Wikipedia talk:No original research

• Wikipedia talk:No original research (archive 1), archive 2
• Wikipedia:No original research (draft rewrite)
• Wikipedia talk:No original research (draft rewrite)

IMPORTANT TEXT CHANGETHAT WOULD BRING IN A NEW POLICY ON WHAT IS ORIGINAL RESEARCH

Original Text

Original research refers to original research by editors of Wikipedia. It does not refer to original research that is published or available elsewhere (although such research may be excluded if editors consider the source to be disreputable or inappropriate).

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".

Proposed text

Original research refers to theories, data, statements, concepts, arguments, and ideas that have not been published in a reputable publication; or any unpublished interpretation, analysis, or synthesis of published data, statements, concepts or ideas that, in the words of Wikipedia's founder Jimbo Wales, would amount to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".

Original research is prohibited when produced by editors of Wikipedia if it has not been published elsewhere. (This prohibition does not refer to research that is published or available elsewhere, although such research may be excluded if editors consider the source to be disreputable or inappropriate.)

In summary, if the facts, opinions, or arguments you want to include have not been published already by a credible or reputable publication, you're engaged in original research.

Comment The proposed text is so very wide that 40% of wiki articles would have to be deleted.

• Vfd could and would placed on many many articles.--Poorman 05:24, 6 September 2005 (UTC)

Grocery-story analogy

Hi Vkokielov, I've partially reverted the changes you made. I've left your addition to the intro, which I didn't mind, though others may disagree, as follows:

In order to become known as a reliable source of information, Wikipedia has to impose this restriction on its content. Another way to consider it: what do you expect from your grocery store? Your grocery store doesn't make anything: everything it sells to you, it buys from someone else. Now ask yourself: would you buy from your grocery store if you so much as suspected that what you were buying came from no one knows where? So, too, here at Wikipedia. It's an obligation of Wikipedia to its readers that the information they read here be reliable and reputable.

I've restored your deletion of the "if you have a great idea that should become part of" section, because I think it explains the position quite clearly, but I've deleted "great" because, as you and El C pointed out, it might be seen as patronizing or sarcastic. I also restored your deletion of the sentence saying we'd have to turn away Pulitzer-prize-level journalism or Nobel-level science if its authors tried to publish it here first, as that gets the point across well.

I deleted your grocery story/FDA analogy, which you had replaced the above with, as I felt it labored the point and that the analogy broke down in places, as follows:

If you ran a grocery store, how would you decide from whom to buy your food? Which questions would you ask Mr. Doe if he came to you and said he will sell you the same food you sell now for less than anyone from whom you would buy? If you have an original idea, Wikipedia cannot - and will not - publish it, unless it is good to eat. The only way to make sure it is good to eat is to send it through the FDA of original ideas - the professional publications. Now, of course not every food the FDA forbids will kill you; and so not every original idea which you can't publish is not worth publication. But it would be unfair to sell you beef from England just because mad cow disease doesn't happen very often. In other words, the fact that we exclude something does not necessarily mean that material is bad – Wikipedia is simply not the proper venue for it. We would have to turn away even Pulitzer-level journalism and Nobel-level science if its authors tried to publish it first on Wikipedia.

Others may disagree with me, so of course feel free to restore if that's the consensus. SlimVirgin ^(talk) 05:05, Apr 17, 2005 (UTC)

Thank you. Don't expect me to argue. I'm not sure what I'm doing, editing a Wikipedia policy page. ...e...what am I doing, editing a policy page?!! ;) --VKokielov 06:08, 17 Apr 2005 (UTC)

A chance to engage in original research, perhaps? Sorry, could'nt resist. :p El_C 06:34, 17 Apr 2005 (UTC)

Yes, VKokielov. Well, I'm glad you said that and not me. ;-) Actually, we are allowed to edit policy pages, but the chances are high we'll be reverted unless it's a minor change. It's always best to argue these things out on talk pages first and get some kind of consensus. SlimVirgin ^(talk) 06:37, Apr 17, 2005 (UTC)

I notice the analogy has something of a loophole, which carries through to the wiki case. Imagine wartime. Goods are scarce and people are desperate to get them. In those circumstances, people will take a chance however dodgy the grocery vendor is. Well maybe the meat was classed as fit only for petfood, but its better than starving, and hey the dog's still alive. This holds true here too. What do you do about a topic on which no guaranteed information exists? People still want to know. Best may not be perfect, but is better than nothing. Sandpiper 19:38, 25 August 2005 (UTC)

Just one of many reasons why I think a clear and straightforward explanation of the policy and why it is important is better than any analogy. Slrubenstein | Talk 19:42, 25 August 2005 (UTC)

An article about finding research?

How about an article that helps people find quality research for their article, especially in controversial areas? Like that Google Scholar is an amazing tool to find free online scientific papers? Or that blogs about particular subjects often summarize and explain that latest findings and studies in a particular field? Or that prior research is often summarized selectively by opponents in controversial areas (like gun control), so it is a good idea to read what the major opposing organizations present as evidence. Ultramarine 20:16, 17 Apr 2005 (UTC)

Usenet as a secondary source

Being bold, I just added a paragraph to the "what counts as a reputable publication?" discussing Usenet as a source. Lay on, Macduff, and damn'd be him that first cries, "Hold, enough!" ➥the Epopt 18:05, 8 May 2005 (UTC)

Please don't be bold. This is a policy page and substantive changes have to be discussed here first. SlimVirgin ^(talk) 18:57, May 8, 2005 (UTC)

I must say that I was expecting something more articulate and polite than simply hitting the "rollback" link. Or is that the sum of your argument? For the record, the paragraph that SlimVirgin so brilliantly refuted read:

Usenet is an example of a source whose "reputability" is highly questionable. Most posters to Usenet are effectively anonymous, unaccountable, have unknown expertise, and often have opinions and beliefs outside the mainstream. However, other posters are well-known and reputable authorities in their fields, and the ability to correspond with them over an extended period of time is valuable. Accordingly, Usenet is no more or less useable as a source (not only as a primary source about itself, but also as a secondary source about any subject) than any other popular medium, provided the usual standards of verifiability are applied. ➥the Epopt 20:39, 8 May 2005 (UTC)

I'll put up some argument when I have time. Just to make it clear: I don't, of course, object to Usenet as a primary source (as a source of information about itself), but I do object to it as a secondary source (as a source of information about someone or something else).

Meanwhile, Slrubenstein has given me permission to post the following from him. SlimVirgin ^(talk) 20:56, May 8, 2005 (UTC)

Are many editors trying to use usenet as an authoritative source? If so, perhaps we do need more explanation of the policy vis a vis usenet ... I think this may be a symptom of a bigger problem: people using other web-pages of other encyclopedias as sources. I see the value in comparing ourselves to other encyclopedias, but it seems to me patently absurd that one encyclopedia rely on another encyclopedia as a source. Similarly, it seems patently absurd for one web-page to use another web-page as a source. It is some kind of incest. Computer-ignorant though I really am, in the 10th grade I took data processing (learning BASIC) and the most important thing I learned from a very good teacher was "Garbage in - Garbage out." Slrubenstein

I find that comment about web pages referencing web pages really odd. By analogy, you are saying that paper pages (books) should not be published if they can only reference other paper pages. So it would be logical to ignore any book which did not mention other media? Anyone writing for wikipedia should by now be aware that the volume of material on the internet increases every second. Most of it is junk, but some is assuredly first class. This is a trend which will continue. The media should not be the issue, the problem is quality. Sandpiper 20:02, 25 August 2005 (UTC)

I know nothing about Usenet, but I find it highly unlikely that anyone will run across a topic that has no sources outside of Usenet posts. To me, this makes the need for a specific policy on the use of Usenet citations less critical. Furthermore, I agree with Slrubenstein's comments above (if I understood them correctly): that maybe it's best for us to try to stay away from websites too much. →Iñgōlemo← ^talk 21:39, 2005 May 8 (UTC)

How about something like Pleasant Street Incline? Yeah, if I was in Boston and went to the State Transportation Library or the Boston Globe archives I could find primary sources. But I'm not in Boston and all I have is this knowledgable person posting on Usenet (as well as whatever else people have posted online - that PDF external link is very useful, but again could be argued to be unreliable). --SPUI (talk) 21:55, 8 May 2005 (UTC)

How do you know that the information that can only be found on Usenet is reliable -- in short, how do you know some bored Usenetter isn't trolling you? If you can offer an objective procedure to determine the reliability of a given Usenet post (other than "I know it when I read it"), I'm happy to reconsider. -- llywrch 23:31, 8 May 2005 (UTC)

And how do you know s/he's knowledgeable if you haven't found his/her claims confirmed elsewhere? SlimVirgin ^(talk) 23:34, May 8, 2005 (UTC)

The fundamental question should be whether there is any reason to doubt the veracity of the information? Is it asserting some sort of dubious theory or promoting a crackpot agenda? If not, then why is there any issue with using a Usenet source or any other sort of publicly accessible resource? The information is inherently verifiable, and if it contains any obvious inaccuracies, it will come out it in the wash, eventually. If you insist on having only reputable (by your standards) sources for every factual assertion before it can be included in an article, then most of Wikipedia would quite simply not be here. And if someone is basing an article on information from Usenet, wouldn't you rather have that clearly indicated so the reader can make their own judgements? I mean if you simply bar Usenet citations (or other internet sources that do not satisfy your criteria for reliability), then people will simply add information without providing any sources. I think it is a much clearer message to emphasize first to always cite sources, whatever they might be and secondarily to make it clear that some sources are less reliable than others and may be subject to greater scrutiny and criticism. I think that is all ➥the Epopt was trying to do with his/her addition. While I'm not attached to any particular phrasing, I find the idea quite reasonable. older≠wiser 01:34, May 9, 2005 (UTC)

If there is no reason to doubt the veracity of the information then there is no need to supply a source, i.e. if it is common knowledge. Wikipedia is mostly built out of people writing what they know without backing it up with references. However, if a piece of information is questioned, then there is a need to provide a reputable source.

Usenet would be a perfectly reasonable place to learn stuff; but it doesn't qualify as a reputable source to substantiate something if used as a reference. You might use a usenet post on the talk page to support an argument; but it wouldn't be authorative. If we used usenet posts as references we undermine our standards of referencing. :ChrisG 12:14, 9 May 2005 (UTC)

Because I've been reading that newsgroup for several years, and I have never known rtspcc to post anything false. As I said, you're free to find better sources. But I'm not flying to Boston just to make you happy. --SPUI (talk) 16:21, 9 May 2005 (UTC)

SPUI: I was asking for a general rule for Usenet, & you have given me a specific example. That doesn't answer my question, & doesn't tell me anything more than you trust what this particular Usenetter contributes.

older≠wiser, as I wrote on Wiki-EN, the problem I have with Usenet in general is that the only difference between performing research from Usenet queries & from asking questions over a beer or two is that the text of a Usenet answer can be verified. However, the intent of the text, & the credibility of its author are not easily determined: an answer can come from an expert or it can come from an ignorant dog (to allude to a famous cartoon). As unreliable as the Corporate Media & published books can be shown to be, at least everything that is published is reviewed at least once for content (& hopefully once for spelling & grammar), thus providing some bare minimum level of accuracy.

Further, the rules of No Original Research as they currently read explicitly exclude personal communications -- the example given is what Stephen Hawkings might say over a beer; there is nothing in the current policy that explains why this would be different than drawing material from Usenet, which is an interactive medium of personal exchanges. If the consensus is to include Usenet as a citable source, then this part of the page should be changed.

I though the problem with stephen Hawking and his beer, was that it was off the record. If he was prepared to stand up in public and support it, then it would be the view of an acknowledged expert. If he just tells you in confidence, then you are left with the problem of finding a different authoritative source to confirm it.Sandpiper 20:02, 25 August 2005 (UTC)

As for your point that if I "insist on having only reputable (by your standards) sources for every factual assertion before it can be included in an article, then most of Wikipedia would quite simply not be here", let me agree with you: most of Wikipedia needs better sourcing. (I thought that Wikipedia:Forum for Encyclopedic Standards was created to help address that.) But that is another issue.

More to your point, I see the point of offering sources & citations as a gradual process: one person adds a statement for which there is no source; another alters it, but add a source; the next changes it again, but adds a better source; & still another changes it to the best possible source. In my eyes, the value between offering material that has no verifiable source & offering material that can only be verified from a Usenet source is minimal, & in some cases I'd sooner believe an unverified assertion to one that references some Usenet sources. The only real advantage I see to quoting a Usenet post might be to carefully side-step the problem of adding a personal opinion to a Wikipedia article -- but frequent reliance on this trick only weakens its use. -- llywrch 19:51, 9 May 2005 (UTC)

So you're saying that you'd rather I just didn't include the source? I'll ignore that bullshit. --SPUI (talk) 00:46, 10 May 2005 (UTC)

Are you bothering to read anything I'm writing? This is twice where you have clearly misunderstood what I have written. -- llywrch 03:10, 10 May 2005 (UTC)

To llywrch, if you really are saying that it would be better for someone who makes an edit to an article based on information from Usenet to NOT cite that as their source, then I competely disagree with you. In practice, I agree that in general using Usenet as a source is pretty much the bottom of the barrel in terms of consistent reliability. However, I think it is a terrible mistake to pre-emptively bar it as a source. There is valuable information to be found in Usenet, although it is easy to miss because of the overwhelming volume of drek found there as well.

I think there may be a bit of a realos/fundis division here between holding to unrealistic standards of ideological purity regarding citations and references and trying to work with the reality of the situation as it is. Wikipedia is a community work of volunteers, open to everyone, some with academic training and many more who care little for academic standards. I think the first hurdle is to promote a culture in which citing sources, whatever they may be, is expected. To my mind, sorting out the quality of references can only be done if references are provided. older≠wiser 01:12, May 10, 2005 (UTC)

As I mentioned above, using Usenet as an authority conflicts with the No Original Research rule, & pointed to one point in our policy as they are written where this is clear. I could expound on others.

I've been with Wikipedia for a few years, & remember when we worried that too many users confused "NPOV" with uncritical use of any source that confirmed a given POV -- no matter how unrepresentative that POV was. Usenet was one of the examples mentioned in the past as a source that does not meet acceptibility. While it's possible that I've missed seeing a sea change in the attitude about Usenet -- & at the same time an understanding of how to properly use other questionable sources -- I still doubt enough Wikipedians have thought thru just how to allow intelligent use of Usenet on one hand, while knowing how to fight abuse of its resources on other than a "I know garbige when I see it, & I will delete it repeatedly" basis. Aggressive POV pushers will exploit any opportunities they find. -- llywrch 03:10, 10 May 2005 (UTC)

SlimVirgin quoted me above and accurately represented a general view I have about Wikipedia and the Web. As to this discussion, there is one point I want to add to this conversation: policies cannot cover all possibilities and indeed are meant to be written in such a general way that they can be applied to differenc cases as they come up. Epopt I believe initiated this discussion with a brief comment about usenet. If he is correct (and I will grant that for present purposes) there is simply no need to mention usenet in this policy. Perhaps there is a need for a general discussion of usenet (I know there was a long one on the mailing list) but honestly I just do not believe it belongs here. A general discussion of the strenghts, weaknesses, and uses of usenet belong on the talk page of the usenet article. As to the question of whether it constitutes original research or is a reputable source, I think that has to be debated on a case-by-case basis. Let's wait until someone wants to use usenet as a source on "race" or "Jesus" or "Creationism" or "Fascism" (and so on). Then, the people working on those articles can discuss whether the incorporation of something from usenet violates our NOR policy, or qualifies as a reputabble and verifiable source for that topic. I think our policies are clear enough, right now, that if two people got into an argument over using something from usenet on, say, the Jesus article, the policies as written will be able to guide them in evaluating the appropriateness of that use. In short, this discussion belongs on the talk page of the Usenet article, and more focussed discussions of usenet belong on the talk pages of articles where editors are actually relying on usenet as sources. I don't see the value of talking about it here. Slrubenstein | Talk 21:02, 10 May 2005 (UTC)

Two comments. I find llywrch's absolutist position rather disturbing (and there is a distinction between what you describe as using Usenet as an authority and someone citing Usenet as a source for their contribution). To Slrubenstein, my comments here were prompted by Slimvirgin's argument on Wikien-L (and I paraphrase from memory, my apologies if I am unfairly representing your views here) that since neither No original research nor Cite your sources mention Usenet, therefore any reference to Usenet as a source outside of the context of a Usenet-related article was categorically invalid. Although I agree that Usenet is in general a lousy source, and that most nearly all of the time better sources can be found (though not without some effort) -- I don't agree with her conclusion to bar it's use in Wikipedia outside of Usenet-related articles. So long as that rather peculiar reading of policy doesn't gain currency, I've no problem with leaving things as they stand. I agree with Slrubenstein that the place to hash out the appropriateness or validity of any reference (not only Usenet) is on the talk pages of the article. I just don't want to see a novel interpretation of policy turn into a blunt object used to bludgeon unsuspecting contributors. older≠wiser 01:52, May 11, 2005 (UTC)

I'm in basically complete agreement with Bkonrad (older!=wiser) here. If Usenet (or a Web page - to me, the difference is minimal) is the source for some of the content of an article, then it should be listed as a source. Both the readers, and other editors, can then make their own value judgements as to whether they trust that particular source. (I assume I don't need to point out that being printed is no guarantee of accuracy either - I can show you all sorts of bogus drivel in widely distributed books.) Yes, if better sources are available, prefer them - but if we don't let people cite Internet sources, they just won't list their sources at all, which is definitely worse. At least when they are listed you can i) verify that the article correctly reports what the source says, and ii) independently decide whether you trust the source. Noel (talk) 01:03, 21 May 2005 (UTC)

Grocery-store analogy

Thodin, I reverted your edit because I felt it labored the grocer-store analogy a little by going into detail about bakery and deli items, and price. If you feel the analogy as it stands isn't helpful, could you make a suggestion for change here? Many thanks, SlimVirgin ^(talk) 23:38, May 16, 2005 (UTC)

SlimVirgin,

It really needs a better analogy, but the best I could come up with was to exclude certain parts of a grocery store. Maybe a bunch of encyclopedias in book or CD form might work better. I was hoping someone would edit to improve it and not just merely erase my changes, but you did start a chat.

Another idea is a bakery. That you have to trust what is produced in a bakery -- that the food is safe to eat and what bread is supposed to be. Or if you think the news on TV is lying, you don't trust it. Personally, I'd say comparing it with an encyclopedia you buy or a respected news source (although a lot of news sources aren't so respected and seem biased). So maybe something like if you buy a bunch of encyclopedias from a salesman you except them to be accurate and not just some made up information, badly researched information, or propoganda.

-Thodin (Forgot to sign my name)

Sure, Thodin, by all means come up with something, but could you put it on this page, instead of directly onto the policy page? I was never hugely keen on the grocery-store analogy, and in fact I'm not sure we need one, as the page seems pretty clear to me, but if an analogy is needed, the grocery-store one did hit home, I'd say. But I'd like to see any other suggestion you have. SlimVirgin ^(talk) 21:41, May 17, 2005 (UTC)

I was frankly surprised that the page was even editable by people who didn't have special privladges, but in the history I see everyone tinkering with it. Thing is when I read the grocery store analogy, I immediately thought of all its contradictions. I'd still say just working it by comparing it with a reputable journal or encyclopedia would do better. -Thodin

We're not supposed to edit it, except for minor adjustments, without discussing it here first with other editors, because it's policy. But don't worry about it, as you seem to be new. Another thing in case it helps, if you type four tildes after your posts, like this ~~~~ that will produce your name and the date and time of the post. And if you fill in your user page, that'll stop your name being a red link. Not that you have to, of course. You can remain mysterious if you prefer. ;-p SlimVirgin ^(talk) 22:05, May 17, 2005 (UTC)

Thanks. How do I instant message other users through this system? Also, another analogy idea is buying a car -- I think that would fit the grocery idea with possibly some word changes as people are more picky about a car they trust and that places that sell cars don't make them. Or maybe extending the analogy to a chosing a hospital? Thodin 00:33, 18 May 2005 (UTC)

Sorry, Thodin, I only just saw your post. I don't think people can be instant-messaged through WP; if they can, I've never heard of it. Regarding your suggestions, by all means compose something if you want to, but it would be appreciated if you could post it here first, so that other editors can comment on it if they want to. SlimVirgin ^(talk) 05:43, May 25, 2005 (UTC)

Hi, this probably isn't the place to post this. Who were you trying to reach? SlimVirgin ^(talk) 05:43, May 25, 2005 (UTC)

Quite right, my apologiesSumergocognito 06:11, 25 May 2005 (UTC)

On talk pages addition

I added a section discussing the fact that this policy does not apply to talk pages. While perhaps this should be obvious, I've found that occasionally people cite this rule to support removal of a new user's ideas regarding a subject, which is not only discouraging them from editing but from continuing to think about new ideas. I don't think it says anything that isn't already true, just a reminder. Deco 02:14, 28 May 2005 (UTC)

Wikipedia:Requests for arbitration/RFC

An important policy discussion has started concerning ways in which our content-related polices, such as NPOV, No original research and Verifiability could be better enforced. I've made a proposal to give the Arbitration Committee the ability to consult Wikipedia users who are knowledgeable in subject-areas that apply to cases before them. Such consultation is needed due to the fact that the ArbCom does not by itself have the requisite knowledge to easily tell what is NPOV, original research, or a fringe idea in every field. Please read my proposal at Wikipedia:Requests for arbitration/RFC#Alternate solution #9 by mav. Content subcommittee and comment. Thank you! --mav 02:47, 7 Jun 2005 (UTC)

Bludgeon

It seems to me that this policy is used as a bludgeon by some people to force their POV on to other people. If primary research was as easy as some people implied then we would all have the title Dr. To be original research it has to be more than one sentence or a list of things!

Let me give you an examples:

1. Several historians regard Dresden more as a cultural landmark than anything else and assert that the number of civilians killed was excessive to a criminal degree.
2. General Sir Alan Flemming Hartley was appointed Commander-in-Chief in India on January 5 1942.

The first sentence may or may not be true. The first thing that someone who disagrees with it will say is "This is a POV without references". When the author of the sentence provides references, then the objector is able to say that "this is original research because no one else has linked those sources together before, provide me with a reference which has done that or remove this original research".

You would have to find a credible, published source who said something like: "Dresden was more a cultural landmark than anything else and the number of civilians killed was excessive to a criminal degree." If you are the one who builds the case (even using sources), it's original research; if the source builds the case, it isn't. SlimVirgin ^(talk) 01:08, Jun 12, 2005 (UTC)

No I do not think it is to be original research it must be substantial not inconsequential. Saying that "3 historians have published statements[1][2][3] which can be paraphrased as 'Dresden was more a cultural landmark than anything else and the number of civilians killed was excessive to a criminal degree.'" should not be construed as original research. To be original research it needs to be something more substantial than counting up sources. Philip Baird Shearer 17:43, 15 Jun 2005 (UTC)

If three sources correctly paraphrase as that then there isn't an issue in the first place. Just point out the source. If you're combining different statments from different sources into a conclusion, that may be a different story. See below. - Taxman ^Talk 19:30, Jun 15, 2005 (UTC)

Why in your view is it not original research to say that "three historians have said" if one can not provide a fourth source which has already counted those sources? Philip Baird Shearer 20:09, 20 Jun 2005 (UTC)

I have a hard time believing that question was posed in good faith, as I believe the answer is obvious. It's the same issue as here and I'd still like a response to that. However, I'll answer anyway. For one, counting, if that is all that is needed, is trivial and certainly doesn't qualify as novel. But if the various sources didn't say exactly what you say they did, that could be considered a novel combination of sources. Notice I didn't tell you to say "three historians have said". If needed, the three sources, and what they do say can be noted separately. Your over-reliance on this issue is troubling. Much, much more common than the issue you are bringing up is the simple lack of reliable sources being cited at all, and you know that. So please, focus your energy on doing that, and adding quality content, instead of arguing on talk pages. - Taxman ^Talk 20:47, Jun 20, 2005 (UTC)

At last something we seem to agree upon. As I said to be "original research it must be substantial not inconsequential" and you say "For one, counting, if that is all that is needed, is trivial and certainly doesn't qualify as novel." So I think that the page include a phrase which makes it clear that the "trivial" research should not be deemed to be original research. This is necessary because people are using NOR as a Bludgeon and not IMHO in the best interests of Wikipedia development. Philip Baird Shearer 14:10, 25 Jun 2005 (UTC)

---

In the second case it is possible to prove from primary sources on the web that General Sir Alan Flemming Hartley was C-In-C of the Indian army for two month, from January - March 1942. But the only source Web source for a list of C-in-Cs for India gets this wrong. So should one leave the known mistake in the list of C-in-Cs even when a primary source document on the web clearly shows that the secondary source is wrong? Because synthesising a new list which fixes this problem could be construed as primary research. If it is original research and so should be be in Wikipedia, should Wikipedia publish information which it can be proven is false?

In Europe any "collections of information" [1], which a list is, can be copyrighted so in theory any list on Wikipedia which is not a copy from somewhere else could be construed as original research because under the EC Databases Directive [2] which defines a protected collection as "a collection of independent works, data or other materials arranged in a systematic or methodical way and individually accessible by electronic or other means". So does this mean that all lists on Wikipedia which are not copied from somewhere else (which would often place them in breach of copyright) are original research and should be removed from Wikipedia?

See also the 1995 ruling in the "Magill case" [3]Philip Baird Shearer 11:40, 10 Jun 2005 (UTC)

Data Mining

If one uses data mining technique where one trawls data to see if any links appear and them fit a theory to explain the links, is not considered to be legitimate scientific research. To be scientific research, one is meant to come up with a hypothesis and then look to see if the data to prove or disprove the theory. (Reminds me of a policeman in South Africa who explained to me "By law you have to fire a warning shot. So the first shot hits the man with the gun and the second shot is fired at the ceiling".) Does data mining count as original research on Wikipedia and if so why? Philip Baird Shearer 11:57, 10 Jun 2005 (UTC)

I think you'll need to provide a concrete example as it pertains to Wikipedia. SlimVirgin ^(talk) 01:09, Jun 12, 2005 (UTC)

Why a concrete example let us use a hypothetical example. Suppose that you are looking for all the references for Gorgon so that you can correct a date on the Gorgon page. Using Google you notice that the pages returned show that there is a correlation between the word Gorgon and the geographical spread of the articles returned by the search, so you add a paragraph to the Gorgon page pointing this out. This new fact has not been discovered using "original research" methods, it been acquired via data mining. Philip Baird Shearer 14:01, 13 Jun 2005 (UTC)

Sorry, Philip, I only just saw this response. I have too many pages on my watchlist and keep missing things I'm supposed to look out for. I'm not sure what to make of your query. The phrase "original research" doesn't refer to any research method, be it data mining or any other. It refers to novel material: unpublished material, new arguments, the editor's opinion, or a new synthesis of published facts. The key is that everything in Wikipedia must have been published somewhere else already, somewhere reputable. If not, it counts as original research. SlimVirgin ^(talk) 20:58, Jun 20, 2005 (UTC)

Are you saying "original research" does not mean original research? As you will know, "research" has a specific meaning in academic circles (scientific methodology etc). Philip Baird Shearer 13:54, 25 Jun 2005 (UTC)

Cited sources =/= original research-free

I'm not sure the page makes it clear enough that citing sources, although necessary, is not a sufficient guard against original research. Any academic dissertation is a piece of original research that's absolutely crawling with cited sources. --Angr/_{tɔk tə mi} 00:53, 12 Jun 2005 (UTC)

Hi Angr, citing published sources is a sufficient safeguard against original research, so long as sourced material is not put together in a novel way by the Wikipedia editor in order to build a case. SlimVirgin ^(talk) 01:05, Jun 12, 2005 (UTC)

Well precisely. The "so long as" clause is absolutely necessary. But all too often I see a page up at VfD on the charge of being original research, only to have people vote to keep, arguing that it isn't original research because it cites its sources, and failing to remember your "so long as" clause. I just want to make it clearer that the presence of cited sources alone is no guarantee that a page isn't original research. --Angr/_{tɔk tə mi} 23:09, 12 Jun 2005 (UTC)

Yes, I would agree. If you can find a way to make it clear citing sources is necessary, but not sufficient, and without being too wordy, that would be an improvement. - Taxman ^Talk 12:11, Jun 13, 2005 (UTC)

How does one judge if the sourced material is put together in a novel way. If it is not novel then it is a copy and liable to copyright violations. I do not think that a sentence or two can be defined as "original research" if it were then Doctorates would be handed out with the rations. Besides I have known people in the UK, who were dismissed from their academic posts for not doing enough original research, if all they did was to collate sources and quote other peoples research. And yes they did draw conclusions from others research but that did not count as original enough for them to remain at a first class institution. Philip Baird Shearer 13:54, 13 Jun 2005 (UTC)

It's fairly simple. If you draw conclusion X and that conclusion is not drawn in any of the sources, you have done original research. Citing a source and even quoting small portions is not a copyright violation in the least. You are missing the point on original research. Something does not have to qualify as ground breaking research to fail the no original research policy. - Taxman ^Talk 15:18, Jun 13, 2005 (UTC)

Now this is foolish! Would you break ground on Wikipedia, PBS?! --VKokielov 23:57, 10 July 2005 (UTC)

Statement of fact "No Allied service men were tried at the Nuremberg war crime trials". Statement of fact "Major Denis Healey was not tried at the Nuremberg war crime trials". Original research: "Major Denis Healey could not be tried at the Nuremberg war crime trials, because he was an allied service man"; because there is no source which says such a thing. (BTW I am sure that Denis Healey might have wanted to "squease the rich till the pips squeak" but I am sure he was talking metaphysically :-) and am not implying that he ever committed any war crimes or crimes against humanity) I think to try to call such a sentence original research is to make "Original research" into a bludgeon. I think that the cry of "original research" is used far too often on Wikipedia talk pages. One could never claim that such a sentence as original research in any historical publication, but you seem to be advocating that it is in Wikipedia article and I do not think that helps with writing concise, precise, and informative articles. Philip Baird Shearer 17:36, 13 Jun 2005 (UTC)

Wikipedia is a reference work. It is very different, in many ways, from other written works. What is fine in other formats is not fine in Wikipedia if we want to have any relevance or accuracy. See Wikipedia:Verifiability. If you think NOR is a bludgeon, you are thinking the wrong way. An appeal to NOR to get sources for a claim is an appeal to improve the quality and reliability of Wikipedia. It is very important that we don't put our own opinions or analysis into article, and the NOR policy is the formal way to avoid that. Instead we should cite the most reliable external sources in the event of a challenge. The NOR policy is strongly supported by the NPOV and Verifiability policy. - Taxman ^Talk 18:37, Jun 13, 2005 (UTC)

Name one credible academic journal which would consider publishing as original research a sentence which says: "Major Denis Healey could not be tried at the Nuremberg war crime trials, because he was an Allied service man". If an undergraduate tried to present such as original research project to their professor, she would die laughing. Just because it is a combination of two know facts does not make it is not original research. I am in favour of a ban on original research but it has to be substantive original research. I may not be able to find any article which says "Major Denis Healey could not.." but I can find two references which say that he was a British soldier during world war II and that allies were not tried for war crimes at the Nuremberg war crime trials. To ban the use of such a sentence because it is original research would mean that huge chunks of this encyclopaedia would have to be re-written into a set of quotations. It text would be even more stilted than it already is (A Camel is a horse designed by a committee) and would then probably run into copyright infringements. Philip Baird Shearer 22:48, 13 Jun 2005 (UTC)

You keep missing the point: Wikipedia is different from other written works. What would apply in other situations does not apply here. Stop trying to act like it is the same. And cited facts can be woven into good prose. It's just a little harder, but certainly very possible. - Taxman ^Talk 23:12, Jun 13, 2005 (UTC)

I do not think that I am missing any point. I do not think that the "original research" ban

You are in that you keep trying to apply the standards of research publishing or employment at a research organization to deciding what is OR at Wikipedia. - Taxman ^Talk 19:25, Jun 15, 2005 (UTC)

should be applied to things like "Major Denis Healey could not be tried at the Nuremberg war crime trials, because he was an Allied service man" even though it is a synthesis of two facts which although they can be found, can not be found in the same document. Given that do you think that:

1. The sentence "Major Denis..." is original research and should not be allowed.
2. The sentence "Major Denis..." is original research but is should be allowed.
3. The sentence "Major Denis..." is not original research.

--Philip Baird Shearer 17:31, 15 Jun 2005 (UTC)

In general direct combinations of facts from two different sources, with no additional novel analysis added is not original research, it's just citing two sources. But your example points out one of the possible pitfalls. You have combined the facts in a way that does not follow from the original facts. "No allied servicemen were..." does not lead to "Major Denis could not be". He could have been, he just wasn't in this example. You may have been thinking of "could not" meaning "was not", but that's part of the danger, it has another meaning, that of "there is no way he could have been". The fact that you have repeated this point multiple times and not reallized you have a non sequitur is worrisome, and points to the danger of doing what you are talking about. And no I don't want to lean too much on this one specific example, so I'll speak more to the general issue. If the facts are combined in a way that is not at all novel and the combination correctly, logically follows from the precedents, then that is not original research. So yes, there is a little subtlety involved in deciding what is novel and what is not, it's certainly not black and white. I would suggest that the more contentious and/or important a point is to a topic the lower the threshhold would be for considering it novel and thus OR. A highly important point, central to a topic, or one that is highly contended, should allow little to no wiggle room in what is considered OR and what is not. In that case any combination of facts into one not found in sources would be OR unless the conclusion is blatantly obvious. To correct your example "No allied servicemen..." and "Major Denis was an allied..." do lead to Major Denis was not tried at Nuremberg, if both of the premises are fully correct. Now that is not very novel if we're just talking about one serviceman that didn't have an impact on later events. But for ex if whether Major Denis was tried at Nuremberg was a central fact causing World War III, then that systhesis of facts may be unnaceptable as OR. Also, if the premises are not fully correct or are questionable, the conclusion of course does not follow. If you've been "bludgeoned" with the NOR policy, it is more likely that your conclusions don't follow from the premises, the premises are not as solid as you'd like, or your conclusions are a lot more novel than you think. - Taxman ^Talk 19:25, Jun 15, 2005 (UTC)

Whoops my mistake so lets rephrase it slightly:

"No Allied service men were tried at the Nuremberg war crime trials because the Charter of the International Military Tribunal stated that only major war criminals of the European Axis countries could be tried."

"Major Denis Healey was a British service man".

Let us also assume that there is no reference which states that he was an Allied service man just that he was a British soldier.

"Major Denis Healey could not have been tried at the Nuremberg war crime trials, because he was an Allied service man".

1. The sentence "Major Denis..." is original research and should not be allowed.
2. The sentence "Major Denis..." is original research but is should be allowed.
3. The sentence "Major Denis..." is not original research.

--Philip Baird Shearer 20:30, 20 Jun 2005 (UTC)

It's not original research if all you're saying is that the Charter of the IMT stated that Allied servicemen could not be tried, and you can source that claim, and you're giving Major Denis as an example. The NOR policy states simply that no unpublished facts, arguments, opinion, analysis, or novel synthesis of published material, may be published in Wikipedia. I see no novel synthesis in the example you gave. SlimVirgin ^(talk) 21:04, Jun 20, 2005 (UTC)

For the most part I would just reiterate what I wrote above. One guy's opinion though, after thinking about it, is that that is borderline OR, and in any case would be easy to avoid. Combining into a conclusion carries both the risk that the precedents are not correct, and that the combination is done incorrectly, so just avoid the issue. Just state the first two points above, note their sources, and let the reader decide if they have enough evidence. Just don't write them in such a way that other possible interpretations are unneccessarily eliminated. For example, don't eliminate potential other explanations, maybe Major Denis was British, but fought for the Germans. - Taxman ^Talk 21:12, Jun 20, 2005 (UTC)

One can be any nationality and a British soldier. If he had fought for the Germans then he would have been a German soldier. Philip Baird Shearer 13:39, 25 Jun 2005 (UTC)

This is becoming reductio ad absurdem, but if he had been acting as an important german spy, then despite also being a British serviceman, he might have been a major war criminal of the european axis. However, I have to chip in on the side that this is splitting hairs. It would seem to be acceptable to write the conclusion on the basis of the two facts, because it is a trivial conclusion to draw which follows straightforwardly in logic, despite the possibility of some unlikely additional information (such as the discovery he was a spy). Surely this would not be original research, rather good faith reporting of bad information? ...so it would not be unreasonable to make the conclusion and it could not be struck out on grounds of original research, only in the event of someone spotting it was bad information? Sandpiper 21:52, 25 August 2005 (UTC)

But since the conclusion can be incorrect, it is much simpler and less error prone to simply state the two facts as the sources have them and let it be. The reader is then able to decide if they have enough evidence or not, and we have simply and easily avoided original research with no loss. - Taxman ^Talk 22:19, August 25, 2005 (UTC)

Except that this is supposed to be a user friendly source of information, not an exercise in applied logic. The statement would only be written as per example 1 because it was part of some article where it made sense to draw the two points together. Then the original question was really, whether it was permissable to express it in that way. I presume it is another unstated condition of the example that there was some reason for wanting to express the facts in that way.Sandpiper 23:43, 26 August 2005 (UTC)

Request for clarification: self-cite

May an editor cite his own original research, as published in a peer-reviewed journal? --Trovatore 28 June 2005 16:56 (UTC)

Most certainly. See '1 What is excluded' on the main page, where the steps for doing this are explained. To attempt to ban expert editors who publish in peer-reviewed journals or who write books is self-defeating and impossible to police. Its self defeating since it would exclude a needed body of expertise from Wikpedia, and impossible to police since few editors log in under their own names, so we would never be able to identify those who write about their own research. Apwoolrich 28 June 2005 18:03 (UTC)

Someone added to a Wikipedia article a reference to an article I wrote that was published in Advances in Applied Mathematics. If I had been the one who added the reference, would that have made any difference? I don't think so, unless someone thought they should disregard it because they thought my adding it was mere self-promotion. I think "original research" in this context means research originally published by putting it on Wikipedia, so citing something you wrote that appeared in a research journal is not the same thing at all. Michael Hardy 28 June 2005 19:49 (UTC)

question about usability

would uploading a copy of a public record tax form obtained from the us irs be an allowable source? I am not able to find a copy of the actual form itself, only references to it, and a tv show about it.... It would not be research, per se, but a reference to a particular expense on the sheet, no analysis needed or give... the claim is that a group gave money to a particular person, and since they did, it shows up on their tax form, along with the exact amount...IreverentReverend 2 July 2005 08:51 (UTC)

Yes, as I've said elsewhere, if you can show that you took it from the IRS or from an equally reliable website, it would be fine. We just need to be able to trace the provenance of it. What surprises me is, if it's legit, why you can't find a good article about it, which would be easier. SlimVirgin ^(talk) July 2, 2005 08:58 (UTC)

It suffices to cite the form specifically. We don't generally upload primary or secondary sources, we just cite them. If it's available online, link to it, otherwise anyone who wants to check it should be able to get a copy. If you think it has good impact, you could take a screenshot, crop it, and use it as an image in the article. Deco 2 July 2005 08:56 (UTC)

actually what you said elswhere was only if i can link to it on the IRS site. Deco was helpful. thank you. IreverentReverend 2 July 2005 09:05 (UTC)

No, that is not what I said elsewhere. I've said maybe six times that all your edits, if challenged, must be sourced to reputable publications. SlimVirgin ^(talk) July 3, 2005 21:07 (UTC)

"Not a copy of the tax form, and not a personal website. We need authoritative third-party sources for your edits " -- slimvirgin

"If you can supply a link to the tax form that is on an authoritative website, then fine, do it. But if you're only going to obtain a copy of it yourself, that's no use to the readers" -- slimvirgin

""They also have to be sources that are credible and accessible to ordinary readers."" -- again slimvirgin

funny how devo seemed to think it was fine... and you repeatedly say otherwise... IreverentReverend 4 July 2005 06:01 (UTC)

I am not an '80s pop group, and I'm not familiar with the context of your situation, so don't ask a misleading question trying to get the answer you want. Citing information does not justify its inclusion by itself if the source is difficult to obtain and difficult to believe; otherwise this would be an easy way to include arbitrary information in any article. 4 July 2005 23:47 (UTC)

Yes, that's the point. If IreverentReverend had himself obtained a copy of someone's tax form to back up a claim, and that form wasn't otherwise available and the information also not otherwise available, it would be a difficult case, because the information would be unpublished and inaccessible to other readers, including other editors who wanted to check its veracity.

The most important point in the NOR policy is that Wikipedia publishes only material that has been published elsewhere. We had a case once of an editor who had an extensive archive of published magazines that were very hard, if not impossible, to get hold of elsewhere. He entered a quote from one of them into an article, and was challenged by another editor, so he scanned the page with the quote on it, and the front page to show the name and date of the publication, onto his computer, then uploaded it onto his website, to satisfy the other editor; and so that a link could be provided in the article for readers to check it. It's important to remember that the criterion for entry into Wikipedia is not truth, but verifiability. SlimVirgin ^(talk) July 5, 2005 00:34 (UTC)

Popular culture?

I think the original research idea utterly fails in connection with popular culture. Consider, for instance, UNIT_dating_controversy. At some point, someone (or several someones) must have sat down and done their own research to find all the UNIT dates in the Dr. Who series. Each individual item in the lists might have a cite, but the combination is a piece of original research used to support the proposition that dating in the series is inconsistent.

The problem is that "original research" is different for popular culture. The "research data" may be available to anyone with a book collection or DVD player, and generally no expertise is needed to verify or refute claims. There are no peer-reviewed journals, and often few relevant published sources of any type.

Ken Arromdee 8 July 2005 20:08 (UTC)

Yes, you're right; it's original research and shouldn't be here. If anyone challenged it, and no citation was produced to support it, it would probably have to be removed. SlimVirgin ^(talk) 16:47, July 10, 2005 (UTC)

I disagree with SlimVirgin. Research to establish facts is what writing an encyclopedia is all about. Researching facts is not 'original' research unless the facts are themselves controversial and questionable.

Someone has gone the trouble in the UNIT_dating_controversy article to identify a number of facts about this controversy; but they have made no attempt to produce a theory that unravels the controversy. Individual facts they have supplied could of course be challenged. As Ken Arromdee points out Doctor Who facts are not controversial because all the research data needed is a book collection or DVD player.

Likewise we can write synopsis' to books, films and episodes, because we are describing the action; not interpreting what it means.

The policy really could do with a sentence in the intro to say that of course you can research facts, it is after all what writing an encyclopedia is all about

I think the first line should read something like this:

"Wikipedia is not the place for original research; research should be limited to the identification and recording of fact." .:ChrisG 22:19, 10 July 2005 (UTC)

Maybe this article wasn't the best of examples, but I consider statements like "the exact years in which UNIT operates are never made precisely clear" as conclusions. You won't find a published source which says "the UNIT dates aren't clear", except maybe in some obscure fanzine that isn't quoted in the article anyway. The nature of popular culture topics is that information about them beyond names, dates, plot summaries, etc. may not necessarily found in published sources at all.

Consider the following sentence from the Dalek article: "In The Dalek Invasion of Earth (1964) a Dalek emerges from the waters of the River Thames, indicating that they are amphibious to a degree." Should the phrase "indicating that they are amphibious to a degree" be removed, on the grounds that it's a conclusion and someone could conceivably argue that the Dalek isn't amphibious at all? Ken Arromdee 23:25, 10 July 2005 (UTC)

I also disagree somewhat with SV. IMO, the simple compilation of data from verifiable sources is not necessarily original research. However, using such data in support of a novel conclusion would be OR. In the specific case mention above, what I find somewhat lacking is citations supporting the assertion that there is "an ongoing debate in Doctor Who fandom" regarding the matter. I'd like to see a better summary and sourcing for what the positions are in this "debate and some verifiable examples of claims that have been put forward. Simply providing a list of "incidents" pertaining to the matter is not necessarily OR -- I mean given the data presented it is almost beyond question that there are some inconsistencies in the timeline. But constructing the list for the purpose of advancing a specific conclusion without providing additional sources for the argument would be quite dubious. older≠wiser 00:22, July 11, 2005 (UTC)

Ken, I'm not sure that really counts as a conclusion, as it's just rephrasing that a Dalek has been seen to emerge from water.

However, the previous point is original research because it draws a novel conclusion, viz. that dating in Dr. Who is inconsistent. SlimVirgin ^(talk) 00:51, July 11, 2005 (UTC)

I disagree again, even more strenuously. That the dating under discussion is inconsistent is merely a fact that can be reported by citing the episodes in question. There are countless instances of similar inconsistencies in fictional worlds and the source materials are the primary documents. Simply reporting the existence of such inconsistencies is not necessarily OR (although it might if if done for the sole purpose of advancing a novel conclusion). What requires citation, IMO, is twofold. First, it would be nice to have some better evidence that there really is any sort of significant debate in Dr. Who fandom regarding this matter (i.e., a couple of Dr. Who fans having an argument is not much of a debate -- while persistent discussions in verifiable fandom sources like newsgroups or listserves or fanzines might amount to something worth mentioning). Second, any accounts that attempt to rationalize such discrepencies would need a verifiable source beyond simply drawing a conclusion from the evidence. older≠wiser 01:18, July 11, 2005 (UTC)

The assertion that there is an ongoing debate in Dr. Who fandom is another example of the same problem. You're not going to find a source that describes the ongoing debates in Dr. Who fandom, in the same way that you might find a source that says "string theory is controversial among physicists." (And "there's a lot of discussion about this topic in newgroups; therefore this topic is controversial" is itself original research which draws a conclusion.) Likewise, you won't find an additional source which says "UNIT dating is inconsistent". Popular culture just doesn't *have* sources in the way that physics or history does; requiring sources for this type of material is unreasonable. The original research ban is ill-suited to popular culture. Ken Arromdee 01:30, 11 July 2005 (UTC)

But you don't necessarily need a source that "describes the ongoing debate". AFAIC, it would be sufficient to point to forums where such debates have been prominent, and ideally give some more specific hooks, such as a time frame, or the "Subject: line" of a particularly pertinent newsgroup thread, or something similar. I think I am in general agreement with you that the standards for citation needs to be something of a sliding scale. The fine points of contention in a particular fandom does not necessarily require the same degree of citation as, say, a discussion of particle physics or the politics of the United States. What is needed is some sort of verifiability. Wikipedia should not be allowed to itself become a primary "source" for expressing original opinions in such matters. older≠wiser 02:00, July 11, 2005 (UTC)

But as I noted, pointing to newsgroups or forums doesn't work. Saying "There is discussion about this in forums; therefore this is controversial" is a conclusion. After all, not everything discussed in forums is a controversial topic; claiming that something is controversial involves making a judgment call on whether the level of discussion is enough to truly call the topic controversial.

Moreover, forums aren't permanent, and it would again be an unreasonable burden to expect people to save copies of forums just so that their statement "UNIT dating is controversial" won't violate official policy a couple of months later.

In order to satisfy the original research ban, the article could only make weak statements like "some people disagree about the UNIT dates". This requirement is useful only to pedants. Ken Arromdee 13:35, 11 July 2005 (UTC)

I would take a different tack than older; I think the point I would make is that references are only demanded when someone finds a point controversial. Popular culture articles seldom attract such controversy.

The suggestion that there is a Unit dating controversy in Doctor Who fandom strikes me as highly likely given what I know about fandom in general. The authors have then gone to the enormous effort to give examples. As such then it convinces me until someone goes to the effort of tellling me otherwise. More importantly another Doctor Who fan reading that article would know if it was controversial point or discover it very easily; because they will know the doctor who websites. If it was controversial it would soon be edited out by someone as incorrect. Likewise for many other points in popular culture articles.

There is a sliding scale of demand for referencing; and it is seldom that a popular culture article needs much referencing, because they court less controversy. The normal wiki process deals with mistakes.

Much of the point of the No original research policy is that it is there when controversy strikes, the more controversy the more stringently it is applied. :ChrisG 22:26, 11 July 2005 (UTC)

I mostly agree with ChrisG. While it is important to always cite your sources, sometimes sources are so diffused as to be difficult to offer up a specific reference. But even so, providing general references that highlight forums in which discussions about the dating controvery have taken place might be helpful for those uninitiated in Dr. Who fandom. I mean, although this doesn't directly tackle the dating "controversy" -- whoniverse has extensive coverage of UNIT materials.

Ken Arromdee wrote But as I noted, pointing to newsgroups or forums doesn't work. Saying "There is discussion about this in forums; therefore this is controversial" is a conclusion. You are correct to a degree, but it is not necessary to frame it in such a manner. Simply saying that there has been considerable discussion on X forum over the dating inconsistencies is more an observation than a conclusion.

IMO, the main point is that Wikipedia should not become an extension of fandom forums. Upon closer reading, there does seem to be a fair amount of inference and conjecture in the UNIT_dating_controversy. But then again, I'm still not exactly clear on what the "controversy" is here. Mostly what I see is an exposition of inconsistencies. That is why I would like to some sort of references (or at least a summary) of just what the controversy is. older≠wiser 03:21, July 12, 2005 (UTC)

ChrisG: There are two problems with putting the references in only if there's controversy. The first is that that isn't what the policy says. Someone could conceivably say 'There's no controversy over this, but it's still original research unless you provide a reference'--and they would be perfectly justified in demanding changes. If original research is permitted as long as there is no controversy, the policy should say so. The second problem is that in popular culture, it's common for statements to be uncontroversial, yet for reasonably accessible references for them not to exist at all.

Unnamed: It is true that many statements could be reframed to eliminate the original research claim. Yes, you could say "there is discussion on this forum about X, so here's some facts" rather than "X is controversial, so here's some facts". But what good does it do to require such a thing, other than to provide an excuse for pedants? (And in this case, to require people to preserve a copy of the forum.) Ken Arromdee 05:50, 12 July 2005 (UTC)

The point is to stop people from making things up. SlimVirgin ^(talk) 06:58, July 12, 2005 (UTC)

I concur with SlimVirgin point, perhaps we should work it into the policy. More generally, I agree with Ken that the policy is overly heavy handed and could do with some softening as we've discussed in this mini-debate and in previous discussions. But the policy has been written the way it has so we can ensure that when necessary the policy can be applied to remove controversy and establish consensus in highly controversial articles, which to be honest often attract various kinds of less than well balanced people. Obviously the policy could be applied to literally hundreds of thousands of wikipedia articles that don't have proper referencing. Is that likely to happen? Of course not, the articles don't attract controversy. But what he have seen is the increasing use of references throughout Wikipedia and that is a process that is increasing in momentum.

Ken, I don't think there is much real difference in point of view between we four discussing this issue; but its remarkably difficult to word the policy in a way that achieves consensus; but by all means if you have suggestions bring it forward.

But the key thing to remember that it is a policy which is applied rigidly in certain specific circumstances and applied far more flexibly elsewhere. Wikipedia would not have 600.000 plus articles if that wasn't the case. :ChrisG 17:48, 12 July 2005 (UTC)

Chris, you're right that this was a hard policy to word, because something that sounds reasonable and workable among a small number of reasonable editors might allow loopholes for hundreds of unreasonable ones, and the loopholes have to stay tightly shut. I'd say the main point is for editors to realize that Wikipedia is not the place for them to write personal essays on any subject. We're not here to present personal opinions or personal knowledge, but to summarize for the readers what has been published elsewhere by credible sources. SlimVirgin ^(talk) 19:30, July 12, 2005 (UTC)

This conversation has gone off on a variety of tangents. Suffice to say that I agree with what SlimVirgin just wrote, and agree with Chris that policie are usually applied most strictly when their are controversies. This does not mean, as I think the anonymous commenter suggested, that policies should be enforced only when there is controversy. It just means that controversy calls attention to a problem in an article that needs fixing. Since Wikipedia is a permanent work in process, this will never stop. Look, all articles should have proper punctuation, and many don't. And there are some noble people here who just go around correcting punctuation. Equally noble people could go around looking for places that need citations. But with a project as huge as ours, few people have the time to do that systematically. Controversies are just an index of how urgent is the need to make an article policy-compliant. As for the real issue of how to handle popular culture, I think the apple pie example covers this, along with SlimVirgin's point. Slrubenstein | Talk 22:28, 12 July 2005 (UTC)

I had forgotten to sign earlier--not sure if that is the "anonymous commentator" you refer to, but I most emphatically am not suggesting that policies should only be enforced only when there is controversy. I think the implication is that the de facto practice at Wikipedia is that controversial topics are subject to considerable scrutiny and there is a very low tolerance for unsourced statements and original research in them; while in less controversial topics many unsourced statements and even a degree of original research slide under the radar and are tolerated by benign neglect, apathy or obscurity.

Ken writes Someone could conceivably say 'There's no controversy over this, but it's still original research unless you provide a reference'--and they would be perfectly justified in demanding changes. Yes, that is precisely what the policy says. And in practice, that is one of the strengths of Wikipedia. Someone may write an article (or portion) based on their own knowledge of a subject and not provide any sources and such an article could sit around all "uncontroversial" for months or even years. Now, if someone later questions the content, not because they doubt the accuracy, but simply because it is unsourced--that alone, IMO, would not warrant wholesale removal. But ideally, it would generate some effort to provide some corroboration of the information.

To Ken's other point: The second problem is that in popular culture, it's common for statements to be uncontroversial, yet for reasonably accessible references for them not to exist at all. Well, perhaps this just means that it is difficult to write good quality articles about popular culture and requires a bit more effort to assemble references for the material.

And although this is tangential to the discussion, can someone explain exactly what the "controversy" actually is in the UNIT_dating_controversy? As I said earlier, there certainly appear to be some inconsistencies in the timeline (big surprise in a show like Dr. Who), but what is controversial in this? older≠wiser 01:29, July 13, 2005 (UTC)

I concur with all older≠wiser said. By normal frames of reference I don't suppose it is a controversy; but we are talking fandom here. I'm something of a Dune fanatic and there are 'controversies' within Dune devotees that wouldn't be worth a mention outside the glass bubble of fandom. :ChrisG 17:52, 13 July 2005 (UTC)

"Well, perhaps this just means that it is difficult to write good quality articles about popular culture and requires a bit more effort to assemble references for the material."

You seem to be suggesting that it can never be so difficult that the policy becomes unreasonable. I disagree. If I wanted to say "the Doctor Who universe does not date UNIT consistently", I cannot provide a list of UNIT dates and say "these aren't consistent". That would be drawing a conclusion and violating the original research policy. Instead I would be required to dig up a reference which states that the dates aren't consistent. That reference is either very, very, obscure or doesn't exist, and in either case is completely useless. Demanding it serves no purpose other than to satisfy pedants.

Telling people that it requires a "bit more effort" ignores the question of whether requiring such effort is a good idea (and "bit" is a serious understatement). I am aware that the policy requires it; I'm questioning the policy. Ken Arromdee 16:29, 14 July 2005 (UTC)

I've already said I don't agree that providing a verifiable list which demonstrates that inconsistencies exist and then labelling it as such should fall afoul of the original research prohibition. If we were to prohibit making such facile, verifiable statements, then Wikipedia would immediately shrink by a significant fraction. Where there would be more of a problem is with presenting novel attempts to rationalize such inconsistencies. As I and others have said here, this policy is intended to provide a bulwark against nonsense presented in the guise of statements that might seem plausible unless one is familiar with the subject. Anyone should be able to request corroboration for unsourced statements which do not accord with other verifiable sources. Challenging otherwise uncontroversial statements simply because they are unsourced is, IMO, not really in the spirit of this policy (although it may be within the letter). But even so, people should not be put off from requesting some corroboration for unsourced statements. Ideally, such corroboration would be published sources. But occasionally, the corroboration comes down to the word of editors familiar with the topic. Depending on the gravity of the subject, the reasonableness or not of the statements, the value of the statements in the context of an article, and the reputation of the editor(s) defending them, the community may decide either to accept the statements, modify them, or reject them. Such things would be very difficult to codify. older≠wiser 01:26, July 15, 2005 (UTC)

Clarifications

Some things need to be clarified. If I conclude, on my own, that I can solve y' = y + x by differentiating and substituting, you can't call it "original research." If I call it the Kokielov Trick (or the George Bush Trick), you can accuse me of assigning undeserved credit, but it still won't be original research. It becomes original research the moment I introduce a hypothesis not for the sake of experiment, but as a foundation. For instance, if I suddenly say that you can chart the trajectory of, oh, say, a comet by solving y' = y + x, then you tell me I'm hypothesizing and throw me out the window. But not a moment before. --VKokielov 16:21, 10 July 2005 (UTC)

Sorry, I didn't understand this point. SlimVirgin ^(talk) 16:49, July 10, 2005 (UTC)

This can't be a slap-stick label. You need to explain that you're not forbidding every kind of conclusion, but only dubious or potentially dubious conclusions. Take the strictest view, but a correct mathematical derivation still won't be a dubious conclusion. --VKokielov 16:51, 10 July 2005 (UTC)

This is important, because correct mathematical derivations, even if they don't lead anywhere, can be instructive. If anyone in the public at which you're aiming can decide whether a derivation is right or wrong, I think you should keep it. --VKokielov 16:56, 10 July 2005 (UTC)

If nothing else, and pertaining to everything rather than only mathematical derivations, it will be foolish to censor anything which is relatively easy to prove and to see proven. --VKokielov 16:59, 10 July 2005 (UTC)

(edit conflict) We don't only exclude dubious or potentially dubious conclusions. As I wrote above, and as I think the policy page says, the criterion for entry into Wikiedia is not truth — that can't be said often enough — it's verifiability. We publish material that has already been published. We don't publish our own views, opinions, experiences, novel mathematical calculations, or any novel analysis or synthesis of known facts, no matter how right we think we are, and no matter how right we may turn out to be. SlimVirgin ^(talk) 17:00, July 10, 2005 (UTC)

"Truth" is a word very easy to inflate. As far as mathematics goes, truth and verifiability are one and the same. That's what I mean to say. --VKokielov 23:53, 10 July 2005 (UTC)

I think what VKokielov is trying to say is that, in mathematics and many of the sciences employing mathematics, if a particular calculation or proof is easily verified by any professional or even a significant segment of professionals, it should be allowable, even if it hasn't been published before. In math, there's an uncountable number of correct, easily verifiable arguments, and to say that you can't include a new mathematical argument which is easily verified by professionals just because it's unpublished is asinine. I myself did this at proof that the sum of the reciprocals of the primes diverges; one of the proofs is a minor modification of a well-known one. The fact is, practically anyone with a BA (or less) training in math can verify in a few minutes that it's correct. Yet, it's not WORTHY of publishing. Not every new argument in math is "original" enough to warrant publication. There should be a specific statement in the main page dealing with this case. Revolver 20:12, 2 September 2005 (UTC)

Confusion over the meaning of original research

We have a discussion ongoing with an editor who seems to have a drastically different interpretation of Original Research. For anyone who either helped draft this guideline or knows it well, you might want to comment :) --kizzle 19:29, July 14, 2005 (UTC)

I agree. This section of the NOR is the most commonly quoted and misrepresented part of the NOR:

Original research that produces primary sources is not allowed. However, research that consists of collecting and organizing information from existing primary and secondary sources is strongly encouraged. In fact, all articles on Wikipedia should be based on information collected from primary and secondary sources. This is called source-based research, and it is fundamental to writing an encyclopedia.

I've seen more than one user quote it as some sort of get-out for using original research in articles. By arguing that their original ideas are backed up by logic applied to disparate sources and citing this section they claim they are not performing original research, even though they clearly are. I would advise this section is either reworded for clarity or removed completely. Axon 15:40, 15 July 2005 (UTC)

This guy's logic was that "Original research that produces primary sources is not allowed" meant you could only use secondary sources, and that "collecting...from existing primary and secondary sources is strongly encouraged" was a logical operator AND, meaning every primary source had to be reference through a secondary source. I have my own feelings about this interpretation, but I feel they would quickly devolve into personal attacks.

What if we change the line

Original research that produces primary sources is not allowed.

to:

Original research that produces primary sources where they did not exist before is not allowed.

Just a thought. --kizzle 15:47, July 15, 2005 (UTC)

Oh ya, also changing "existing primary and secondary sources" to existing primary and/or secondary sources" --kizzle 15:48, July 15, 2005 (UTC)

I agree with you point, although I think I'm complaining about something else. I would argue that this line is best re-written as:

Original research that produces new primary sources is not allowed.

or:

Original research that creates primary sources is not allowed.

Which would resolve the issue. I also agree with the "and/or" which is a no-brainer modification to clear up pedantic misunderstandings. Not sure how one goes about modifying Wikipedia policy though. Any thoughts on my problem? Axon 16:27, 15 July 2005 (UTC)

Ya, I understand what you're complaining about, which is different from what this user's misinterpretations consist of. However, I think what you're talking about is drawing original conclusions from existing primary sources and justifying it based upon logic. We used to have a policy at Wikipedia:Spoon_Feeding for witholding such original conclusions by editors that I created but it got VfD'ed with one vote to merge into NPOV, which I am currently drafting. --kizzle 18:34, July 15, 2005 (UTC)

I'd be interested to know exactly what you are drafting for the NPOV. Is there any other reference I can point editors to when the above situation arises again in the meantime? Can you suggest any modifications to the above section to avoid such mis-uses of this section in future? Axon 19:22, 15 July 2005 (UTC)

Just incorporating what's at Spoon Feeding currently into WP:NPOV. Just need to find a good place and make it work with the hierarchy of info there. There needs to be policy specifically tailored towards drawing conclusions for people (spoon feeding), which is a daughter of OR and NPOV, but still unique. --kizzle 19:58, July 15, 2005 (UTC)

Oops, didn't know it got deleted already. Try User:Kizzle/Spoon_Feeding --kizzle 20:12, July 15, 2005 (UTC)

Examples are original research

ie. the examples of Polymath's and Nerd's are obviously individuals interpretations (data, etc...) of these terms and the people who belong to them. Baring objective testing, no amount of group-agreement will fully make these non-original (baring a defining number of agreements -- where defining number is a proportion of the population "randomly" agreed to be the minimum amount necessary to form a definition -- ie. the circular arguments inherently necessary to define terms -- group-aware non-individual lexicons).

Tell me how to get published!!! (I have no credentials - the entirety of my formal schooling in psychology being Psych 101, collegiate education in general being sophomore/junior status as of 2000 - , though my understanding of certain topics borders on the post-graduate.) If you tell me how to get published, or show me someone/something which publishes and has the intellectual capacity to understand what I am talking about I probably still wouldn't do so, though it would be nice to know. A significant amount of my ability to understand the world is predicated on me not being able to effectively interact with or interject myself into the world. How the h*** do I get my ideas and research published in a peer-reviewed journal when I see so very few people as my intellectual peers you **** untermenschen who write these broken rules. How the h*** does one with no external credentials find the relevant journal to publish in????? I seriously want to know. Your encyclopedia, by referencing all and sundry (no matter how flawed/error ridden -- and many of them are), lacks rigor, lacks exactitude, and to a strong extent lacks scientific and intellectual suitability. But that's the point of an encyclopedia isn't it!?!?!?

Aahhhhh...... It sucks to every few months or years actually want to interact with other humans, but finding myself unable to do so. If i fail one more time, that will be the last time I attempt to communicate with those who lie on the other side of the cognitive divide. This was a minor rant brought on by profound frustration in both others, and in my inability to engage others. Disregard or not, though my first paragraph actually does stand on it's own as a critique (note: As a means toward clarification and disambiguation I would rather the examples be deleted or explicitly labeled as idiosyncratic and thus potentially wrong/misleading than have this rule modified to explicitly allow them -- especially for terms such as Nerd that have such diverse and mutually contradictory meanings (thus diverse and potentially contradictory examples) -- either that or segregated via what particular aspect of the article they're referring to).

Abandon all hope, all ye who enter here: unless you're sufficiently similar to those who create it that you can effectively use it, and be used by it. This is not a place where dialecticals are allowed to form, except possibly as examples defining the term "dialectic". --24.22.227.53 20:56, 26 July 2005 (UTC)

Unless the specific images and examples are of characters/people who have been specifically referenced as nerds in an external source (ie. if fictional have been referenced as Nerds within the fictional world, or by the creators/actors of the fictional world). --24.22.227.53 23:11, 26 July 2005 (UTC)

Proposals for exceptions to the policy

The Otherkin article is one of a few who are special in that its not particularly easy to come across information that isnt "original research" as termed in this policy, so i propose an exception, an article specific bending of the rules, because otherwise there wont really be an article, except sources that state that the otherkin believers are insane. Gabrielsimon 23:26, 26 July 2005 (UTC0

(The above text was also posted to the policy page.) There are some guidelines about "obscure topics" at Wikipedia:Verifiability#Obscure topics. They don't draw any clear bright lines, though. I tend to think that subjects that are obscure enough that there aren't any reliable sources are obscure enough that they don't need detailed Wikipedia articles. FreplySpang (talk) 00:02, July 27, 2005 (UTC)

its not what id call an obscure topic, i just wish to obtain permission for the ability to use internet based research for that article. Gabrielsimon 00:05, 27 July 2005 (UTC)

There are no exceptions to the NOR rule. We don't want original research in Wikipedia: that's the point of the policy. But that doesn't mean you can't do Internet-based research. What kind of thing are you trying to track down? SlimVirgin ^(talk) 23:59, July 28, 2005 (UTC)

data on the otherkin article, which a rather deletionist user caled friday is seemingly trying to have either gauged or deleted. Gabrielsimon 00:01, 29 July 2005 (UTC)

I've just glanced at this. The medical perspectives section is original research, unless you can find a source that has said these things of the otherkin movement. And the explanation about schizotypal personality disorder is likewise original research unless, again, you can find someone who has said these things specifically about the otherkin movement. As for the rest, it might be OR too, depending on the quality of the sources, which I haven't checked yet. SlimVirgin ^(talk) 00:05, July 29, 2005 (UTC)

Why is your data being deemed original research? SlimVirgin ^(talk) 00:08, July 29, 2005 (UTC)

its difficlt to find "scholarly" refernces forthis subject, that dont deal with medical " problems" etc, i wish to find a way to allow the bulk of otherkin sites that otherkin put up to be used as reference material, because of the lack of other possible resources forhte time being. Gabrielsimon 00:05, 29 July 2005 (UTC)

You probably can use them as sources. What is the argument against them? Please link to the arguments or give more details here. SlimVirgin ^(talk) 00:08, July 29, 2005 (UTC)

The discussion is at Talk:Otherkin#Verifiability and sources onwards. The problem that has arisen with the article is that it uses various websites and message boards as support for the fact that self-identified otherkin have made these claims about themselves. I am not, myself, clear on why this is considered original research, and am not happy with the explanations that I've been given.

With regard to the medical perspectives section, by the way, most of this discussion can be sourced to a website that has made similar claims. However, is it the case that journal references would be required for a medical discussion, even in cases such as these where the allegations are commonplace? Vashti 07:43, 29 July 2005 (UTC)

Part of our definition of original research is putting facts together to build a case, rather than to present a neutral picture. The medical section of that page looks as though the editors of the page have decided to turn membership of the otherkin movement (if that's the right way of putting it) into a medical complaint, and specifically a consequences of schizoptypal personality disorder. A medical reference would therefore be required, and the otherkin article would have to stick closely to what the medical references said, and not elaborate. If the reference to members of the otherkin movement as mentally ill is being inserted only because the allegations are commonplace, that counts as original research, because it's unverifiable gossip.

Regarding the sources for the rest of the piece, message boards are best avoided. But a website about the otherkin movement should be acceptable, though again, the article must stick closely to what it says and not elaborate using personal experiences or opinions. That is, it should be an encyclopedia entry, and not a personal essay. It's hard with this kind of subject, and what that probably means is that the article should be a short one. SlimVirgin ^(talk) 09:08, July 29, 2005 (UTC)

I'm sorry, but I'm afraid I'm going to have to disagree with you here, quite strenuously. There are plenty of medical sources in existence about clinical lycanthropy and other related disorders. Considering that otherkin evolved from therianthropes which came from lycanthropy, and it is just a matter of them having purposefully changing the name they referred to themselves as -- in early cases specifically to try to distance themselves from the name of the mental disorder -- while retaining the same features, it seems pretty bizarre to know be insisting that the medical literature mention them by the name they invented up for themselves. So, if some psychiatrists sneaks into the Internet's small corners and discovers the term otherkin, specifically says that otherkin meet the features of clinical lycanthropy, and then the otherkin go and change their names to beastlies, are you going to be arguing that the new beastlies article cannot mention the medical disorders because no study or sources specifically mention beastlies? I think you need to reassess your argument here, quite drastically. Yes, the section of the article in question could have used some clean up, but an overly zealous editor whose POV wishes to remove any mention of any disorders on that page (even in passing) used your claim to justify erasing the entire section. They are also trying to claim that you being an admin means that you have more authority to speak to this topic than anyone else, which, as you know, is not something admins gain by accepting their positions. The fact is that there are plenty of sources to back up the information in that section, and even for being a little sloppily put together that section linked to articles where these sources were discussed. Your claim that someone was trying to label them as having a disorder by even mentioning the topic seems highly POV itself, because mentioning that claims in no way endorses them. In fact, not mentioning them at all endorses the idea that there is nothing to them, which is far more POV-loaded than the slight you imagined by its inclusion. DreamGuy 09:29, August 1, 2005 (UTC)

Hi, Dreamguy. Firstly, unless you have some sources that I've missed, I believe your claim that otherkin evolved from the therianthrope community to be demonstrably incorrect; please see User:Vashti/Otherkin for more on this. Given that, I think you are also mistaken about the "beastlies" case here, because a published paper dealing with otherkin would still be valid even if the community then renamed itself - because it would be verifiable as the same community. Vashti 10:38, 1 August 2005 (UTC)

First, I haven't looked at the article closely, so I don't know the details. The important point is that medical sources must be quoted if medical claims are being made. And the article must stick very closely to what the sources say, and mustn't elaborate. For example, I saw there was a long section explaining what schizotypal personality disorder is. That is original research because it was lifted (as I recall) from a DSM, not from an article about otherkins, or whatever they used to call themselves. If doctors have written about schizotypal personality disorder in relation to otherkins (or their old name), you can repeat what those doctors say about it, but nothing else. Also, you would have to be very sure that the otherkins really are the other people with a new name, and not in fact a different group. In summary, you mustn't insert your own opinions or arguments into the piece, or insert facts in order to build a case. Any of those things counts as original research. You may only repeat other people's published facts, opinions, and arguments. SlimVirgin ^(talk) 10:00, August 1, 2005 (UTC)

without turning the article into a poorly written stub, i think its already as short as its gonna get. Gabrielsimon 09:12, 29 July 2005 (UTC)

The situation we were faced with was that the original page had some heavily POV material stating that otherkin beliefs were a mental illness. A number of editors were pushing strongly for the retention of the material and for the viewpoint in general. Since the medical section had become larger than the section actually giving information on the page topic, we dealt with the problem partly by adding more relevant material, and by adding balancing material for the medical section. While I personally felt that the various allegations of insanity were unjustified, I figured that in the interests of neutrality, they should at least be dealt with in the entry and not just ignored. Is it your opinion, then, that the section should just be cut altogether because the allegations are gossip? Even if nine-tenths of the people who read the entry are likely to think "these people are nuts"? Vashti

The claims that "the original page had some heavily POV material stating that otherkin beliefs were a mental illness" seems completely at odds with what the article actually had, unless there was something quite severe put into the article when I wasn't looking and then later removed (which is plausible)... and then the severe part is what had to be removed, not the entire thing. DreamGuy 09:29, August 1, 2005 (UTC)

Hi Vashti, the section should definitely be deleted if there are no medical sources for it. If readers want to think the otherkins are nuts, that's fine, or if Dr X has said so in the Lancet, or wherever, we can quote Dr X, but we can't have Wikipedia saying it. That counts as original research. SlimVirgin ^(talk) 10:22, July 29, 2005 (UTC)

It never did say they were nuts. And if it did say that in some part because some editor sneaked it in when I wasn't looking, that part needs to be removed, not the whole section. Plus there were sources. You have a straw man argument going here.. DreamGuy 09:29, August 1, 2005 (UTC)

The thing is, as I understand it, that the article was drawing conclusions from the sources that were original research, because the people making them weren't competent to do so. And yes, I'm as guilty of that as anyone else working on it. But taking a page from the DSM-IV and saying "look, these people have a couple of symptoms of this disorder" isn't any less OR than saying "look, these people don't have all these other symptoms of the disorder, plus they're a subcultural group", because nobody competent has made the observation about them. That's why it's gossip. And if a consultant psychologist came along and said "actually, they are nuts", it would *still* be original research.

To draw a comparison, a lot of people suggest that politicians and celebrities suffer from narcissistic personality disorder, but without a journal source saying that, it's not something that should be on Tony Blair, for instance. Vashti 10:38, 1 August 2005 (UTC)

Well, the section appears to have disappeared on its own ... although I took an archive copy of it for later on, when the stuff inevitably hits the fan. Thanks. :) I'll look at rewriting the main body of the article. Vashti 11:11, 29 July 2005 (UTC)

The difficulty as I see it, is the conflict between somewhat strict "encyclopedic standards" and the ability to provide an article on an emerging or otherwise obscure topic. Maybe I'm actually more concerned with verifiability than OR, I'm not sure. They seem related to me. I've been thinking that if you collect information from various unverifiable sources and compile it together to get facts, that's original research. I could be wrong.

At any rate, Otherkin (where I've been heavily involved in the verifiability/original research question) seems to me like an interesting test case on this issue. On the one hand, we don't want Wikipedia to look silly. We also don't want misleading or POV information in an article. On the other hand, if Wikipedia had no "Otherkin" article at all, that doesn't really help much either. Nobody that I've seen is trying to suggest that Otherkin do not exist, so why shouldn't there be an article on it?

What if there was a way to tag the article, so readers would understand that the sources were the best available on the subject but weren't neccessarily considered reputable? We already have the OriginalReseach tag, but I'm not sure it's appropriate because to me this implies a problem that's going to be fixed. What if there's no fixing it? What if editors want the article to stay, but feel the sources are questionable? Maybe this is a silly idea, since articles should either be verifiable or not, I don't know. Maybe it even came up years ago and was shot down. I just wondered if anyone else thought this was a fair summary of the issue at hand, and/or is a new tag a useful idea. Friday 03:26, 30 July 2005 (UTC)

I think the NOR policy is absolute for a very good reason. If there is no verifiable information about a subject, then it probably shouldn't be in Wikipedia, and that's ok. We don't need to have articles on everything. What verifiable information there is, great, include it, but speculation, etc, is specifically what we don't want, unless we can refer to another reliable source making the speculation. Now there being only low quality sources for a given subject, perhaps there should be a way to note that. But that's hard to do without being POV. Many readers can discern a low quality source when they see it, but then again, many don't understand the difference. Articles with only low quality sources are a problem in my opinion because they get a sort of stamp of legitimacy from Wikipedia which is what the NOR policy needs to avoid. - Taxman ^Talk 03:42, July 30, 2005 (UTC)

I agree with Taxman. SlimVirgin ^(talk) 21:01, July 31, 2005 (UTC)

Wikipedia being place for peer-reviewing

From the project page:

If you have an idea that you think should become part of the corpus of knowledge that is Wikipedia, the best approach is to arrange to have your results published in a peer-reviewed journal.

Getting a journal to peer-review your statement is pretty hard. Actually, Wikipedia itself is a place of peer-reviewing, therefore original research is acceptable at Wikipedia, as there are a million eyes reading what you wrote, isn't it? For example, the possible reason given for the video game crash of 1983 in that article is purely based on original research, yet it is a great of source of information and a good consideration on why the crash happened. --Abdull 13:50, 31 July 2005 (UTC)

It shouldn't be there if it's original research, Abdull. And it's because it can be hard to get a journal to peer-review academic papers (though it's not that hard) that Wikipedia regards peer-reviewed journals as good sources. SlimVirgin ^(talk) 21:10, July 31, 2005 (UTC)

Please use talk page before adding content

Please use this page before adding any substantive content that might contradict the policy. Many thanks, SlimVirgin ^(talk) 21:01, July 31, 2005 (UTC)

Well, I added some points on why this policy is important. I didn't think I was adding anything that could contradict the policy (I did emphasize "citable sources" because I believe that citing sources is very closely linked to this policy. My main motive was to rewrite the section to be clearer (than the grocery metaphor) and more inclusive. But of course, if any one has objections I hope they will bring them up. I just didn't think any of the points were really contentious, Slrubenstein | Talk 21:14, 31 July 2005 (UTC)

I didn't mean your additions, SR. I deleted a couple of points earlier that seemed a bit odd or potentially contradictory, but I don't think I touched any of your additions. I'll go back and check. SlimVirgin ^(talk) 21:48, July 31, 2005 (UTC)

No, I didn't touch your edit, [4] because it clarified things. I was never keen on the grocery analogy. SlimVirgin ^(talk) 21:52, July 31, 2005(UTC)

Thanks! Yeah, the grocery business — well-intentioned, but ... Anyway, I just modified the heading to "what makes this important" in a way that I hope makes the nature of the points that follow more understandable within the parameters of this policy. Slrubenstein | Talk 22:50, 31 July 2005 (UTC)

Francis, could you say what your aim is with your changes, as they seem pretty extensive? SlimVirgin ^(talk) 22:35, July 31, 2005 (UTC)

I suppose the aim was pretty self-evident: clarify the line between "research" projects going on (in wikipedia/wikimedia context) and the "no original research" policy.

The Slrubenstein text contained some pretty "original" phrases, for instance

Relying on citable sources also may encourage new contributors. For example, if someone knows of an important source that the article has not drawn on, he or she may feel more confident in adding important material to the article.

Unconfirmed speculation I would say! Tedious lists of references may as well discourage new contributors. Who can tell? Has there been any kind of research about that?

And the "citable source" concept is also new, and not explained, as I noted in the edit summary (which you probably ignored, while it clarified my "aims").

General style recommendations are to avoid long introductory paragraphs (etc), so I added subtitles.

--Francis Schonken 23:05, 31 July 2005 (UTC)

Overly long intros aren't good in articles, but this isn't in the main namespace. I like this long intro, because it means that people reading only the intro will still get the main points.

Overly long intros are to be avoided from a web-usability viewpoint too (which can be externally referenced), so it's only logical this ends up in wikipedia policies & guidelines. I don't understand that such an "important" guideline should be low-quality from a readibility viewpoint...

Not sure what you mean by a "web-usability viewpoint." SlimVirgin ^(talk) 23:38, July 31, 2005 (UTC)

Never heard about web usability?

SR meant that new users might be encouraged to edit by adding a source, which is a good point.

But not discussed here. I don't like your "IS" a good point. I happen to think it's not a good point w.r.t. explaining the rationale of the guideline.

You're right: it doesn't explain the reason for the policy. But it does no harm, doesn't take up much length, and it remains a good point. If we can see, from an article's cited sources, that a particular source's viewpoint has not been included, we know to include it. SlimVirgin ^(talk) 23:38, July 31, 2005 (UTC)

The "harm" if you must call it that way is lengthening the article without particular relevance to tha article itself. And no, I don't think it is a "good" point, because it makes questionable assumptions (which is particularily bad taste on a page explaining to avoid "original ideas").

A citable source is one that accords with the NOR policy and with Wikipedia:Cite sources. SlimVirgin ^(talk) 23:13, July 31, 2005 (UTC)

It was not defined on the page. Try to avoid "wikipedia-specific" terminology for guidelines, and in any case when not explaining it. --Francis Schonken 23:22, 31 July 2005 (UTC)

A source that can be cited is a published source as opposed to, for example, my friend who told me something over lunch. It can be cited = citable. To clarify what we mean by a reputable source, I added a link to that part of the discussion. SlimVirgin ^(talk) 23:38, July 31, 2005 (UTC)

yes, might have a look at that first --Francis Schonken 00:00, 1 August 2005 (UTC)

proposed changes

I would like to change the subtitle:

On talk pages and project pages

to:

Original research on talk pages and project pages

and the first paragraph from that section from:

Like most Wikipedia policies, No original research applies to articles, not to talk pages or project pages.

to:

Like most Wikipedia policies, No original research applies to articles, not to talk pages or project pages. Note however that referring from an article page to such pages as a "source" would conflict as well with the "Avoid self-references" guideline.

I disagree. There's no need to say that, and it's not the only reason it would be unacceptable. You're needlessly complicating things. SlimVirgin ^(talk) 23:53, July 31, 2005 (UTC)

What "other" reasons do you see besides "avoid self-references" and the reasons already mentioned (or linked to) on the "no original research" page? --Francis Schonken 00:44, 1 August 2005 (UTC)

It wouldn't be a reputable or credible source. SlimVirgin ^(talk) 00:57, August 1, 2005 (UTC)

Needing to be a "reputable or credible source" is indeed already mentioned on the "no original research" page. So I repeat the question: What "other" reasons do you see besides "avoid self-references" and the reasons already mentioned (or linked to) on the "no original research" page?

further the expression:

Wikipedia namespace

to:

Wikipedia (or "project") namespace

why? SlimVirgin ^(talk) 23:54, July 31, 2005 (UTC)

Because one doesn't know at what stage a wikipedian arrives at this page the "namespace" concept might strike as confusing, especially the expression "wikipedia namespace" which might be confused with "wikipedia's main namespace.

I don't think adding "project" will clarify it. SlimVirgin ^(talk) 00:57, August 1, 2005 (UTC)

It's about adding the piped link - the official name of the intended namespace is "project namespace"

and the paragraph:

Meta-Wiki allows original research, see for instance m:research, m:Wikiresearch, m:Wikimedia Research Network, m:wikiversity, m:category:research, and m:statistics.

to:

Meta-Wiki allows original research, see for instance m:research, m:Wikiresearch, m:Wikimedia Research Network, m:wikiversity, m:category:research, and m:statistics. Whether Meta-Wiki is a "reputable" enough source to be cited in Wikipedia has yet to be established (...to be continued).

It is not established as a reputable source for WP, so why bother mentioning it as such? SlimVirgin ^(talk) 23:53, July 31, 2005 (UTC)

Well, that's of course the question I'd like to see answered, putting it here might speed up the process. --Francis Schonken 00:08, 1 August 2005 (UTC)

This is a policy page. We can't insert things here in order to speed up things elsewhere. It isn't established as a reputable source for WP. SlimVirgin ^(talk) 00:57, August 1, 2005 (UTC)

No, it's not about "speed up elsewhere", it's about how wikipedia assesses the reliability of meta-wiki as a "source", so that is a question only wikipedia can answer, and as far as I can see this is the most appropriate place to do so, while this is the place where sources for wikipedia content are qualified w.r.t. their "reputability".

Further I think it only correct not to avoid the question "Is meta-wiki a reputable source?" on the page that tries to define "reputable source", and sends people to "reputable publishers" if they want to publish some original research for the first time: it makes a lot of difference whether at that stage meta-wiki can be included or not. People should at least be warned if they shouldn't have hope to acquire respectability through meta-wiki. --Francis Schonken 00:44, 1 August 2005 (UTC)

Sorry, I didn't understand the above. SlimVirgin ^(talk) 00:57, August 1, 2005 (UTC)

Anyone having a problem with any of these proposed changes? --Francis Schonken 23:43, 31 July 2005 (UTC)

You seem to be wanting to add things that have nothing to do with the policy, and which are potentially confusing. SlimVirgin ^(talk) 01:00, August 1, 2005 (UTC)

Well, you look very confused anyway, which I think very remarkable for someone stating over and over again he knows the "no original research" policy thoroughly. You defend to add things with as reason "that they don't harm", instead of whether they're relevant and consistent with wikipedia policies. Seems, anyway, the policy has no answer to some annoying questions related to it. Seems also you can't answer them while being too "confused". So my plan is to continue to upgrade the policy to a higher level of efficiency. --Francis Schonken 07:57, 1 August 2005 (UTC)

I'm asking you again not to do that.

???? What kind of a personalised instruction creep is that? As long as the policy page is eligible for improvement as far as I'm concerned it could be anybody's plan to upgrade the policy to a higher level of efficiency.

If I'm confused, others might be too.

That's why (among other improvements) a clearer structure of the page might be better. Or did you intend to say that you want to prevent others from being less confused than you are?

Perhaps you could try to explain again what you feel your edits add.

Why should I explain "again"? I propose you read "again" what I already wrote.

For example, your wanting to use this page to speed up discussion of whether meta-wiki is a reputable source is, in my view, an inappropriate use of a policy page. SlimVirgin ^(talk) 09:51, August 1, 2005 (UTC)

This talk page is as good for me as the project page. But this far on the talk page the question has only been avoided. --Francis Schonken 07:21, 2 August 2005 (UTC)

The argumentation of the importance of this policy

The grocery store analogy

Till a few days ago the paragraph pointing at the importance of this policy read:

This policy is important because Wikipedia must impose certain restrictions to its content if it is to become known as a well-respected and reliable source of information. Another way to consider it: what do you expect from your grocery store? Almost everything it sells to you, it buys from someone else. Now ask yourself: would you buy from your grocery store if you so much as suspected that you couldn't trust what you were buying? So, too, here at Wikipedia. It's an obligation of Wikipedia to its readers that the information they read here be reliable and reputable.

I suppose it's safe to state that Slrubenstein, SlimVirgin and Francis Schonken agree that the "grocery store analogy" was not the best available solution to tackle this topic.

Note also that the "grocery store" paragraph already has been discussed a few times on this page. --Francis Schonken 07:21, 2 August 2005 (UTC)

Slrubenstein's solution

Presently this part of the policy text reads:

This policy is a counterpart to several core policies, and is thus important for several reasons:

1. It's an obligation of Wikipedia to its readers that the information they read here be reliable and reputable. Published sources are generally reliable and reputable.
2. Of course, not all published sources are reliable and reputable. But by relying on citable sources, we help readers of Wikipedia evaluate the quality of our research. See Wikipedia:No_original_research#What_counts_as_a_reputable_publication? for a discussion on how to judge whether a source is reliable.
3. Relying on citable sources also may encourage new contributors. For example, if someone knows of an important source that the article has not drawn on, he or she may feel more confident in adding important material to the article.
4. Citable sources provide readers with resources they may consult to pursue their own research. After all, some people turn to encyclopedias as a first step in research, not as a last step.
5. Relying on citable sources helps clarify what points of view are represented in an article, and thus helps us comply with our NPOV policy.

The text is by Slrubenstein, with minor adjustments by SlimVirgin. Francis already produced some comments on this text above, here is a more comprehensive list of comments:

Ad #1

• Don't like the expression "It's an obligation of wikipedia to...". I didn't think wikipedia works with this kind of "obligation" scheme. As far as I experienced policies, guidelines & the whole of wikipedia, I think wikipedia works with things at which wikipedians aim, trying to make that a communal goal. Not the concept of a tamed herd working under an "obligation". --Francis Schonken 07:21, 2 August 2005 (UTC)
• "Published sources are generally reliable and reputable.": What kind of a subjective POV is that? Note that "published" includes under most law systems, for example, internet from the most subjective blog to the NASA website, from the Microsoft website to one or two day websites claiming autorship over terrorist attacks: generalisations claiming "reliability" and "reputability" for all that is published are quite stupid if you'd ask me. Even for "printed" publications the generalisation wouldn't work as far as I can assess. --Francis Schonken 07:21, 2 August 2005 (UTC)

There is also a problem with political content and a misunderstanding of the word reliable. For example I have recently been in a discussion about PLO Allegations of Terrorism. Both the lawyers of the PLO and lawyers Israel are likely to be reliable (in the sence of consistently biased) with the analysis and information which they provide. But their sources are tainted by the Mandy Rice Davies test "Well he would [say that] wouldn't he?" because they are not disinterested parties to the conflict. In many articles where there is a difference of opinion, a source from a disinterested party ought to carry more weight than one from a intrested party. Philip Baird Shearer 09:29, 2 August 2005 (UTC)

If I'm not erring any reliable version can be presented in wikipedia (note that I'm not saying either of the parties you mention or both of them are "necessarily" reliable - I'm not that acquainted with the topic you mention), and it is best to mention all "equally reliable" sources in "equal measure", including a reference to the source of the information, that's NPOV policy (see also NPOV tutorial for practical help). So, no, I don't think there's a problem with the interpretation of the word "reliable" here. It means the same as in everyday use. Further, regarding "disinterested parties", well of course, if they're more reliable (for example generally acknowledged to be more reputable for being unbiased or whatever other reason), their version of the story should get a greater share, in the "space and balance" sense of the NPOV tutorial. But being "disintrested" is not by definition being more reliable, example: gay marriage is often discussed by a by definition "disintrested" party called "heterosexuals" - would that party "by definition" be the most "reliable" on that topic? I don't think so! So, really, being "reliable" is the adequate terminology as far as I'm concerned. Note that the Mandy Rice Davies test can be part of an overall assessment of reliability of a party. So, yes, this shows probably at least "something" is lacking in the reliability of both PLO and Israel lawyers (which you forgot to mention), but Mandy Rice Davies test is far from the unique & only test for reliability: Oxford press boasting about international prizes won by their professors is equally "Well they would say that wouldn't they", but I suppose other factors indicating "reliability" would have a greater significance in this case - in the mean while I would however check if the Cambridge press had anything to mention about these Oxford professors and their prizes. --Francis Schonken 12:18, 2 August 2005 (UTC)

Isn't that a rather doubtful analogy regarding 'disinterested'? To say that a heterosexual is not interested in entering into a marriage with someone else of the same sex (which is what I think you meant by 'disinterested'), is hardly the same as meaning that person does not have personal involvement in the subject. Disinterested means not being involved. In your example the heterosexual might be happily married, miserably divorced 10 times, or hate homosexuals and anything concerning them. Any of these might give the heterosexual an 'interest'. Sandpiper 20:15, 25 August 2005 (UTC)

Ad #2

• Already discussed above on this page. Trying to equal "citable sources" with "reliable sources/reputable publications" just doesn't work: it lifts the "citable sources" concept way beyond its literal meaning - which is confusing. Either there is an understood assumption that "citable sources" has a specific meaning in wikipedia context: as this is not explained or clarified, the least that can be said is that it is confusing too. While there is no need to use a specific wikipedia-definition for "citable sources" in order to clarify the importance of the NOR policy, I really don't see what the "citable sources" expression is doing there. --Francis Schonken 07:21, 2 August 2005 (UTC)

Ad #3

• "citable sources" concept, see Ad #2 --Francis Schonken 07:21, 2 August 2005 (UTC)
• As mentioned above on this page the claim that citing sources "encourages" more contributors to add "important" material is POV, not backed by evidence, and thus a quite questionable assertion. SlimVirgin's defence that this statement should remain nonetheless on the policy page on the grounds of "doing no harm" appears refutable by me (see above). --Francis Schonken 07:21, 2 August 2005 (UTC)
• Speaking as someone who has only been here a few months, but who feels they have actually added a few worthwhile elements here and there, I must say that any rule saying I am expected to cite sources absolutely does not encourage me to contribute, completely the reverse. Where I have seen an article with an error I have two choices, either change it or search for a reference and then change it. It is very much more likely I will do the former, or just forget it completely if I have to go and find something to prove I am correct. I have even added a few references, though quite a few people seem to feel that no reference is better than a poor one available on the internet. Ah well, why bother? Sandpiper 20:28, 25 August 2005 (UTC)

Surely you would agree that citing sources is something we should aspire to? In any event, the way this works out, practically, is peoplesometimes do not put in sources and putting in sources only because an issue when the edit is controversial (i.e. other editors think you are wrong) which seems reasonable to me. Be that as it may, you should make this comment on the talk page fore the Wikipedia:Cite sources policy, not here, if you want to generate useful discussion concerning your point, Slrubenstein | Talk 21:31, 25 August 2005 (UTC)

well, no, i think writing here to discuss an element of this policy (section3 above) is more sensible than writing it on a different page. Someone else just protested that citing sources does not encourage contributors (despite what is contended in the section under discussion), and I was agreeing. I do agree that citing sources is a good idea. But on balance I think the content is more important than the reference. If I had to choose between a book with articles and no reference, or references and no articles, then I would take the one with articles every time. The result of this policy is in effect that you have an encyclopedia which values the reference more than the article, and are discouraging people who have the knowledge but strangely do not carry references in their head. In fact, now I think about it, I do not remember seeing references at all in most encyclopedias. So it is rather 'unencyclopedic' to have them. Sandpiper 00:05, 27 August 2005 (UTC)

Well, I didn't invent the policy; it was here long before I arrived. But I support it because I think it promotes verifiability — this is essential given that this is a wikipedia. Some encyclopedia's do not cite sources, but they also have editorial boards with experts policing the quality and accuracy of articles. We do not have that, nor do we want that, so we need other mechanisms. Also, sources point people in the right direction if they want to do more research. I know it is well-intentioned, but I reject the two choices you suggest (all references, no article; all article, no references) because we are not chosing between these two. Nothing is stopping someone from contributing even if they don't provide sources, indeed, it happens all the time. But if another editor says "That is not true!" ... well, this is the real dilemma we face. What happens when two editors disagree as to the validity or accuracy of a contribution? Being able to cite sources is one way of dealing with this kind of problem, and a good one, I think. Slrubenstein | Talk 00:51, 27 August 2005 (UTC)

No, I see the justification for the rule. However, the main guarantee of article quality on wiki is nothing to do with references. References are just used as a tool by people who want to object to an article, in support of their case. So there has to already be some editor who believes there is a problem with a page before references have any effect. The most direct action an editor can take in that situation is to simply change the article himself, or draw attention to the errors. So the primary way articles here get altered is peer review, exactly the same as a published book.

The problem with the rule is that it allows a consensus to be steamrollered by whoever can find the best quote. Now this may not matter so much in a situation where a determined person can go out and find references, but it does become a problem when there is dispute about the reliability of sources, or essentially they are non-existant (especially on the internet, where they are available to us). In that case we are back to the problem of valuing the reference more than the article, and jusifying wholesale deleting. I am struck by the fact that most pages I have seen do not have fully supporting references, or any, and that the encyclopedia would disappear if this rule was rigidly enforced. The question then is whether it is a useful tool to encourage accuracy, or simply a way of furthering arguments. People are liable to claim something is 'not original research' on the basis that references are unavailable, even when their own edits suggest they believe the thing is not 'original' and is even correct, but they want it deleted. The tail is wagging the dog. Sandpiper 07:34, 27 August 2005 (UTC)

Ad #4

• "citable sources" concept, see Ad #2 --Francis Schonken 07:21, 2 August 2005 (UTC)

Ad #5

• "citable sources" concept, see Ad #2 --Francis Schonken 07:21, 2 August 2005 (UTC)

General

• Too long for an intro section; where not questionable it consists of redundant repeats of what is elsewhere on the policy page. The least that can be said is that it contributes to the information flow on the page being disorganised/fragmentary, and thus confusing for the inexperienced reader. --Francis Schonken 07:21, 2 August 2005 (UTC)

Short version

I proposed following short version of the argumentation of why this policy is important:

This policy is important because Wikipedia wants to be a reliable source of information, simply said: reliability above novelty.

Other comments than "reverts" where not yet provided re. the content of this version of the argumentation. --Francis Schonken 07:21, 2 August 2005 (UTC)

I wish you'd stop being so rude. It detracts from your arguments.

Agreed! Please also try not to detract readers of my arguments by your rudeness.

I'm not being rude, Francis! SlimVirgin ^(talk) 08:43, August 2, 2005 (UTC)

I agree you probably weren't aware. Try to read the "Same for you, of course,..." - "Well, no. ..." conversation, or the "my plan is to continue to upgrade the policy to a higher level of efficiency." - "I'm asking you again not to do that." conversation, from where I am, and with some empathy & consideration of wikiquette you might see my point. --Francis Schonken 09:45, 2 August 2005 (UTC)

Regarding "Wikipedia wants": it isn't good English. And I don't understand this sentence at all: "Other comments than "reverts" where not yet provided re. the content of this version of the argumentation."

Yeah, I'm not a native English speaker, I know, it's on my user page: care to help me with reformulating? For instance, is this better English: "This policy is important because Wikipedia aims at being a reliable source of information, simply said: reliability above novelty."? (Anyway I think this far better than rephrasing towards "It's an obligation of wikipedia to...")

I don't understand enough of the sentence "other comments than 'revert' ..." to be able to rephrase it. Regarding "This policy is important because ...," we could try: "This policy is important because if Wikipedia is to become a reliable and respected source of information, it must impose certain restrictions on its content. Simply put: we prize reliability over novelty." (Or: we want reliability, not novelty.) SlimVirgin ^(talk) 09:33, August 2, 2005 (UTC)

About the "other comments..." sentence: I had posted the short "This policy is important because..." version on the policy page (which I shouldn't have done without prior talk, agreed). It was removed ("revert"). The only "comment" I received regarding my short version is that something was probably wrong with the English (which is not a remark regarding the "content" of what I had proposed to put on the policy page, it's only a remark regarding the "form" in which it was presented). Does this clarify? So, indeed, I'm inviting you, like any other wikipedian, to comment on the content of my proposal/version of the argumentation of why this policy is important, that is, including explaining why you think this formulation better/worse than other possible versions. --Francis Schonken 09:45, 2 August 2005 (UTC)

The one thing I think should be tidied is the bit that says published sources tend to be reliable. SlimVirgin ^(talk) 07:48, August 2, 2005 (UTC)

Have any propositions? --Francis Schonken 08:21, 2 August 2005 (UTC)

I'll take another look tomorrow. It just needs a tweak. SlimVirgin ^(talk) 08:43, August 2, 2005 (UTC)

tx! --Francis Schonken 09:45, 2 August 2005 (UTC)

Data about planets — is some calculation allowed?

If we know a planet's (or more specifically a Kuiper belt object's) mass and diameter, does it count as original research if I calculate its density, surface gravity and escape velocity? The formulae for that are quite well-known, trivial to derive, and definitely verifiable. However, I haven't yet found a source where such per-calculated data is given for, for example, 2003 EL61. The reason for that is, in my opinion, that the scientific community does not need such plain figures, and if they do, they can always calculate them as I've done. The only places where such data is useful are encyclopedias and, for example, popular science magazines, but the subject is too new for them (except Wikipedia) to include them yet. Personally I think that it counts as plain synthesis of known sources and not an original research as no research paper would consist of just applying three well-known formulae to well-known data in a totally non-original and routine way. What do you think? — Pt _(T) 11:02, 2 August 2005 (UTC)

FYI: another thing I've calculated from given data is the mean orbital speed for 2005 FY9. The formula for that is perhaps not so easy to derive, but anyway it isn't anything very special and I'm quite sure it has been published somewhere. Also I've checked my formulae with the planets where more data are known. Anyhow, may I do that or not? — Pt _(T) 11:12, 2 August 2005 (UTC)

I don't think NOR policy says anything about whether or not calculation by standard formulae is "allowed". There are other policies/guidelines advising wikipedians to perform calculation every time (e.g.) speed data are mentioned (km/h <-> mph), also Kelvin/Celsius/Fahrenheit conversions, etc. If the formulae are somewhat trickier to find or to apply unambiguously you might help your fellow-wikipedians who might want to check the calculation, by giving the source of the formulae and/or by mentioning which parameters were used (e.g. mentioning with how many decimals gravity acceleration constant(s) were implemented,...). And keep an eye on when such data might get published elsewhere, which might help in double-checking your work. --Francis Schonken 13:11, 2 August 2005 (UTC)

Calculations are by their very nature verifiable by someone with the appropriate maths background, providing you are using accepted math; so I don't think it can be considered to be original in any shape or form; since anyone else can derive the same data. What would be helpful to others would be if you put the calculations in the talk page; so they can be checked by someone else in the future. :ChrisG 17:29, 2 August 2005 (UTC)

Simple derivation of new facts doesn't count as original research; it's typical in many theory-related articles, for example, to give trivial results without a reference. It's a good idea to put derivations on the talk page, to help fact-checkers out. However, be wary of creating misleading figures or statistics, and don't add anything that's nontrivial enough to be publishable (publish it first :-). If an article is describing a proof however, I think it's alright to take a proof or proofs from other sources and simplify/combine them. OR is a subjective thing and it's hard sometimes to tell where the line is. Deco 18:35, 2 August 2005 (UTC)

I don't see any problem with that Pt. We're not supposed to produce a novel synthesis of facts in order to build case, but you're not trying to build a case, and a calculation isn't quite what's meant by a synthesis of facts as we use it here — so long as it isn't a controversial calculation. SlimVirgin ^(talk) 19:16, August 2, 2005 (UTC)

It surprises me, how often questions like this come up (although I take it as a positive sign that people take this policy seriously). I thought the 'apple pie" example would suffice to make clear what is allowable. However, I added to that paragraph this: "(1) makes descriptive claims that are easily verifiable by any reasonable adult" based on ChrisG's very clear and simple explanation above. I do think it is evident that we need to work a bit more on the policy to make it less likely that people will raise these kinds of questions on the talk:page. I am not sure if this means adding even more to the apple pie/current events paragraph in the intro (I am inclined not to), or to add another section in the main body of the article, perhaps including a checklist, that will help people assuage their doubts about whether they are or are not complying with the policy. I consider this one of the most important policies we have, but if many people interpret it in a way that discourages them from contributing to articles — especially in the way Pt raises (which to me sounds more like elaborating on or clarifying or developing a point that has already been established), then we are not done with our work of crafting this policy into something really useful. Slrubenstein | Talk 15:33, 4 August 2005 (UTC)

Thank you all very much for the feedback! I think that the policy should really be a bit clearer about that. — Pt _(T) 18:22, 4 August 2005 (UTC)

What of describing assertions as non-sensical and other simple logical conclusions?

We cannot make an assertion in an article if "it introduces an original argument purporting to refute or support another idea, theory, argument, or position described in the article;"

What if someone makes claims that have no valid semantic interpretation, and they have not provided a full explanation? I'm trying to come up with examples, but it's difficult because I tend to think too logically. The one I've come across is "the documentary was a conspiracy theory". That is nonsense as "conspiracy theory" is not an applicable category to "documentary". Can one not say so in an article where this assertion is quoted? (I guess the similar sentence "the documentary described a conspiracy theory" is the one intended, but what if the other's a construct that's used consistently? Common sloppy use of language should be highlighted in articles. This is perfectly normal in an encylopedia.)

Editors of Wikipedia ought not to editorialize in articles. Period. If an editor writes something that complies with our policies but is unclear, of course ediotrs can improve it. All of this is already in written policies. Slrubenstein | Talk 22:22, 5 August 2005 (UTC)

But accusations of editorialising are too easy to make. It is failing to assume good faith. Making an unattributed assertion is not editorialising per se, but an accusation can be made and cannot be simply refuted in all cases. This highlights the falacious nature of this restriction. Mr. Jones 06:39, 6 August 2005 (UTC)

What of other basic (valid) logical derivations? e.g. X was in Texas, Y was in Kansas, so X could not perform some action with Y. Is logical derivation restricted to the talk pages unless one can show that someone else has made the derivation? Is it prohibited even there? This seems like a very bad policy, and not what was originally intended when the concept of original research was introduced.

What are you talking about? Please read the above (immediately preceeding) section, which responds directly to your question; the policy as currently written answers your question.Slrubenstein | Talk 22:22, 5 August 2005 (UTC)

What do you mean by "what are you talking about?" Don't you understand my question?

The sections on citation, verifiability and NPOV do not cover the notion of simple logical deductions as placeholders. If you feel they do, please indicate where. OTOH, I think that the original statement has some relevance: "synthesize[ing] work in a non-standard way... [is] original research": elementary logical inferences are not (or should not be) "non-standard" That this is the case is far from being made clear by the current policy text, however. Mr. Jones 06:39, 6 August 2005 (UTC)

A further problem is with assertions that cannot be immediately sourced. Placeholders should be allowed for a duration based on good faith (a more fundamental wiki principle that NOR) and the nature of the previous edits of the contributor. Assertions that are not present cannot be sourced.

I hereby object to this part of the policy as it stands. Please sign below if you agree. Suggestions for alternatives welcome.

Otherwise, please explain why you think it's necessary.

Mr. Jones 21:49, 5 August 2005 (UTC)

Hi Mr Jones, I didn't understand the first part of your post, but the example of "X is in Texas, Y is in Kansas, therefore X couldn't touch Y" would either be unnecessary, or not allowed if being used to try to build a case. The point is that we're not allowed to deploy our own synthesis of facts, or our own argument, in order to advance a particular position. SlimVirgin ^(talk) 23:21, August 5, 2005 (UTC)

I do not think you understand the policy. Slrubenstein | Talk 22:22, 5 August 2005 (UTC)

I think I do. Why do you think it's necessary? Mr. Jones 06:39, 6 August 2005 (UTC)

This policy states that there are some things that you cannot do on wikipedia. If the policy page does not mention something, thenn obviously, this policy has nothing to say about that. If you really are doing original research, then do not do it. If you think you are not doing original research, then you do not think you are violating this policy. You make it sound like youthink you are violating this policy and you think you are not. Make up your mind. If you are not doing original research, why are youmaking such a big deal about this? What is the point? Go work on an article. Research a topic, and contribute something. Slrubenstein | Talk 03:04, 8 August 2005 (UTC)

Related policies section

Why were several other policy pages re-hashed here? Linking is sufficient. Effectively creating forks of existing policies here isn't a very good idea. It's also an example of instruction creep.

It is important not to distract from what this page is about: No original research. All this page should do is explain that policy, and no more. The Related policies section seems more a general primer for Wikipedia policies.

A user coming to this page may well have looked at the huge size of it and walked away. With a smaller, tighter page hopefully more people will read it, and take it on board.

I must also say, I prefer the original version best :-). Dan100 (Talk) 13:16, August 11, 2005 (UTC)

In fact I've got stuck in and removed a lot of what seems to have been classic instruction creep: the same thing being said many different ways, which actually makes it harder to comprehend.

I have not, at least not intentionally, changed what the policy says. I have, hopefully, made it much easier to read. As a result, editors are much more likely to read the whole thing through properly, understand it, and take it onboard (rather than being turned off by its great length) - surely the aim of this page.

I shall now run and hide from the inevitable flaming :-) Dan100 (Talk) 13:38, August 11, 2005 (UTC)

Hi Dan, I've restored the deleted material, because the contents of this page were worked out over several months as a draft with a number of people. The reason for discussing the need for sources is that citing sources and avoiding original research are inextricably linked: the only way to show you're not doing OR is to cite your sources. SlimVirgin ^(talk) 19:29, August 15, 2005 (UTC)

SlimVirgin is right, Dan — in general, no one should make changes, let alone major changes, to content before discussing it first. As for the specific material you removed, well, I personally don't think you have a strong case. Wikipedia is not a paper encyclopedia so there is no need to be skimply; the policies are, as SV points our, very much linked. I don't think reviewing or summarizing other policies is "instruction creep." Nor di I see how it makes it harder to follow, ipso facto. Indeed, I think there is a real virtue to showing how, exactly, this policy fits in with others. I thinkk we have to do this here. Of course, if you think it is poorly written i.e. not clear that is a matter of style and we should discuss how to make it clearer. Slrubenstein | Talk 20:04, 15 August 2005 (UTC)

Well, my changes lasted four days, which is longer than I thought they would.

SV, your explanation there could be simply all that needs to be on the page. Why make it harder than that?

Slrub, wiki is indeed not paper but I'd hope our policy pages were intended to educate users, rather than be used as text dumping grounds. Try and think like a newbie: would you really read this massive page? Are you going to come away with a crystal clear understanding? The answer has to be, honestly, no. And therefore this page fails in it's purpose.

Finally, I'd like to say please don't revert on-sight or simply on principle. Try to understand why the other person makes the changes they did, and recognize that they may have a point. Dan100 (Talk) 16:45, August 20, 2005 (UTC)

And what strikes me as being plain stupid is the fact it states the same thing over and over again. We only need to be to be told once, you know! For example: the numbered list just above "What is excluded from articles" is very clear. However it only repeats exactly what's just been written immeadiately above. What't the point in that?! The reader is only going to go "eh?" when confronted with such a muddle.

The "What counts as a reputable publication?" is also nothing more than a simple fork of Wikipedia:Reliable sources. Again - why?

This is classic instruction creep. Dan100 (Talk) 17:04, August 20, 2005 (UTC)

And why do we need such a long list of what primary sources are? If the reader doesn't know, they can just click the wikilink. Dan100 (Talk) 17:06, August 20, 2005 (UTC)

Dan, a couple of points. First, this is a policy page, the draft of which was discussed by several people over several months, so being bold is controversial. ;-) There are only three content-related policy pages: this one, WP:NPOV, and WP:Verifiability. All the rest e.g. Wikipedia:Reliable sources are either guidelines or style guides, and aren't mandatory. This one is, so we have to make sure we cover every eventuality, because this policy is often the only thing that stands between us and nonsense. Also, WP:RS is a fork of this (and not a very good one), not the other way round.

Second, this page isn't written only (or even mostly) for newbies. It's written for seasoned users, because many editors have difficulty understanding the no-original-research rule — especially the issue about a new synthesis of established facts or arguments not being permitted, which lots of editors have trouble with. That's why certain points on the page are labored and are approached from different angles. The intention is to close loopholes. (Having said that, improvements to the writing are welcomed.)

If you think it's too dense or repetitive in places, I'd favor having a longish intro that acts as a summary of the main points so that people (e.g. newbies) needing the quick version can read that and ignore the rest. But the rest should be there too (in my view) in order to address more complex enquiries. I'd like to have a go at writing a succinct but comprehensive intro next week, if no one minds, and put it up for review. SlimVirgin ^(talk) 02:41, August 21, 2005 (UTC)

As there's been no response, I'm going to revert back to the agreed version, but that doesn't mean I don't agree with some of your changes, Dan. It's just that it's going to be easier to edit from the old version than from the new one. And there are sections removed that need to be there. For example, you removed the references to NPOV, but the NPOV and NOR policies make most sense together, and say: We publish only majority and significant-minority views that have already been published by credible sources. That's the core of our content policies right there: everything else — in NPOV, NOR, and Verifiability — is just an explanation of what that sentence means. But without the explanations (long-winded as they sometimes are), some editors won't understand it, and people editing in bad faith will exploit the loopholes.

I'm also going to work on the lead section as I said I would; just haven't had time yet. SlimVirgin ^(talk) 01:26, August 24, 2005 (UTC)

I've retained Dan's intro; restored the deleted material; reordered some sections to improve flow; and created new headers to address Dan's concern about readability. SlimVirgin ^(talk) 03:01, August 24, 2005 (UTC)

With all due respect to SlimVirgin, I have done some more rearranging and editing. I added back the material on "linked oilicies" that had been deleted, but I also pared these sections down to the bare minimum and did some rewriting to show how, exactly, they link to this policy. I also tried to make the organization a little more logical (well, my kind of logic, at least), Slrubenstein | Talk 19:22, 24 August 2005 (UTC)

Thanks for doing that. The reason I deleted the linked policies bit is that Dan got the impression from that that we were needlessly mentioning other policies. So I tried to incorporate them more into the text, I think (I'm speaking from memory and I don't remember much of what I did). But if you've done that, that's fine. I think I spotted that you deleted my Stephen Hawking example, and if you did, I may have to start crying hysterically. ;-O

Just kidding: I wasn't keen on it myself. It's just that so many editors don't get NOR. An example of some kind might be useful. SlimVirgin ^(talk) 19:30, August 24, 2005 (UTC)

Now you know I'd never want to make you cry! In part, my editing was defensive (against anyone who would continue to complain about instruction-creep). Q: would the Hawking example be better placed in the Verifiability policy page? Slrubenstein | Talk 19:39, 24 August 2005 (UTC)

Ok, I'm somewhat recovered from my distress at losing Stephen Hawking, but comforted to see Uncle G has transferred him to Verifiability. I did re-add to this page the one sentence (originally written by Maurreen, and I've always liked it): "The criterion for entry into Wikipedia is verifiability, not truth," because it sums up the NOR policy, and it's one of the things editors have trouble with, especially newbies. Steve, I like what you've done with the page. It should go some way toward addressing Dan's concerns. SlimVirgin ^(talk) 20:12, August 25, 2005 (UTC)

Some concerns with the current NOR article

I have real concerns over this page. While I do not want to see research papers placed into the Wikipedia data because they can not make the grade to get into a peer review journal, I think that this page is very badly worded and lends its self as being used as a bludgeon in arguments over the contents of Wikipedia articles.

For example instead of using outside definitions of what original research is, this article states: "Original research 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". as it is a statement which has not been published in a reputable publication it can be argued that it is Original research because "original research refers to...statements...that have not been published in a reputable publication". Now I know that one can argue that it is covered by the statment lower down the page under "'A few pages have been created devoted to research into issues related to Wikipedia" but that is tenuous to say the least.

It seem to me from the above NOR statment that any statement which is not a quote with a reference, can be construed as original research. For example if one makes the statement in an article "The Thames flows through London" without citing a source then it can be argued that it is original research, because the NOR sentence "n some cases, where an article makes descriptive claims that are easily verifiable by any reasonable adult...Wikipedia article may be based entirely on primary sources" still means that a primary source ought to be cited. But if every statement in every sentence has to be sourced then Wikipedia will become stilted to the point of being unreadable.

See the section above on #Data Mining. In that User:SlimVirgin states ""original research" doesn't refer to any research method", yet as I pointed out to her "research" has a specific meaning in academic circles. I think the term "Original Research" should be based on external to Wikipedia definitions, like one from the OED (if it exists) or those used by respectable peer review journals like Nature or European Constitutional Law Review (EuConst) to take two which are in different fields and would cover diffrent types of Wikipedia articles.

This of course lends its self to a potential paradox. If one tries to publish an article in Nature which is rejected because it is not "original research" could it be published here?

Never the less the defintion for "original research" should be based on external definitions it should not be based on one make up by a few wikipedia editors who happen on this wikipedia page.Philip Baird Shearer 13:28, 27 August 2005 (UTC)

Wikipedia should have its own definition of "original research"; because Wikipedia defines original research far more strictly than academia. I take your point about statements above and I've amended the policy. Statements and facts are governed by the verifiability policy, not the NOR policy, and are dealt with in far greater detail there. :ChrisG 16:30, 28 August 2005 (UTC)

Also, I think that the apple pie and current events examples respond to most of the concerns expressed here. The Thames example is an inappropriate one, given the text of this policy. SR

Lists

"In summary, if the facts, opinions, or arguments you want to include in an article have not been published by a credible or reputable publication, you're engaged in original research." What about lists? As I pointed out above #Bludgeon any list which is not already in the public domain and appears in Wikipedia as a copy of list generated in a EU memeber country is either already protected under EU directives and so is a breach of copyright or the database directive [5] [6] [7], or if written by a EU resident is original research by the definitions given in this article and those of European directives.Philip Baird Shearer 13:28, 27 August 2005 (UTC)

Links to other Wikipedia pages with sources

"Secondary sources present a generalization, analysis, synthesis, interpretation, or evaluation of information or data." By definition if any Wikipedia page fulfils the earlier criteria of no NOR, then surly are they not a secondary source because they "present a generalization... of information or data." ?

The phrase "people who do not rely on Wikipedia" can cause a problem. There needs to be something added to this article about following wikipedia links to sources. For example the article Levellers used to contain this reference to the Putney Debates[8] as does The Agreement of the People[9]. However when the article Putney Debates was written user: DanKeshet removed the first referece (I think correctly) stating in the history of the Levellers article that the "external link already present in our article on putney debates, no need to cite it here, esp. not in a footnote". Yet the wording on the NOR makes no allowance for this very sensible edit. However one has to guard against expecting a person to wade through lots of links to get to an external source which may not exist if the Wikipedia pages go in a cirlcle: e.g. States of Germany->State->State (non-sovereign)->States of Germany! (In this case province would be a better article to link to, but that is a seperate issue.) Philip Baird Shearer 13:28, 27 August 2005 (UTC)

Specific applications

I've done some redrafting, which includes removing two paragraphs which relate wholly to the NPOV policy and add nothing to the the NOR discussion. I like the new specific applications sections and suspect the policy could be best improved with the addition of additional specific applications that keep coming up on the talk pages. For example:

• How to deal popular culture articles
• Calculations
• Philip Baird Shearer's points immediately above

We can probably identify more by going through the talk page discussions in detail. T:ChrisG 16:22, 28 August 2005 (UTC)

Chris, I restored SlRubenstein's version as he had just redrafted it, and I couldn't see what you were trying to do. Could you say here in detail what your objections are before changing the page, please? SlimVirgin ^(talk)

I explained as I went through the edit. I do not believe I have made any substantial changes to the policy, I was simply trying to simplify and clarify, and thus it was just normal editing.

• As a clarification, NOR is not just about untested theories, it is about all theories.
• The paragraph concerning expert editors is duplicated.
• The paragraph concerning how to deal with controversy and debate is not relevant to NOR. It is about attaining NPOV. We have the related policy sections to comment about NPOV.
• The section on how to deal with theories is also classic NPOV subject matter and tells us nothing about NOR.
• I also tried to remove the confusion about facts and statements. While the NOR policy makes it a requirement that statements be verifiable, the NOR policy is mainly about how to properly handle opinion and arguments. The verifiability policy discusses in detail how to handle factual statements. :ChrisG 12:57, 29 August 2005 (UTC)

Over the next few days, I'll be adding a section about the need for cited material to be directly relevant to the article, and not to be an original synthesis of established claims, as lots of editors have trouble with that point. I haven't quite worked out how to express it yet, but I've started by adding to the intro that source material has to be directly related to the article. For example, today we have a new user trying to add an explanation of what the term "fair comment" means in an article saying that a court in New York judged that to call Lyndon LaRouche an anti-Semite was fair comment. The new editor is trying to argue that what fair comment means in law is not what it means in everyday language, and in so doing (whether he's right or wrong), he's trying to build a case, which is not allowed. I'm going to try to come up with a paragraph explaining why this is original research. SlimVirgin ^(talk) 21:48, August 28, 2005 (UTC)

Primary and secondary sources

I'm a little confused about the section that deals with primary and secondary sources. I removed secondary sources from: "Original research is research that creates primary sources or secondary sources." If we quote or discuss a court transcript, for example (a primary source), then we become a secondary source, but that's not an example of original research. If anyone objects to this being removed, feel free to restore. SlimVirgin ^(talk) 02:01, August 30, 2005 (UTC)