Wikipedia talk:No original research/Archive 14

Specialist knowledge
Slrubenstein recently struck out a word in the policy Such an article or section should (1) only make descriptive claims the accuracy of which is easily verifiable by any reasonable adult without specialist knowledge, and (2) make no analytic, synthetic, interpretive, or evaluative claims. Contributors drawing entirely on primary sources shoud be exceptionally careful to comply with both conditions. I believe the conditions are appropriate for articles or sections that rely entirely or predominantly on primary sources, due to the danger of original research. I believe it is overly strict for articles that rely predominantly on secondary sources. I go along with no analytic, synthetic, or evaluative claims in any article. But some simple interpretation should be allowed. Also, some acceptable paraphrases may require some degree of specialist knowledge to verify that the WP article says the same thing as the source; typically about the same degree of specialist knowledge needed to read the article without struggling.

Some might say the entire paragraph is about articles or sections that rely entirely on primary sources, and the last sentence should be interpreted in that context. But because a few words are changed at a time in Wikipedia, I am concerned that any alteration to the first sentence could change the meaning of the last sentence. Therefore I'm puting entirely back in. --Gerry Ashton 18:57, 15 September 2006 (UTC)

my revert
Gerry, after considerable discussion on the talk page you proposed the following:
 * Although most articles and sections should rely predominantly on secondary sources, there are relatively rare occasions when they may rely entirely or primarily on primary sources (for example, current events) or Braunfeld v. Brown. Such an article or section should (1) only make descriptive claims the accuracy of which is easily verifiable by any reasonable adult without specialist knowledge, and (2) the editor makes no analytic, synthetic, interpretive, or evaluative claims. Contributors drawing on primary sources shoud be exceptionally careful to comply with both conditions.

Other people accepted this, and I moved this version to the policy page, just fixing the parenthasis that followes "pie" and should have followed "brown." I repeat: (1) this was your version and (2) there was considerable discussion and support and no objection to it. Because of this, at 16:27, I added it to the policy. I repeat: I added your proposed version, which others had accepted. Then, at 16:32, Wjhonson, without any discussion at all changed your version by ading the word "entirely". So please, Gerry, do not accuse me of deleting a word without any discussion, when I was restoring the version that had support following considerable discussion, reverting the addition of a word that had been added with no discussion. May I add that I prefer your version because the word "entire" is unnecessary. We allow the use of primary sources but they must be used carefully, even in articles where they are not the only source. All editors deserve "good faith" but I had hoped that given our recent interactions I had actually earned some good faith from you. Slrubenstein  |  Talk 19:58, 15 September 2006 (UTC)


 * FYI: it was proposed before and rejected. Second-hand information is certainly not preferred over first-hand information, as hear-say about information is generally less reliable and accurate than the original information. Reverted. Harald88 15:50, 16 September 2006 (UTC)


 * I apologize for my confusion; there have been so many small edits I had trouble remembering which version was which. After studying the history, I agree it was the additon of "entire" which was the undiscussed change.
 * Nevertheless, I interpret the current version to mean that the two conditions apply to articles that are based predominantly on primary sources. In articles based predominantly on secondary sources I would allow more leeway for paraphrases that require some expertise to compare the paraphrase to the original source. I would also allow more leeway for interpretations; of course interpretations should not be original research, but they need not be instantly aparent either. --Gerry Ashton 20:29, 15 September 2006 (UTC)

Thanks Garry. Can you explain a bit more what you mean in your last sentence? When would an editor interpreting a primary source not be doing original research (or be inserting his or her own POV into the article)? Also, concerning the modifier "predominantly" while this was not in the version that had support on the other talk page (which I think we can pretty much archive now, unless you or others object) - I have an open mind but would really like to hear others weigh in before we decide anything. Slrubenstein  |  Talk 22:49, 15 September 2006 (UTC)
 * Interpreting a primary source beyond the standard concise phrasing sometimes necessary (and often even then) is by definition original research. I don't think it matters much what Gerry meant; we must not allow original research or we open the door to everyone with an interpretation, which would be chaos at the very least. Gerry's constant bickering about established policies is becoming disruptive. His position that interpretations should be allowed is contrary to the NOR policy, and is ipso facto not acceptable. One puppy's opinion. KillerChihuahua?!? 23:12, 15 September 2006 (UTC)


 * I resent the implication that I want to allow any and all interpretations. --Gerry Ashton 23:29, 15 September 2006 (UTC)
 * Original research is creating a new (and possibly wrong) idea that people in a field did not know about previously. Interpreting is explaining the meaning of. Not all interpreting rises to the level of creating new ideas. For example, if the author of a primary source makes an argument that alludes to a statement in an earlier chapter of his book, I might combine the argument and the earlier statement into one sentence. This could be considered an interpretation. It might take a little work for another editor to verify that the paraphrase is correct. But is would not be original research. Another example would be adding a Gregorian date in brackets for a quote that uses the Julian calendar; that hardly rises to the level of original research. --Gerry Ashton 23:26, 15 September 2006 (UTC)
 * The problem, Gerry, is that this "caveat" you propose opens a huge can of worms: Now we will have edit wars about what is considered a "valid" interpretation... Thus, it would be better to leave this as is: no interpretations of primary sources allowed Simple and effective policy anyone can understand and apply. ≈ jossi ≈ t &bull; @ 01:12, 16 September 2006 (UTC)
 * I guess we will just have to agree to disagree. --Gerry Ashton 02:18, 16 September 2006 (UTC)

Without a qualifier, the paragraph does not clarify but confuses. The entire paragraph is stating the case of the predominant use of primary sources. Wjhonson 05:42, 16 September 2006 (UTC)


 * I'd say instead that whoever started this stuff about primary, secondary and tertiary sources opened a can of worms. No new interpretation of any source is allowed - what's so difficult about that? Harald88 15:55, 16 September 2006 (UTC)

As explained below, articles should not rely predominanatly on secondary sources but on all notable sources. Sorry for not having noticed it when a corresponding phrase was added. Any serious encyclopedia must strive to rely on primary sources for primary information and on secondary sources for secondary information. Harald88 11:50, 21 September 2006 (UTC)

secondary sources preferred?
Two days ago the following text was added:

Articles which draw predominantly on primary sources are generally discouraged, in favor of articles based predominantly on secondary sources

1. I know nothing about such a policy, nor does it seem to depend on WP:NOR. 2. Personally I disagree, since high quality depends on taking into account all relevant sources for their different qualities; but if anything, IMO such a detailed advice should be motivated and does not belong in policy but in a guideline.

A good alternative that avoids that issue (and to which, as it happened, I reverted):

 Articles that draw predominantly on any one single source are generally discouraged, in favor of articles based on a number of primary and secondary sources.''.

Harald88 16:30, 16 September 2006 (UTC)


 * Note: as Jossi reverted my version on the basis of "no consensus", I now reverted further back to the consensus version of 3 days ago by Jossi. That older version is consistent with the above alternative, as it has that:
 * All articles on Wikipedia should be based on information collected from published primary and secondary sources. 
 * I now note that the useful precision of "published" had disappeared in later versions...
 * Harald88 18:30, 16 September 2006 (UTC)
 * Just saw this change -- I've been going by an earlier version of this policy that allowed primary sources as long as it is verifiable -- so I've used letters, journals or memoirs that are not published but are deposited in archival centers such as Duke University Special Collections Department, or National Archives, etc. I do not interpret these, but only relate facts given in these documents, i.e. such and such did this --plange 16:15, 18 September 2006 (UTC)
 * Can someone get back to me on this? It's a huge change and AFAIK, unpublished primary sources ARE allowed as long as they meet WP:V --plange
 * For now I'll protect the version that does not nor include "published" as it may have been removed by consensus, I don't know. For sure the claim that secondary sources are preferred is non-consensual. Harald88 06:48, 21 September 2006 (UTC)
 * Harald, it is the consensus that secondary sources are preferred; consensus in the sense that that's what all good editors that I know do: they rely on secondary sources wherever possible. Sometimes it isn't possible, of course, and sometimes it isn't desirable either, and knowing how to judge that is something good editors have no problem with. Others do have a problem with it, however, and for that reason, secondary sources are preferred.
 * I've currently got someone trying to use the Bhagavad Gita to show that Hindus have always believed in the concept of animal rights. How do I explain to him that he needs to find a secondary source that says this, if I can't point to this policy that says secondary sources are preferred? SlimVirgin (talk) 12:44, 21 September 2006 (UTC)


 * May I remind you of what you yourself emphasized higher up on this page (and on which I commented): it's not so much primary or secondary stuff that matters but not doing original research that matters; moreover there already is a proper warning statement (although of terrible grammar and even with a spelling error) about the use of primary sources in the same section.
 * Apart from the fact that most sources are mixed, one should always rely for primary information (the data) on primary sources and for secondary information (opinions about the data) on secondary sources. Basing onseself on hear-say is an absolute No-No! But that is not really appropriate for this Original Research Poilicy page: it belongs in part in a guideline for editors, how to write a good quality article; in part in WP:V where it should be stressed that one should rely as much as possible on the original sources (which may be "primary", "secondary"or whatever!).
 * Similarly, WP:NPOV demands to take into account all relevant, notable opinons and there can be no doubt that usually the original source belongs to that group; the original source may well be a "primary" source which - against policy - this policy now suggests should be less relied on.
 * Based on WP:V as well as on WP:NPOV, we are not allowed to base ourself more on secondary sources than on primary sources for primary information.
 * An old secondary source might state (possibly due to ignorance) that So-and-so does not state anything about animal rights [= second-hand primary info], therefore we conclude that [first-hand secondary info], while the nowadays easily verifiable source itself might state Animals should be treated well, they have rights too.
 * Clear, unambiguous statements can stand on themselves, while vague, ambiguous statements (of primary or secondary sources!) don't belong in Wikipedia.
 * Cheers, Harald88 18:36, 21 September 2006 (UTC)


 * Except that you could also point out that they are making inferences from a Primary source, which is explicitly not allowed. Unless the Bhagavad Gita actually says "we support animal rights", making inferences and interpretive stances is clearly not allowed per WP:OR. What my question is about is the insertion of the word "published" in the definition of what primary sources can be used. --plange 14:54, 21 September 2006 (UTC)
 * It's this sentence: "All articles in Wikipedia should be based on information collected from published primary and secondary sources" which is what I'm referring to --plange 14:56, 21 September 2006 (UTC)
 * It seems correct to me. It means that you can't go to Stanley Kubrik's house and quote here from a document you found there in one of his boxes; but you can quote from Jon Ronson's article about his doing just that:


 * Dear Mr Kubrick,


 * Just a line to express to you and to Mrs Kubrick my husband's and my own deep appreciation of your kindness in arranging for Dimitri's introduction to your uncle, Mr Günther Rennert.


 * Sincerely,


 * Mrs Vladimir Nabokov


 * qp10qp 16:18, 21 September 2006 (UTC)


 * Yes, but you should be able to quote from primary material that is housed in a special collections department. That meets WP:V, but yet it's not published. For instance, the article I just recently expanded on John W. Johnston uses secondary sources, but it also pulls from a couple of letters housed at Duke University's Special Collections department as well as a memoir there. However, I also happen to have a primary source in my own possession of something he wrote about his life, but I cannot use it, and am not asking to, because it doesn't meet WP:V. I bring it up as it's a good illustration of the difference between a primary document that isn't published but is verifiable (and so is useable), and another that isn't published and is not verifiable (and isn't useable). --plange 17:45, 21 September 2006 (UTC)


 * That goes beyond the issue that I brought up. The newly added phrase (perhaps it was discussed on a new subpage of which I now discover its existance, and which of course is not tracked by my watchlist of OR discussions?!) already flies against other policies as I stressed here above. It would also be illustrated here by the hypothetical case that the manuscript has been photocopied and is published online; and that it flies in everyone's eyes that a claim made by the secondary source about primary information is erroneous. WP:NPOV demands inclusion of both and WP:V effectively degrades the secondary source ("the burden of evidence", "reliability") - which is what the disputed sentence discourages to do. Harald88 18:47, 21 September 2006 (UTC)
 * Okay, I'll start a new thread as I wasn't commenting on your original issue (and can't quite understand what you just said) but rather it flagged for me that the word published was in there. Sorry to have hijacked yours. --plange 19:00, 21 September 2006 (UTC)
 * Perhaps and perhaps not. It will depend on exactly how they are using the Bhagavad Gita to state this.  If they say "Give respect to all animals" (Bhagavad Gita, Sura92.3) that is perfectly acceptable as a quote from a published source.  If however they say "Well the Gita expressed in 42 sections how valuable animals are." that would be their own statement which is subject to debate.  As I've said for a while, direct quotes are not original research, and a perfectly valid use of primary material. Wjhonson 16:48, 21 September 2006 (UTC)

Original research
Prior to my edit "Original research that creates primary sources is not allowed." sounds to my ear stilted and unsatisfactory. It gives the impression that possibly there is "original research" which "does NOT" create primary sources. And as we all know, all original research creates primary sources. So I've edited it to say "Original research, that is research which creates primary sources, is not allowed." which, to me, sounds much more clear and precise. Wjhonson 17:02, 16 September 2006 (UTC)
 * Ga... The fact is, all works are primary sources, so as I've always said, this expression is currently poorly worded. What we are wanting to prevent is work which speculates upon new concepts or ideas (the primary source concept being used in this case) or performs non-trivial analysis or any deduction of any other work (a secondary source form of original research). LinaMishima 17:20, 16 September 2006 (UTC)
 * Sorry, I know that was off-topic from your post somewhat :/ LinaMishima 17:22, 16 September 2006 (UTC)


 * Various definitions of secondary source exist. Three descriptions from the James Cook University Library web site are


 * "describe, interpret, analyze, and evaluate the primary sources
 * comment on and discuss the evidence provided by primary sources
 * are works which are one or more steps removed from the event or information they refer to, being written after the fact with the benefit of hindsight"


 * Some of these actions constitute original research, so sometimes creating secondary sources is original research. --Gerry Ashton 18:56, 16 September 2006 (UTC)


 * Yes. It's not clear what you try to argue. In case you suggest that basing oneself on secondary sources is "safe": Any creation of sources of n+1 from sources of n carries the risk of original research, as also Slimvirgin has pointed out here above.
 * Moreover, the existence of secondary sources does not liberate editors from the obligation to consult primary sources that are available as well. Harald88 22:37, 16 September 2006 (UTC)
 * PS, for a very recent example of OR with secondary sources see Talk:Global_Positioning_System Harald88 00:19, 17 September 2006 (UTC)


 * It should be noted that Wjhonson never intended for a debate here over a line that's been (unfortunately) happily in NOR for quite some time. By the nature of wikipedia, we serve as both a secondary and tirtary (correct spelling if you can) source on subjects and so the line's intent, to prevent new facts being introduced, was good (if badly worded, as everything is a primary source for itself by definition). In being a secondary or tirtary source, we have to by definition describe, paraphase and summarise sources of a lower order; but in doing so we should not perform any non-trivial analysis or interpretation. As a good secondary and tirtary source, we have, to a certain degree, to perform some evaluation of the sources being used - mainly to allow WP:NPOV to be fully met..
 * Ironically, all this puts wikipedia into a strange position, as crafting a good article will mean carefully summarising sources and standpoints, indicating their stated allegiences (detailed investigations are out of bounds, though), judging the due weight of the varies views and discuss (by comparison with other sources) the views. However unless this is done very, very carefully, it quickly becomes OR and unverifiable, and it is easier to enforce competance via strongly prohibitive rules. Interestingly, I've just proven why WP:IAR exists :P LinaMishima 23:31, 16 September 2006 (UTC)


 * Interesting, I didn't even know that contrary policy! Harald88 00:19, 17 September 2006 (UTC)

Nonsense. Summarizing and paraphrasing is not synonymous with interpreting or creating. LinaMishima, you really don't understand policy. You think 3RR applies to groups and now you think there is a contradiction where there is none. I suggest you read the policies carefully, and ask questions of experienced users, rather than making such mis-informed and contentious posts on WP namespace. KillerChihuahua?!? 09:22, 17 September 2006 (UTC)
 * KillerChihuahua, if you had read this talk page, you will see that other editors are equating summarizing and paraphrasing (allowable things) with interpreting or creating (not allowed things). And to a certain degree that is understandable, as these good things when done poorly do fall easily into the realm of the bad things. I advise that you assume good faith and that I need a gentle detailing of were I have misinterpreted matters or the dialogue on this page, rather than something that does not feel friendly to read LinaMishima 14:47, 17 September 2006 (UTC)
 * Splitting hairs. Policy cannot and should not cover each and all eventualities. ≈ jossi ≈ t &bull; @ 15:22, 17 September 2006 (UTC)
 * KillerChihuahua suggested that LinaMishima "read the policies carefully, and ask questions of experienced users" in order to understand whether interpret and create are synonymous with summarize and paraphrase. I suggest this is a straw man argument, because the words need not be synonymous, it is only necessary that the meanings overlap to create confusion. I also suggest that since none of these words are defined in the policy, the proper place to understand their meaning is a dictionary, not the policy. If the meaning within the policy differs from the dictionary definition, the policy should be changed to either use other words, or define the specialized meaning. --Gerry Ashton 18:53, 17 September 2006 (UTC)
 * I thought she was making a point that "we should not perform any non-trivial analysis or interpretation". How is that a lack of understanding of the policy? Also please avoid ad hominem arguments. They don't help. You have caricatured her 3RR points, which were argued on a sensible basis. Tyrenius 00:41, 18 September 2006 (UTC)
 * Tyrenius, a lack of understanding policy so clearly in one place can be indicative of an overall unfamiliarity of policies in general. In what way can that possibly be an ad hom? Ad homs are directly about the person - "you're an idiot" not about their actions - "you clearly do not understand policy". LinaMishima: you say "if you had read this talk page" - how, precisely, do you support your allegation that I have not? Perhaps you are claiming you are a psychic? KillerChihuahua?!? 11:08, 18 September 2006 (UTC)
 * "You...do not understand" and "if you had read", same difference. I disagree with your evaluation of her understanding of policy. I think she showed a deep understanding of the purpose of that policy. Best to stick to the actual point presented on this page and address that. Tyrenius 23:31, 18 September 2006 (UTC)
 * Nonsense. They are very different. KillerChihuahua?!? 23:34, 18 September 2006 (UTC)

We had a couple of weeks worth of discussion concerning edits to the primary and secondary sources section. I broke down the discussion to specific edits. As we reached consensus on each edit, I made the edit on this page. If there was significant opposition to any proposed edit on that talk page, I did not make the edit to this page. Now I see that one or two editors reverted, unilaterally, those changes. This is not how we do things here. We have discussion, and when a number of people reach agreement, we make the change. That is waht we did. Slrubenstein  |  Talk 10:33, 18 September 2006 (UTC)


 * I notice that here you comment on another subject than "creating primary sources".
 * The way things were done here - the only way that works - is that any significant change or addition is discussed for weeks and next after implementing it, one waits another week or so for other editors who missed an essential change to make a motivated revert before assuming real consensus. We can't expect that the whole community reads all the discussions every day, or claim when for example some editor is on vacation that a consensual edit was made without him/or that can't be reverted. Apart of that, it's Wikipedia practice to move disputed sentences to the Talk page and only reinsert them only after new consensus has been reached. Harald88 10:57, 21 September 2006 (UTC)


 * Hi Harald, you left a note on my page about an incomprehensible edit. All I did was revert your change. SlimVirgin (talk) 11:07, 21 September 2006 (UTC)


 * Hi Slimvirgin, all I did was to revert to a consensus version, as specified. I could find no discussion with opinion poll that warrants the addition that I reverted, while I clearly objected to it with motivation oñ this page. Harald88 20:59, 24 September 2006 (UTC)


 * In my opinion, the reverts by three editors yesterday were a mile away from best practice. A reversion should be explained on the Talk page, or, better still, proposed there. Editors should have the patience to allow good-faith edits they disagree with (I don't mean little things like spelling mistakes etc.) to hang there for a few days while they address the issue on the Talk page. Wikipedia won't explode in the meantime, honest.qp10qp 11:16, 21 September 2006 (UTC)
 * This is a policy page. We need consensus before making major changes, and there's been far too much reverting back and forth. I suspect most people have lost track so it needs to stop. SlimVirgin (talk) 12:38, 21 September 2006 (UTC)


 * OK, We have a recent starting point after the page had been frozen because of edit-warring: "marked as consensus version" by Jossi. It does not contain the disputed sentence which certainly was added recently.
 * As the sentence in question has been added without a clear consensus, and is in apparent conflict with the other non-negotiable policies, I'll remove that sentence again and we can discuss to either put a revised version back, or simply leave it out as already a sentence that warns for the use of primary sources is included.
 * Harald88 20:59, 24 September 2006 (UTC)

Consensus
Removed post from banned user. SlimVirgin (talk) 15:49, 18 September 2006 (UTC)


 * Removed my post as well, as now my response makes no sense. KillerChihuahua?!? 14:46, 18 September 2006 (UTC)


 * Sorry, KC, I didn't mean to cause chaos. :-) SlimVirgin (talk) 08:01, 21 September 2006 (UTC)

Suggested change to opening paragraph
I suggest changing the opening paragraph to read:

"Wikipedia is not the place for original research. Citing sources and avoiding original research are inextricably linked: the only way to demonstrate that you are not doing original research is to (a) cite reliable sources which provide information that is directly related to the topic of the article and (b) adhere to what those sources say."

The only change is to clearly separate the two requirements. I suggest this change because it makes more clear these two requirements. I'm usually happy to directly make such changes myself but this is a very high-profile policy and I'm hesitant to make any changes in the lead paragraph without consulting others. --ElKevbo 15:03, 18 September 2006 (UTC)
 * the two requirements are already separated by a comma. It is not a very complex sentence. Slrubenstein   |  Talk 15:20, 18 September 2006 (UTC)
 * Picking nits: In the current version of the sentence, they shouldn't be separated by a comma. It's not consistent with standard English as I know it to place a comma between only two items separated by a conjuction.  The most relevant section of the MOS seems to be the section entitled "Serial commas" which would appear to support this assertion but it's not completely clear since it only addresses comma usage in lists of three or more.   --ElKevbo 15:42, 18 September 2006 (UTC)
 * I agree with ElKevbo entirely. In my opinion, the present opening paragraph is written in bad English verging on wrong English. ElKevbo's rewrite is an improvement, but it still doesn't get round the oddity of saying that there's only one way and then giving two ways. Semantically it makes sense (the only way consists of doing this and this); syntactically not.qp10qp 23:53, 18 September 2006 (UTC)

Although I disagree, I won't fight over it - if a number of people prefer ElKevbo's suggestion, fine. If Qp10qp wants to propose an alternative, fine. I do think we should allow a couple of days' discussion and see what others think before rewriting it though. Slrubenstein  |  Talk 09:44, 19 September 2006 (UTC)
 * To be honest, trying to think of a rewrite hurt my brain. I think it needs redrafting and can't be fixed by relocating a word or two and I doubt that anyone would be in the slightest bit interested. I'm a little surprised though, that you disagree that there's something wrong with the following:


 * the only way to demonstrate that you are not doing original research is to cite reliable sources which provide information that is directly related to the topic of the article and to adhere to what those sources say.


 * Firstly, the readers may assume that they are being told about one way and follow that way through only to be surprised at the end by the second way tagged on at the end. This wouldn't be so bad if the first way was short (is to cite reliable sources and to adhere to what those sources say), which would establish matching parts. The trouble is that the long clause describing the first part of the "way" creates a partial miscue for the reader by appearing to govern the final clause (which provide information that is directly related to the topic of the article and to adhere to what those sources say). qp10qp 11:38, 19 September 2006 (UTC)

Okay, here is what I propose as an alternative:
 * Original research is not permitted at Wikipedia.  Original research refers to unpublished material, especially arguments, concepts, data, ideas, statements, or theories, or any new analysis or synthesis of published material that appears to advance a position — or, in the words of Wikipedia's co-founder Jimbo Wales, that would amount to a "novel narrative or historical interpretation".  Wikipedians are instead expected to do draw on relevant, reliable and verifiable sources, and to cite sources in articles.

and then delete the "definition" that appears shortly thereafter. Slrubenstein  |  Talk 11:53, 19 September 2006 (UTC)


 * Well, that's bold; but I have a hunch it mightn't be accepted.


 * I've been trying to think of a grammatical improvement that doesn't affect the content and sense: what about this?


 * Citing sources and avoiding original research are inextricably linked: the only way to demonstrate that you are not doing original research is to cite and adhere to reliable sources which provide information that is directly related to the topic of the article. qp10qp 20:34, 19 September 2006 (UTC)
 * Looks good! I don't really like the colon (should it be a semi-colon?) but otherwise it's much better.  --ElKevbo 22:41, 19 September 2006 (UTC)


 * The emphasis should remain on "to adhere to", see the example of.
 * Harald88 22:43, 19 September 2006 (UTC)
 * Are you suggesting that the order of the words be changed or would more substantial changes be required to answer your objection? And, more philosophically, how can we know if an editor is adhering to a source if they don't cite it?  --ElKevbo 22:52, 19 September 2006 (UTC)


 * Elkevbo: the colon was in the original; but anyway, I think it's fine (Fowler said that a colon "delivers the goods that have been invoiced in the preceding words"). Harald88, I don't quite understand what you mean; please explain more. qp10qp 00:51, 20 September 2006 (UTC)


 * We already have WP:V, thus it's superfluous (and-off-topic!) to state that one should cite sources. Citing sources is not sufficient: what should catch the attention of the editor who reads this policy is that the expressed opinion and claims must match that of the used sources. That's made clear by splitting it up and ending the sentence with "adhere to" for emphasis, like ElKevbo suggested.
 * Maybe simply like this, reworking the last version above:
 * Citing sources and avoiding original research are inextricably linked: the only way to demonstrate that you are not doing original research is to cite reliable sources which provide information that is directly related to the topic of the article, and to adhere to them. Harald88 10:43, 20 September 2006 (UTC)
 * How about:
 * "Citing sources and avoiding original research are inextricably linked: the only way to demonstrate that you are not doing original research is to adhere to cited, reliable sources which provide information that is directly related to the topic of the article."
 * At a bare minimum, the comma near the end of your proposed sentence has to go. I also ask if the phrase "sources which provide" is correct.  Should it be "sources that provide?"  The which/that distinction is one with which I still struggle...  --ElKevbo 14:17, 20 September 2006 (UTC)
 * I hope that version will satisy Harald88's wish for adherence to be emphasized ahead of citing. (Harald88, you must listen to ElKevbo when he says that your comma is wrong. He is spot on.) qp10qp 15:43, 20 September 2006 (UTC)

On the "which/that" issue, they are interchangeable in this construction, though not in some others. And the sentence is thatted up enough already, in my opinion (in fact, it would be elegant to drop the final "that is" entirely: information directly related to...). qp10qp 15:43, 20 September 2006 (UTC)
 * Okay, so removing that "that phrase" and adding the wikilinks (to get the full effect) would leave us with:
 * "Citing sources and avoiding original research are inextricably linked: the only way to demonstrate that you are not doing original research is to adhere to cited, reliable sources which provide information directly related to the topic of the article.
 * How's that look? I think it retains the spirit and letter of the original sentence but states it in a more concise and grammatically correct manner.  --ElKevbo 16:12, 20 September 2006 (UTC)
 * OK insofar as the grammar goes; but now the emphasis on what NOR is about (to adhere to) is gone, while already the current version apparently lacks emphasis as I illustrated. People already tend to confuse WP:NOR and WP:V. Perhaps add a sentence about the difference? Harald88 06:39, 21 September 2006 (UTC)
 * The emphasis is gone? How is adherence being the first injunction in the first sentence of the whole policy not emphasis enough? In my opinion, the above version is a real improvement, and I would vote for it to go in. We need to wait a while first, of course. qp10qp 10:58, 21 September 2006 (UTC)

Adhere is much more emphasized when mentioned separately, in the end - which is the case now. That's perfectly in line with ElKevbo original proposal for improvement. For now, my proposal for improvement is to simply put adhere in bold in the existing phrase, which aids with Elkevbo's objective:

Citing sources and avoiding original research are inextricably linked: the only way to demonstrate that you are not doing original research is to cite reliable sources which provide information that is directly related to the topic of the article and to adhere to what those sources say.

In fact either this or splitting it up is necessary for the same effect now that the comma has been removed. In other words, I disagree with removal of the comma without a consequential rephrasing: removing a comma is not always harmless. Harald88 11:58, 21 September 2006 (UTC)


 * I'm sorry; I don't follow you. Removing the comma doesn't change the meaning of the sentence at all.  It only brings it into compliance with Standard English.  I can understand and agree with a general argument that minor changes to a sentence (such as removing a comma) may sometimes have a large impact on its meaning but I don't think that's true in this case.
 * In any case, I think your formulation with the comma removed and the critical words bolded is acceptable. --ElKevbo 18:28, 21 September 2006 (UTC)


 * A comma splits a phrase in two, allowing to more clearly distinguish two parts. Removing the comma as was done removed that clearness, while your original proposal above was to enhanced it. That's all. Harald88 19:47, 21 September 2006 (UTC)


 * A comma should not be used to separate verb phrases in a complex predicate. It's poor grammar and completely unnecessary.  --ElKevbo 06:32, 22 September 2006 (UTC)


 * We all agree on the first part; while your last claim is obviously false by the very fact that the comma was intentional. Harald88 20:47, 24 September 2006 (UTC)

Punctuation of current version
I suggest one of these satisfies requirements: Tyrenius 21:50, 22 September 2006 (UTC)
 * 1) the only way to demonstrate that you are not doing original research is to cite reliable sources which provide information that is directly related to the topic of the article (and to adhere to what those sources say}.
 * 2) the only way to demonstrate that you are not doing original research is to cite reliable sources which provide information that is directly related to the topic of the article &mdash; and to adhere to what those sources say.


 * Two more good versions. So, when is someone going to take the plunge and replace the bad sentence presently in the text with one of these better versions?


 * Harald88, I'm disappointed that you're still arguing with ElKevbo about the comma. If you need convincing that it was bad English, here are some references for ElKevbo's point:


 * "If there's a new subject—or the old subject is restated—use the comma. If the second clause shares a subject with the first (and that subject is not restated), don't use the comma. (Bill Walsh, Lapsing into a Comma, Contemporary Books, 2000, ISBN 0-8092-2535-2: p76)


 * "In compound sentences, an unnecessary comma is sometimes inserted before a second independent clause when the subject is the same as in the first clause. (As some grammarians put it, a comma shouldn't appear before the second part of a 'compound predicate'.)" (Bryan.A.Garner, A Dictionary of Modern American Usage, Oxford University Press, 1998, ISBN 0-19-507853-5: p539)


 * "When the subject is the same for both clauses and is expressed only once, a comma is useful if the connective is but. When the connective is and, the comma should be omitted if the relation between the two statements is close or immediate." (Strunk and White, The Elements of Style, 4th Edition, Allyn and Bacon, 2000(1979), ISBN 0-205-30902-X: p5)


 * qp10qp 17:49, 25 September 2006 (UTC)

The solution seems obvious; this reversal:
 * the only way to demonstrate that you are not doing original research is to adhere to reliable sources which provide information that is directly related to the topic of the article, and then to cite them to show that you have done so.

Editors should adhere to sources before they cite them, shouldn't they? Septentrionalis 20:31, 25 September 2006 (UTC)

Several changes per day - please stop!
Gentlemen, may I remind you of the following:

''When editing this [article], please ensure that your revision reflects consensus. When in doubt, discuss first on the talk page.''

In the past it was customary to discuss a significant modification of text (not spelling or grammar errors) for several weeks, thus giving aother editors the time to have a look at it, think about it and comment on it; and only when a true consensus was reached, a change was made.

In contrast, now several changes per day occur in full defiance of the above rule, and without a clear correspondence to the recent discussion threads. I'm afraid the page needs protection again. Harald88 22:38, 19 September 2006 (UTC)


 * Agree. If this situation persists, I will protect the policy page. ≈ jossi ≈ t &bull; @ 00:23, 20 September 2006 (UTC)


 * One of the changes was to fix a problem I noticed in Wikipedia_Talk:Ignore all rules. This policy says that no rules supersede original research, but IAR says that IAR does supersede it.
 * It's a contradiction. Either you have to change this rule to say that it can be superseded by IAR, or you have to change IAR to say that some rules may not be ignored.  Leaving it the way it is makes the policies logically inconsistent. (And there was in fact discussion, but in IAR, not here.) Ken Arromdee 05:02, 20 September 2006 (UTC)

Ignore all rules is part of the wiki process (foundation issue #3 "wiki process" ).

We're not going to protect this page unless there's vandalism or editwarring (foundation issue #2 "anyone can edit").

This rule helps us maintain NPOV (foundation issue #1 "NPOV").

None of these are up for debate today.

If the current version of this interpretation of these basic policies causes a contradiction; I suggest you rewrite the interpretation, or revert these guidelines to versions which do not contradict each other. Those versions exist, I assure you. Kim Bruning 09:37, 20 September 2006 (UTC)


 * Right now, the article says that "The principles upon which these three policies are based are non-negotiable and cannot be superseded by other policies or guidelines". This is slightly better in that it can be interpreted to mean the *rule* can be superseded, but that requires a rather strained reading of it.  The most straightforward reading of the phrase still appears to directly contradict WP:IAR--you can't say a policy may not be superseded and then in the other policy say that it sometimes supersedes the first policy.
 * Unless we're going to change "ignore all rules" to say that it doesn't apply to original research, this should be changed in order to avoid the contradiction. Ken Arromdee 14:01, 25 September 2006 (UTC)

User:Tyresias
Just a note that Tyresias, who has edited the policy page, has been indef blocked, and also that he should not be confused with me. Tyrenius 23:54, 19 September 2006 (UTC)

Lost clarification against editing policy
With editing the following clarification was edited out:

''Their policy pages may be edited only to improve the application and explanation of the principles. ''

However, the remark about possible change on foundation level is much less clear and could easily be misunderstood (or simply not understood). BTW where was the discussion?

Thus I disagree with the removal and propose to put it back in, so that we obtain:

'' In practice, the principles upon which these three policies are based are only negotiable at the foundation level. Their policy pages may be edited only to improve the application and explanation of the principles.

Harald88 10:53, 20 September 2006 (UTC)


 * No. The pages may be edited by anyone for any purpose that improves the encyclopedia.


 * The discussion you seek was in 2001. You are expected to accept the foundation issues before you start editing wikipedia on a regular basis. This is the basis for Jimbos remarks about certain things being "non-negotiable". I'm not going to discuss this with you here and now, because this subject can effectively only be discussed at foundation level, if at all. If you don't accept the foundation issues, you are quite welcome to join a different project, no one will hold it against you. Kim Bruning 13:32, 20 September 2006 (UTC)


 * Please summarize that discussion, as the removal of the abovementioned phrase was apparently against consensus and IMO the precision is agreement with your above remark; on top of that it appears from Jossi's remarks below that all the changes were in fact erroneous. Harald88 06:43, 21 September 2006 (UTC)
 * The Foundation issues page is that summary. Kim Bruning 12:16, 21 September 2006 (UTC)


 * Fine. As Jossi tried to make clear, the foundation issues don't really matter here (WP:NOR is not mentioned); and as I tried to make clear, even if it was mentioned, it's not the point that was essential - what is essential is the above phrase which you apparently deleted without consensus. Harald88 20:45, 24 September 2006 (UTC)


 * FYI, foundation issues does not address WP:NOR, WP:V or WP:NOT, nevertheless, the principles upon which these policies have been written are non-negotiable. ≈ jossi ≈ t &bull; @ 21:57, 20 September 2006 (UTC)


 * And yet the foundation issues are beyond debate here. There are two outcomes.
 * * Either you show how these three items are tied in with the foundation issues, so that you can safely say that these are based on principles that are beyond debate.
 * * Or you are making statements about policy that flatly contradict the foundation issues.
 * (page up a section or two for more detail).
 * Kim Bruning 08:53, 21 September 2006 (UTC)


 * Restored foundation issues after 6 days. If you want to negotiate "non-negotiable" issues, I strongly reccomend you take it to m:Talk:Foundation issues. Kim Bruning 20:15, 27 September 2006 (UTC)


 * Those foundation issues are a recent non-consensual revision; and on top of that you deleted a long-time undebated clarifying sentence. Reverted. Harald88 20:23, 27 September 2006 (UTC)


 * That is false. The foundation issues have very wide consensus at the foundation level, so much so that they are essentially beyond debate, especially on wikipedia. If you wish to override them, I suggest you take it to the foundation level. Kim Bruning 20:43, 27 September 2006 (UTC)

Please don't start discussions on other pages without repeated mention on this page!
A technicality of importance about, in this case: /Primary v. secondary sources discussion

I just noticed that a discussion was going on of which I completely ignored its existence because I trust in the overviews of my watchlist to see what discussions are going on about subjects on their Talk pages. It turns out thatSub-pages are not automatically included.

Thus in case such separate discussion pages are set up, regulary summaries and/or remarks need to be given here for the wider audience, otherwise it becomes de facto off-the-record for the larger group of generally interested editors. In my case, I'm not at all interested in primary/secondary as it is irrelevant for this policy (but possibly many interested participants think differently and made a corresponding edit!), and anyway I did not notice the creation of that page on my watchlist.

Harald88 19:00, 21 September 2006 (UTC)


 * Which was my exact objection when the subpage was created. I found the entire episode really troubling on many levels, this being one of them.  It completely splits off the discussion and de facto hides it from the rest of the editors. Wjhonson 06:15, 22 September 2006 (UTC)

Only published primary sources?
I'm starting a new thread as apparently I hijacked another thread (didn't mean to) and it was confusing the issue. So am restating my concern. I notice that the policy now says (it didn't use to) that primary sources must be published. The problem with this is that there are primary sources that are deposited in Special Collections departments and archives that are NOT published but do satisfy WP:V since they are accessible and verifiable. For instance, the article I just recently expanded on John W. Johnston uses secondary sources, but it also pulls from a memoir housed at Duke University's Special Collections department. However, I also happen to have a primary source in my own possession of something he wrote about his life, but I cannot use it, and am not asking to, because it doesn't meet WP:V. I bring it up as it's a good illustration of the difference between a primary document that isn't published but is verifiable (and so is useable), and another that isn't published and is not verifiable (and so isn't useable). With the new wording on WP:OR, the first instance is now not useable. --plange 19:04, 21 September 2006 (UTC)


 * Your personally owned document could not be included in any manner whatsoever in Wikipedia until or unless it is published (in some manner) by a reliable source. The same would hold true for family owned civil war letters which have never been published.  The question, "may a Wiki article be written about such a document" falls under WP:NOR but the reason it does is because WP:V requires previously published by a reliable source.  So, were you to photograph your document and put it on a personal webpage for all to view, it could not be used by Wikipedia.  But, if the New York Times (or your local newspaper) published an article about it, then a Wiki article could be built around the newspaper information and an "exterior link" could be placed in the article which linked to your document on your personal website. Terryeo 20:28, 25 September 2006 (UTC)


 * IMO this is all about WP:V and therefore this issue even doesn't belong on this Talk page!
 * Insofar as this WP:OR policy proposes new WP:V policy it should be deleted from this policy page and proposed on the corresponding policy page. Ouff what a mess!
 * But off-the record: such a verifiability sounds very limited to me, I guess editors should physically go there to verify it? Harald88 19:43, 21 September 2006 (UTC)
 * Yes, and on WP:V they maintain that WP:V is not about ease of verifying, but that it ultimately IS verifiable. I agree, WP:OR should only be about parts that have to do with OR. It seems like that was the intent, as there's a line at the beginning of the section in question that says "All sources must be verifiable" and links to WP:V, but then later on the word "published" is in there which is the realm of WP:V and contrary to my understanding of WP:V. I'll scurry over there and make sure... but I'm pretty sure I'm right because I recently (in the last week) had a discussion there that explored just this issue. --plange 19:57, 21 September 2006 (UTC)
 * Plange, when you said that you should be able to use primary sources that are in official collections, I entirely agree with you. In fact, it sounds to me as if you are doing excellent work, and it makes me want to go and read your contributions. But I have always thought that the word "published" meant literally that (and I am sure that requirement has been about before and not just in recent additions to this page).
 * On the other hand, I think that the collections you use probably have a published catalogue: if you quoted the catalogue in the references, I am sure that no one would object. There are so many unreferenced articles in wikipedia that the work you are doing is highly unlikely to draw comlaint. The issue will arise on wikipedia more now with the growing importance of footnotes, which are usually the province of academic articles and books rather than encyclopaedias; and we all know that primary sources are not only used in academic articles and books but hold a high premium there. I suspect you are actually one step ahead of the policy, and I hope you will continue what you are doing.qp10qp 22:34, 21 September 2006 (UTC)
 * If you use an unpublished, yet public, memoir, you should at least cite it fully in the sections you use and probably with quotes as well. In other words, you should make it clear that that is what you're doing.  I think it probably does run afoul of the requirement that something be "published".  I'd suggest you type out the relevant parts of the memoir and submit them to wikisource at least.  Then you could even link them from the biography. Wjhonson 06:14, 22 September 2006 (UTC)


 * I guess that the above implies that "published" does not belong in that sentence. Does anyone disagree? Harald88 21:28, 24 September 2006 (UTC)

I would be very careful here. I am persuaded that plange is acting responsibly, largely because he she has gone to so much trouble to discuss his her privately owned memoir, when he she could probably have slipped much of the data in without challenge even in a FA. But archival research does very often require the sort of interpretation and selection that are OR. More later. Septentrionalis 05:59, 25 September 2006 (UTC)
 * Actually I'm of the fairer sex :-) The privately-owned one I'm not pursuing as I've realized I need to get that published (which I should be doing anyway). The above is about publicly held documents at Duke University which I already used in the article. My thoughts are that these are a reliable source, but that I should still be subject to the same rules of WP:OR as someone using a secondary source. I brought the question up over on WP:RS and haven't gotten much feedback, but so far no one's disagreed. --plange 14:11, 25 September 2006 (UTC)
 * My apologies; I've commented there. Septentrionalis 17:26, 25 September 2006 (UTC)


 * WP:NOR is an appropriate place to begin, until you have satisfied yourself about the applicibility of a Wiki article to your document. Then the next stop (were you to wish to publish it) would be WP:V, which would tell you what level of reliable publication would be needed for your document's content to be considered published.  A personal website won't do it for a Wiki article.  But WP:RS really should have responded to your question in the first place, heh. Also worth mentioning, there are some history majors who edit Wiki regularly.  I can't recommend anyone, but there are history buffs who have experience in the area. Terryeo 20:59, 25 September 2006 (UTC)
 * Thanks Terryeo, they did answer and so far agree it is reliable... It's not on a personal website (which I agree is a no-no) but are public documents deposited at reputable archives (National Archives, Duke Univ, etc). I also have a Master's in Heritage Preservation and so am used to being able to use primary sources (and how to cite them), but just wanted clarification on what was allowed here. By the wording on this page, it says I cannot use primary sources that have not been published, so that's why I asked. I'm thinking for the purposes of OR, it should refer people to RS and leave it at that... Thoughts? --plange 21:10, 25 September 2006 (UTC)
 * I think some of will recall a *very* long conversation about WP:V being easy vs. difficult. The price the Mayor of San Francisco paid for his townhouse, exists in an archive in the county land records office.  Some editors were very adament that such an item is hard to verify and so should not be used.  The same applies for any primary item not published, but still public.  My own contention was WP:V only requires it be *possible* to verify, not that it should be easy.  One difference however, is that microfilmed records can travel by ILL or similar sort of requests.  Microfilming is a process of "publishing".  If your primary items are microfilmed and can be ordered through ILL, I'd say they would pass WP:V, otherwise the relative inaccessibility of those items, might be an issue to some editors. Wjhonson 04:53, 26 September 2006 (UTC)
 * Plange's involvement brings an interesting question foreward. If WP:NPOV, WP:NOR and WP:V were really well written the questions our fairer sex is asking would be spelled out fully and I say this because they are good questions.  Of course the discussion at WP:V should have been responsive.  If you are satisfied of the intent of WP:NOR, Plange, then Wikipedia would be best served by asking at WP:V, I believe.  The reason I say this is becuase WP:V does not yet spell out real clearly, what exactly consitutes "verifiable" for the situation that you are describing, and really should.  I hope you understand, a lot of the discussion has been on the most common difficulties (personal websites) (emotionally charged politics) (religious issues), rather than on historical documents.  So, if you are satisfied about the intent of WP:NOR, that might be the place to go. I think it would improve Wikipedia policy to be reworded a little so that future questions of the sort you bring up, are addressed within policy. Terryeo 14:41, 26 September 2006 (UTC)

Feedback Please
I am hoping some expert WP:OR people can give some feedback on an article I stumbled on. Its a heated arguement of which I just tossed my coin into and walked away. I think the participants can use some outside advice on WP:OR or I can learn something if I am indeed mistaken. The article in question is Talk:Allegations of state terrorism by United States of America and the issue seems to be if editors are allowed to cast events as terrorism or terrorist acts by comparing them to popular definitions of terrorism found in dictionaries. I concluded from the title that an allegation would have to be made directly, meaning a person has to actually say the word "terrorism" for their to be an allegation of terrorism. However its being argued that if what the person is alleging falls within a definition of terrorism then its the same thing. Example: John Forman (made up) states that "the United States took part in an intimidation campaign that was illegal and heinous." Would this be permissable? Any feedback given here or on the talk page is greatly appreciated. --NuclearUmpf 17:20, 26 September 2006 (UTC)


 * This is just my two cents and perhaps loads of people will disagree with me on this. But it seems to me that John Forman's quote should not be attributed to any popular definition of terrorism.  Afterall, other dictionaries can use terrorism to be whatever they want regardless of NPOV.  Just use the quote as is without interpretation.  The word "terrorism" is a politically charged, loaded word.  It should be used sparingly, IMHO.  I'm not even sure why we would be TRYING to use it.  If someone uses the words "terrorism", then that is fine.  But putting words into people's mouths (especially living persons) is dangerous and should be avoided. -- RM 17:47, 26 September 2006 (UTC)


 * I'm forced to agree with Ram-Man. Generally speaking, quotes such as the one your describe should only be interpreted to the extent that such interpretation is obvious and beyond dispute. The assignment of a word that, regardless of its denotation, carries a loaded negative meaning with it, may distort the speaker's intentions. Deco 22:03, 26 September 2006 (UTC)


 * I thank you all for your feedback, hopefully someone from that page will see this discussion. I left a link pointing them here. I will post a more obvious link and hopefully they can ask the experts regarding WP:OR questions. --NuclearUmpf 15:33, 27 September 2006 (UTC)


 * Hi Nuclear, it's difficult to comment meaningfully because this is a bad article that probably shouldn't exist. But given that it does, it should use only reliable published sources who have alleged the United States engages in acts of state terrorism, and who actually use that term. No sources should be interpreted as meaning "terrorism" unless they use that word. It's a loaded term and we shouldn't attribute it to sources who chose not to use it themselves. SlimVirgin (talk) 15:36, 27 September 2006 (UTC)


 * As I read the article I see it presents a lot of references. Yet there is very very little hard information in it, its more like a pile of rising bread dough than a baked piece of bread.  The very first reference seems to be to a commercial site which seems to have that information on its site just to generate web traffic.  The article is so far from encyclopedic as to be useless to a reader.  It might make fair, inflammatory reading in a newspaper or something like that. Overly referenced articles make me suspicious, anyway, heh. Terryeo 18:12, 27 September 2006 (UTC)
 * Just to clarify many of the references are not of people alleging terrorism, but just supporting that the events happened, leaving the allegations themselves widely unsourced. That is a whole new basket to tackle I believe though. --NuclearUmpf 18:50, 27 September 2006 (UTC)
 * The article was originally titled something along the lines of "State Sponsored Terrorism and the U.S.", or "American Terrorism" or something like that. By and large, the title "Allegations..." is widely acknowledged among the posters as a bad title, but because of edit-wars and other vandalizations of the page it has been hammered out as a sort of temporary compromise.  Moreover. given that  A) the U.S. State Department uses the term, and B) it's widely heard and read in the U.S. media, then there is no possible way to argue that it's a "bad" article simply because it uses the word "terrorism";  in fact, the idea that an encyclopedic article should *avoid* loaded terms is peculiar to the Wikipedia crowd alone.  The main use i put encyclopedias to -- and most other folks, as well -- is to try to get a starting handle on difficult terms, and by avoiding such issues one winds up avoiding a large part of what encyclopedias are supposed to do.  Finally, the situation referred to above occurred because one poster objected that definitions which were introduced did not "adhere to wikipedia standards" because they introduced "neologisms" and "presumes new definitiosn of pre-existing terms".  It was only within that context that the dictionary definition was introduced:  as a demonstration that the definition being proposed was a) conservative, and b) widely accepted.  218.160.178.169 21:53, 28 September 2006 (UTC)
 * Wikipedia should support such conservative usage of terms as far as possible, precisely because by not doing so one winds up with a mash of rising bread dough instead of hard information. The page in question has been frozen now for three weeks, without any return to normal editing status.  The only person who is allowed to edit it is an "approved" wikipedia "negotiator", who has done little besides excise entries to the article and accuse its editors of "not fulfilling Wikipedia standards".  When new information is introduced, it is immediately vandalized by contrarians who object that the word "terrorism" should not be associated with the United States -- even while the United States has itself created a defintion and defacto standard by which the suitability of the term can be measured.  The problem with the page is, in short, that the editors who created it are not being allowed to shape it into an objective and fair page, but are instead being repeatedly hampered by cynical and unrepentant "negotiators" and "administrators" who are a wee bit too politically delicate to deal responsibly with the issue.218.160.178.169 21:53, 28 September 2006 (UTC)
 * @SlimVirgin: your assertion -- that no sources be allowed unless they explicitly use the term "terrorism" -- winds up conveniently excising 300 years of American History from any possibility of inclusion in the article.  I really don't think anyone would object to someone including a passage that speaks of a "battle" or "raid" in an article about warfare, nor does anyone here object to the inclusion of a "carrack" or "caravel" in the Wikipedia article on battleships.  That's because carracks and caravels are relevant to the history of battleships, and battles and raids are readily identifiable as forms of warfare.  If there is clearly documented evidence that the United States was, let us say, supporting squads of heavily armed people who would purposefully and methodically murder large groups of unarmed people in an otherwise peaceful or undefended place, then i really don't see how it matters whether or not some commentator does or doesn't use the word "terrorism" -- the example clearly falls under the standard, most conservative definition of the word (i.e. -- the use of violence to spread terror amongst civilians, specifically as a military tactic).  Moreover, by adhering to such a restriction, the considerable commentaries of people like Mark Twain, Ralph Waldo Emerson, and Henry David Thoreau are all suddenly rendered mute on the subject, even though it is obvious from any cursory reading of their works that they were steadfast figures who vocally and brashly condemned the U.S. for its use of such methods. Of course, the word "terrorism" not having yet been invented, they didn't use it;  yet to say that they had nothing to say on the modern issue of terrorism, and that their works do not hold some important insight into how the topic relates to the U.S. in our current day, is by no means "scholarship" or "adherence to standards", but rather mere obstinate rejection of sources whose relevance is indisputable.218.160.178.169 22:13, 28 September 2006 (UTC)
 * I think its libel to say "Mark Twain alleged the US commited terrorism in relation to the Cuban missle crisis" if Mark Twain merely said "The US commited terrible acts during the Cuban missle crisis. The situation is made up but the point remains. To take his comments and label them I think is against Wikipedia rules and possibly the law. I didnt see the arguement over the definition. I just seen a user arguing heavily that they can add events that fall under one of many definitions he found and even one general one that was similar to "acts of violence" or something far to the extreme. But their extreme example is exactly the slope people go down when we are allowed to change peoples words to things we feel they said, and not what they actually said. --NuclearUmpf 10:27, 29 September 2006 (UTC)

No original reverts?
Er, why is there a revert war here? I don't even see that much of a difference between the two versions.  &gt; R a d i a n t &lt;  21:27, 27 September 2006 (UTC)

Wikiversity
Wikiversity does not exclude original researches. We should note this.--Jusjih 10:10, 28 September 2006 (UTC)
 * We should note that on the policy pages for wikiversity. There are plenty of sites that allow OR.  &gt; R a d i a n t &lt;  11:06, 28 September 2006 (UTC)
 * I think the point here is as follows: "We do not allow Original Research on Wikipedia, so if you are looking to post original research, consider this other wikimedia project." It would be helpful to state this on our page, although perhaps not until Wikiversity has a firm policy on OR, rather than the proposal it currently has. -- RM 12:55, 28 September 2006 (UTC)
 * Is there a peer-review process for OR posted in Wikiversity? I am concerned that adding the above language could allow someone to post their OR in Wikiversity and then immediately come into Wikipedia and reference their newly-posted OR, thus circumventing this policy.  --ElKevbo 14:25, 28 September 2006 (UTC)
 * First thing is that Wikiversity does not yet have a policy on OR, so at the moment there is no way of circumventing policy. Also, I'm sure someone else will provide the link, but I'm pretty sure that it is forbidden to reference Wikimedia publications as sources.  All sources must be external sources, so no, this can't be used to circumvent policy, at least AFAIK. -- RM 14:35, 28 September 2006 (UTC)
 * Excellent. I knew Wikipidia could not be used as a source but I have never looked into whether or not Wikimedia could be used as a source.  Makes sense.  Thanks!  --ElKevbo 14:43, 28 September 2006 (UTC)
 * Perhaps we should make sure someone else can verify this first though. -- RM 14:45, 28 September 2006 (UTC)
 * I can think of no problem to link from Wikipedia to Wikiversity (a new site) but I do not think using Wikiversity as a source a good idea.--Jusjih 15:00, 28 September 2006 (UTC)

As an example, wikinews explicitly drops the no original research requirement. They're a journalistic source, and they do original research. Even so, one of the objectives of wikinews is to serve as a source for wikipedia.

Anytime you do a page move (trivial example), or do a translation from another project (slightly less trivial), you are using wikipedia as your source. Generally people don't check all the way back to original sources, though perhaps at times it'd be handy if they did. Kim Bruning 08:10, 29 September 2006 (UTC)
 * Technically Wikinews is not the source for Wikipedia, but Wikinews gets their material from other sources and passes that along. If WN makes a conjecture on its own, we don't usually want to copy it unless it's corroborated by facts or external sources.  &gt; R a d i a n t &lt;  08:28, 29 September 2006 (UTC)
 * Wikinews is intended to be used as a source for wikipedia on current events, so that wikipedia articles don't have to be in constant flux at the time of the event. In fact, if you are writing about current events, you should write about them on wikinews, not on wikipedia (though this is not always what happens in practice). Kim Bruning 08:34, 29 September 2006 (UTC)

There's also. Kim Bruning 08:31, 29 September 2006 (UTC)


 * While if you use NOR policy on its own, you could perhaps make the argument that other Wikimedia sites could be used as a source, this falls apart on Verifiability grounds. The sources on Wikinews or Wikisource must be available from third-party sources in order to merit inclusion.  Original research on either of those sites would not satisfy the NOR and verifiability requirements (the two go together).  It also fails the reliable soure, since there is no way that original research on Wikinews or Wikisource could be considered verifiable unless it was backed by a reliable third party source or, in the case of Wikisource, a physical copy of a document. -- RM 15:15, 29 September 2006 (UTC)

Wikis aren't reliable sources by our standards, so we should use neither Wikiversity nor Wikinews as a source. I'm not saying I totally agree with this, but that's what the guideline says. --Conti|&#9993; 17:12, 29 September 2006 (UTC)


 * That's why it is a guideline. No one would/should think twice about citing the text of U.S. constitution or GFDL from WikiSource, since that information is obviously from a legitimate third-party source.  In fact, I'd rather link to another wikimedia project in that case because then the data is "closer to home" and easier to access and link to.  Don't have to worry about links being broken, etc.  But again here, the point is that the information is backed by another reliable source. -- RM 18:04, 29 September 2006 (UTC)

Carrying NPOV foreward into NOR
"WP:NPOV says that the article should fairly represent all significant viewpoints that have been published by a reliable source, and should do so in proportion to the prominence of each". Therefore, in regards to the idea of Original Research; unpublished information is always original research and defined to be original research by NPOV. This is why I propose we place "unpublished" in the first sentence. I would propose the first sentence of this policy change from:
 * Wikipedia is not the place for original research. To:
 * Wikipedia is not the place for unpublished original research.
 * Academia people will immediately recognize they can not publish their unpublished ideas at Wikipedia. Also, including "unpublished" in the first sentence carries the idea presented by NPOV into this policy in an obvious way.  I think it makes this policy easier to understand and harder to misunderstand. Terryeo 17:58, 29 September 2006 (UTC)

Math
Does math count as original research?

An example: A reputable source says that X = 10. A different equally reputable source says that Y = 5. Is it OK to say that "X is two times larger than Y"?

More sophisticated example:

Car A has 2.0 liter engine and produces 240 hp. Car B has 3.0 liter engine and produces 215 hp. Is it OK to say "Car A is far superior to car B in terms of horsepower per liter of displacement"? --Itinerant1 22:36, 29 September 2006 (UTC)
 * No, stating the obvious is not original research, and that includes basic math (after all, addition is hardly original, now is it). In the above case, saying it's "far" superior may be somewhat opinionated, but it's not OR to state that Mount Everest is higher than the Eiffel Tower.  &gt; R a d i a n t &lt;  23:19, 29 September 2006 (UTC)


 * I would avoid the "superior" as a value judgment. Numerous bio articles refer to "older" or "younger" siblings on the strength of verifiable birthdates -- this is source-based writing in action.  In the same vein, suppose that a source states, "The Prussians approached from an unexpected direction, almost invisible in the setting sun."  It is a valid use of the source to write, "The French did not expect the Prussian approach from the west, which took place near sunset."  Reducing an overwrought victorian writer to the facts is not OR either.  Is someone actually arguing that it is? Robert A.West (Talk) 16:23, 30 September 2006 (UTC)


 * No. If an editor places two different examples close to each other then the reader can apply his own good sense and this is how we serve.  We don't draw the conclusion for the reader, instead we place information (which the reader might not become aware of otherwise), place information in such a way that the reader becomes informed. Terryeo 17:04, 30 September 2006 (UTC)


 * Actually, we do (and should) present conclusions to the reader all the time, provided that those conclusions come from a reliable source. Placing two examples close to one another to encourage a particular (nontrivial) conclusion on the part of the reader is sailing very close to the wind when it comes to NOR and NPOV. Robert A.West (Talk) 23:48, 30 September 2006 (UTC)
 * Maybe he has an example in mind. I was speaking of the theory of it and haven't run into an example like that. Terryeo 14:51, 2 October 2006 (UTC)

OR in articles vs OR in talk pages
An issue that I've seen a couple of times lately (Talk:South_Tyrol & Talk:H._P._Lovecraft) is the question of whether it's appropriate for original research to influence how an article is written, without actually appearing in the article itself.

My view is that this is not actually prohibited by the wording of WP:OR, and that a certain amount of OR at a meta-level is not just acceptable but necessary in order to comply with other Wikipedia policies.

For instance, Naming conventions (places) says: "article naming should give priority to what the majority of English speakers would most easily recognize..." This is something that would usually impossible to determine except by methods (synthesis from a variety of sources, straw-polling editors, interpreting Googlefights etc.) that would clearly violate OR if used for actual article content. Similarly, whether an image can be counted as 'fair use' often comes down to an editor's judgement based on synthesis and interpretation from previous examples.

In these cases, while it would violate WP:OR for an article to say "the majority of English speakers prefer $NAME" or "the precedent of $CASE1 combined with criteria established in $CASE2 indicate that such-and-such would be fair use", it would be quite appropriate to name an article or include an image on the strength of that reasoning (provided other editors find it reasonable) and indeed this is often exactly how we handle it. But since there seems to be some confusion on this account, perhaps it would be useful for WP:OR to explicitly state that it only covers claims presented within articles, not meta-considerations? --Calair 07:19, 4 October 2006 (UTC)
 * Hm, tricky issue. With fair use, we tend to stay on the side of caution, so if a fair use claim is dubious, WP:CP or WP:IFD would be the places to ask. The place naming issue, I've seen debates of; usually such debates are not based on OR but on stating the obvious (Deutschland is called Germany in English, that's not OR). If they're not, they tend to head towards WP:LAME. That is rather unfortunate. We do, however, have some sources for common names, e.g. what the CIA factbook uses.  &gt; R a d i a n t &lt;  15:23, 4 October 2006 (UTC)


 * We do approach fair use cautiously, but even there it tends not to be as stringent as WP:OR is on article text. If I present my arguments as to why an image is fair use, and none of the other editors see a hole in those arguments, that's probably enough; OTOH, in article space, even a very reasonable argument that seems plausible to other editors is quite likely to be deleted on the grounds that it's an original argument.--Calair 01:23, 5 October 2006 (UTC)

Nutshell
I object to the current nutshell on this page because it is too long.  &gt; R a d i a n t &lt;  15:34, 5 October 2006 (UTC)


 * Too long for whom? ≈ jossi ≈ t &bull; @ 16:28, 5 October 2006 (UTC)


 * Too long for a nutshell. In my opinion, lengthy texts defy the point of nutshelling, and would be better explained in the first paragraph or two of the page. This nut simply feels like an attempt to put too much words in a single sentence.  &gt; R a d i a n t &lt;  16:46, 5 October 2006 (UTC)


 * I agree. Since it's a nutshell, it could dispense with the examples, I think (examples are going to crop up in the article text soon enough). The word "moreover" is deprecated in many style books; it's pompous and doesn't really mean much more than "and" or "also". "May not" is rather weasely here; I think it means "must not" or "should not", which would be stronger, in my opinion ("may not" can be misread to assume "might not" or "but may"). The top of this article looks about as elegant to me as a man wearing three hats.qp10qp 04:05, 6 October 2006 (UTC)
 * Ah, someone has improved the text of it now, thank goodness.qp10qp 09:55, 6 October 2006 (UTC)

I tried to improve it but Wjhonson reverted me saying my edit didn't represent "the accurate position."


 * It said: "Articles may not contain any previously unpublished arguments, concepts, data, ideas, statements, or theories. Moreover, articles may not contain any new analysis or synthesis of published material (such as arguments, concepts, data, ideas, or statements) that serves to advance a position."


 * I tidied the writing to say: "Articles may not contain any unpublished arguments, ideas, data, or theories; or any unpublished analysis or synthesis of published arguments, ideas, data, or theories that serves to advance a position."

Wjhonson, can you explain how the second does not represent the accurate position, and what the difference is (apart from tidier writing) between the first and second? SlimVirgin (talk) 22:02, 6 October 2006 (UTC)