Wikipedia talk:No original research/Archive 17

Archive 10

←

Archive 15

Archive 16

Archive 17

Archive 18

Archive 19

Archive 20

removing {{historical}} from talk page

It would appear that the policy page has been returned from Historical to active status or at least that there is a movement afoot to do so, so I have taken the liberty of commenting out the historical tag here. If this movement carries on it may be appropriate to unprotect the page as well in accordance with common practice. ++Lar: t/c 16:44, 20 March 2007 (UTC)

Protect

I have protected this page to maintain stability while the WP:ATT situation is being sorted out. Crum375 00:34, 22 March 2007 (UTC)

How on earth do you have power to do this?01001 05:07, 28 April 2007 (UTC)

WP:ATT: Join the discussion at

Wikipedia talk:Attribution/Community discussion

≈ jossi ≈ (talk) 23:12, 22 March 2007 (UTC)

An inquiry on the synthesis policy

The current policy currently states that if A and B are both reliable sources, they still cannot be used together to advance opinion C. Specifically, "In other words, that precise analysis must have been published by a reliable source in relation to the topic before it can be published in Wikipedia."

But I have question on basis of logical reasoning. In philosophy, we call the idea hypothetical syllogism. The argument goes as follows: If Q, then R. If P, then Q. Therefore, if P, then R. This is a deductively valid argument. If P and Q are both true, then R is logically true as well. For example, consider the following argument: All cats are mammals. All mammals are warmblooded. Therefore, all cats are warmblooded. If we provide a reference for the first premise (all cats are mammals) and different reference for the second premise (all mammals are warmblooded), would we be be allowed to use that as justification for writing within an article "all cats are warmblooded"? ~ UBeR 01:24, 23 March 2007 (UTC)

Answering on Wikipedia talk:Attribution. SlimVirgin ^(talk) 01:26, 23 March 2007 (UTC)

Whereabouts? ~ UBeR 01:29, 23 March 2007 (UTC)

here. :-) SlimVirgin ^(talk) 02:09, 23 March 2007 (UTC)

Thanks. ~ UBeR 02:21, 23 March 2007 (UTC)

That is really the exact thing that the synthesis part of NOR is prohibiting. We want to avoid making these new connections (or, to put it another way, if someone wants to publish such connections, this is the wrong place to do so). Jkelly 01:28, 23 March 2007 (UTC)

I quarrel with that idea, however, if this what the policy is actually stating. If we, Wikipedia, treat both the former premises as fact, quite logically the conclusion based on the previous premises is true. The mathematical equivalence could be as follows:

1 + 1 = 2

2 + 1 = 3

Therefore, (1 + 1) + 1 = 3

Now assume we have independent references that confirm 1 + 1 = 2 and 2 + 1 = 3, but we do not have a source saying 1 + 1 + 1 = 3. CLEARLY, based on the previous two equations Wikipedia treats as truths, the third equation can only be true. ~ UBeR 01:48, 23 March 2007 (UTC)

I suspect that one would encounter little resistance to that in an article on math or the sciences, although I would still hesitate to draw our own conclusions. If we've come up with a conclusion no one else has, it may not be as obvious as it appears. In an article on the humanities it works especially poorly. For instance, if we have a source saying "Jkelly says that all basket weaving is a deeply creative process" [ref] and "Jkelly says that creativity in basket weaving should be rewarded with cash prizes" [ref], we cannot then say "Jkelly thinks that all basket weaving should be rewarded with cash prizes" [ref][ref]. Jkelly 01:57, 23 March 2007 (UTC)

This is a lousy rule then, I must confess. It's the simple idea of substitution. "J says A is B. J says B warrants C. Thus, J says A warrants C." Saying otherwise seems to be self-contradictory. Admittedly, the current dispute I am is a little more complex. Arghhh. ~ UBeR 02:12, 23 March 2007 (UTC)

NOTE: There is a proposal to merge this page into WP:ATT.

Resolved

 – Merge tag installed.

There should be a prominent merge tag posted on this page, directing people to the discussion, which is supposed to be trying to attract as many people as possible. This is the very problem that led to the current dispute: that there had been no merge tag on this and similar pages while discussions about possibly merging were going on. Would someone please correct this and put up the merge tags? --Coppertwig 00:13, 24 March 2007 (UTC)

The project page needs {{Mergeto|Attribution}} (see Wikipedia:Attribution/Poll - merge has been proposed). — SMcCandlish [talk] [contrib] ツ 19:37, 25 March 2007 (UTC)

{{editprotected}}. There is an enormous amount of discussion going on about these pages. There are many users with sysop rights who are actively involved, and I trust them to add appropriate templates when/if they are appropriate. There is no need to request assistance from uninvolved administrators using the editprotected tag.CMummert · talk 20:17, 25 March 2007 (UTC)

I'm not sure if you read all of the WP:AN discussion you seem to be responding to. That users with sysop rights have been partisan, obstructionist and wheelwarring (i.e. not being trustable to add the appropriate templates) was the entire issue. I don't see anyone here suggesting that the text of this page should be modified, simply the proper merge tag installed. — SMcCandlish [talk] [contrib] ツ 22:01, 25 March 2007 (UTC)

I suggest the following template is added along with {{protected2}}:

A discussion regarding merging Wikipedia:Verifiability, Wikipedia:No original research, and Wikipedia:Reliable sources into a single page at Wikipedia:Attribution is in progress. Pending long-term resolution of this situation, the page has been protected from editing. This is not an endorsement of the current version.

You may have seen something similar on your watchlists. Would anyone object if I added this or something similar? -- zzuuzz^(talk) 18:52, 26 March 2007 (UTC)

New combined merge/prot/community discussion header tag

Resolved

 – Merge tag installed.

Wikipedia:Verifiability/Header. I was asked by an admin from WP:RFPP to come up with a combination of the merge header from WP:RS and the protection header from WP:ATT, to be used on both WP:V and WP:RS, and propose it here. My take at this is located at the link above (it has code in it so it can be used on both pages). By belief is that this version will satisfy everyone. It has the text (with twiddles that make it apply to these pages instead of ATT) from ATT's tag, including the protection discussion, with the merge tag formatting of the one at RS. So, it would replace the {{Protected}} on this page, and obviate the need to continue editprotecting about the need for a merge tag. I think it also absorbs all the ideas of the template in the topic above, too. Any objections? It looks like this:

— SMcCandlish [talk] [contrib] ツ 00:44, 27 March 2007 (UTC)

This is becoming insane.... ≈ jossi ≈ (talk) 00:48, 27 March 2007 (UTC)

Could you elaborate, substantively I mean? And did you have any sort of understandable reason for attacking someone who is trying to get consensus on a simple notification-header issue, at the request of others? — SMcCandlish [talk] [contrib] ツ 18:59, 27 March 2007 (UTC)

Page rename?

This should absolutely be renamed to Wikipedia:No original thought because of the problem in wikilawyering where ANY research, including organizing information logically or seeking out citations, is seen as "original research". Mermaid from the Baltic Sea 01:53, 27 March 2007 (UTC)

The title you suggest is no better at avoiding misuse. "You're not allowed to think about if that is a reliable source!" -Amarkov moo! 01:55, 27 March 2007 (UTC)

Jimbo Wales's requested poll nearly done - please see

Jimbo Wales requested a poll to gauge community thoughts on the Wikipedia:Attribution merger. A poll for this is being crafted, and is somewhat close to done. Concensus for the past 24 hours (with the occasional dissenting voice of course) that the thing is close to done. Only the main question is still heavily debated. A pre-poll straw poll is here:

Wikipedia_talk:Attribution/Poll#Q1_Straw_poll_duration

To sort that out. Accepted group concensus seems to be to pre-poll to 4/1/07 22:00 and then launch a site-wide poll (again, as implied/requested by Jimbo) at 4/2/07 00:00. Please help hash out the wording for that last quesion. - Denny 13:31, 29 March 2007 (UTC)

Slow down, slow down! I thought you'd give me time to talk with people? --Kim Bruning 13:44, 29 March 2007 (UTC)

Kim, I wouldn't mind, but I don't know if there is overall concensus. Jimbo was online when I posted to his talk page last night; he almost certainly saw my message. - Denny 13:49, 29 March 2007 (UTC)

Wikipedia:Miscellany for deletion/Wikipedia:Attribution/Poll

Resolved

 – 'Speedy keep' result.

Per comments on the Talk page here, and in other locales, it appears groups of editors are specifically against Jimbo's specifically requested public poll to gauge thoughts/support on the idea of the ATT merger. As it has been stated that the Poll is "dead" per users such as User:WAS 4.250, I am nominating this. If there is wide spread support to run this poll, this page should be kept. The MfD is here:

Wikipedia:Miscellany for deletion/Wikipedia:Attribution/Poll

Thank you. - Denny 16:10, 30 March 2007 (UTC)

The result of the MfD was a Speedy Keep - Let's get back to work on the Poll folks... New voices are needed to break the current deadlock over language. Please help out. Blueboar 17:42, 30 March 2007 (UTC)

Refrencing a T.V

Is referencing a T.V show original research?--Lucy-marie 10:08, 2 April 2007 (UTC)

Are you referring to fact or fiction?

With facts, you can reference anything that somebody said or claimed in a documentary or a news show just like you would reference them if they wrote what they said in a book.

With fiction, it's a little more complex. You can directly cites something as written in a show's script or shown on screen. For example "Doctor Drake showed a more sensitive side in episode X of Y when he.....". However, you can't take it any further than that by attributing something to a character if it isn't directly stated. For example you can't say "Doctor Drake's emotional breakdown in episode X of Y was possibly a sign that he was abused as a child" unless abuse during his childhood was specifically raised in the show.

perfectblue 09:54, 1 May 2007 (UTC)

Making your own sources/citations

What are the views of making a page/site or having someone else make a site just to use said site to have a source for wikipedia? I want to point out the fact that if this happened i am speaking as if the information posted in said sight really was true and legitimate. Would That be original research >:]? Just trying to point out the fact that you can't go and mark out someones information they post on here just because they don't have a source, yes sources are very nice and you should have one none the less. --NekoD 11:26, 2 April 2007 (UTC)

The questions are :

• If you or a person you know has a paper/book published at a JOurnal and you cite him..would it be 'Original research'

• If someone in 'Science' publish he/seh has discovered martians in Saturn..is he/she serious to be included the resarch into Wikipedia even though is supported (his/her job by a journal like Science or Physical Review )

• Arxiv.org or similar..count as a verifiable or reliable sources?

• If someone cites a source of a not-widely-known journal at the web or of a small country (let's say someone publishes a math result in the LItuanian journal of math) could it be considered as original research

• FOr me a 'peer-reviewed' just no means verifiable/exact since at last editor publishes whatever he wants, i could be the son of the editor of Physical review and have my papers published for example, of course it doesn't mean i'm a hoaxer of that they're wrong.

?

I get the idea of no original research because the information must be verifiable. However, how can any "new" research be done without it being (somewhat) original? Think about it.--Sportman2 15:08, 8 April 2007 (UTC)

Excellent question! When people on wikipedia say "No original research" they don't quite mean what it seems. The idea is that you (as a Wikipedian) shouldn't post your own conclusions, ideas or theories on Wikipedia. Here's how it does beyond verifiability: you shouldn't put sources together in a way that advances a position that is no longer attributable to the original sources. Some editors prefer the term "No personal research" and that's much better, because you can indeed rephrase the material and draw from a variety of sources to give a "new" presentation, as long as the sources are fairly represented, and are not used to advance a position that can't be explicitly and directly attributed to the given sources. I hope this makes more sense. --Merzul 01:27, 9 April 2007 (UTC)

importance-s template

I just recently ran across the {{importance-s}} warning, and am a bit confused by it given WP:N#Notability_guidelines_do_not_directly_limit_article-content. Isn't importance of sections more an issue of original research, sources, and undue weight than notability? --Ronz 00:01, 10 April 2007 (UTC)

Absolutely, isn't it interesting that WP:N starts off saying it is not synonymous with "fame" or "importance". --Merzul 11:24, 11 April 2007 (UTC)

Exactly. So why does the warning refer to WP:N? Shouldn't it refer to NOR, SOURCE, and WEIGHT? --Ronz 22:13, 11 April 2007 (UTC)

Because it used to point at Wikipedia:Importance, which was redirected to Wikipedia:Notability. —Centrx→talk • 22:57, 11 April 2007 (UTC)

Thanks. Makes much more sense why it's in such a state. Not so sure what to do about it though. --Ronz 23:15, 11 April 2007 (UTC)

I like the idea of WP:WEIGHT being the relevant policy for importance related issues, but again, I don't know what to do about it. If I was king of Wikipedia, I would deprecate these importance templates, and create undue weight template "This section or article lacks reliable sources justifying the weight given to it." or something like that. --Merzul 16:46, 12 April 2007 (UTC) --Merzul 16:46, 12 April 2007 (UTC)

Great idea! I've copied your comment to Template talk:Importance-s. Hope you don't mind. --Ronz 16:56, 12 April 2007 (UTC)

Novelty in Synthesis

Wales' quote specifically mentions novel deductions, and my recollection is that this policy used to say the same - but it doesn't seem to now. Anyone know what's going on? I ask because I'm currently involved in a dispute where I'm accused of original research by synthesizing a reliable source and a thesaurus, essentially, as I'm using a synonym of the topic in the source, rather than the topic itself. I can't imagine, for example, that sources that talk about HD 39060 but not Beta Pictoris can't be used at the later, but a strict reading of this policy could be seen to say that. WilyD 21:52, 11 April 2007 (UTC)

Well, the key idea is still that we don't want to publish original thought, so I'm often opposed to articles that string together an entirely new narrative without following at least a few sources that are explicitly about the subject at hand. When at least a few sources are available about the general idea expressed by the article, I see no problem in adding detail using sources that are more specific or don't even mention the exact same topic. However, synthesis is the most difficult policy, and replying like this without looking at the case in point is very difficult. I often accuse articles of synthesizing material, but whether something really is an improper synthesis depends on the consensus reached at the talk page. I don't think I know exactly where to draw the line between source-based and improperly synthesized research. (The "serves to advance a position" isn't very helpful, everything we do advances some position, the question is I suppose when it advances a position other Wikipedians object to.) --Merzul 17:01, 12 April 2007 (UTC)

We should defiantly keep the words "novel synthesis" and not "synthesis" on its own, because without the word "novel" many articles would have sections reduced to quotes, as editors who dislike a POV inevitably insist on no words but those from sources, as they will argue (correctly) any others are a synthesis of the source(s) and breach OR.--Philip Baird Shearer 18:35, 12 April 2007 (UTC)

Okay, I'll be specific. I used a dictionary to conclude that racism is Discrimination or unequal treatment based on ethnicity or race. I then concluded it was appropriate to use sources which talk about discrimination based on ethnicity without explicitly using the word racism were appropriate for Racism by country because discrimination based on ethnicity is a kind of racism, per my reliable dictionary. This is being accused of being OR because I'm synthesising a dictionary and a UN high commision on human rights report. WilyD 22:38, 12 April 2007 (UTC)

The "serves to advance a position" clause ought to be bolded if not printed in big block letters. The purpose of SYNT is to keep people from doing "term paper" style of writing in the articles, and that's it. Squidfryerchef 04:15, 13 April 2007 (UTC)

This remains a problem. I'm unwilling to change this because I'm involved in a relevent dispute, but right now the policy as read literally strictly forbids the use of multiple sources in an article, and implies it forbids anything but straight copying. Using a thesaurus to reword a statement is synthesis, it's just not original research, nor is it novel synthesis, which is what used to be prohibited. Is there a consensus to make this clear? Will someone fix it? WilyD 20:11, 14 April 2007 (UTC)

The key surely is "acceptable only if a reliable source has published this argument in relation to the topic of the article". If the UN is clear that they mean discrimination based on ethnicity, then using another form of words to say the same thing is surely acceptable. If the UN's not clear, but another source has provided that clarification in relation to the topic, then that other source can be cited in addition to the UN. If the UN is unclear and the only other source is the dictionary which of course isn't related to the topic, then you shouldn't try to clarify it for them, but should represent their statement as accurately as possible. Multiple sources are fine as long as they're attributed, putting things into other words is fine as long as it represents the source accurately. Does that help, or do you think my interpretation is wrong? ... dave souza, talk 21:28, 14 April 2007 (UTC)

I think in a very strictly literal sense, if I use my desktop dictionary to conclude racism is discrimination based on race or ethnicity and the UN report to conclude ethnic minorities may be discriminated against in Iran, I am engaging in synthesis to conclude ethnic minorities may face racism in Iran. This policy is clearly not designed to prevent that kind of synthesis. For instance Every featured article in Wikipedia engages in this kind of synthesis. Furthermore, this article used to include the idea that any synthesis had to result in a novel idea or conclusion, but somewhere that's been lost. The issue is that what I'm doing here clearly isn't original research, it's just speaking english. In a very literal way, everything that isn't a straight copy of something has some synthesis in it. If we look at this version of the page (which is older), we see 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". Here's the word novel.

For instance, suppose that I have a reliable source that says the Germans lost 300 000 men at the Battle of Verdun, and a second reliable source that the French lost 150 000 men. It would be synthesis for me to write The Germans lost more troops than the French - but there's nothing novel about concluding that, and it would be downright silly to forbid writing that - but as written, this policy does forbid that. I'm fairly sure it's not supposed to. WilyD 21:58, 14 April 2007 (UTC)

My comment is that we have to be sure that the source is using the same dictionary definition, but regarding the second example it's clear that 300 is more than 150, and the source can be attributed with the detailed numbers in a footnote. Whether rephrasing the guidance would help is another question which I've not fully considered. .. dave souza, talk 22:18, 14 April 2007 (UTC)

I agree with Dave. I think some people are using "synthesis" too broadly. Synthesis does not just mean using two sources; it means using two sources in a very particular way. Rewording something without changing its meaning is not a synthesis - it is good editing. According to Kant, synthetic judgments are those which are not analytic, meaning that the terms alone are insufficient to determine the truth of the proposition, and this gets to the heart of what we are talking about. If a source says "x is afraid" and (thanks to Roget) we write "x is fearful" we are not synthesizing. A synthetic statement is more like this: The ICC defines genocide as doing x; according to Joe Smith the United States did x, therefore the US is guilty of genocide. To comply with this policy, we need a different source that says the US is guilty of genocide. Slrubenstein | Talk 13:02, 15 April 2007 (UTC)

Slrubenstein, that's a very good example of the kind of inference SYNT is meant for. I like it a lot better than the controversial "plagiarism" example that's in the policy now. People keep trying to remove the plagiarism example because they either find it confusing, or, IMHO, that it's not a cut and dried example of OR and banning everything like it promotes a deletionist viewpoint. Perhaps yours should be the new WP:SYNT example. Squidfryerchef 15:39, 15 April 2007 (UTC)

It isn't "synthesis" to say the Germans lost more troops than the French. The new "attribution" policy specifically allows simple deductions. But we do need to do something about the overuse of SYNT in wikilawyering. I've heard all kinds of bizarre interpretations of SYNT, that instead of saying they don't agree with something in an article, they will mention SYNT and think they have the last word. I'd suggest bringing up these situations on the Attribution talk page as an "Is this OR?" question. I don't know if the main editors of these policies are aware of how much they are being misused, but if we bring it up they may add some counterexamples of what is not SYNT to the policy. Squidfryerchef 15:31, 15 April 2007 (UTC)

Yeah, this is my point. That SYNT (and I think this is the generic problem with ATT, though I haven't been watching the whole mess) is that it's so vague it can appear to mean whatever you like, and requires you to be intellectually honest in applying it. The problem is that it's very straightforward for anyone to use the policies to dispute anything at all because they're so unclear. WilyD 19:20, 15 April 2007 (UTC)

In a line of thought that I've been working on recently, I tend to see the prohibition on synthesis as a corollary of neutral point of view. Synthesis is about producing a new point of view, which is contrary to NPOV because our objective is to present existing points of view. Naturally there will be many Wikipedia articles which present information in a form that it hasn't been presented in before, but as long as it's merely putting things alongside each other that's ok. So the question to be asked is, am I using these sources to produce a new point of view? --bainer (talk) 16:27, 15 April 2007 (UTC)

I do not think that a neutral point of view covers this. There is a short example on a NPOV in Allied war crimes:

• Atomic bombings of Hiroshima and Nagasaki: In 1963 these were the subject of a judicial review in Ryuichi Shimoda et al. v. The State. (source given 1) The District Court of Tokyo declined to rule on the legality of nuclear weapons in general, but found that "the attacks upon Hiroshima and Nagasaki caused such severe and indiscriminate suffering that they did violate the most basic legal principles governing the conduct of war."(source given 2) Nevertheless, the prevalent international legal opinion is that these bombings were not a war crime. (Source given 3)

If the last source (3) had not been given, then the last sentence would have been original research, but it would still have given a balanced POV. I agree with others here that "novel" in front of synthesis. But in the case given on racism I am not at all sure that it is correct and that it is not OR. I think before it is used, something like the judgements on the Race Relations Act 1976 need to be consulted, because it is an old parlour trick to chase a word through a dictionary, (one which most good English teachers warn their pupils about). --Philip Baird Shearer 19:17, 15 April 2007 (UTC)

You really think using dictionary definitions of words is original research? The implication is that we can never paraphrase any source, or adjust its wording to fit the article flow. Under this kind of interpretation, every article can be nothing but quotes - that certainly isn't the standard, and I think it'd be even more contraversial a proposal than WP:ATT WilyD 19:23, 15 April 2007 (UTC)

If (3) wasn't sourced, it might be OR, but not by SYNT. PS. what is "chasing" a word through a dictionary? 141.154.24.225 20:10, 15 April 2007 (UTC)

SYNT is not what I was talking about I was making an example of NPOV for Thebainer's "Synthesis is about producing a new point of view".

Camouflet->cavern-> lage cave-> "A storage cellar, especially for wine". So chasing wordsmeanings thorough dictionaries can lead to this: "The Tallboy bomb was designed to create a storage cellar, especially for wine.", instead of "The Tallboy bomb was designed to create a camouflet." --Philip Baird Shearer 21:22, 15 April 2007 (UTC)

That's not synthesis, that's stupidity. --bainer (talk) 03:23, 16 April 2007 (UTC)

• For instance, would it be appropriate to include some information on Apples that come from a source that doesn't mention that apples are Fruit in the Fruit article, by synthesising that first source with a dictionary that defines apples as a kind of fruit (or for that matter, any reliable source that defines apples as fruit). It's clearly synthesis, but this sort of thing does go on all the time. WilyD 17:19, 16 April 2007 (UTC)

To my mind, the trivial replacement of definitions, if that action does not violate logic nor advance a ideological cause is certainly acceptable. "Apples are good" cannot be replaced by "Fruits are good", but *may* be replaced by "Some fruits are good". To my mind that is not synthesis, it's a trivial action of "like-definition replacement" or "category replacement". An apple is a fruit, a fruit is a type of food, food is organic matter... These sorts of category replacement do not constitute synthesis to my mind. Similarly replacing "Marilyn Monroe had platinum blonde hair" with "Marilyn Monroe had blonde hair" is not synthesis even though it's not a direct quote. Wjhonson 20:26, 16 April 2007 (UTC)

I agree with you, but the policy page doesn't. Apples are red + apples are fruit => some fruit are red is a textbook example of synthesis. It just isn't novel, because no real thinking is involved. The point isn't whether we think it's appropriate or not, but what the policies say - I've already entered an agreement to go into mediation over two issues, one of which is whether combining a dictionary definition with a source is original research. I'd like it if the policy was clear one way or the other before I got into it. WilyD 20:43, 16 April 2007 (UTC)

I disagree that your example is "synthesis", rather it is deduction. Synthesis would take two disparent ideas and merge them into one structure. Such as "Liza Minelli was born in New York" and "Liza Minella was born on Tuesday" to form the sentence "Liza Minella was born in New York on Tuesday". It's trivial, but it's synthesis. Wjhonson 06:01, 17 April 2007 (UTC)

To your specific point of whether we can use a dictionary definition, I would say "Yes but". There are quite a lot of examples of dictionary abuse. For example "George Bush has directly caused the deaths of over ten thousand Iraqis" and "Genocide is the act of causing the deaths of a large number of people of one race or peoples" to match the sentence "George Bush has committed genocide." I'm sure you can follow why such a construction is problematic from our wikiview. Now if someone is debating whether "Apples are red" and "fire engines are red" can be used to make the sentences "fire engines are the same color as apples" then they are spending too much time on silliness. I'm sure you can see why these examples are disparent. One quite clearly "advances a cause" the other does not, at least not clearly. Wjhonson 06:06, 17 April 2007 (UTC)

Genocide is more narrowly focused than that, and I'm not even sure anyone who'd say point blank George Bush has directly caused the deaths of over ten thousand Iraqis would not end up say George Bush is guilty of genocide. For what it's worth, Genocide gives the definition: "any of the following acts committed with intent to destroy, in whole or in part, a national, ethnic, racial or religious group, as such: killing members of the group; causing serious bodily or mental harm to members of the group; deliberately inflicting on the group conditions of life, calculated to bring about its physical destruction in whole or in part; imposing measures intended to prevent births within the group; and forcibly transferring children of the group to another group." I'm fairly sure George Bush hasn't done that, which is why we're uncomfortable calling him a Genocide. I can try to devise a more "politically charged" example. WilyD 16:24, 17 April 2007 (UTC)

WilyD is providing a virtually textbook example of a synthetic statement, it is virtually right out of Kant. Slrubenstein | Talk 10:46, 17 April 2007 (UTC)

• So I then ask: Is that kind of synthesis allowable? If it is, shouldn't this page reflect that? WilyD 16:26, 17 April 2007 (UTC)

I think it is forbidden. But see also my comment about rapid transit, below. Slrubenstein | Talk 16:34, 17 April 2007 (UTC)

• The issue, then, is that essentially every source article on Wikipedia does this (excluding, obviously, copyvios & public domain texts). As a test, I examined the first reference of today's featured article and found it used dictionary/thesaurus type interpretation of the source. I'm fairly sure we could find it in every featured article if we had the free time to search. WilyD 16:52, 17 April 2007 (UTC)

Maybe I am confused. I am talking about this "Apples are red + apples are fruit => some fruit are red" - I do not see this as an example of the dictionary/thesaurus type interpretation, which I believe the policy fully allows. Slrubenstein | Talk 17:28, 17 April 2007 (UTC)

• Hmm, okay maybe I'm confused. What appears to be the case to me is that historical versions of this policy allowed that kind of synthesis, but that the policy doesn't anymore. Is this a deliberate change? I haven't seen yet, but I don't want to do a thousand diffs to find who changed it. WilyD 17:39, 17 April 2007 (UTC)

Feel free to help us word the policy more carefully. What is essential is that editors cannot forward their own argument that involves drawing on two different sources, neither of which make the actual argument at issue. If source A says "X was a terrible person" and source B says "X was a horrible person" and a thesaurus says "horrible" is a synonym for "terrible" and an editor makes an edit "Several people have claimed that X is a horrible person, including A,B, C, and D" the edit does NOT violate our policy. If source A says "men are horrible" and source B says "X is a man" and an editor writes "According to A, X is horrible" the edit DOES violate the policy. Is this clear? Does it make sense? If the answer to both questions is yes, and you feel the policy can explain this more clearly, go ahead (I am about to go off-line) Slrubenstein | Talk 17:55, 17 April 2007 (UTC)

• Thanks for the invitation Slrubenstein, but I do not feel it would be appropriate for me to make any edits of this sort while I'm actively involved in a dispute on this issue. Specifically, I used this definition of racial discrimination: 1. In this Convention, the term "racial discrimination" shall mean any distinction, exclusion, restriction or preference based on race, colour, descent, or national or ethnic origin which has the purpose or effect of nullifying or impairing the recognition, enjoyment or exercise, on an equal footing, of human rights and fundamental freedoms in the political, economic, social, cultural or any other field of public life. from the International Convention on the Elimination of All Forms of Racial Discrimination to conclude that material which talks about discrimination based on ethnicity would be appropriate as references for an article about racism, and this is actively being disputed by a couple of editors as OR/Synthesis. It certainly wasn't when I was getting a grip on policies, but it may be now, which is why I'm looking for clarification - and a clearer policy to help avoid this kind of thing in the future. Before this ends up in mediation (which may be soon - the article's been locked for over a month), it'd be nice if policy was clear one way or the other. WilyD 18:05, 17 April 2007 (UTC)

I appreciate your honesty and sense of propriety. As to your case, I have a comment: the question is, who is authorized to apply/execute/enforce the convention? My understanding is, not individuals. Perhaps states or international bodies like the UN or the ICC. In this case, it is a violation of our policy for you to use this Convention to decide who has committed racial discrimination. Someone else has to do it. Let me give an analogy: rape is a crime in most states (all?). If X believes s/he was raped by Y, according to the statute, that is just not enough for someone to add to the article on X or Y (if either of them are famous enough to have an article here). Only if X actually accuses Y of rape, and this accusation is recorded in some public document - for example, if the police arrest Y or the DA indicts - can the article report that "X accused Y of rape" - even now, we still wouldn't write, Y raped X. If there is a trial and Y is found guilty, we can write "Y was found guilty of rape." Do you see my point? We are reporting on what an appropriate authority does, we are not actually making the decision as to whether rape occured. Same goes for genocide. No matter how we read the conventions on genocide, wikipedia editors cannot say "Genocide = X" and "X happened to Armenians" therefore "The Armenians were victims of genocie." All we can do is report "The UN (or the ICC or whomever) concluded that the Armenians were victims of genocide." Lacking that, we can also say "historians x, y, and z have concluded that the Armenians were victims of genocide." But we cannot make this conclusion. Does this help? Slrubenstein | Talk 18:33, 17 April 2007 (UTC)

• In the specific case I'm talking about Racism in Iran (see Racism by country) and Iran is a signatory to the convention. In terms of the section I proposed, I follow the word usage pretty closely, and just say basically "Iran says there's no racism in Iran, but there hasn't been a good study. Other organisations alledge that ethnic minorities face persection (such as AI, US Department of State, UNPO), and reported events include some stuff. All of it's very well sourced, although there's also a bit of a dispute about whether the US Department of State and Amnesty International are reliable sources and so on. WilyD 18:38, 17 April 2007 (UTC)

(edit conflict) The question remains, who, according to the convention, has the authority to enforce or apply the convention? Whatever the convention says, our policy is this: it is not for a Wikipedia editor to make these judgements. This is also an NPOV issue, by the way. If any Iranian institution, state, international organization, real (meaning, not consisting of one or two people who just call themselves this) NGO - or if a journalist, historian, or social or political organization, accuses Iran of having violated the convention, we can report this accusation. And if the international body empowered by the convention to enforce the convention decides that Iran has indeed violated the convention, then we can report that too. But an editor cannot add to an article his/her own views on this matter. Slrubenstein | Talk 18:48, 17 April 2007 (UTC)

WilyD, given what you just added, you may be doing yourself a disservice if you are quoting the convention - it means you are adding your own argument, and that violates our policy. What is important is that the US DOS and AI and UNPO have alleged that ethnic minorities face persecution. It doesn't matter whether they are biased, you are talking about a major NGO, a superpower, and an international organization - NPOV does not prevent us from adding sources that are biased, indeed, we assume ALL sources are biased; NPOV insists we correctly identify the POV and provide multiple and if they exist opposing views. So the thing to do is to make clear that these entities have accused Iran of persecuting ethnic minorities - this is a verifiable fact and must be included in the article, no one can remove it. Have any of these entities explicitly cited the convention? If so, you can say so and provide the source - what is important is that a verifiable source is quoting the convention, not you. If none of these entities quote the convention, then you cannot introduce it, that violates policy. Slrubenstein | Talk 18:48, 17 April 2007 (UTC)

• Okay, this is kind of unclear to me. In any event, my only agenda is finding a paragraph about Racism in Iran that all the editors of Racism by country will find acceptable so we can get the article unlocked and begin working on it again.

That said, this is not a particularly peculier definition of racism, it's fairly generic. The sources below talk about persecution of ethnic minorities, but they never explicitly label it racism. The section also never explicitly labels it racism, but it's in an article titled racism by country, so the claim of the opposed editors is that any source needs to explicitly use the word racism before it can be used. The use of the convention is a bit of a seperate thought. I only mean that I use a definition of racism in order to judge what is or isn't a relevent fact for the article. I can cite that Babe Ruth hit 60 homeruns in 1927, but as far as I can see, that doesn't belong in Racism by country. So to write the article, I need to know what the word racism means.... WilyD 19:00, 17 April 2007 (UTC)

For instance, I use this sentence The Iranian regime continues to repress its minority ethnic and religious groups, including Azeris, Kurds, Bahai, ethnic Arabs and others. from a US State Department Report to talk about persecution of ethnic minorities - nowhere does the word racism appear. WilyD 19:02, 17 April 2007 (UTC)

Oh here, since I don't seem to be making myself clear, here's the section I proposed. The objection is basically talking about ethnic discrimination in an article about racism at all.

The following section appears to me to be compliant with all of our policies. I do not think that you need to find a source that uses racism since you are not claiming racism - This sentence - "The 1993 review of Iran's compliance with the treaty by the United Nations Committee on the Elimination of Racial Discrimination found that although the government of Iran condemned racial discrimination in public statements, insufficient information was provided by Iran to properly assess how the convention was being implimented in Iran, and whether Iran was fufilling its obligations under the treaty" - justifies inclusion in the "racism by country" article, and the sentences on persecution of minorities are appropriate. I would not change the wording to add the word "racism" I would just say that the issue of persecution of minorities is relevant enough to merit inclusion as long as you do not change the wording. That is my view. What you need to do is take a break, use our discussion to gather and clarify your thoughts, and file a "request for comment" in which you provide the most concise summary of what is going on and what the key points of contention are. Slrubenstein | Talk 19:09, 17 April 2007 (UTC)

Well, okay thanks - I think I basically agree with you here - maybe more what I was looking for is a way to justify by policy that including the stuff about persecution of minorities was not original research, or a sanity check telling me I was crazy and it was. I already filed a request for comment, with - ambigious - results, and have an agreement to enter into formal mediation with the other editor who was involved the in "edit war" - although we haven't filed a request. Anyways, thanks for the help. WilyD 19:23, 17 April 2007 (UTC)

Sample Potentially Problematic Section

According to article 19 of the Iranian constitution:

the people of Iran belonging to whatever ethnic or tribal group shall enjoy equal rights and colour of skin, race, language and the like shall not be considered as a privilege

Iran is a signatory to the International Convention on the Elimination of All Forms of Racial Discrimination. The 1993 review of Iran's compliance with the treaty by the United Nations Committee on the Elimination of Racial Discrimination found that although the government of Iran condemned racial discrimination in public statements, insufficient information was provided by Iran to properly assess how the convention was being implimented in Iran, and whether Iran was fufilling its obligations under the treaty. The Iranian representative respond to the committee saying that there has not been a census of racial demographics in Iran, that the government of Iran does not collect or use racial information in hiring government employees or university admissions and that Iran is not a multiracial society.

Other agencies have alledged that ethnic minorites such as Azeris, Kurds and ethnic Arabs face persecution. Reported issues include Arab land being purchased at low prices or confiscated and the violent repression of Kurds. Iranian President Mahmud Ahmadinejad blamed unspecified "enemies" for increasing ethnic tensions in Iran, while interior minister Mostafa Pour-Mohammadi claimed the United States plans to increase ethnic violence in Iran.

Synthesis example

I've proposed (at WP:ATT) that the plagiarism example about synthesis is removed, and propose that it is removed from this page also. I've also proposed a replacement example at Wikipedia talk:Attribution/FAQ#Example(s) of unpublished synthesis and at Wikipedia talk:Attribution#Concerning "Unpublished synthesis of published material", and have had no objections since. There is also some earlier discussion at Wikipedia talk:Attribution/Archive 13#Synthesis section. The change was put on hold a bit because of the merge debate on the attribution policy, I'm raising the issue here again.

To summarise a few of the reasons why I think the current example should be removed:

• The whole background to the case is complicated and not explained. What does "copying references" mean? The reader is left guessing as to what the case is about.
• The claim that "The whole point of this paragraph is the conclusion that, given the Chicago Manual of Style's definition of plagiarism, Jones did not commit it." does not follow from the text quoted in the example. The paragraph cites a definition of plagiarism, and makes no comment on whether or not Jones committed it or not.
• The example is supposed to be illustrating synthesis, i.e. putting together two sourced ideas A and B to come up with an unsourced conclusion, C. None of the conclusions in this passage appear to rely on more than one source.
• As well as being misleading to the casual reader, it doesn't stand up to close scrutiny either. Where the source is used, it is totally misquoted. The example claims that the CMS defines plagiarism as "using a source's information, ideas, words, or structure without citing them", and implies that this is a properly sourced statement used to build the unsourced conclusion. However, if you check the CMS, there is no such definition there at all, or any that resembles it.

For our policy pages to do their job, they need to be clearly understood by the reader, and clear, unambigous examples help with this. But this example is unclear and ambiguous, and has fundamental problems that mean that it can't be fixed just by changing a few words. I propose that it is removed from here and WP:ATT (and perhaps replaced by an alternative example at WP:ATT/FAQ).

Does anyone have any comments, objections or alternative suggestions before making the change? Enchanter 00:50, 16 April 2007 (UTC)

I think it should be repaced with a clear example. --Philip Baird Shearer 10:53, 17 April 2007 (UTC)

I liked Slrubenstein's "genocide" example in the "Novelty in Synthesis" discussion section ( it's about a page down from the top of section; I bolded the word "very" in my rejoinder ). I also liked the reference to Kant. Now I don't go particularly deep in philosophy, but if "synthesis" in policy is based on Kant, it would be awesome if policy could link to the article on Kant which might give a good explanation. Squidfryerchef 00:52, 18 April 2007 (UTC)

I like the genocide example too. As it's an emotive subject, I suggest it refers to a fictional country rather than the US. Here's some suggested fuller wording:

"The ICC defines genocide as doing x_{Reliable source 1}. Northland did x_{Reliable source 2}, therefore the Northland is guilty of genocide_{No source}.

This unsourced conclusion that Northland was guilty of genocide is not appropriate for Wikipedia, because it is not the role of an encyclopedia to judge controversial issues such as whether or not a country committed genocide - even if the two sources are good and the logic is sound. Instead, Wikipedia describes what other reliable sources say in a neutral manner. Therefore, to comply with this policy, we need a different source that says the Northland is guilty of genocide, and attribute the conclusion to that source."

I had proposed another example at Wikipedia talk:Attribution/FAQ#Example(s) of unpublished synthesis. That was addressing a different sort of example - interpreting scientific data. I think different types of synthesis means that more than one example might be useful, and I suggest that we put all the examples in WP:ATT/FAQ, with a note on this page saying something like "For examples of this policy, see the attribution FAQ".

Any comments? Enchanter 21:01, 23 April 2007 (UTC)

Suggested clarification: No original research rule applies to the article

As far as I have understood, it is perfectly all right to use a bit of original research in the discussion pages as an argument for a point (unless the point is that original research should be included in the article itself). For instance, when I wrote this, I looked at a few discussion pages and found that most of the discussion was actually more or less original research. Let me give you a fictional example:

On a discussion page reagarding a show that takes place in medieval times A says: "Do you really think we need a 'historical inaccuracies section'?" and B replies: "Well I showed it to three people who watched the show and they thought it was interesting."

B's reply is of course original research. It is perfectly all right for B to write this on the discussion page. It is only when he tries to include his little (three-person) inquiry to the article itself that he commits an error.

This may seem obvious but I have had people complain that an argument on the discussion page was original research and I think this clarification is needed.

Sensemaker

I agree that NOR applies strictly to the article. Sometimes it can be very useful for people to bat around their own ideas on the talk-page, and we shouldn't ban people from doing this. But there is a serious risk of people using talk pages as soap-boxes, which is wrong. Talk pages are for improving articles and since OR cannot go into articles if people spend too much time batting about their own explanations and sytnthetic judgements, it can and often does reach a counter-productive point as talk is being generated that cannot go into the article, and thus is not helping improve the article. So your distinction is correct but people need to use good judgement. Slrubenstein | Talk 11:39, 16 April 2007 (UTC)

I've never seen someone using talk pages as a soap-box for some personal issue that has nothing to do with an article. However, judging from your self description you have more wiki-experience than I have so I presume you know what you are talking about. I shall take your word for it. Since you say this is a problem, I shall gladly make sure to point out that original research is OK on a talk page as long as (1) you are not arguing for including the original research in the article itself and (2) follows other wikipedia guidelines for talk pages, particularly the guideline of sticking to the subject (how to improve the article) -wikipedia is not your personal soap box. Would this be OK with you? -Sensemaker

WP:OR and images of non-existant things

I've got a small problem with a user who keeps insisting on deleting 2 user generated images because they think that they violate WP:OR. Their argument is that because the pictures are "artists renderings" based on witness statements they violate WP:OR, my standpoint is that because the creature that the pictures are depicting doesn't actually (it's a paranormal creature) there is no way that you can possibly have a real photograph.

An outside opinion on this would be useful.

The page in question is Shadow people.

perfectblue 14:48, 16 April 2007 (UTC)

In those cases, aren't the "artists renderings" based on things the artists claim to have seen themselves, and not "witness statements" taken from a reliable source? And one of the images has text, which is subject to OR and a bad thing in an image since the text can't be edited by other wikipedia editors. It also has some OR in the text of the image page. --Minderbinder 14:55, 16 April 2007 (UTC)

I don't really care about the second picture, but the first picture is of good quality, is an represent of shadow people as commonly described, and we have permission to publish it.

What would you consider to be a non WP:OR picture? Honestly? If you're holding out for a real photograph, you're simply not going to get one. Shadow people are basically fictional so an artists impression is permissible. They are the result of an overactive imagination and people's eyes playing tricks on them, and you can't photograph something that doesn't exist.

I can fake a picture of my own based on the description on the page, would you RV that too?

perfectblue 15:09, 16 April 2007 (UTC)

See the Original Images section on [[WP:OR} Pictures have enjoyed a broad exception from this policy, WilyD 17:11, 16 April 2007 (UTC)

It also says "Images that constitute original research in any other way are not allowed" which is what applies in this case. The images in question are based on the personal experiences of those who created them, as said on the image pages. This page certainly doesn't declare all original images to not be OR, it just says they are allowed as long as they don't violate OR otherwise - these pictures do violate it. The intention of NOR isn't to allow using images as a loophole, is it? --Minderbinder 15:58, 19 April 2007 (UTC)

The question specifically says witness statements - if the statements are verifiable, then I believe the art should be fine. Take a look at all the art in Category:Sexual acts, for instance. From a practical perspective, there's a recognised need to allow some latitude in the production of public domain or GDFL images for articles - every photo I take, for instance, isn't really verifiable. I just say it's an image of X, and then in it goes. For instance, I took this photograph of Highway IIa in Toronto - there's no way any user can verify that without going to the highway themselves, but we allow it anyways, otherwise we'd be hard pressed to add most images. WilyD 16:05, 19 April 2007 (UTC)

In this case the "witness statements" are from the people who did the art. Just as text of an anecdotal story of the experience of a wikipedia editor isn't verifiable, a drawn picture of an anecdotal of the experience of a wikipedia editor isn't verifiable either. I don't take issue with pictures of overpasses, since they can be verified, the problem is drawings of "this is what I saw" when the subject is something generally considered not to exist. --Minderbinder 16:33, 19 April 2007 (UTC)

Please cease twisting things. I'm referring to the first picture, not the second one. The first one is based on outside descriptions. perfectblue 12:24, 27 April 2007 (UTC)

Please WP:AGF. I was talking about the second picture. If you only think the first picture meets OR, why did you keep trying to add both back to the article? --Minderbinder 13:43, 27 April 2007 (UTC)

I was reverting to the previous version, the same as you. Did you think that I typed all that code in by hand? The second image is completely superfluous as far as I'm concerned. Especially since the page now has an infobox which leaves room (sensibly speaking) for only one image.

Seriously, though "generally considered not to exist", what am I supposed to make of this? We're talking about "shadow people" (scare quotes intentional). There's no general about it. There is no room for shadow people in parapsychology, ghost hunting or even pseudoscience. They are an extension of the bogeyman left over from when people feared the dark.

perfectblue 20:12, 27 April 2007 (UTC)

Which is why you should only revert the content you're concerned about and not do blanket reverts. If there's no room for it, then why are you putting in personal accounts of people describing "I saw this!"? --Minderbinder 20:15, 27 April 2007 (UTC)

Is using a definition original research?

According to many reliable sources ([1][2][3] and others), rapid transit is "rail or motorbus transit service operating completely separate from all modes of transportation on an exclusive right-of-way". Is it original research to say that since AirTrain JFK meets this definition, it is rapid transit? Does [4], which says that automated light rail (which AirTrain JFK is [5]) is a type of rapid transit, help? --NE2 07:19, 17 April 2007 (UTC)

I think this is a bad example because you do have the press release so your bases are covered. But let's say you didn't have the press release. I think here is a case where we need to use common sense. The more controversial the example, the more strict we need to be about compying with the policy. The less controversial, the less strict. What happens if two people disagree over the level of controversy? well, if two people disagree, I would say that is proof enough that it is controvesial. So what would I do? I would call the Airtrain rapid transit even though this is a synthetic statement. I bet no one will object. If someone objects and they refuse to consider this so obvious and uncontroversial a case that we can let it slide, then I would try to find a compromise. But my guess is no one will object. Now, look at it the other way - I am sure you can imagine all sorts of synthetic claims that would be very controversial (and that will be challenged by many other editors). That's when we need to follow the policy strictly. Slrubenstein | Talk 10:54, 17 April 2007 (UTC)

Actually the only press release is a Canadian one that says that a new Vancouver ALR line is rapid transit; it doesn't say that AirTrain JFK, a line in New York City, is. There have been related issues with light rail, essentially a marketing term for some tram systems. --NE2 11:11, 17 April 2007 (UTC)

Sorry I misread it. Well, my appeal to common sense stands. Slrubenstein | Talk 11:16, 17 April 2007 (UTC)

is this original research?

On Talk:Ancient_Egypt_and_race#remove_OR disagreement over whether a certain paragraph is Original Research. Urthogie 19:35, 18 April 2007 (UTC)

Most scientists

Regarding this addition of Pmanderson's, [6] is this something we want to include? Sometimes it really is obvious to specialists that "most" of them believe X, even if there's no reliable source that says this explictly. For example, it seems obvious to me that most scientists agree with the man-made global warming hypothesis, although I suspect the ones who don't would say it's not true that most scientists agree with it. I feel we should leave this issue to the actual editors, rather than issuing mandatory advice here out of context. We already have some advice about it in Wikipedia:Words to avoid. Any thoughts? SlimVirgin ^(talk) 22:21, 18 April 2007 (UTC)

Unless I am missing something, this example has nothing to do with our concept of 'synthesis'. In this example the topic appears to be how to determine consensus among published sources about a given issue. This has to do with NPOV, not with synthesis or NOR. Crum375 22:33, 18 April 2007 (UTC)

Semi-hypothetical query

This is moot now as one of my sources has disappeared, but:

If I have a primary source statement that says "If A then B", and I have another source that states "A", then is it acceptable to state these, and then say that "B" is likely? Another editor opined that this might be original research, but it seems that logically I'm creating no new information. scot 22:27, 18 April 2007 (UTC)

This was regarding an article we were both involved in, but I might as well reply (even though it's now out of date) - see WP:SYN. -Halo 23:15, 18 April 2007 (UTC)

The original research there is introducing the idea that B is likely. The non original research way to present this would be simply "A is so, according to Z. Y says that if A then B." and nothing further. --bainer (talk) 05:34, 19 April 2007 (UTC)

I think you're right--and Halo has provided me with an even more explicit source for "A", and I have "If A then B" from a primary source. I'm still not certain that "B" follows, based on potential ambiguity in the definition of "A" used by both sources (hence my desire to state that "B" was likely) but I suppose that can be left for readers to deal with--since there are other sources which state "not B" (sources I obviously question in light of "A") then perhaps those sources should be repeated in the same paragraph. scot 13:50, 19 April 2007 (UTC)

Well, now the "not B" sources appear to have disappeared, and someone else did as Thebainer suggested and just stated "A" and "If A then B", so I guess I leave it at that. scot 13:56, 19 April 2007 (UTC)

Citing Oneself

I think this section needs to be worded to more strongly discourage against citing oneself. As it reads, if one does not follow the link to Wikipedia:Conflict of interest, it sounds like citing oneself is not a problem. I propose replacing this:

If an editor has published the results of his or her research in a reliable publication, then s/he may cite that source while writing in the third person and complying with our NPOV policy. See also Wikipedia's guidelines on conflict of interest.

With something like this (the bold does not go in, I've just put it here to emphasise change):

If an editor has published the results of his or her research in a reliable publication, then s/he may cite that source while writing in the third person and complying with our NPOV policy. However this is discouraged - see Wikipedia:Conflict of interest

How does that sound? Rocksong 03:12, 20 April 2007 (UTC)

Sounds like an improvement, but is it necessary to use bold text? -Sensemaker

May I suggest this tweaked version? "Editors who have had their material published by reliable publications may cite themselves as sources, while writing in the third person and complying with our NPOV policy, but doing so is not encouraged. See Wikipedia:Conflict of interest." SlimVirgin ^(talk) 07:51, 20 April 2007 (UTC)

• This seems a bit better - legimately using your own reliable sources shouldn't be discouraged - this just isn't a license to self-promote. As long as editors disclose that they're citing themselves, I don't see why this is a problem. Maybe it's better is they offer it on the talk page and get someone else to do it. (Okay, definitely). WilyD 21:09, 20 April 2007 (UTC)

I've just noticed that Wikipedia:Conflict of interest has a specific paragraph Wikipedia:Conflict of interest#Citing oneself which puts it fairly well. (Doesn't actually discourage it, but suggests using the Talk page like WilyD says). Perhaps we could just add one of those "See also" templates to point specifically to that paragraph, i.e.:

See also: Wikipedia:Conflict of interest § Citing oneself

Rocksong 00:27, 21 April 2007 (UTC)

Article with a LOT of Synthesis

Could someone please check out List of United States Presidential religious affiliations. It seems to me that significant sections of this article amount to an original Synthesis ... drawing conclusions about the religious affiliations and beliefs of the Presidents based upon primary sources. This isn't a POV issue. My sole concern is with such a massive WP:SYNT violation. Unfortunately, based upon my discussions about similar issues at related articles, I am not sure if the principal editor understands this, or sees what is wrong with the article. I am not sure how to fix this... If we cut the synth, we would basically gut the article (which is definitely noteworthy and should not be deleted). I could use some help trying to explain things and figuring out how to fix them. Thanks Blueboar 14:09, 20 April 2007 (UTC)

Original images - Need an admin to assist

I need an admin to take a look at Shadow people and to rule whether the original images that were included there (see history) violate WP:OR.

I've tried to explain to another user how images are mostly exempt so long as they represent a something that has previous been established, but they won't listen.

perfectblue 18:28, 20 April 2007 (UTC)

• See WP:ADMIN - but Admins don't really do this. Try WP:RFC as your first step in dispute resolution. WilyD 19:18, 20 April 2007 (UTC)

The idea that there is leeway with respect to images is so that someone can go take a photo of, say, a galah, and give it the caption "this is a galah", or draw a diagram of a spark plug and give it the caption "this is what a spark plug looks like". It's not a free licence for speculation. --bainer (talk) 01:01, 21 April 2007 (UTC)

When there are clear descriptions, it's not speculation.

perfectblue 13:19, 23 April 2007 (UTC)

Dispute regarding publication of original research in article

Please see here. Thanks, --Sagie 18:46, 22 April 2007 (UTC)

Creation-evolution_controversy article (clarification and guidance sought)

Regarding On using primary sources, I contend that by taking the side of scientists (or creationists) in a socio-political article regarding a controversy involving scientists (and creationists), it is POV--while taking the side of the scientists on a pseudo-scientific article is NPOV. Please either comment here or at the Creation-evolution_controversy talk page. This is an important issue requiring the guidance of disinterested wikipedians as the article attracts a lot of well meaning partisans. Thanks. ImprobabilityDrive 07:06, 25 April 2007 (UTC)

"Taking the side" of either side is, by definition, POV in any article. However, pointing out facts, such as that creationism has very little standing in the scientific community and hardly qualifies as a scientific theory, is NPOV in any article. However, verifiable sources for the facts should always be added to the article. -Sensemaker

There is no rule in Wikipedia that an editor can not "take sides" on an issue. What we have is a rule that says we should write from a Neutral point of view. Please see WP:NPOV. And wouldn't this be more appropriate to ask at that policy's talk page?... it really doesn't have anything to do with original research. Blueboar 17:11, 25 April 2007 (UTC)

Remove two words?

There is a sentence in this policy that reads "Wikipedia articles should rely on reliable, verifiable, published secondary sources wherever possible." What are people's thoughts about removing the words "wherever possible?" Isn't it precisely the csaes where reliance on secondary sources is NOT possible that are most important to avoid in WP, in order to follow the NOR policy? Note that the sentence would still have the word "should" and not an alternative like "must." Thoughts? UnitedStatesian 17:04, 25 April 2007 (UTC)

To give an example to illustrate this issue: suppose there is a major current news event. A Wikipedia article might be started that relied mainly on news reports of eyewitness accounts - i.e. primary sources. This is not original research provided that the article is not interpreting the primary sources and drawing novel conclusions from them. Months or years later, historians and experts would write secondary accounts of what happened, and as and when these good secondary sources become available, the Wikipedia article should rely on them. I think the "where possible" wording is emphasising that where good quality secondary sources exist, we should use them, but that when they don't, it could still be possible to write an article based on good quality, relevant, primary sources.

I think it's also important that we don't elevate secondary sources to being automatically "better" than primary sources. Sometimes the reverse is true. A common example is when a scientist publishes a scholarly paper (primary source), and lots of clueless journalists write headline grabbing stories that misrepresent the findings (secondarly sources). Often, the best articles use a combination of primary and secondary sources.

That said, I have no strong objection to removing the "wherever possible", provided that we retain the "should" rather than "must" wording. Enchanter 18:48, 25 April 2007 (UTC)

Well, I hate to quibble when you've agreed with me, but I have a different understanding of primary and secondary: I would call the unedited interviews (in the case of the news event) or the scientist's data (in the case of the paper) the primary source - neither of which should ever be used in the WP article. This makes the news report and the scientific paper both secondary sources, and yes, later there may well be other secondary sources that may be better (in the case of the news event) or worse (in the case of the scientific paper), and the WP article should always contain the best in either case, but they are all secondary sources. UnitedStatesian 19:33, 25 April 2007 (UTC)

I think there's a grey area on whether you call a newspaper report that is very close to the original story (such as, for example, a TV news story that is mostly interviews) a primary or a secondary source; the same goes for some scientific papers which give the results of experiments without necessarily interpreting them. However, I think your more narrow interpretation of what a primary source means is more natural and more common (particularly outside Wikipedia!). Enchanter 22:49, 26 April 2007 (UTC)