm glad to know that this isn't what you were trying to do, but I don't think it changes the fact that right now it's worded in a way that is misleading. john k 20:51, 15 May 2006 (UTC)

I don't think any part of it says or implies that, and even if it seemed to, we have to assume people have common sense. As I said, this sentence is an unpublished synthesis, as are all yours. Therefore, how could all unpublished syntheses possibly be prohibited? SlimVirgin ^(talk) 00:24, 16 May 2006 (UTC)

Example

Suppose X is an advocate of an ethical and political philosophy based on evolutionary psychology. X is not one of the foremost advocates of such a philosophy (though he has published a couple of articles in reputable journals), so there is no critique specifically addressing his work which can be cited. However, in the exposition of his ideas, it would not be NPOV to fail to mention that his philosophy is somewhat controversial. It would be appropriate to cite more general critiques of moral philosophies based on evolutionary psychology and briefly explain how they apply to his ideas, without taking a stance on whether these critiques were correct. This would appear to violate "[the]...precise argument, or combination of material, must have been published by a reliable source in the context of the topic the article is about." The word "precise", especially, is getting in the way of writing a good, informative article. WP:OR should not be a catch-all policy as it is at the moment -- we ought to exercise our judgement in deciding where source-based synthesis is appropriate. To my mind, it's appropriate when it doesn't violate NPOV, or introduce patently nonsensical arguments. Cadr 01:42, 16 May 2006 (UTC)

That would clearly be OR, because if no one has written about it, where is the evidence that it's controversial? SlimVirgin ^(talk) 02:02, 16 May 2006 (UTC)

I think it might help to cite actual examples, because with these hypothetical ones, we'll always end up with "it depends whether this or that." If you cite a real one, we'll have all the variables. SlimVirgin ^(talk) 02:04, 16 May 2006 (UTC)

The only scenario under which your example arises is where you can reference sources that state that X is an advocate of evolutionary psychology, but none of these sources state that X's philosophy is controversial. It would be inappropriate for Wikipedia to draw the conclusion that X's philosophy is controversial when the available sources have declined to do so. Snottygobble 02:38, 16 May 2006 (UTC)

I admit that it's not a particularly good example, and as SlimVirgin points out, hypothetical examples are always difficult to deal with because there are so many "depends on". But I don't see that this is true in the general case. Maybe X advocates some novel theory of Holocaust denial (it's all a conspiracy involving the Martians, etc. etc.) If there is some fairly detailed explanation of this theory on the page about X, for balance it would only be appropriate to say something like "the general consensus is that the Holocaust really happened and wasn't fabricated by Martians; see Holocaust". The evidence that the theory is controversial is, in this case, simply that 99.99% of people believe in a different theory -- there need not be any specific reference saying "yeah, X is a lunatic". Usual problems with hyothetical examples apply. We must assume, for example, that X, despite being a complete lunatic, is somehow notable enough to have a Wikipedia page, but that's not soooo unlikely. I'm afraid I can't think of a real example right now.

Anyway, unless anyone should miraculously agree with me, I'm going to give up at this point, since it's clear that I'm in a minority. As you can see from my first post, my reasons for bringing this up were partly just that it wound me up to see that the debate on the Finkelstein page had been (in my view) misrepresented as an example on WP:OR, and this lead me to question the interpretation of that policy which the example was intended to support. The policy as it stands seems to have a consensus at the moment with one or two exceptions (judging by the pattern of responses and non-responses to my comment), so there's nothing much I can do. I hope there are no (overly) hard feelings. Although I was certainly annoyed with SlimVirgin, this wasn't meant to be a personal attack/vendeta/whatever. Cadr 14:31, 16 May 2006 (UTC)

No hard feelings at all, Cadr. :-) SlimVirgin ^(talk) 03:19, 17 May 2006 (UTC)

Definitions

For editors to propose new definitions of established terms on talk pages, and to edit the article according to these new definitions, violates the "provides" clause in spirit and in practice. Thus, I've changed "provides" to "provides or presumes." Timothy Usher 06:39, 19 May 2006 (UTC)

Question

I have a question about a concrete implication of the NOR policy. On a page about Anti-X[nation]-ian sentiment, would it be OR to add, in the form of examples with quotes and short comments, links to Internet sites, forums, books etc. which, according to the editor, manifest that sentiment?--85.187.44.131 14:51, 19 May 2006 (UTC)

This seems more like a question of quality of sources than original research. I guess it comes down to what your "short comments" might contain: that could fall under original research. I would recommend adding the links (if they are reliable) and quotes but not adding your own commentary. If the quotes demonstrate your point they don't need editorial comment, readers will make up their own minds. If they don't demonstrate your point, you shouldn't use them in the first place. You may also want to review WP:NPOV when considering which links/quotes to include. Gwernol 15:06, 19 May 2006 (UTC)

I was wondering if collecting these statements and classifying them as X-ophobic (a classification which others might dispute) isn't, respectively, a kind of "field work" and a kind of "original analysis" of primary source material? --85.187.44.131 15:45, 19 May 2006 (UTC)

This is just my opinion. This feels like a borderline case. Are you providing sources to show that X-ophobia exists by doing this, or are you providing original analysis? It will come down to the specifics of the links and the article, I suspect. If the links and quotes are out of context and disputed, then you may be pushing a point of view and violating original research. On the other hand you may be providing reliable sources for an article. Without know the link(s) and article(s) involved its impossible to say for sure. Gwernol 15:53, 19 May 2006 (UTC)

Well, thanks for the reply. I kept silent, because I was waiting to see if anyone else would like to comment. In case you're interested, the page I was wondering about was Anticroatian sentiment (sic! bad spelling for technical reasons). Thanks once again. --85.187.44.131 22:09, 20 May 2006 (UTC)

Question 2

What happens if someone starts a stub giving the bare details, and contributors between them add more, creating what is in effect Original Research? Jackiespeel 16:02, 24 May 2006 (UTC)

IMHO this would be treated in the same way as any other case of OR. Add citations if you can (which shouldn't be too difficult if the information in the article is correct), or if it's absolutely unavoidable, remove the uncited material. Cadr 17:34, 24 May 2006 (UTC)

Every single article on wiki began as a stub, then different contributors added more detail. And between them, it is to be hoped, they did indeed create an original article, and this is exactly what was supposed to happen. Sandpiper 15:16, 26 May 2006 (UTC)

Question 3

Can it really be labeled as Original Research if the majority of wiki users accept it and only a small minority, such as one stray admin, does not agree with them?.....I believe that original research was set up so that one wiki user would not put ideas that have no basis. In Real Life, the rule is that if enough people accept it, it becomes a law or the truth. —Preceding unsigned comment added by Lord Hawk (talk • contribs)

"Can it really be labeled as Original Research if the majority of wiki users accept it and only a small minority, such as one stray admin, does not agree with them?" Yes. The majority of Wikipedia are readers, not editors. We don't know if they agree, because they don't post to the article's talk page. Besides, if a reader either doubts the accuracy of an article, or wants to explore it more deeply, the reader should read the references, not try to figure out what the majority of Wikipedia users think. If there are no references, the article is unverified. Gerry Ashton 20:37, 3 June 2006 (UTC)