Wikipedia talk:These are not original research

calculations

The text here about avoiding statistics, algebra, or calculus calculations doesn't fit actual practice. In practice, calculations that would be clear to someone with a first course in the area are used with no complaints. The idea that calculations must be clear to someone with no training also seems to contradict the permission on translations; if everyone could translate already, we could just quote the text in its original language. — Carl (CBM · talk) 15:41, 13 December 2007 (UTC)

Maybe in some limited areas it's done, but my impression has been that the broader community disapproves of complex calculations as not any different than other synthesis, particularly in crunching statistics. The translation matter is just an exception around since the early days of Wikipedia. I understand the two do not seem balanced, but it's just how the overall community has treated matters. Vassyana (talk) 17:22, 13 December 2007 (UTC)

Trouble arises not so much when someone introduces "advanced mathematics" to a subject where such is the norm; one would expect an article on thermodynamics or econometrics to be full of calculus. Trouble arises when people introduce "advanced mathematics" to a study in which such is novel -- or introduces novel mathematical or statistical material to an existing study. Thus an attempt to use a mathematical formula to determine what creates human happiness would likely be "original research".

It is often not so much the mathematics that creates problems; a typical college graduate probably knows his first-year calculus if he has kept up with it; it is with the formulas that one creates. Few have a problem with the perfect gas formula

pV = nRT

(p is pressure, V is volume, n is the measure of atoms of gas, R is the gas constant, and T is absolute temperature), and it proves useful in many physical contexts. For example, what happens to a gas as it expands or contracts without radiating off heat?

or the quantity theory of money in economics:

MV = PQ

M is money supply, V is velocity of money (rate of spending), P is the price level, and Q is the quantity of transactions. It too proves useful in explaining that if the money supply increases that the price level can be expected to rise some -- slightly in a time of monetary insufficiency (1933, USA) or roughly proportionally in an overheated economy (1923 in Germany).

Neither the perfect gas law nor the quantity theory of money is original research in their respective fields.

Application of new formulas to a study or novel interpretations of existing formulas is original research. If someone contends that the rock surface of Venus is hot and the rock surface of Mars is cold in contrast to that of the Earth's water, ice, and land/life surfaces entirely because of differences in pressure, then such is likely original research even if one uses the perfect gas law.--Paul from Michigan (talk) 15:55, 20 February 2008 (UTC)

One remark - there is a clear difference between scientific calculations and economical. The MV=PQ economic formula is not proven in the same sence as a mathematical formula. All math formulas with the exceptions of a few very simple

statements (such as A+B=B+A) can be proven step by step. Doubt if that's the case with all economical math. Boeing720 (talk) 05:00, 3 April 2013 (UTC)

Can I confirm if things like counting or aggregating statistics 'count' as simple calculations. E.g. if my source is a list of a the winners of Formula 1 Grand Prix, with their nationalities, and I was to say "x% of all winners are German" even if I didn't have a source that mentioned the % of winners by nationality? The article says "If the source says 75% you can say 'most'" it doesn't say what to do in the event that the source does not discuss relative numbers. (this is just an example, please don't respond with a source that says the % of German winners!)

JeffUK (talk) 13:14, 24 March 2018 (UTC)

Complex mathematical derivations should be legalized

As I explained here we need a rule that allows for unsourced arguments/derivations. I suggested this formulation:

• Detailed arguments/derivations don't have to be sourced if included to make statements verifiable. In that case they are not included as encyclopedic content whose veracity should be verifiable to non-experts who are not able to understand the derivation/argumentation.

This violates the ban on complicated derivations. But if you read this, you have to agree that only allowing simple arguments, referencing statements to the literature, did not work well for the thermodynamics articles. Of course, one can point out that in theory it could have worked. But that's irrelevant. My point is that in practice it didn't work (like communism, great in theory, in practice it didn't work well). Huge errors remained uncorrected for two or three years in the articles and that is a very serious matter.

In the many other articles that have been free of these problems, the ban on complex derivations is not adhered too. So, it would be a good thing if Wikipedia legalizes this successful practice. Count Iblis (talk) 15:43, 26 August 2008 (UTC)

This is really something that would need to be addressed on the policy level over at WT:NOR. This essay is just a bulwark against overly strict interpretations of NOR, based on long-standing community consensus. I do agree, to a certain extant, with your statements that a large part of the problem is people with limited to no knowledge of a topic mucking it up. This essay allows for calculations that are understandable to reasonably educated editors. To me, that means if someone with basic working knowledge of calculus and physics can verify the calculations with relative ease, then it is permissable. This should be relative to the topic, to some degree. For example, proper calculations for the propagation of electromagnetic waves or fluid mechanics pretty much require an ability to work with partial differentials. As another example, working with alternating current circuits mathematically requires a functional knowledge of serious algebra and imaginary numbers. While I would not support unsourced complex equations to prove an argument, I would entirely support a rephrasing to reflect my statements here. To me, there is a world of difference between the complex math simply required in some topics and using complex math to build an argument within the topic. The former is a common circumstance and mostly understood by the broader community. The latter would require a fundamental shift in NOR and the community consensus around it.

Also, that all said, I tend to think the problems on the articles you raise are not so much rooted in wiki policy. Rather, they are rooted in people with no real familiarity with the topic working on such articles. Philosophy articles have similar accuracy and readability issues, but it is a result of fauxlosophers with fundamentally flawed superficial misunderstandings of the topics, rather than adherence to sources. Vassyana (talk) 16:17, 26 August 2008 (UTC)

I would tend to agree. I don't think that a change in policy is needed, just a clue about the nature of the topic. While Democratic peace theory is apt to interest readers without algebra, let alone a fair grasp of statistical inference, anyone who ventures into Quantum gauge theory should have at least one 300-level math course under his or her belt. I think it better to make one's case article by article, or even to ignore the rules where needed, than to open the floodgates by trying to alter policy. Robert A.West (Talk) 21:47, 26 August 2008 (UTC)

If both the start and end points of the mathematical calculation are properly sourced and the intervening calculation is what would be expected of any person who works in that field, I see no problem with such calculations. Martinvl (talk) 15:57, 18 July 2011 (UTC)

What is much more important is that any analysis or conclusions based upon a calculation be sourced. The point behind WP:CALC is to allow using standard, easy to follow calculation to restate what sources say... so, for example, if a source gives a measurement in metric units, we can restate that in standard non-metric units. Taking this concept to more complex topics (and tossing in some made up technobabble to illustrate the point)... if "Quantum tachyon parameter units" are commonly converted to "Einstein-Bose wave functions" using a known and WP:Verifiable formula, it should be OK for us to perform this conversion in an article... It's not OR no matter how complex the math involved because the concept of using the formula to convert from one of these two measurements to the other is not original.

However... if an editor has discovered that you can convert "Quantum tachyon parameter units" into "Einstein-Bose wave functions", and uses complex math to explain (or "prove") his discovery in Wikipedia... that definitely is OR.

To put this all another way: it is OK to you to use a mathematical formula (no matter how complex) to calculate something if (and only if) that formula is commonly used by sources to calculate the same thing. It is not OK to invent/create your own formula... even if the math is accurate. Doing this is Original Research. Blueboar (talk) 13:08, 19 July 2011 (UTC)

Reading WT:NOR the I get the impression that the basic requirement for NOR is that the entry must be verifiable, not that either the final result or the method must already exist. For instance, it is permitted under the rules to translate a piece of text into English from another language, when an existing translation cannot be sourced. Translation is not an exact science, however: two people translating from the same source will tend to product two different results and only those who know the original language may check the translation. Any new translation is thus OR, if we apply the rules you give, but it's not OR according to the guidelines. Isn't there some double standard here? What applies to language does not, it seems, apply equally to maths. Were we to apply the rule that the entry must be verifiable by anyone with the necessary skill, as I think we should, then complex calculations would be allowable, as Count Iblis suggests. Dendropithecus (talk) 02:19, 2 October 2011 (UTC)

Reasonably educated

I have removed the second paragraph from the "Simple calculations" section pending discussion. It read:

• "Reasonably educated" is directly relative to the topic involved. For example, for someone working with fluid mechanics topics, many common calculations require a working knowledge of partial differentials. As another example, someone working with calculations involving alternating current circuits would need a working knowledge of imaginary numbers and advanced algebra.

I think this paragraph implies that we allow a level of calculation that we don't actually allow. There is no way that an editor can or should use partial differentials to build a statement. That simply does not pass WP:CALC, no matter what the topic. A particle physicist probably finds differential calculus "Simple"... but a particle physicist is more than just "reasonably educated" he/she would be classified as highly educated. I think this needs to be reworked. Blueboar (talk) 14:00, 16 July 2011 (UTC)

I think one should define "reasonably eductated" as having the minimum amount of knowledge to be able to understand some explanation in sources. Wikipedia can have that argument in a Wiki article, but we are required to present that in our own words. Then someone reading that who doesn't have the knowledge to follow that argument cannot verify that the argument is consistent with the cited references. The issue with CALC is then that the difference between a literal reading of the sources and what is written in the Wiki article may involve non-trivial mathematical manipulations that should be routine for someone who is able to understand the text. Therefore it is necessary to expand CALC and say that this is indeed allowed.

E.g. some time ago, I wrote the section on the quantum mechanical version of the Poincaré recurrence theorem. To understand this, you need to have at least a basic understanding of the mathematical formalism of quantum mechanics. The paraphrasing also involves math at this level, so much more than just "simple calculations" are actually used to build the argument. But since there is nothing wrong with having this text on Wikipedia, something is wrong with the current formulation of CALC. Count Iblis (talk) 15:22, 16 July 2011 (UTC)

"Reasonably educated" has to be understood as context dependent,i.e. on the topic of the article. So it is more like "reasonably educated in the field" or "possessing an reasonable amount of domain knowledge".--Kmhkmh (talk) 18:59, 16 July 2011 (UTC)

I understand that... The problem is that the way this paragraph was worded it goes way beyond the original intent of WP:CALC. The entire point of the WP:CALC exception was to allow very very basic math... math that a reasonably intelligent average reader could do... It was to allow things like converting square meters into square feet. The mathematical equivalent of simply swapping one word with a synonym. the way this paragraph reads (at least to me) it would allow a very highly educated editor to come up with a completely original analysis or conclusion... and justify it by saying "the math is too complex for the average reader to follow... but trust me its right." What the reader needs is a citation that will show that the conclusions based on the higher math are not original. Blueboar (talk) 20:03, 16 July 2011 (UTC)

I agree it can be abused in the form of "the math is too complex for the average reader to follow... but trust me its right.", but for the average educated reader simple mathematical paraphrasing of abstract material might be already too complex to follow (similarly to paraphrasing a in language that reader doesnt speak). In other words the notion, that the average educated reader needs to be able to verify the article/recognize for sure that there is no OR at all is nonsensecial to begin with. Certain articles on advanced topics can only be seriously proofread/verified by people having a reasonable degree of domain knowledge, that just comes with the nature of the material. In some cases we simply need to rely on our editors with domain knowledge to keep potential absuses in check.

Now you could argue of course that WP:CALC fails somewhat in that regard, but I thought it point of this essay would be to address the shortcomings of a "too narrow" or "false" interpretations of our poliocies. If it however just repeats the views/interpretations given those policies already, I see no real point in having this essay in the first place.--Kmhkmh (talk) 10:49, 17 July 2011 (UTC)

So we should re-write the paragraph...to allow what should reasonably be allowed, but at the same time avoid the abuse. Blueboar (talk) 11:12, 17 July 2011 (UTC)

Yes imho the original text was ok, but it could be amended by a comment regarding abuse and possibly ways to handle it (in particular getting people with better domain knowledge to have a look it). It could also be made clear that this is restricted to cases, where the nature of the material really requires it for an intellegent description and that it cannot be used as an excuse for overly abstract paraphrasing or advanced calculations where the material doesn't really require it.--Kmhkmh (talk) 12:42, 17 July 2011 (UTC)

Criteria of "difficult to detect errors" ignored with regard to translation from other languages.

I respect the desire to use foreign language material as sources, but having seen dozens of sources misinterpreted in a simple summarizing reorganization (i.e., from English to English), I believe the potential for mistranslation and misinterpretation is even higher for a Wikipedia editor-translated source. Also, even assuming a professional translation, the problem cited for any mathematics other than the most simple applies -- checking a source other than in English requires "skills that common educated readers do not possess, or involve a large number of steps that may not be obvious, making it difficult to detect errors." I do not believe that the desire for a less English-centric encyclopedia overrides the need for editors to be able to verify sources without special training. At the very least, the translator should be cited along with the original source, following the principle outlined in Wikipedia:CITE#Say_where_you_found_the_material. Blackworm (talk) 06:09, 13 June 2008 (UTC)

On the other hand if the article is about something in a foreign country, then the best references might well be in the language of that country and a significant number of readers would also have a knowledge of that language. I have had a few instances where I have corrected an article because it was erroneously translated into English by a native speaker of the foreign language concerned. Martinvl (talk) 21:15, 3 November 2009 (UTC)

NOR and first use of quotations

OK. I am curious. Articles on specific phrases often include the "first use". In many cases, this is a well-known phrase. However, what if a Wikipedian finds- and cites- an earlier use of said phrase? (For examples, in the talk page on Where no man has gone before, I cited a use prior to the date of said article, which someone else added in later. A similar situation may occur with Behind the sofa or Faith-based (See the talk pages of said articles for my contribution).) Orville Eastland (talk) 03:35, 14 November 2008 (UTC)

I've seen this a number of times. Read the bit about primary sources in the No Original Research policy. Using a primary source- Lovecraft'ss books when discussing Lovecraft's books- is NOT original research so long as you are drawing interpretation from them. That may be unclear in some cases, but as far as "this quote is in this book", there is no interpretation involved. It's on the face of the work. Though wiki may prefer as secondary or tertiary source, even the letter of the rule does not require us to be blind to what is patently on the face of the book. I would include original copyright dates as patent to the book. If there is any issue, it's more verifiability, and really, if there's no specific reason to doubt it, there should be no further sourcing needed. In fact, years of release throughout most author, series, and book articles is based off the books themselves from what I've seen. IMHO (talk) 08:02, 14 November 2008 (UTC)

Making this article a guideline??

As an extension of the "Policy" and "Kudos" discussions above, whose resolution is unclear to me...I think a) this "essay" has to be explicitly referenced by a short summary paragraph in the WP:NOR article - and b) additionally -- upgraded to a guideline. With the holidays here, I'm just throwing this out there. I'll come up with specific examples in a few days. CarolMooreDC (talk) 15:09, 23 December 2008 (UTC)

syllogisms

i wondered how editors felt about moving what i think is the least controversial section to NOR. as NOR allows routine calculations, routine syllogisms should be included there not here because arithmetic implicitly contains logical "calculations". in fact language itself is fundamentally built on (among other things) syllogisms. we make them all the time, we just don't realise. for example, using a pronoun involves a syllogism--Mongreilf (talk) 09:40, 4 April 2009 (UTC)

more on logical deductions

The issue of whether basic logical deductions (and especially syllogisms) should be allowed or not has come up again at the main NOR article... the consensus seems to be fairly clear that a logical deduction can be OR... and that we should not exempt such material from the policy. I have therefor removed listing them in this essay. Blueboar (talk) 17:11, 3 November 2009 (UTC)

I've reverted, as just because some people don't think it should be explicitly stated in the policy doesn't mean it should be removed from this essay. I haven't had time to read the huge discussion at WT:OR, but from what I skimmed the major objecting arguments are: (1) someone could potentially reason from faulty premises and/or using invalid logical reasoning, (2) no one has shown that the lack of such language in the policy has actually caused any problems, and (3) "the proposer is trying to legitimize his POV-pushing in some unspecified manner". I don't see any real argument that all logical deduction, including that using correct reasoning from well-supported premises, is or should be disallowed. Anomie⚔ 18:14, 3 November 2009 (UTC)

You know what? Screw it. If people would rather make undiscussed edits to gut the essay rather than actually discussing anything, I'm not going to worry about it (and just WP:IAR if the situation ever comes up). Anomie⚔ 19:10, 3 November 2009 (UTC)

I originally wrote this essay to start a discussion over what original research should, & should not, cover. Since then it's been embraced as policy, & rewritten beyond what I intended it to be. (Essays don't create policy, they are meant only to make people think.) I've preserved my version here. If you want to quote this essay, quote my version; this version should simply be deleted as being redundant, & disputes over what "original research" means should be centralized at the NOR talk pages. -- llywrch (talk) 22:32, 3 November 2009 (UTC)

Good point... let us all remember that the actual policy is still at WP:NOR. This page is not policy. Blueboar (talk) 13:11, 20 July 2011 (UTC)

Another kind of simpe or direct deduction

I've come into a discussion on Talk:App Store which involves the limits of WP:OR and WP:WEASEL, which might be of relevance to this essay. I added to the article App Store a sentence saying that "some have started using 'app store' as a generic term referring to any similar service", and then giving three references to such use in notable sources. Another editor claims that this is synthesis, since none of the sources say that the term is used in this way - although he agrees that the sources are using the term in the way I say. My point here is, that maybe this is an example of something that should be addressed either in the Simple or direct deductions or Compiling facts and information section in this essay? To me, the edit in question seems to represent the same kind of thinking that lies behind the "Simple or direct deductions" section, but in a different field of knowledge (language rather than mathematics or logics). Could something be added to this section to include similar kinds of deductions that don't necessarily belong strictly to the fields of mathematics or formal logic? --Anderssl (talk) 21:32, 23 June 2009 (UTC)

In situ recording

If I record something that I see and report it in Wikipedia without any additional comment, is this research or not. The item that I have in mind is in the article M3_motorway. Towards the end of the article is an exit list with headings "Northeast exits", "Junction" and "Southwest exists". Recently roadsigns have been erected which identify the Northeast carriageway as being the "B" carriageway and the Southwest carriageway as being the "A" carriageway. I cannot find any documentary evidence about this, but I frequently drive on that motorway and I can verify the fact for myself (as can anybody else who is able to drive there).

Since I describe it in an unambiguous manner without interpretation I have recorded the source as "Verified ... in situ ... on <date>". Comments? I am assuming of course that the signs will be there for the foreseeable future. Martinvl (talk) 20:18, 14 July 2009 (UTC)

Have you tried contacting the government body responsible for those signs (would that be the Department for Transport?) to see if they have any official map you could cite? Such a map would also be useful for citing the rest of the table. Otherwise, you could word it as something like "Driver location signs erected by the Department for Transport" (or whoever) to make it sound a bit better. Anomie⚔ 18:12, 18 July 2009 (UTC)

Anomie, thank you for your comments. I have located one map which I have used for the M25_motorway, but unfortunately the Highways Agency (who erected the signs) are trying to get maps done on the cheap, so they are not readily available (apart from the M25). What I will do is to write and article on the driver location signs and cross-refernce that article. —Preceding unsigned comment added by Martinvl (talk • contribs) 20:09, 18 July 2009 (UTC)

Page move

I don't think moving this page was a good idea. For setting that this or any page page is an essay/policy/guideline or what it comes, the template in the begining is enough, it's not needed to tell it in the page name. If someone takes an essay as policy, he should be explained better, and if someone tries to enforce an essay as if it was policy, it should be pointed that way.

Besides, if someone attempts to turn a lowly watched essay into a policy or guideline, for later enforcing it, there are bots that would warn so at the village pump. MBelgrano (talk) 21:33, 3 November 2009 (UTC)

Agreed. The move doesn't make much sense, considering this essay uses Template:Essay. --Conti|✉ 21:52, 3 November 2009 (UTC)

Translation and contextualizing

An issue has arisen here Talk:Anti-Americanism#Latin_America over whether this text [[1]] should be retained as resonable contextualisation, or does it constitute OR.Slatersteven (talk) 15:48, 6 July 2010 (UTC)

Use of maps

Material that I posted here was on the wrong notice-board. I am moving it. Martinvl (talk) 12:30, 4 May 2011 (UTC)

The material that I originally posted has resulted in an essay written mainly by by User:Moabdave and myself. That essay has now reached a relatively mature stage and I feel that we need to bring it to the attention of the user community. I have chosen to do it via this article. Martinvl (talk) 16:42, 28 July 2011 (UTC)

Where is your essay and what are you doing? User talk:Bluerasberry (signature added by User:Martinvl)

It is here. Martinvl (talk) 19:49, 28 July 2011 (UTC)

Simply syllogisms

SlimVirgin removed this:

• Simple logical deductions. For example, if A is in district B, and district B is in province C, then A is in province C. This is a simple syllogism. Included are all of the other simple deductions. More complex logical deductions should not be used unless cited to a reliable source. The concerns are similar to the issues with complex mathematics.

Now I seriously doubt that SV thinks that if Town A is in District B, and that District B is in Province C, that Town A is anywhere other than in Province C. But perhaps there is a more significant concern, e.g., people trying to pass of complex "logical" deductions as being acceptable? WhatamIdoing (talk) 07:36, 20 July 2011 (UTC)

I should be edited back. Moreover from my perspective the point of this essay (rather than policy or guideline) to add a particular perspective to the grey border areas of the existing policy to show in which cases their rigid interpretation is counterproductive. Now if the differing views about the policy (in particular the literalist or rigid interpretation) get mixed into this essay and turn it into a text somewhat mostly redundant to the policy, then there is no point in having this essay in the first place. If there are some editor who feel the need to formulate the "literalist" interpretation, they should restrict that to the policies discussion or write their own separate essay.--Kmhkmh (talk) 11:54, 20 July 2011 (UTC)

Accuracy conflict

Bruce's ongoing effort to get his findings on the phrase conspiracy theory presented as the Truth™ is getting a little old.

Y Yes: Sources may be wrong.

Y Yes: If you can produce published evidence that the source is wrong, then you may omit the erroneous information. You may use WP:Editorial discretion to prevent WP:Inaccuracy in articles.

However, N No: If you cannot produce a published source that says some other source is wrong, then you may not say, or imply, or even hint, that the other source is wrong.

For example, imagine a typo in a statistic. 99.999% of sources say that 5% of American women die from breast cancer. One lonely source says that 55% do. We know from both personal experience and the overwhelming balance of sources that this one source is factually, obviously, definitely, absolutely, unquestionably wrong. But we do not say, "One source says that 55% of American women die from breast cancer, although all other sources say that only 5% do." We silently omit the erroneous information as WP:UNDUE and against our better editorial judgment. We do not hang the bad source out as proof that someone screwed up. We just say, "5% of American women die from breast cancer" and stop there.

Which is exactly what Bruce ought to be satisfied with in the dispute that he's wrapped up in, rather than trying to re-write all of the advice pages to support his side in the dispute. WhatamIdoing (talk) 16:37, 28 September 2011 (UTC)

I agree. Although I reach a different conclusion on how to deal with the issue... I will note that WP:OR includes the line: "If you discover something new, Wikipedia is not the place to premiere such a discovery." This includes discovering something about when a phrase was first used. If reliable sources say that the phrase was first used in 1980, and you (a Wikipedia editor) discover that the phrase was used before the date ... it certainly is OR to premiere your discovery in an article. Instead you need to write a monograph or something that "corrects the record"... and (if it gets published) that monograph might be used in the article.

That said... we are allowed to venture into OR on talk pages. There is nothing wrong with raising the fact that you have discovered the pre-1980 usage and then argue that the 1980 date is inaccurate. Your discovery presents a very good argument in favor of making the editorial decision that the article should not say that the fist use was in 1980 ... However, since we can not give the newly discovered previous date either... it probably means that the correct solution is for the article to simply omit any discussion of when the phrase was first used. Blueboar (talk) 20:38, 28 September 2011 (UTC)

Or you could say "The phrase has been used at least as early as X", without mentioning that any source came to a different conclusion. WhatamIdoing (talk) 21:05, 28 September 2011 (UTC)

The problem with that itself is OR. Let look at part of the actual example presented on the TALK page shall we?

"The first recorded use of the phrase "conspiracy theory" dates back to a history article from 1909." (Knight, Peter. "Plots, paranoia and blame". BBC News 7 December 2006) met Verifiability by any reasonable standard--a direct quote by a Senior lecturer in American Studies from the University of Manchester in a well respected paper.

"I must content myself with saying that the class conspiracy theory of economic development may generally be considered false,..." (The Economic review: Volume 1 Christian Social Union (Great Britain) Oxford University Branch 1891 Page 540.)

It is clearly Verifiability that an 1891 Oxford University Branch article does use the phrase conspiracy theory before 1909. Trying to handway this away just hampers understanding of what Verifiability and OR really are.--BruceGrubb (talk) 06:34, 29 September 2011 (UTC)

It may be a verifiable fact, but it is still OR to discuss that fact it in Wikipedia. I still think the best solution here is to simply omit all discussion of when the first recorded usage was. It is never OR to not talk about something. Blueboar (talk) 12:49, 29 September 2011 (UTC)

Again I point to this part of this article:

"Comparing and contrasting conflicting facts and opinion is not original research, as long as any characterization of the conflict is sourced to reliable sources. If reliable references cannot be found to explain the apparent discrepancy, editors should resist the temptation to add their own explanation."

By WP:NOTOR discussing a conflict in the article in of itself is NOT OR only presenting unsourced explanations for said conflict.--BruceGrubb (talk) 04:11, 30 September 2011 (UTC)

Yes, that's exactly the line in NOR that you're violating. Presenting two sourced facts so that the reader sees that one of them is wrong is an implicit "characterization" of the fact as being wrong. You may not do this unless you have a source that directly says that the fact is wrong. It is not necessary to present an "explanation" for why the fact is wrong to violate the rule against providing "any characterization" of the sourced material as being wrong. WhatamIdoing (talk) 14:32, 30 September 2011 (UTC)

WP:NOTOR expressly states that presenting "characterizations" of conflicts is NOT OR. The idea that presenting a source meeting Verifiability showing other source may have an accuracy problem (unless people are suggesting Oxford University Branch doesn't meet Verifiability which is a real huh?)

This is a variation of the same kind of nonsense the Jesus myth theory article went through (Talk:Christ_myth_theory/definition Talk:Christ_myth_theory/POV_tag give good overviews to some of that nonsense). Anything that showed there wasn't a single unified definition of "Christ-Jesus myth theory" was called OR by some editor and it took a two year battle before the conflict of definition was put into the article space.

That WhatamIdoing wants to unilaterally remove the following from this article space shows that he does not understand what OR and Verifiability really mean:

An accuracy conflict is when a source presents something that is not accurate (Example: one source states that a certain phrase first appeared in 1920 but other sources show the phrase was used before that year.) [...] For example, you may not, with these sources, say "Source A gave the town's population as 5,000; however, this is disproven by the following sources and circumstances, and the true population was at least 7,000 in 1990." but you can in the talk page point out "Source A claiming this phrase was first used in 1920 is wrong as there are sources before 1920 that use the phrase" (it is not a good idea to do this in the article page however).

I would like to point out that administrator User:Will Beback stated regarding this: "When equally reliable sources give different accounts we should include them both. In the words of the great American philosopher Yogi Berra, "When you come to a fork in the road, take it.""

When I presented how conflict between Schweitzer and Sir James George Frazer was resolved in the Jesus myth theory:

Despite Sir James George Frazer stating "My theory assumes the historical reality of Jesus of Nazareth" in volume 9 of The golden bough in 1913, [22] Schweitzer continued to group him with John M. Robertson, William Benjamin Smith, and Arthur Drews in subsequent printings of his The Quest of the Historical Jesus and repeated this classification in his 1931 autobiography Out of My Life and Thought with the statement "I especially wanted to explain late Jewish eschatology more thoroughly and to discuss the works of John M. Robertson, William Benjamin Smith, James George Frazer, Arthur Drews, and others, who contested the historical existence of Jesus." [23][24] (22) Frazer, Sir James George (1913) The golden bough: a study in magic and religion, Volume 9 pg 412 (23) Schweitzer (1931) Out of My Life and Thought page 125 (24) Bennett, Clinton (2001) In search of Jesus: insider and outsider images page 205

one editor flew off the rails claiming 'yes this is OR via SYN' even though the conflict itself was referenced (Bennett)

Lambiam over as Wikipedia_talk:Verifiability/Archive_51#Contradictory_sources gave examples of Verifiability but inaccurate information to which I add the following examples:

"That's one small step for man, one giant leap for mankind." - Neil Armstrong July 29, 1969. (Multidisciplinary units for prekindergarten through grade 2 ISTE (Interntl Soc Tech Educ)

"On July 29, 1969, over 600 million people watched their televisions. They saw astronaut Neil Armstrong become the first human to step on the moon." Greathouse, Lisa E. (2007) The Wonder of Our Solar System Teacher Created Materials pg 15)

(The date for both these events is July 20, 1969)

The conflict between the Encyclopaedia Brittanica and Vrej Nersessian, in Essays on Armenian music regarding the birthdate of Komitas Vardapet (October 8 vs September 26) is a prime example as it could be a correction between the Rumi calendar and Gregorian calendar or a miscorrection of an already Gregorian date.--BruceGrubb (talk) 06:05, 1 October 2011 (UTC)

WhatamIdoing has not "unilaterally" removed the material... I completely support the removal. Let me try to explain why... Let us take the situation where we have two (reliable) sources that both discuss when word was first used:

• Source A notes that "the word was first used in 1920", Source B notes that "the the word was first used in 1918"

The two sources disagree over a fact (when the word was first used). As editors we should... 1) check other sources to see if a significant majority agree with one or the other of these dates; If so, we can 2a) discount the extreme minority view according to WP:Due weight. If not, we should 2b) mention what both sources say (with both views discussed in the article as being opinions. We could say: "Source A states that the word was first used in 1920 <cite A>, however Source B disagrees, and states that word was first used in 1918 <cite B>".) In this case Wikipedia is not claiming that either opinion is "correct"... we are simply presenting the disagreement. This is not OR.

Now let us take a look at your situation...

• Source A states that "the word was first used in 1920", however you have discovered that Source B used the word in 1918 (or when ever).

In this case, we do not have a difference of opinion between the two sources ... what we have is the discovery that that Source A is inaccurate. Since we don't have a difference of opinion to compare and contrast, we can not present the information as being disagreement between sources.

More importantly, you (a Wikipedia editor) were the one who made the discovery that Source B used the word in 1918. No reliable published source has taken note of this fact before. Wikipedia is not the place to premier an editor's discoveries. Thus, it is OR to mention your discovery in an article... and it is especially OR to contrast your discovery with what Source A says.

However, what you can do is discuss your discovery on the article's talk page (we are allowed to discuss OR on talk pages if needed)... you can argue that Source A is not accurate, and thus not reliable for this specific bit of information (it may be reliable for other things, but not this). You can use your discovery to argue that we should give Source A significantly less weight, and even omit what Source A has to say all together.

Alternatively, you could write a monograph or paper, noting your discovery (that the word was used in 1918), and try to get that monograph or paper reliably published somewhere. If successful, then we would have two reliably published opinions to contrast in the article. Blueboar (talk) 13:52, 1 October 2011 (UTC)

Blueboar, OR states "The term "original research" (OR) is used on Wikipedia to refer to material—such as facts, allegations, and ideas—for which no reliable, published source exists." Source A presents as a fact that a phrase existed no earlier then 1920. Source B from 1910 demonstrates the fact that the phrase did indeed exist before 1920.

Discovery is NOT OR--if it was then looking for any source meeting Verifiability would be OR and that is just plain nuts. This is the same argument that made the Jesus myth theory a two year exercise in head banging and making the Weston Price article a similar exercise in frustration.

Looking for a source should NEVER be considered OR and that is what claiming the discovery of a source is! How a source meeting Verifiability is use could be OR but most of the time it is a in realty a NPOV.--BruceGrubb (talk) 04:01, 2 October 2011 (UTC)

I can't speak to the wording of the specific case the BruceGrubb is concerned with since I'm not familiar with the details. I do however disagree pretty strongly with WhatamIdoing and Blueboar on one particular of what is allowed under WP:V and WP:NOR. In Blueboar's analysis wikipedia has greater freedom to contrast a difference in opinion than it does to point out an outright factual error, even when the impeachment is well sourced. This doesn't make sense to me, even in the context of our policies. I agree that evidence of factual errors can and should be used on talk pages to exclude erroneous sources, but editorial judgment is required as in some circumstances it would be a constant struggle to exclude the incorrect information from the article if the error is found in a well known source. It can be especially odious when the error is a simple typo in an otherwise impeccable source, although that doesn't seem to be the case here. I do not see any impermissible OR in a statement like: "Hochstetter (1971) claims that the first use of 'gonkulator' was Burkhalter (1968), but Crittendon (1966) is an earlier example." I think WP:UNDO [oops] WP:UNDUE can also apply, so I would recommend that these details be put in a footnote and that the main text simply say "The term 'gonkulator' was used as early as 1966.^[1]". Quale (talk) 07:05, 2 October 2011 (UTC)

I would disagree with "The term 'gonkulator' was used as early as 1966" footnote idea as in the course of doing more research an editor could stumble on even earlier use for 'gonkulator'.

One of the problems Jesus myth theory article had was conflict of definition and ambiguity between sources as to what it even was. When you get into subjects like that the footnote situation can get a little out of hand with many points that really should be int he article text itself dumped into the footnote section.--BruceGrubb (talk) 15:02, 2 October 2011 (UTC)

Quale, In Blueboar's analysis wikipedia has greater freedom to contrast a difference in opinion than it does to point out an outright factual error, even when the impeachment is well sourced. Yes, in some cases, but not all, exactly. We are free to note difference of opinion found in reliable sources, especially if they are secondary sources. If A says Y and B says X, we can note each, assuming each is reliable. But if a most reliable source contains a statement that is in error about the date a term was first used, and we cannot find any secondary sources that say that statement is in error, but rather go about and do our own research and discover that the statement is in error, we should not, by NOR, make the statement that "C says Z, but it's really S"--that is, strictly speaking, SYNTH. Your point about UNDUE (I assume that's what you meant) is well taken, as if we cannot find secondary sources contradicting the source we have, we should be asking ourselves whether the information is significant to be included--after all, if the error is not noteworthy by those in the field, it's probably not worth including in an encyclopedia. --Nuujinn (talk) 18:59, 2 October 2011 (UTC)

I don't agree. In this case we do have a source contradicting the incorrect source. The first source we have is "the first usage of 'foo' was in some particular year" and the second source directly impeaches the first as it is a usage of 'foo' that was published earlier. For a somewhat different example, I have seen some otherwise very reliable sources that have an incorrect date of birth that is given correctly in other reliable sources. The first source with the incorrect date of birth has other valuable information we include in the article so it is cited multiple times. This makes it very difficult for the article to ignore the mistake without noting it. The other sources that are correct on the date do not say "source X has the wrong birth date", and it's unrealistic to expect that these sorts of factual errors will always be noted explicitly. This is a real life example of an important factual error that is contradicted by other sources in precisely the same way as the original example. Quale (talk) 05:25, 3 October 2011 (UTC)

I think we're close to saying the same thing--for me the key issue is the nature of the sources. If reliable secondary sources disagree, we can note the disagreement. If one reliable secondary source differs from some number of other reliable secondary sources, and the disagreement is not controversial, such as a birthdate, I myself would not feel a need to even note that difference unless the data were challenged by another editor. But I contend that if an editor uncovers an error in an otherwise reliable secondary source through their own research in primary sources, we cannot say in the article "C says Z, but it's really S"--that's synth, because we're not just saying S is true, we are impugning C by implication and correcting someone who we regard as otherwise reliable. We could say "S", but we'd have to be careful, because S comes from a primary source, as you correctly suggest in the usage of "Gonkulator" in main article space. We could leave C out, because we know it is incorrect (which is what we've done thus far in the article whence this tempest in a teacup first sprang), and that would make good sense if Hochstetter's mistake is not noted in other secondary sources--if it is not, how important is that particular datum in the first place? But for me, "Hochstetter (1971) claims that the first use of 'gonkulator' was Burkhalter (1968), but Crittendon (1966) is an earlier example" even in a footnote crosses the line into OR, since that synthesis is the result of our original research. And I think that the more controversial the information, the more we'd want to hold that line, and the less controversial the information, the less reason there is to cross it in the first place. --Nuujinn (talk) 11:53, 3 October 2011 (UTC)

I agree. Personally I'd either silently drop the conflicting information, if I have very good reasons to believe its wrong, or I use a footnote to outline/explain discrepancies in greater detail. We have to walk a fine line here to assure that our content is as accurate as possible and to keep authors from abusing the guideline to push "real" WP:OR. But in but in cases where a mistake in secondary sources is "obvious" and "factual" (in a sort of scientific sense and there's no real disagreement regarding the mistake) we should not copy the mistakes for the sake of a(n imho pointless) policy literalism. In some of the cases discussed above you probably could simply use WP:IAR to assure the accuracy of the WP article.

Whether the policy needs to state how to handle such rare cases explicitly, is however another matter. In any case we should be clear what's to be done in such cases to keep our content accurate, i. e. if a policy literalism conflicts with the "obvious" and "objective" (scientific) "truth" (not your personal one!), then overrule it. Depending on the exact situation you might justify that with WP:UNDUE, WP:IAR or just plain common sense.--Kmhkmh (talk) 14:32, 3 October 2011 (UTC)

I agree, but note that WhatamIdoing and Blueboar claim that using a footnote to describe the conflicting sources is a violation WP:NOR. I don't think that policy needs any explicit statement about these uncommon cases, but I am surprised by the opposition as I think their interpretation of WP:NOR is incorrect on this point. The situation is uncommon but not necessarily extremely rare in my experience. I've encountered it about three times in chess biographies, where the standard basic biographical reference Chess Personalia by Jeremy Gaige contains information that is not correct. (These errors generally do not seem to be the author's fault, as they were not apparent in the sources available to him in 1987.) The incorrect information has been dates of birth and international titles, both things that are not easy to omit in a chess biography. To make matters worse, because Chess Personalia is the standard reference in English, its rare mistakes can be widely copied by other sources. This means that it's likely that we will never catch most errors in this source, but when they are apparent we should not propagate them. Talk:Yakov Estrin is an example where one intrepid editor had to work too hard to convince me that I was mistaken to rely on a title claim from Chess Personalia. Quale (talk) 04:11, 4 October 2011 (UTC)

I just read through Talk:Yakov Estrin and that situation is not parallel to the one we're describing--there you have a mix of primary, secondary and tertiary source that differ, and you all have chosen to document the disagreement. Personally, I would remove "The 1985 article by Nikolai Krogius, "At the FIDE Congress", 64 - Chess Review, , 2/85, pp. 2-3, lists all Soviet players awarded titles at that event and Estrin is not mentioned. Another report of the Congress mentions honorary GM awards only for Stojan Puc and Eero Böök (BCM, April 1985, p.159). " from the footnote, since there is a conclusion being drawn there, and I think that is SYNTH, but aside from that I see no problem there--sources disagree, you all list the sources and note the disagreement, and the sources directly concern the subject.

In the case we're talking about, there is one secondary source that gives 1909 as the date of the first usage, and a number of primary sources which do not treat the subject "Conspiracy Theory", but only use the term prior to the date given in the secondary source. We can use those primary sources to make a determination that the secondary source is incorrect, (which we have done), or we could mention that the term was in use "as early as x", but what we cannot do is link the secondary source and the primary sources, even by juxtaposition--Do not combine material from multiple sources to reach or imply a conclusion not explicitly stated by any of the sources. Does that make sense? --Nuujinn (talk) 10:28, 4 October 2011 (UTC)

As I pointed out over at Talk:Conspiracy_theory#The_first_recorded_use_of_the_phrase_.22conspiracy_theory.22_dates_from_1909.3F_WRONG.21 Knight states "The first recorded use of the phrase "conspiracy theory" dates back to a history article from 1909." He does NOT say that the phase "conspiracy theory" was used in any particular way--only that it was first used in 1909. The statement can be shown via WP:RS to be incorrect--the phrase "conspiracy theory" was used before 1909 and that fact can be proven via reliable sources.--BruceGrubb (talk) 14:32, 4 October 2011 (UTC)

Yes, BruceGrubb, that's why we removed it. What's your point? --Nuujinn (talk) 23:17, 4 October 2011 (UTC)

(@Nuujinn concerning Talk:Yakov Estrin): No, that doesn't make sense to me. I'm not well informed on all the particulars of the situation that has BruceGrubb displeased, but under those circumstances simply omitting the incorrect claim seems sensible and I agree that this will usually be the best approach if it is feasible. The nature of the Estrin title claims is puts it in far greater danger of running afoul of WP:SYNTH than does a simple mistaken claim about first usage. We do not currently have any source that says "Gaige and others were incorrect to state that Estrin was awarded the GM title". In particular note further that we don't have even a single source that directly says that Estrin was not a GM. Our best sources say that he was an IM (a lesser title), and we infer that that was the highest (over-the-board) title he was awarded. That inference is synthesis. Because of this, one could even attempt to claim that the Estrin sources are not even in conflict. Note also that the answer to the question of whether Estrin was a GM is a matter of fact, not opinion, even if we don't know the answer with absolute certainty. This means we don't have the out of claiming "in the opinion of some sources he was a GM, other sources opine he wasn't". (Blueboar placed emphasis on the difference of the greater latitude he says SYNTH allows contrasting opinions, compared to contrasting an erroneous factual claim with a simple and direct counterexample.) What to do with the GM claim for Estrin affects more than just his bio, as we have Category:Chess grandmasters and List of chess grandmasters to consider too. Switching back to the case of the publication dates the impeaching example is prima facie evidence that the claim was wrong, and I don't see any synthesis involved in pointing that out if necessary. This should be far safer to do than trying to handle conflicts like the Estrin GM title. (I waited a long time before putting anything in the article about it, and I never felt that what I wrote handled it quite as well as it should be.) I understand that we want to be very careful to not encourage editors to synthesize their own claims, but wikipedia should be able to correct simple factual errors. This is not something I would ever approach lightly and I think it always deserves discussion on the talk page, but I think there is a little bit more freedom here than some claim. Quale (talk) 04:17, 5 October 2011 (UTC)

In regard to freedom, one can always invoke IAR, and I appreciate this is a subtle issue. Also, thank you for continuing this, it is helping me refine some of my notions in this area.

It seems to me a key issue is phrasing, and another is purpose. The titles that a chess master holds or held is an important matter. In Yakov Estrin you all noted that some reliable sources claimed he was a GM, and that other reliable sources "report only that he was an IM and ICGM". The position of "only" there is critical, as you all are not saying that these sources report that he was "only an IM and ICGM", but rather that the only thing they say is he was an IM and ICGM. I see no conclusions being drawn there, implicitly or explicitly. I agree that the phrasing is a bit awkward, but I think it is about as good as can be managed given the sources--you really have to say something about the titles given the subject, you have high quality reliable sources which contain a implicit contradiction, so you list what the sources say and you leave it up to the reader. You all did make an inference in preparing the article, but you did not imply a conclusion in the text of the article or the footnote--it's a fine line, but I don't think you crossed it, and if you did, you didn't go far over the line (which isn't bright in any case).

In the Conspiracy Theory article, Knight is a reliable source for the subject in general, but not an expert in etymology. He makes a statement about the earliest usage of the term here, and that was found to be an incorrect statement. The statement there just contrasts that the phrase has been around a while but entered common usage in the 1960s. The latter part of that is admittedly interesting, especially since the term is largely been used since then as a pejorative, but I would argue that the particular year mentioned in the first part isn't really important but for the contrast it established--if Knight had said the earliest usage was 1805, or 1700, or 345 BCE, it doesn't really change the meaning of what he is saying in that article. And Bruce Grubb, to his credit, dug up another bit by Knight that suggested his source for that date was the 1997 Oxford dictionary, which suggested the first recorded usage was in a 1909 article in the American Historical Review. So it appears that Knight made an assumption about the accuracy of the dictionary, which is understandable since he's not an etymologist, but a lecturer in American studies, and turned out to be incorrect.

If, having found that Knight is incorrect about the first recorded usage, we simply said "the term was in use as early as date X", based on a primary source that simply uses the term, that seems fine to me. If there were a debate between him and another reliable secondary source about first usage, documenting that disagreement would also be fair game, but that's not the case here. Removing the material that is incorrect also seems fine to me, and that's what we did about a month ago. But to juxtapose the known incorrect statement by Knight with primary sources showing him up as wrong does not simply provide accuracy and correct the mistake, it also implicitly brings into question his reliability as a source by correcting his error through our own research. And, although we know he's wrong about that particular datum, I would suggest there's no value created in the article by saying he's wrong--really, all we're doing there is saying he said X, but he's wrong, as if we were Hollywood in 2 Stupid Dogs. And I think that's what bothers me about this particular example. And I think that does clearly cross the line into OR, and we cannot, I think, justify that action with IAR as the datum in question is of very limited or no substantial value (which, arguably, one could in regard to Estrin, since the issue there is important). I hope that makes some sense. --Nuujinn (talk) 09:32, 5 October 2011 (UTC)

Substantial edits to conflict between sources

I reworded most of this, please take a look and correct me as necessary. I think I improved it, but others may likely disagree. --Nuujinn (talk) 16:32, 8 October 2011 (UTC)

Inaccuracy

Seems there is a dispute over adding Wikipedia:Inaccuracy and mentioning it in Wikipedia:These_are_not_original_research#Conflict_between_sources. Maybe someone could start by trying to summarize the lengthy "Accuracy conflict" section above, or is the just more WP:DEADHORSE? — Preceding unsigned comment added by Ronz (talk • contribs)

More DEADHORSE. As far as I can tell, Bruce is trying to "game the system" by making the advice pages reflect his view of what should have happened in disputes that he lost, e.g., that he should be permitted not just to use his WP:Editorial discretion to completely and silently omit Knight's apparent error about the origin of the phrase conspiracy theory (and to substitute correct information), but that Bruce should be permitted to publicly embarrass Knight over the error. WhatamIdoing (talk) 16:10, 10 November 2011 (UTC)

Publicly embarrass Knight?!? You're kidding right? The general community reception of Wikipedia_talk:Verifiability/First_sentence/Archive_1#Verifiability_Fact_vs_Truth was supportive. It is not a deadhorse to point out that you can have sources conflict in terms of accuracy. This is akin to saying I am trying to Publicly embarrass the Interntl Soc Tech Educ or Lisa Greathouse by showing they messed up on the date of Neil Armstrong's famous speech and about as credable.--BruceGrubb (talk) 17:48, 10 November 2011 (UTC)

Again: Unless and until you have an actual, published, non-Wikipedia source that specifically says there is disagreement between other sources, then you can't say that Knight was wrong.

And I disagree with your characterization of that discussion as "supportive". You've been told over and and over and over that it's not okay for you to de-bunk sources in articles. You are permitted to silently omit erroneous information, but you are not permitted to say that a source is wrong. WhatamIdoing (talk) 18:05, 10 November 2011 (UTC)

So you agree with with following? "User BruceGrubb is for example arguing that we have to remove a statement by a recognized professional making a claim about the first usage of the word, because he himself has found an earlier usage. I say if the statement is significant we include it attributed to its source, regardless of whether BruceGrubb's or another editors original research suggest that the statement may be factually incorrect." (User talk:Maunus) I hope to heaven that this accuracy be damned approach is NOT what you are saying because Maunus effectively got my original goal perfectly.--BruceGrubb (talk) 18:59, 10 November 2011 (UTC)

• If there is a consensus that the interview with Knight is not a significant source (I think it probably isn't - since interviews do not necessarily require the same degree of accuracy as published research - and because in this case he seems just to be citing OED) then the claim about the earliest mention can be left out (I would suggest phrasing it as "the earliest entry in the OED is..." or "ealierst uses appear around the turn of the 20th century" ). What we don't do is publish OR/Synth (Knight is wrong) - nor do we beg the question (Knight says X BUT Y).·ʍaunus·snunɐw· 19:32, 10 November 2011 (UTC)

The problem is that Plots, paranoia and blame was NOT an interview but an article Knight himself wrote. S Marshall agreed with Mystichumwipe that putting in both pieces of information was a good (?) idea ("My opinion is that it would be reasonable, in that particular case, to say: Peter Knight states the first recorded use of the phrase "conspiracy theory" dates from 1909,(ref) but the phrase "conspiracy theory" also appears in Garrison, George Pierce (1906). I feel that it can't possibly be original research to identify what's said in a published work with citations." (S Marshall T/C 18:56, 8 October 2011 (UTC))). Several other editors wanted to go with the questionable idea of using the earlier source as evidence the phrase existed at least as early as that and others agreed with the remove the inaccurate information that actually was done.

The ironic thing is in his Conspiracy theories in American history: an encyclopedia Knight wrote that was vetted by ABC-CLIO said it was implied by the Oxford Dictionary that the earliest date was 1909.--BruceGrubb (talk) 23:53, 10 November 2011 (UTC)

That's not a scholarly article. If there is an encyclopedia that mentions this correctly then why are we even considering using the blurb at the BBC site?·ʍaunus·snunɐw· 00:00, 11 November 2011 (UTC)

Also you are interpreting his phrase "first recorded usage" to mean "first published usage" when he obviously means "first usage recorded in the OED". Your OR is not correcting anything or showing anyone wrong. If the 1891 example is relevant for anything then write an article about it - then maybe the next edition of OED will adopt it. ·ʍaunus·snunɐw· 00:02, 11 November 2011 (UTC)

If you didn't know of the Conspiracy theories in American history: an encyclopedia work how could you interpret Knight saying that? All Knight does say in the BBC2 article is "The first recorded use of the phrase "conspiracy theory" dates back to a history article from 1909." That is it. No information as to where the information is coming from or even what history article he is referring to.--BruceGrubb (talk) 00:35, 11 November 2011 (UTC)

The BBC article isn't a scholarly work and of course cannot be assumed to have authority or accuracy as if it were. The first error was that someone used it as a source. The second was to do original research to contraqdict it instead of just mnetiong wht the OED says.·ʍaunus·snunɐw· 17:09, 12 November 2011 (UTC)

The point getting missed in all this is this is actually covered under the Caveats about expert material section ("Experts are human, and can publish statements that are contradicted by known facts, or otherwise erroneous") Claiming that this is a deadhorse when the point is already in the article is curious to say the least.--BruceGrubb (talk) 01:20, 12 November 2011 (UTC)

Yes, experts can make mistakes.

No, it is not Wikipedia's job to say, "This expert said X, but Y is actually true." WhatamIdoing (talk) 02:18, 12 November 2011 (UTC)

Exactly... if we have reason to believe that an expert has made an error, best practice is to discuss the situation on the talk page - with the suggestion that the erroneous information be omitted. We should not try to "correct" the expert by pointing out the expert's error in the article (The one exception to this is when some other reliable source has explicitly commented upon the expert's error). Blueboar (talk) 13:41, 12 November 2011 (UTC)

Phrased that exact way you are correct but Mystichumwipe pointed out the editors in Talk:Conspiracy_theory/Archive_15#The_first_recorded_use_of_the_phrase_.22conspiracy_theory.22_dates_from_1909.3F_WRONG.21:

"I don't get it. BruceG has AGAIN provided numerous verifiable reliable sources that demonstrate that something we have in the article is clearly and unequivocally incorrect. And the response is to argue "it's beyond us to counter the author's claim because doing so would be original research." ! and weakly suggest "we could consider trimming that line out of the section." :-o

Only "...could consider"? What is there to discuss? I would say OF COURSE, we just take it out. And put in these new verifiable, reliable sources from Bruce G stating that here are some of the earliest examples of its usage." (sic; Mystichumwipe)

Until that last sentence Mystichumwipe was describing my goal perfectly but several editors wanted to argue nonsense:

"So it seem to me we have two choices, either to remove that particular bit or to see if we can work in the material without violating OR. The former is not particularly attractive and the latter would be difficult as some of the sources BruceGrubb has found are primary sources, and I'm not sure that the secondary sources are about conspiracy theories per se but may be passing mentions." (Nuujinn)

"Looking for sources using the phrase "conspiracy theory" is indisputably original research, and we simply don't publish our own work here." (Rklawton)

Simply LOOKING for sources is OR?!? REMOVING a proven inaccurate statement "is not particularly attractive"?!? THAT IS INSANE and THAT is what was really being argued.--BruceGrubb (talk) 04:54, 13 November 2011 (UTC)

You weren't looking for sources, you were looking for evidence to prove Knight wrong. Looking for sources would have been to look for sources that stated that "the first documented use of "conspiracy theory" was in 1891". ·ʍaunus·snunɐw· 15:23, 13 November 2011 (UTC)

And that evidence would require looking for sources to support it would it not? Face it your efforts to salvage what was a ridiculous position is getting more ludicrous by the moment.--BruceGrubb (talk) 03:40, 21 November 2011 (UTC)

No, looking for sources is not OR... but drawing a conclusion based on those sources is. In this case the conclusion is that a cited source is "inaccurate". It does not matter whether the conclusion is correct or not... if the conclusion is invented by a Wikipedia editor, it is OR. Now, OR conclusions like this can be discussed on the article talk page (Content policies such as WP:NOR do not apply to talk pages), but they can not be added to an article.

In this case, the discovery of an earlier usage for the term can be used to challenge the reliability of the cited source (and to argue that the cited source should, therefor, not be used)... but the discovery of the earlier usage can not be used to support adding a statement designed to counter or correct the cited source. Blueboar (talk) 14:29, 13 November 2011 (UTC)

This is the other part of the problem--accuracy and reliable are NOT often connected together. For example I argued a long time ago in the Weston Price article that "(i)n the 1930s, editorials and research refuted the theory of focal infection" referenced to Ingle's Endodontics 5th edition (2002) was unreliable precisely for this reason even using another textbook by the same publisher the exact same year that said the exact opposite thing but many editors could not grasp the idea that Ingle's could be wrong in this one regard and ruled it was reliable even though it was demonstratively inaccurate. Eventually the inaccurate material was removed but NOT for accuracy or reliable reasons. There is NOTHING about accuracy regarding reliably currently in WP:RS.--BruceGrubb (talk) 15:15, 13 November 2011 (UTC)

I am sure you don't mean to suggest that unreliable sources are more often accurate than reliable ones. This leaves me at a loss to what you are actually suggesting.·ʍaunus·snunɐw· 15:21, 13 November 2011 (UTC)

I think Bruce is suggesting that sources do not necessarily need to be accurate to be considered reliable. If so, I agree. Lots of reliable sources contain inaccuracies and errors. Blueboar (talk) 19:00, 13 November 2011 (UTC)

All reliable sources contain inaccuracies and errors - an "untruthes".·ʍaunus·snunɐw· 04:01, 21 November 2011 (UTC)

Actually, I am just pointing out that accuracy is NOT alway- and is a consideration regarding reliability a problem in more controversial or poorly explained topics where inaccurate information is put is just because is from publisher X and writer Y.--BruceGrubb (talk) 09:18, 15 November 2011 (UTC)

As far as I know accuracy is never a consideration regarding reliability. ·ʍaunus·snunɐw· 04:01, 21 November 2011 (UTC)

Actual accuracy (especially in the opinion of a Wikipedia editor) does not matter. A reputation for fact-checking—that is, a systematic effort by the publication's staff to be accurate, even if that desire is imperfectly fulfilled in any given instance, matters a great deal.

I'd have to go look it up, but if memory serves, Bruce's dental sources were materially different. The one basically said "Weston Price's theory was rejected in the 1930s" and the other said something about a current idea that has some similarities, but doesn't make any reference to Price at all, and for all we know is an idea that is materially different from Price's theory in important respects. Thus declaring the new theory to be the same as the old theory would very much be a NOR violation. WhatamIdoing (talk) 04:57, 21 November 2011 (UTC)

This was argued at length in Talk:Weston_Price/Archive_2#Modern_focal_infection_theory_passage and Talk:Weston_Price/Archive_2#Clarification_of_focal_infection_theory and several sources talking about "modern" oral FIT being a revival of the version seen during Price's active research on the matter in the 1920s were produced. Contrary what WhatamIdoing is remembering the 2002 version of Ingel's actually states "In the 1930s, editorials and research refuted the theory of focal infection and called for a return to constructive rather than destructive dental treatment." on Pg 63 while Sol Silverman's Essentials of oral medicine printed the exact same year by the exact same publisher states "The detrimental effect of focal infection on general health has been known for decades" on page 159 and that "the dental profession readily accepted the notion that bad teeth were one of the major sources of infection".

Note the bad teeth reference. Price himself stated "Each dental caries, dental abscess, gingival and alveolar inflammation and necrosis, has been interpreted as essentially infective processes, and hence their extent is essentially a measure of the infection."

Ludwig's angina, septicaemia, brain abscess, and meningitis are all modern version of a dental abscess based FIT something Price himself noted in his self published 1923 work: "These, incidentally, appeared simultaneously with a very severe overload occasioned by the illness and death of her sister, whose fatal illness had been caused by a septicemia resulting from a dental infection which had remained in a chronic condition for many years..."

Furthermore RS sources after 1940 in support of FIT including Grossman were found clear into the 1950s showing the Ingel's reference to be in error and in fact the 2008 version is worded slightly differently: "In the 1930s and 1940s, editorials and research refuted the theory of focal infection and called for a return to constructive rather than destructive dental treatment." The 2009 PDQ Endodontics companion to the 2008 edition gives thanks to Pallasch who makes it clear that the "modern" oral version of FIT is simply a revival of the oral version seen in the 1920s and has essentially the same problems.

The point that keeps getting lost in all this is focal infection was NOT Weston Price's theory. At best all he really did was present evidence others used to justify what was already being done (extractions and surgeries due to FIT were said to be out of control as early as 1918...some five years before Price even published his book) His work on nutrition seems to be of the same stripe--not a pioneer but a follower of ideas prevalent for his time.

I don't blame the editors but the quality of the material they can easily get a hold of. The enormous amount of effort to get a demonstrable inaccurate statement referenced to Stephen Barrett removed (see Wikipedia:Fringe_theories/Noticeboard/Archive_21#Weston_Price_and_Quackwatch and Wikipedia:Reliable_sources/Noticeboard/Archive_79#Is_a_paper_.28possible_blog.29_by_a_psychiatrist_valid_regarding_old_claims_regarding_dentistry.3F for that mess) demonstrates the problem. Price's own words showed Barrett didn't know what he was talking about with regards to Price but it took a ridiculous amount of arguing to get that nonsense removed from the article.--BruceGrubb (talk) 19:19, 22 November 2011 (UTC)

WP:DUCK

Recently BruceGrubb changed WP:DUCK, an essay about Wikipedia internal procedures, to make it apply to article content: [2]. Slightly before those changes BruceGrubb inserted links to WP:DUCK in this essay, so that it reads "An unpublished synthesis or analysis should not be presented for the readers' "benefit" unless it falls under WP:DUCK." This seems to be saying that an unpublished synthesis can be published if the synthesized argument seems obvious to the editor who puts it in. Well, editors who insert synthesis always think it's obvious, don't they? So this change seems to be opening the door to putting SYNTH in articles.

I think this is a bad change in and of itself, but it's even more troubling because BruceGrubb and I are in a content dispute at Christ myth theory over this very issue: he's inserted original analysis of sources that he thinks are conflicting into the article, and in the discussion of that issue at Talk:Christ myth theory#Removal_of_synthesis_tags he's cited this very essay—in other words, he's edited this essay to give himself ammunition in a content dispute. I don't think this is a good way to go about resolving a dispute. For these reasons I'm going to remove the changes. --Akhilleus (talk) 15:07, 18 April 2012 (UTC)

Thanks. Changes to help in a dispute are never a good idea, and this change was a bad one in any case. Dougweller (talk) 15:34, 18 April 2012 (UTC)

• Bruce Grubb has been capmpainging in favor of allowing certain kinds of original research (I guess the kinds that suit him) for a long time now - several time involving unilateral undiscussed changes to core policies. That is not good.·ʍaunus·snunɐw· 15:43, 18 April 2012 (UTC)

I added a new sentence to that essay to further clarify that: "If there was an animal that "looks like a duck, swims like a duck, and quacks like a duck", but zoologists agree that it does not belong into the Anatidae family, then it would not be a duck, period". Cambalachero (talk) 15:56, 18 April 2012 (UTC)

The inherent flaw in this logic is clear to anyone who bothered to READ the Anatidae article--geese and swans are part of the Anatidae family but they are clearly NOT ducks. The WP:DUCK article deals one aspect of with the blatantly obvious. If you are told that 7+7 = 16 and that 7+1 = 10 then clearly per WP:DUCK you are not dealing with base 10 math but base 8 math.--BruceGrubb (talk) 07:12, 20 April 2012 (UTC)

The logic does not say that all Anatidae are ducks, but that all ducks are Anatidae. So, if some animal is not an Anatidae, it can not be a duck. Simple logic. If (Duck) -> (Anatidae) (if it is a duck, then it's an Anatidae), then it follows that NOT (Anatidae) -> NOT (duck) (if it's not an Anatidae, then it's not a duck) Cambalachero (talk) 12:01, 20 April 2012 (UTC)

Is this synthesis?

I've started a new discussion here to ask about whether a particular edit constitutes synthesis. Could interested, experienced editors offer their viewpoint? Nightscream (talk) 23:24, 12 November 2012 (UTC)

Accurately contextualizing quotations

I added a section, Wikipedia:These are not original research § Accurately contextualizing quotations. There are probably simpler examples, or we could use a hypothetical, but I liked this one because it illustrates not only how the surrounding context of the article shows that the quotation needs to be clarified with a [bracketed insertion], but also that misinterpreting it as applying to a broader context would very radically change its meaning (i.e. there's a danger of OR in the opposite direction), and further illustrates why we could want to use it despite it needing a contextualization, since it's a secondary source interpreting a primary source for us.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  22:52, 29 May 2015 (UTC)

Sagan quote?

I was reading the following:

(section title) Removing incorrect claims and pointing out errors

Further information: Wikipedia:Inaccuracy

"Experts are human, and can publish statements that are contradicted by known facts, or otherwise erroneous."

And then, there was a reference to Carl Sagan's Demon-Haunted World. If the above is a quote, we should note it as such. If its a paraphrasing, its unclear what the original intent of Sagan's comment was. Can someone explain? - Jack Sebastian (talk) 02:51, 7 November 2016 (UTC)

Cherry-picking Sources?

Hey!

I'm doing an article on a train accident, a first for me. I have pulled up several newspaper articles days within the incident. They state a lot of inconsistent numbers. The NTSB report states one number, and another newspaper article 8 years after the incident reports another number inconsistent with the NTSB report!!

Is it ok if I:

• use the NTSB numbers and information, and still use these sources for the gaps in-between? or...
• report on the wild discrepancies on the numbers over the years (which I think might be giving the situation undue weight)? or...
• report the "first estimates" as they are, and report the NTSB numbers as the final counts?

Thanks in advance! I haven't edited on Wikipedia in like TEN YEARS! lol. I ^help_dןǝɥ I 04:22, 24 December 2021 (UTC)

Welcome back, @IHelpWhenICan. I think you might get more useful assistance at Wikipedia:WikiProject Trains. In the meantime, I'd generally suggest that you follow "official" (e.g., NTSB) numbers or the most recent numbers. Sometimes it's useful to provide more than one number ("First estimates suggested X, but the NTSB report said Y"). WhatamIdoing (talk) 00:34, 9 January 2022 (UTC)

New section

I've added a new section at the top, about the definition of what OR actually is. If it doesn't fit the definition, then it's just not OR, full stop. I hope that will eventually reduce the number of people who claim OR on the basis of a single already-cited source. (Of course, Wikipedia:Nobody reads the directions, so maybe it won't actually help.) WhatamIdoing (talk) 00:36, 9 January 2022 (UTC)

Old section

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

---

@Huggums537: its actually you who needs to get consensus to include disputed content. Please do so. Horse Eye's Back (talk) 14:33, 20 March 2023 (UTC)

This page doesn't seem to have that many watchers, and since we are both claiming it is the other one who needs to get consensus, we are going to need the help of other editors to decide. I happened to notice on my watchlist that you are involved in a related discussion, and that seems like a rich source of editors interested in this topic, so I will post a notice there about this discussion to see if we can get the help we need for consensus. Thanks. Huggums537 (talk) 21:12, 20 March 2023 (UTC)

Huh, I though that it was always the person who wanted to keep/add the content who needed to get consensus. WP:ONUS, WP:BURDEN, etc... No? If you intend to duplicate the discussion that is happening there here that would be inappropriate. Horse Eye's Back (talk) 21:13, 20 March 2023 (UTC)

ONUS and BURDEN apply to articles, not to stuff in Wikipedia: space. Jc3s5h (talk) 21:21, 20 March 2023 (UTC)

And in your opinion what does apply here is? What rules govern dispute resolution in wiki space? I'm happy to argue against a policy based argument for the text, but nobody has offered one and the original author was blocked soon after making it for just this sort of thing (the addition appears to have been the result of a wish to win a content dispute). Horse Eye's Back (talk) 21:27, 20 March 2023 (UTC)

I think the fact he was blocked and administration never decided to remove his "disruptive" edit in over a decade says everything we need to know. Huggums537 (talk) 21:34, 20 March 2023 (UTC)

There is no formal process to scrub edits after a block (nor should there be). Disruptive edits are often addressed slowly, as is happening here. This page isn't maintained by "administration" it doesn't even necessarily reflect community consensus. Horse Eye's Back (talk) 21:34, 20 March 2023 (UTC)

I've reverted back to the version before you additions. This is the same situation as you additions to Wikipedia talk:Verifiability/FAQ that were without any backing. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 21:55, 20 March 2023 (UTC)

I see I'm in fact mistaken about the scope of this thread, but beleive it should cover the larger part that I've reverted. I've also placed a notice at Wikipedia talk:Verifiability about it. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 22:00, 20 March 2023 (UTC)

I support the removal of WhatamIdoing's addition as well. Horse Eye's Back (talk) 22:03, 20 March 2023 (UTC)

I've undone the removal for the moment, as I've mixed up WhatamIdoing and Huggums537 actions (for which I apologise to both). I'll get myself straight and start a new thread specifically about WhatamIdoing additions tomorrow. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 22:06, 20 March 2023 (UTC)

(edit conflict) Er, it might be helpful if someone could explain to the community which has been called in just what the dispute is about. I believe that is traditional. Because, having read this section twice, I have no clue. Nor any inkling as to why I should care about the wording (I am guessing that it is the wording which is in dispute?) of a non-policy, non-guidance, ignore it if you want to, essay. Gog the Mild (talk) 22:10, 20 March 2023 (UTC)

The short version is there was a paragraph of text in this essay that has been in place for several years (10 years has been tossed around). This essay is being cited in the !votes at a currently active discussion. An editor deleted this paragraph after participating in said discussion on the grounds that it was added by a now banned user. Another editor restored it on the grounds that something that has been in the essay for years should not just be deleted without discussion. So now given the circumstances comes the debate over which action is appropriate and which isn't.Dave (talk) Dave (talk) 22:58, 20 March 2023 (UTC)

Given that I now know this page was cited in that discussion, I will add to my rationale for restoring that the other thread relies upon having this page in the original form to make an informed decision, and it was not appropriate for the other editor to alter the page while the discussion was going on at the same time the page is being cited without consulting the community. Huggums537 (talk) 23:12, 20 March 2023 (UTC)

Map of the Loire River in France

I have a general query. We allow foreign-language sources. This page says so. Assume the map here is contained in an atlas and considered a Reliable Source. Someone could use that map to cite the statement: "The Loire River flows through Orleans, Tours and Nantes before emptying into the Atlantic Ocean." That is a straight-forward description based on the map.

Now assume instead I used a source in French that states:

La Loire prend sa source sur le versant sud du Mont Gerbier-de-Jonc au sud-est du Massif central et coule vers le nord jusqu'à Orléans avant de tourner vers l'ouest à travers Tours, Angers et Nantes et de se jeter dans son estuaire près de Saint-Nazaire.

As someone who understands French, I can translate that statement to:

The Loire originates on the southern slope of Mont Gerbier-de-Jonc southeast of the Massif Central and flows north to Orléans before turning westward through Tours, Angers, and Nantes and emptying into its estuary near Saint-Nazaire.

which then backs the sample statement above. Perhaps I could find a similar source in English, but I'm not required to do so. This theoretical French source would be accepted.

So my query: why are we allowed to translate non-English sources into English, but "translating" this map in the hypothetical atlas into English is problematic for straight-forward facts illustrated therein? That's all the disputed passage says to me. Imzadi 1979 → 22:44, 20 March 2023 (UTC)

There is a disputed passage? Please share it. Gog the Mild (talk) 22:49, 20 March 2023 (UTC)

@Gog the Mild, I restored a removed passage that was on here for about a decade, which was removed by the other editor on the basis of "disruptive" since the editor who added it 10 years ago was blocked for disruptive editing. Huggums537 (talk) 22:56, 20 March 2023 (UTC)

Do you understand that the other thread is about whether the map is RS at all? You may be ahead of yourself. Sammy D III (talk) 23:02, 20 March 2023 (UTC)

As I stated in my request for comments, I never read the discussion, but I do realize what it is about, and given the fact that it was recently shared in this thread that this very page was cited in that discussion, believe it was the other editor who removed the text who got ahead of themselves. Huggums537 (talk) 23:07, 20 March 2023 (UTC)

OK. I just wasn't sure that you "realize"ed. Have a nice one. Sammy D III (talk) 00:00, 21 March 2023 (UTC)

You too, but I also just realized I didn't give you a completely correct answer since jumping back and forth between conversations got me confused, and by the time you asked me the original question I forgot I actually did go ahead and read most of the first section in the other discussion, and cast a support vote for it. The resulting answer which I first gave to you that I DO in fact realize what the discussion was about still remains intact so I'm not entirely sure it matters to you whether I was mistaken about if I had actually read any of it or not, but I like to correct myself when I realize I've made an error anyway. Thanks. Huggums537 (talk) 03:31, 21 March 2023 (UTC)

On the one hand, if the discussion determines that there is no consensus to include the bullet in question, then it should be removed at when the discussion reaches such a conclusion (either no consensus or consensus to exclude).

However, as long-term stable content on a policy page, it should be retained in its current form pending the result of the present discussion. At least, this seems to be the most widely supported practice, e.g., in recent discussions concerning WP:ONUS. Newimpartial (talk) 23:36, 20 March 2023 (UTC)

The text that was removed by Hors Eye's Back on 19 March 2023 was

* Source information does not need to be in text form—any form of information, such as maps, may be used to provide source information. Interpretation of such media is not original research provided that it is done in a routine manner observing any limitations usually associated with the medium concerned, and such interpretations are readily verifiable by anybody who has access to the same source.

It was in the "Translation and contextualizing" section. I support using information from reliable sources that is in a form other than text, including maps and graphs. But I have problems with the disputed text. It isn't exactly clear what "source information" is.

Also, using the word "interpret" is used at WP:PRIMARY:

Any interpretation of primary source material requires a reliable secondary source for that interpretation.

It's likely that some editors will decide the essay and the policy contradict each other. I feel the word "interpretation" in WP:PRIMARY is unfortunate because it fails to distinguish the allowable process of drawing information from a primary source in exactly the way the author intended, versus whatever WP:PRIMARY is really getting at. Jc3s5h (talk) 23:06, 20 March 2023 (UTC)

Agreed. If it were "Summaries of such media are not original research provided that it is done in a routine manner observing any limitations usually associated with the medium concerned, and such summaries are readily verifiable by anybody who has access to the same source." I don't think I'd have a problem with it. But for interpretation there would really need to be more explanation. An example of summary vs interpretation would be for example the caption for an image of room full of manure, to write that it depicted a room full of manure would be a summary to write that it depicted a room which smelled bad would be an interpretation. Horse Eye's Back (talk) 23:15, 20 March 2023 (UTC)

Either way, I think the text should remain at least until the discussion is concluded at the related thread since the page is being cited in that discussion. Huggums537 (talk) 23:18, 20 March 2023 (UTC)

And what basis in policy and/or guideline does that opinion have? I've never seen someone suggest something like that before. Horse Eye's Back (talk) 23:19, 20 March 2023 (UTC)

I guess I don't really care whether it is in the essay right now or not. The fact that it was there for over 11 years is unchangeable, and the RFC will moot it. --Rschen7754 00:07, 21 March 2023 (UTC)

This should remain at least until the current RFC at VPP is over. It's apparent from that RFC that whatever the result there will need to be further discussions about what the level of what is acceptable, but the outcome of that RFC will determine where to start that discussion. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 14:33, 21 March 2023 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

A comment

I wrote the original draft of this essay back in 2006, but have stayed away from it for at least 15 years. Coming here now I'm amazed how it's changed, & how one hope I had for the essay seems to have been lost: "This essay is offered as a contribution the ongoing discussion -- and as an attempt to help avoid Wikilawyering. For if we have a consensus on what this prohibition is meant to cover, as well as what it was not meant to cover, then we do not need worry about quibbles over the precise wording", even though I had cited an authority in a footnote for my statement. I guess the Wikilawyers will prevail in every instance, no matter our hopes. -- llywrch (talk) 23:21, 19 April 2023 (UTC)