Misplaced Pages

Revision as of 10:29, 6 August 2012

The project page associated with this talk page is an official policy on Misplaced Pages. Policies have wide acceptance among editors and are considered a standard for all users to follow. Please review policy editing recommendations before making any substantive change to this page. Always remember to keep cool when editing, and don't panic.

Template:NORtalk

This page has been cited as a source by a notable professional or academic publication: Stvilia, B. et al. Information Quality Discussions in Misplaced Pages. University of Illinois U-C.

Spoken Misplaced Pages

This page is within the scope of WikiProject Spoken Misplaced Pages, a collaborative effort to improve the coverage of articles that are spoken on Misplaced Pages. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.Spoken WikipediaWikipedia:WikiProject Spoken WikipediaTemplate:WikiProject Spoken WikipediaSpoken Misplaced Pages

Archives

Index 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 21, 22, 23, 24, 25, 26, 27, 28, 29, 30 31, 32, 33, 34, 35, 36, 37, 38, 39, 40 41, 42, 43, 44, 45, 46, 47, 48, 49, 50 51, 52, 53, 54, 55, 56, 57, 58, 59, 60 61, 62, 63, 64

Rewrite 2004-2005: Draft • Talk "Primary v. secondary sources" subpages "Change needed": Proposal • Talk "Various examples": Examples • Talk "Transclusion example": Example • Talk List all subpages

This page has archives. Sections older than 60 days may be automatically archived by Lowercase sigmabot III when more than 2 sections are present.

• WT:OR
• WT:NOR

Original Research from reliable primary sources for exclusion, not inclusion

Please discuss -

“When a secondary source is in conflict with a reliable primary source, then the reliable primary sources can be used to exclude in information, but not include it.”

An example is when a line stating living person’s age has a reliable mainstream newspaper as the source, but the very reliable primary source of government birth records indicates the newspaper made an error, then the information about age should be removed from the article based on the reliable primary source, but not included from the primary source which is original research. PPdd (talk) 13:47, 4 April 2012 (UTC)

When reliable sources conflict, the simplest solution is for the article to say "source A says X, whereas source B says Y". Deciding whether A or B is more likely to be wrong cannot be reduced to a simple rule like one being primary and the other secondary. jnestorius 16:50, 4 April 2012 (UTC)

PPdd your assumption that government records (like birth certificates) are inherently more accurate than newspapers is flawed. This is not always the case. Let me present an example: If you were writing an article about me (assuming I was notable), and you looked at my birth certificate, you would read that my mother was born in Manilla, RI (ie in Rhode Island)... in fact, she was born in Manilla, PI (ie in the Philippine Islands). What obviously occurred was that somewhere in the process of filling out my birth certificate, someone swapped an R for a P (easy enough to do). Thus, my birth certificate contains erroneous information. Now, if I were to be interviewed by the newspaper, and they asked me where my mother was born, I would say "in the Philippines". The newspaper account would be accurate.

Of course, there is no way for you (the person writing an article for Misplaced Pages) to know that the newspaper is accurate and my birth certificate is not. All you know is that there is a discrepancy. So you have two options a) Look for for further sources that will clarify the facts, or b) present the discrepancy by mentioning what both sources say. Blueboar (talk) 18:16, 4 April 2012 (UTC)

To Blueboar, to make sure I am understanding your explanation correctly; in the specific example of confusion over a living person's age or birthplace (or something similar), an outright statement by the living person in question, intentionally given to debunk any false or misleading statements about the issue in dispute (DOB, birthplace, etc.), would trump any inaccurate legal documentation, incorrectly sourced media statements, deliberately-made slanderous or libelous statements, or any other source that conflicts with their own from-the-horse's-mouth statement? (Assuming, of course, that the Misplaced Pages author/editor was the person this statement was made to) Blozier2006 (talk) 22:31, 19 April 2012 (UTC)

Not necessarily. I certainly tend to give a lot of weight to published statements by the person in question, but to know comparative weight between sources (ie which source outweighs or "trumps" the other) I would have to look at the specifics of each case separately. Most of the time, neither source "trumps" the other... so we present what both have to say. In my example, I would say something like: "Blueboar's birth certificate gives his Mother's birth place as Manila, RI (Road Island) <cite birth cirtificate>. In an interview given in the Boston Tribune, Blueboar claimed that his birth certificate contains a clerical error and that his Mother was actually born in Manila, PI (Philippine Islands) <cite Tribune interview>. Blueboar (talk) 23:25, 19 April 2012 (UTC)

OK, thank you for the clarification. Blozier2006 (talk) 23:50, 19 April 2012 (UTC)

And just to complicate things... we also need to keep in mind that there might be other sources that confirm (or debunk) either source. This too impacts how much weight to give each source. Blueboar (talk) 00:21, 20 April 2012 (UTC)

A birth certificate is not a reliable source unless it has been published, as for example the American president's birth certificate. But that is rare. If a newspaper adds ten years to your age, you should get the paper to issue a correction, rather than to correct it here. The typical conflict between a primary and secondary source comes when a secondary source misrepresents a primary source, in which case we could just use the primary source. For example, a newspaper reported that x said y, while a published transcript says that he said z. TFD (talk) 00:38, 20 April 2012 (UTC)

A birth certificate is published if it is available to any member of the public who wishes to pay the fee; this is the case in some US states. A birth certificate that is only available to certain select persons is not published. Jc3s5h (talk) 00:49, 20 April 2012 (UTC)

Blueboar is correct that birth certificates often contain errors, but newspapers contain errors vastly more often. So, if we prefer Misplaced Pages to be right more often than wrong, while maintaining our ban on original research, we ought to allow birth certificates that are publicly available and not allow those which are not publicly available. Of course, if a secondary source has examined the birth certificate and declared it mistaken, we have to go with that. Zero 14:31, 28 April 2012 (UTC)

We do allow birth certificates and other such publicly available documents ... however, we should also remember that such documents are primary sources, and as such are subject to the limitations and cautions of all primary sources. They can be used, but should be used with caution and care. Blueboar (talk) 14:47, 28 April 2012 (UTC)

I believe the point is whether a source explicitly states what is claimed by wikipedia. The problem with primary sources is that they usually need to be synthesised to be able to draw useful conclusions. If the primary source says: the Mona Lisa has black hair and there is no discrepancy between sources, than you can state that she has black hair. If however one source says grass is green when it's alive and a second source says that it's brown when it's dead than you may not conclude that it turns brown when it dies unless you reference a secondary source which makes that conclusion for you. That being said: a birth certificate is often not a primary source, I don't know how it works in the USA, but if the data has to be reported to an official by a parent or a doctor, it's a secondary source (the primary source being the report from the doctor or the parent). PinkShinyRose (talk) 00:10, 27 June 2012 (UTC)

Which is why we don't allow synthetic OR from primary sources—encyclopaedists lack the capacity to make expert and original judgements about primary material. If you want to publish biographic compendiums, then the free online publishing industry is that way (but our reliable sourcing standards are high). Fifelfoo (talk) 00:46, 27 June 2012 (UTC)

A birth certificate is not a secondary source. It's not a matter of counting up links in the chain. WhatamIdoing (talk) 03:06, 27 June 2012 (UTC)

Exceptions for primary sources

The policy states that "Do not base an entire article on primary sources, and be cautious about basing large passages on them." Can an article be majorly based on primary sources which have received coverage in tons of news reports from highly reputed newspapers i.e. do reports in loads of reliable secondary sources over-rule the fact that the site is a primary source and can the policy be flexed for that reason? Secret of success (talk) 12:06, 20 May 2012 (UTC)

Could you give us more specifics? A lot depends on what the article is about... For example, the "plot summary" section of an article about a book, would be appropriately based entirely on the book itself... but other sections of the article (such as a section on how the book was received) would not. Blueboar (talk) 14:00, 20 May 2012 (UTC)

The policy doesn't really distinguish between primary sources used a lot secondary and those which are used less. But if a primary source is evaluated/analyzed in some secondary source, there is usually no reason to cite primary source at all, but you simply use the secondary source instead.--Kmhkmh (talk) 22:09, 20 May 2012 (UTC)

I was specifically referring to box office figures for films, which are without doubt, one of the most sensitive issues across Misplaced Pages. I have seen articles like this being sourced mostly to primary sources. It has had two AfD's in the past and both of them have resulted in a "keep". Now, I do not want to discuss that specifically, my comment above was more general.

If content from a primary source is put up in a secondary source with attribution and some amount of analysis, does it mean that the secondary source endorses the content? I don't think that is the case, right. Secret of success (talk) 09:13, 21 May 2012 (UTC)

The mere use of a primary source does not necessarily mean the secondary source "approves it", you need to read secondary source to see what it does with the primary source. I was saying if you have secondary sources using a primary source for specific information, then usually you get that specific information you want from the secondary source anyway, hence you don't the primary source anymore and arguing whether you could use or not is a bit of a moot point. As far as box office figures are concerned, I see no problem with citing (standard) primary sources for it, but they should be clearly attributed. However these hinges a bit on the question as how reliable you assess a primary source in the first place and you could consider its general use in secondary sources as one factor (of many) to assess the reliability of the primary source. If BoxOffice.com figures are widely used in media and literature, than you consider that as hint that their data is generally accepted/considered reliable. -Kmhkmh (talk) 09:40, 21 May 2012 (UTC)

Fringe, established facts and synthesis

The WP guideline on synthesis may have a major weakness when it comes to attacks by fringe authors on established facts. Image the following: some recent author A makes the obviously ludicrous assertion that Tower Bridge was constructed by 15th century Chinese and he is cited so in the WP article. No other reliable author ever cared to address and refute this statement, hence it remains undisputed' so to speak. Now some editor takes an older book B which states that the bridge was built in the 19th century by the British and adds this statement as a rebuttal. He is, however, reverted on the grounds that this would be synthesis because B did not refer to A explicitly. Consequently, the statement of B is removed as refutation, even though it is an established fact, while A's statement remains in the text as undisputed. Was this right? This case does not seem to have been covered by WP:SYN properly, a kind of loop-hole which favours outlandish and more recent claims. Gun Powder Ma (talk) 15:25, 6 June 2012 (UTC)

That isn't synthesis. That is a source simply being wrong and there being a lack of other sources which is a different problem. An example of synthesis would be 'The Chinese made the earliest iron bridges. Tower Bridge is one of the most famous iron bridges made before the twentieth century'. Dmcq (talk) 17:04, 6 June 2012 (UTC)

So even though the source isn't discussing Menzies, we can use it? At Where Troy Once Stood this edit of mine was, I think correctly, removed. Can I reinstate it? The actual edit that Gun Power Ma is discussing is . Dougweller (talk) 18:11, 6 June 2012 (UTC)

Who are you replying to? If you could state what you see as the problem in a straightforward way it would help. Personally I can't see why anyone would worry about some fantasy pseudohistory like those but if people have gone out of their way to write refutations I suppose they should be in too. Dmcq (talk) 18:27, 6 June 2012 (UTC)

Oh no no MEnzies stuff please:-).--Kmhkmh (talk) 18:30, 6 June 2012 (UTC)

The point is that none of these sources were written as refutations, they are simply about the same subject. The question seems to be should our policy allow us to use them, even though in these two examples they weren't written in response to the authors' books. Dougweller (talk) 20:27, 6 June 2012 (UTC)

Synthesis is using two or more sources to advance a position not advanced by any of the sources. If all the sources are being considered to support an article, there is no problem with ignoring a fringe source and not mentioning it in the article. It's a different story if the article is about the fringe source. Setting out "facts" purported by the fringe source and listing sources that contradict the fringe source could be viewed as a Misplaced Pages editor creating a novel review of the fringe source. If no reliable sources can be found to review the fringe source, the article about the fringe source should probably be deleted. Jc3s5h (talk) 20:48, 6 June 2012 (UTC)

SYNTH would be something like "Source A says the Tower Bridge was built by Chinese laborers. Source B says it was built by the British. Therefore (here's the SYNTH part), China and the UK are the same place." When different views exist indifferent sources, simply stating what the different views are is not a SYNTH violation, because yu're not adding up the views to get something entirely new that can't be found in any of the sources. WhatamIdoing (talk) 14:16, 7 June 2012 (UTC)

You don't have to even say the conclusion, just imply it as in my example near the start implying the Chinese built Tower Bridge using two perfectly okay and true statements. Dmcq (talk) 14:29, 7 June 2012 (UTC)

Does it matter that these are book reviews? Dougweller (talk) 16:40, 7 June 2012 (UTC)

If they are published in a reliable source like a newspaper of journal I cant see why not and they'd be good secondary sources. Dmcq (talk) 22:37, 7 June 2012 (UTC)

No, it is the articles that are book reviews. The authors of the sources may not even have ever heard of the books in question and they certainly aren't commenting on them. Dougweller (talk) 13:01, 8 June 2012 (UTC)

Sounds very strange, but if a book review has been commented on in a number of other ways in some notable way then I guess so. I suppose someone could write a notable book review if they were very good authors themselves or particularly biting critics who attracted attention or something like that. Dmcq (talk) 13:13, 8 June 2012 (UTC)

Let me try again. There are no book review involved here. We are talking about Misplaced Pages articles about books, or sections of a biography that discuss a book, where sources are being used that do not mention the book or the author or even the subject of the book and are being used to refute the book. As I understand it, this is not acceptable according to our NOR policy. Dougweller (talk) 14:47, 9 June 2012 (UTC)

Perhaps you should just point at the discussion on the point that concerns you. Dmcq (talk) 20:30, 9 June 2012 (UTC)

I think my answer to the original question is that author B should not be mentioned. However the article should not state what A says as fact, it should only be stated as what A asserted. It is not up to us to write refutations but we don't have to write rubbish up as truth. Dmcq (talk) 20:37, 9 June 2012 (UTC)

Thanks, I agree entirely. Dougweller (talk) 06:02, 10 June 2012 (UTC)

Well... I don't believe it's necessary for the "refuting" source to mention the book by name. Let me give an example: some <insert disparaging term here> in Hawaii have come up with this silly theory that breast cancer is caused by wearing bras. They wrote a book about it. Every single mainstream source that addresses the question says that their idea is wrong. Almost none of those sources mention their names or the title of their book. They just say "Hey, if you heard this rumor that clothing causes breast cancer, it's wrong". But we use those sources to present information about their silly idea, because there's really no other way to meet NPOV and FRINGE requirements to present a disproven, tiny-minority position as being a silly, disproven idea.

So I'd say that if New Book says that the Tower Bridge was built by 15th century Chinese, and mainstream experts have a very different idea, then you can safely present the mainstream position as being the mainstream view. You can't say "Alice Expert rejects this book's claim", but you can say "Mainstream expert opinion holds that the bridge was built by British laborers" (or whatever). WhatamIdoing (talk) 19:30, 16 June 2012 (UTC)

Academic question: Does recognizing something count as OR?

There's a discussion going on at Talk:Blackwater (Game of Thrones) about whether or not certain types of content from primary sources count as original research. I refer to the discussion as academic because there does not seem to be any disagreement (at the moment) about what the text of the article should read. In these cases, the episodes themselves are considered primary sources and off-Misplaced Pages articles about the episodes are considered secondary sources.

If a work of fiction has a character say, "To be or not to be. That is the question," is it original research for the Wikiarticle to say, "Character Bob Smith quoted part of Shakespeare's Hamlet"? If so, which part of WP:OR covers this?

If a work of fiction has a character hum or sing a song, is it original research for the Wikiarticle to say, "Foghorn Leghorn can be heard humming 'Camptown Races' in the episode 'Fearsome Foghorn'"?

Does it make any difference if the content is well known or obscure? The statement that set off the debate in question was, "Tyrion Lannister can be heard whistling 'The Rains of Castamere' in a previous episode." (This statement was later backed up with a secondary source.)

There seems to be an unspoken consensus that no secondary source is needed to identify characters by their faces ("Bob Smith appears in a scene with John Jones") or to identify words by sound ("Bob Smith says 'Hello, I'm Bobby' to John Jones in the second episode") or to describe events ("Bob Smith can be seen to smack John Jones in the shoulder in the third episode"), and most plot summary sections consist primarily of these kinds of statements. If identifying melodies is different in some way, then which part of WP:OR covers this? Darkfrog24 (talk) 05:45, 10 June 2012 (UTC)

It isn't OR. It is a simple descriptive statement about something that occurs in the work of fiction that a reader can verify by looking at the work of fiction itself.

That said, the fact that the statement isn't OR does not mean we necessarily must (or should) include it. A good plot summary will not include every tiny trivial action made by every character, it will only mention the important ones. So, the next question is... is the fact that the character sings that particular song at that particular point in the plot really worth mentioning in our article? Is this action central to the plot, or is it trivial? Blueboar (talk) 11:43, 10 June 2012 (UTC)

I agree that just because it's not OR doesn't mean it merits inclusion; that's a separate issue. The article in question has a small section about music. The song "Rains of Castamere" features rather heavily in the episode, and the article says where the lyrics came from, who wrote the song, and who performs it in the credits. I figured it was relevant to mention that this was not the first time that it appeared in the series. Feel free to see for yourself. The section is quite short.

The guy I've been talking to was pretty adamant that listening to the melody and concluding that it was the same song as another string of pitches was OR. Darkfrog24 (talk) 14:10, 10 June 2012 (UTC)

Doesn't WP:MOSFICT#Plot summaries describe about plots and refer to some essay and help about them? I think it describes the general consensus about this area saying what people agree is about okay as a straight factual description. Dmcq (talk) 13:31, 10 June 2012 (UTC)

New edits to WP:CALC

A recent edit was made in a good faith attempt to clarify WP:CALC (see this diff). I have reverted because I think the edit needs some discussion, especially the idea that pulling numbers from multiple sources is in all cases an improper synthesis - and therefor the numbers need to come from a single source. I am not sure this idea is completely accurate. Suppose I want to calculate the basic land area of the State of Virgina is 1850. We know that, at that time, Virginia included what is today West Virgina, but otherwise its boarders were the same as today. Now, suppose I have a source that gives me the modern area of Virginia, and another source that gives me the area of modern West Virginia... simple addition will give me the area of 1850 Virgina. This is exactly the sort of basic calculation that CALC was intended to allow. However, the edit would disallow it. Blueboar (talk) 11:48, 16 June 2012 (UTC)

I think that as long as its clear what the units/scale are and how absolute/unquestionable they are. Take an example where I attempt to add 1 million homes with Tivo boxes and 2 million homes with DVRs (from two different sources) and attempt to say that a total of 3 million homes have some type of digital recorder. The problem is that Tivo is a type of DVR and unless source 2 is clear that they didn't count Tivo in their DVR #s, then that 3 million may be double counting. --MASEM (t) 12:51, 16 June 2012 (UTC)

My guess is this is to counter the business where editors agree a calculation is okay in consensus but the calculation is not some standard one for the application. There was quite a row over various editor own calculations in the articles on usage share of web browsers and operating systems putting in a median of the various figures published by outside agencies. The articles no longer have the median figures they calculated so it doesn't look like the change is necessary for that purpose, and I'm not altogether sure the ones pushing the calculation in would have agreed with the interpretation even so anyway as they were arguing that what they did was useful and a fairly standard technique. Dmcq (talk) 12:58, 16 June 2012 (UTC)

The edit would often disallow presenting material in the same units of measure so that numbers in the article could easily be compared. For example, if an article discussed energy usage, home-heating sources might give energy usage in therms and barrels of oil, and electricity-related sources might use kilowatt hours. To make these diverse units comparable, the article might use megajoules. The conversion factors might be found in a different source from the sources that give the energy usage figures. So under the proposed change the unit conversions would be disallowed. Jc3s5h (talk) 13:40, 16 June 2012 (UTC)

I think we all agree that it certainly is possible to create an improper calculation by taking numbers from multiple sources... I hope that we all agree that it is also possible to create an absolutely appropriate calculation by taking numbers from multiple sources. My concern about the edit was that it went too far... phrasing what probably should be a "sometimes" caution as if it were an "always" rule. WP:CALC is one of those things where we simply can not make an "always" rule ... Ultimately, we have to examine each calculation on a case-by-case basis and determine whether that specific calculation is proper or improper. Blueboar (talk) 13:52, 16 June 2012 (UTC)

I would go along with Blueboar. In essence, the paragraph that was reverted didn't add anything to the article. A note describing the pitfalls of using numeric data from multiple sources might be more appropriate, but as an essay, not as part of this partiuclar article. Martinvl (talk) 14:23, 16 June 2012 (UTC)

I am the one who made the bold edit. It was an attempt to avoid situations like the one Dmcq was referring to. I do not agree that the edit would in any way disallow presenting material in the same units of measure as other sources. Converting units of measure of one source is clearly just a single source calculation and does not involve multiple sources, even if the goal of the calculation is to converts numbers from multiple sources into comparable units. I agree that there are some calculations which are relevant across multiple sources. But if they are relevant, surely the editors should be able to find a source which supports that calculation. That was the reason why I put in the phrase about a source directly supporting the calculation involving the sources. Useerup (talk) 16:34, 16 June 2012 (UTC)

Well, let me give you an example, and you tell me whether this is improper OR in your mind:

• Source 1 says that 100,000 white Americans will get Scary Disease this year.
• Source 2 says that 100,000 non-white Americans will get Scary Disease this year.
• I add it up and say that 200,000 Americans will get Scary Disease this year.

Is that okay in your mind? WhatamIdoing (talk) 19:34, 16 June 2012 (UTC)

The proposed change disallows unit conversions unless the conversion factor is contained in the same source as the unit to be converted. I reject Useerup's claim to the contrary. Jc3s5h (talk) 20:03, 16 June 2012 (UTC)

@WhatamIdoing: Yes, that is OR. Even though you used perfectly complementary concepts (white and non-whites) it is not at all clear that the sources could be used that way. Rarely are multiple sources directly comparable that way. The sources could use different standards for accepting a diagnose or use different standards for "Americans" (native, immigrants, natural born or naturalized?, North Americans or just US citizens). In those situations you do *not* add up the numbers, you report on each of the sources. Useerup (talk) 01:01, 17 June 2012 (UTC)

Would you feel the same if the example was "males" and "females" or "adults" and "children" rather than race? What if the sources were actually papers published by the same people, just split up so they could get two publications on their CVs?

Also, do you think that the typical editor would object to my example? WhatamIdoing (talk) 04:20, 17 June 2012 (UTC)

If the papers were published by the same people I would assume that they would reference one paper from the other. That would actually open the door for combining over multiple sources. If the 1st paper was referenced from the 2nd paper in such a way as to compare them or include the conclusions it opens the door for calculations. But think about the opposite: What if two papers were written by the same people, but did not reference each other. Would you then be comfortable drawing conclusions by calculating numbers across the papers, given that it would have been obvious to the authors? Useerup (talk) 06:56, 17 June 2012 (UTC)

Yes, I'd feel perfectly comfortable combining these sources. I'd feel comfortable doing this even if the sources were written by different people. I am, for example, happy to take three solid sources on epidemiology, each of which provide an incidence for one of the three countries in North America, and do a simple addition to produce a statement about "The incidence in North America" rather than "The incidence in Mexico... in the US... in Canada".

You have not answered my second question: Do you think that the typical Misplaced Pages editor would object to adding up such numbers? WhatamIdoing (talk) 16:45, 25 June 2012 (UTC)

@Jc3s5h: Can you propose a change which would in your opinion allow unit conversions? I certainly agree that unit conversions should be considered allowed under the policy; they are (usually) non-controversial and straightforward. Useerup (talk) 01:01, 17 June 2012 (UTC)

I think the policy was fine before the edit. When doing calculations involving different sources (or even different pages within the same source) it is important to make sure one is comparing apples to apples, but I don't think there is any concise formulation that will disallow the incorrect calculations and allow the correct calculation. But a service that an encyclopedia should perform for its readers is to collect information from diverse specialized sources and present an overview that is easier to quickly understand, and this requires activities such as unit conversion and other calculations so that data from diverse sources is more readily comparable. Of course there will be times when the exact nature of the data in various sources is not clearly enough stated to decide if data may be combined, in which case some other approach will be needed. But I think this should be looked at on a case-by-case basis on each article talk page. Jc3s5h (talk) 12:55, 17 June 2012 (UTC)

I think that the bold changes that Useerup made were absolutely appropriate. Those edits are just an explicit restatement of the principles in WP:SYNTH - specifically "Do not combine material from multiple sources to reach or imply a conclusion not explicitly stated by any of the sources". The principles in WP:CALC are extremely useful because they empower users to restate quantitative claims made by a source into the form demanded by the article. The use of routine calculation is analogous to indirect quotation - you are simply restating verifiable claims in the matter most appropriate to the text. However the principles of WP:SYNTH should always be observed to avoid novel claims not made by any single source; and calculations made with data from multiple sources to generate new data not directly attributable to any given source is clearly such a violation. I think, as is pointed out in the discussion above, it certainly is possible to correctly do calculations from multiple sources to reach a true conclusion; however whats at issue is not whether the calculations are correct or the whether the new number is true, its whether it is verifiable and directly attributable to a reliable source. Solid State Survivor (talk) 07:47, 26 June 2012 (UTC)

You then run into the problems that exist with the present text of the verifiability policy page that are as of yet not settled. One of the problems is that you can have verifiable facts that in the relevant field are entirely uncontroversial to state, yet they cannot be attributed to a single source. E.g., in physics and chemistry, it is routine to combine information from sources like reference books to do computations. Then if e.g. we have some list of physical properties of materials here and we want to present that in some way, we should be free to do the relevant computations to extract the correct figures. As long as what we do doesn't deviate from how every professional in the field uses the sources, there shouldn't be a problem here. Count Iblis (talk) 16:21, 26 June 2012 (UTC)

Is simple logic a SYNTH ?

1. If one reliable source say A:X->Y and an other reliable source say B:Y->Z, is the obvious conclusion C:X->Y->Z or C:X->Z a synth?

2. If one reliable source say A:Y happen when X is happen, and an other reliable source say B:Z happen when Y happen, is the obvious conclusion C:Z happen when X happen a synth?

I saw "SYNTH is not explanation - SYNTH is when two or more reliably-sourced statements are combined to produce a new thesis that isn't verifiable from the sources. If you're just explaining the same material in a different way, there's no new thesis".

79.182.215.205 (talk) 23:58, 18 June 2012 (UTC)

Very very few sources meet the specification requirements of the logic you're using. It is usually SYNTHetic and Original Research for an editor to determine that Y(a) = Y(b). Fifelfoo (talk) 05:56, 19 June 2012 (UTC)

In A:X->Y, in B:Y->Z you only have Y in common. Then you derive X->Z which does not mention Y. Aside from it probably being synthesis I'd wonder what the topic was, it would need to cover at least two of X, Y and Z. And why hasn't somebody else pointed out this obvious bit about the topic - so what would be the weight? And without weight why would you be putting it in? Dmcq (talk) 07:13, 19 June 2012 (UTC)

Fifelfoo, Y(first source)=Y(second source)

Dmcq, because there are many (things)->Y, where X is only one such (thing). Thus researchers of Y->Z are not interested in what (thing) caused Y. They are just interested in investigating if Y->Z.

79.182.215.205 (talk) 08:07, 19 June 2012 (UTC)

I guess from that the topic involves X and X implies Y but that Y implies Z does not explicitly involve X. Then yes it is synthesis. And if it was any sort of mentionable weight they would have figured that out without needing editors on Misplaced Pages to do that for them and we'd have a citation pointing out the connection between X and Z. Dmcq (talk) 08:12, 19 June 2012 (UTC)

There are very few reliable sources about Y->Z, maybe because it is a lengthy experiment. One primary source though did state the conclusion X->Z, however, I understand that primary sources are not accepted.

It is true that Y->Z does not explicitly involve X. Suppose there are X->Y, X2->Y, X3->Y. So Y could be the result of X,X2, or X3. But why does it matter?

79.182.215.205 (talk) 08:19, 19 June 2012 (UTC)

It matters because Misplaced Pages is an encyclopaedia and is in the business of summarizing what people have said. They haven't said X implies Z. That is original research. That's not Misplaced Pages's job. We should be approaching the inclusion of stuff in Misplaced Pages with a neutral point of view and putting in the stuff that has some weight. Reliable sources are what give weight. Editors own conclusions have no weight. If an editor wants to popularize their ideas then they should do it outside Misplaced Pages and then eventually perhaps they might gain enough weight to be included. Dmcq (talk) 09:02, 19 June 2012 (UTC)

IP User 79.182 Y(a:Communism) almost never equals Y(b:Communism). You'll find that in actual writing, almost no terms are defined consistently between sources; such that any attempt to use syllogistic logic in the real world will fail, as sources do not use identical definitions of terms. Fifelfoo (talk) 08:24, 19 June 2012 (UTC)

This editor (79.182) is engaged in a debate about X-ray computed tomography, and has been forum shopping on WT:MED and now here about how to circumvent rules about original research. JFW | T@lk 08:34, 19 June 2012 (UTC)

Medicine is an area where we have to be especially careful not to start putting in synthesis. Apart from the general policy medicine is just rife with examples where one needs to check things are really so rather than applying logic, never mind anything about possible real life consequences. Dmcq (talk) 09:11, 19 June 2012 (UTC)

Dmcq,JFW, your reference to other discussions is off topic here in this theoretical discussion, which may or may not be related. Please remove these off topic comments.

Fifelfoo, In this case lets assume it is possible it is evident that both Ys are the same.

Dmcq, I understand that Misplaced Pages summarize what other people say, and other people say X->Y and Y->Z (and even X->Z in a primary source). In the article there is a discussion about stuff=(?) such that Y->(?) and stating there (?)=Z based on sources indicating Y->Z seem to me appropriate, and on topic.

79.182.215.205 (talk) 10:21, 19 June 2012 (UTC)

What's the problem with the X->Z primary source? Dmcq (talk) 10:29, 19 June 2012 (UTC)

Dmcq, other editors indicated that they want only a review from the best of journals, I think.

79.182.215.205 (talk) 10:39, 19 June 2012 (UTC)

I'd guess because the other citations in the article are of about that WP:WEIGHT. That's more of judgement call but mixing things that are properly reviewed along with results in a primary source that people haven't commented on can certainly be problematic. User's own conclusions with SYNTH have much lower weight than that!, so the primary source would be a far better one to look at, maybe someone has talked about it? Dmcq (talk) 10:52, 19 June 2012 (UTC)

Why is this generating this much text? It's clearly synthesis. Taemyr (talk) 11:44, 19 June 2012 (UTC)

I didn't get a specific comment about that primary source yet. I asked in the article's talk page. I did write it in the format of 'A 2004 cohort study concluded ....', so one can see it is a study. In my view it is very important to include it in the article.

79.182.215.205 (talk) 11:54, 19 June 2012 (UTC)

Consider the case - Source A says that the area of a piece of land is 347 ha. Source B says that there are 100 ha in one square kilometre. It is not synthesis to state that the area of the land is 3.47 km, yet this is what various people are saying (or is it synthesis to make this assumption?). Moral of the story - use common sense. Martinvl (talk) 13:35, 19 June 2012 (UTC)

Martinvl asks us to "consider the case". No. This question has been asked within the last few weeks, perhaps by the same person. There is a good chance the original poster is forum shopping. In any case, it almost always turns out that the premise "one reliable source say A:X->Y and an other reliable source say B:Y->Z" is false because Y₁ is not identical to Y₂, or "->", which in math is spelled out as "implies", is not what either source actually says. In math, "X->Y" means that in every single instance throughout the past, present and future, from the big bang to the heat death of the universe, whenever X is true Y is true. When an everyday non-mathematical source makes a statement using the English word "implies" or similar wording, it is hardly ever meant in such absolute terms. Jc3s5h (talk) 14:27, 19 June 2012 (UTC)

Jc3s5h, this is the first time I ask this question here. Thus, if the question has been asked before, I guess that means that other editors have also received comments, that they make a SYNTH due to this simple logic, and they too thought that it doesn't make sense, that obvious logic can't be used.

Jc3s5h, I have already wrote above "In this case lets assume that it is evident that both Ys are the same.". I.e. I am asking only about the case in which Y₁ is identical to Y₂, and both sources do say and mean imply.

Martinvl, in my actual situation, the other editors are saying something like - the second source doesn't say 'land' anywhere, so you can't use it in an article about land.

79.182.215.205 (talk) 14:45, 19 June 2012 (UTC)

79... request "In this case lets assume that it is evident that both Ys are the same." No. Frankly, I don't believe you. This is because many editors have come here with such over-simplified descriptions of the situation, and I have never seen a case where the situation was actually as simple as described. Jc3s5h (talk) 14:53, 19 June 2012 (UTC)

Jc3s5h, What I ask about, is of such a simple case.

79.182.199.172 (talk) 21:09, 19 June 2012 (UTC)

Jc3s5h, not believing violate WP:AGF.79.182.199.172 (talk) 23:07, 19 June 2012 (UTC)

I believe you have had a consensus answer to the question you have posed. It is synthesis as far as Misplaced Pages is concerned and should not be done. If you want to discuss some other case or divulge more details that's fine but as far as the question is posed just arguing isn't going to change peoples minds about it as it seems a pretty strong consensus. Dmcq (talk) 00:33, 20 June 2012 (UTC)

Yes, very strong consensus, two editors vs one. --Nenpog (talk) 13:45, 4 July 2012 (UTC)

Hypothetical case - Source1:"Kabusha drink include 40% alcohol", Source2:"Drinking 40mL alcohol can cause a hangover", Contested text:"Drinking 100mL of Kabusha drink can cause a hangover". Synth? OR? --Nenpog (talk) 15:11, 4 July 2012 (UTC)

Definitely should not be included as it is synthesis. We should not be making up our own deductions and sticking them into articles. We are supposed to be summarizing what other people have thought worth writing about. Dmcq (talk) 15:27, 4 July 2012 (UTC)

Do the Misplaced Pages policy writers expect that researches will think that it is worth writing, again and again, about every alcoholic drink, that it too can cause a hangover? --Nenpog (talk) 16:29, 4 July 2012 (UTC)

Just don't try sticking it into an article if it isn't something people think is worth writing about the topic. How much simpler can anyone put it to you? If researchers don't think it is worth writing again and again it is not worth us writing again and again. Misplaced Pages summarizes so it should have less not more than what is written about in reliable sources. Dmcq (talk) 17:21, 4 July 2012 (UTC)

But researchers don't write again and again about each alcoholic drink, that it too can cause a hangover, because when they write that 40mL alcohol can cause a hangover, they know that everyone can make the above simple deduction, and realize that if the certain drink contain 40mL alcohol, then that drink can cause a hangover. You see researchers like to summarize too, and when they write 40mL alcohol can cause a hangover, they mean it for every drink that contain 40mL alcohol. The fact that they didn't write a research paper for each drink doesn't imply that the fact is not important enough to be mentioned, but rather that fact was already pointed out in a generic form. If the researchers would have thought that their audience is so stupid that it can't perform a simple deduction, I think that they would have kept a list, and update it with every alcoholic drink. Do Misplaced Pages policy writers expect researchers to write this important fact again and again explicitly for each drink, just for people who are too thick to understand a simple deduction? --Nenpog (talk) 18:16, 4 July 2012 (UTC)

It's no more appropriate to go around to every article on an alcoholic beverage and say that it might cause hangovers than it is to go to every article on something that causes dsDNA breaks and say that it might cause cancer. Even if it's not original synthesis, it's undue weight. The outcome is the same using either policy: If you don't have sources for the claim you're making, you are not aloud to make it. Someguy1221 (talk) 04:10, 5 July 2012 (UTC)

Did you read the page of undue weight? Please read/reread it. It is a bit confusing, because in the Misplaced Pages definition the term undue weight refers only to the weight that is given in the Misplaced Pages article, e.g. by depth of detail, quantity of text, prominence of placement, and juxtaposition of statements. The Misplaced Pages definition of weight doesn't refer to the weight given in the sources. Instead in the sources what is important is the degree of acceptance of that view e.g. a majority view, a minority view, a tiny minority view or fringe. The due weight policy calls for a correspondence between the weight of the text in the Misplaced Pages article, and the degree of acceptance of that text in reliable sources. Thus, according to the Misplaced Pages definition, if the view, that dsDNA break might cause cancer is the majority view (and it is), then that view deserves a high weight in Misplaced Pages articles. Likewise, in this hypothetical example, if it is the majority view that 40mL alcohol can cause a hangover, then that view deserve high weight in Misplaced Pages articles. Thus inserting that view into the Kabusha drink article, would not be of undue weight, as it doesn't give prominence to a minority view, as it isn't a minority view, but the mainstream view. Please see also this discussion at the NOR talk page, in which it was suggested, that determining the degree of acceptance of a view is performed by confronting supporting vs. objecting reliable sources and evaluating from that which of the views is more accepted. At this case that would be reliable sources supporting the view that 40mL alcohol cause a hangover, vs. reliable sources objecting that view. At that case there would be only supporting reliable sources, hence the view is the majority view, hence WP:WEIGHT can not be used in order to exclude that view from any Misplaced Pages article. --Nenpog (talk) 06:15, 5 July 2012 (UTC)

This started as a content dispute on Talk:X-ray computed tomography where Nenpog faced a lack of consensus (every other editor opposed the changes he wished to make.)
He was then blocked for edit-warring and tendentious editing.
He then started Misplaced Pages:Forum shopping, taking his dispute to:
Misplaced Pages talk:WikiProject Medicine,
Talk:Ionizing radiation,
Misplaced Pages talk:No original research,
Misplaced Pages:Dispute resolution noticeboard,
User talk:Elen of the Roads,
Misplaced Pages talk:Neutral point of view
Misplaced Pages:Conflict of interest/Noticeboard,
And now he has come back to Misplaced Pages talk:No original research. --Guy Macon (talk) 05:29, 5 July 2012 (UTC)

And Guy Macon has been with this off topic message in all (or most, didn't bother counting) of these forums, so that everyone would be able to avoid a theoretical discussion, and discuss the situation only with the prejudice of a specific content dispute in mind. --Nenpog (talk) 06:15, 5 July 2012 (UTC)

Please join us at WQA and share your view there, if the above disclaimer from Guy Macon is related to the discussion or is off topic. --Nenpog (talk) 13:57, 6 July 2012 (UTC)

And the point that was made is that undue weight is not a valid argument here, and that the interpretation of the WP:SYNTH that was written above is like expecting that after researchers have determined that 40mL alcohol can cause a hangover, the researchers will continue to publish papers about each drink that include alcohol, and state if it too cause a hangover. That just don't make any sense, and thus that WP:SYNTH interpertation doesn't make sense either. --Nenpog (talk) 15:40, 6 July 2012 (UTC)

Fine that's your opinion. However it is not in line with the consensus on Misplaced Pages as acceptable grounds for inserting a reference which does not mention the topic into an article or for putting a conclusion based on that reasoning into an article. If you wanted to change that you'd need to raise a WP:RfC and to be frank I believe you're more likely to win the lottery than get your idea accepted. For far less money than the lottery you could set up your own website or publish your own book pushing your idea, and who knows you might eventually become a recognized authority and so make them acceptable references. Dmcq (talk) 11:44, 7 July 2012 (UTC)

This is now at Misplaced Pages:Administrators' noticeboard/Incidents#User:Nenpog. --Guy Macon (talk) 01:40, 10 July 2012 (UTC)

List articles and NOR

Upon reading WP:SALAT I came to a conclusion that there is some contradiction between those guideline are the NOR policy. The guideline say: "The potential for creating lists is infinite. The number of possible lists is limited only by our collective imagination. To keep the system of lists useful, we must limit the size and topic of lists." The guidelines contain no reference to the NOR policy, and the restrictions applied to the size and topic are dictated only by the usefulness of lists.
However, I see a serious potential of original research here. By arbitrarily choosing the list's subject, a user may create new topics. By expanding the annotations and the lede of the list article users can create de facto full articles about the topics that do not exist in literature, because the difference between the article and extended lists is not obvious.
In connection to that, I propose to add to WP:SALAT a link to WP:NOR, and to add to the policy a notion that, when no sources exist that draw more than a trivial connection between the items listed in the list article, such a list must contain just a very brief description of each item (not more than one short sentence), and the lede of such list must be totally descriptive.--Paul Siebert (talk) 21:43, 26 June 2012 (UTC)

I agree. We should only have finite lists of clear populations, e.g., the list of current members of the US Senate. A list of famous graduates of Oxford for example would be a nightmare, but we could mention in the article - if sourced - that many PMs were graduated, then have an article that lists them. TFD (talk) 22:36, 26 June 2012 (UTC)

We've long had lists of "People of X" (where X may be from a certain city, country, college, etc. etc.) Almost always these lists are strictly limited to those people with non-redirect bluelinks - aka notable people, and generally stated in the lede of the list. Are these lists OR? Not really - we know people must originate from some city, or graduate from some college, or the like, and we do fully expect these claims to be sourced.

I think what we need to be careful on is when the list is construed in a way to push a point. "List of people believed to be homosexual" would never fly regardless of the quality of the sourcing, for example, though on the other hand "List of homosexual people" (where a criteria requires self-assertion of this fact in a reliable source interview) would be ok. This is a whole issue we had discussed in trying to describe the notability of lists at WP:N a while back, but we never came to a clear answer to clearly saying when lists topics are notable, and I think the same problem extends to NOR and NPOV. Basically it is a point we all should be aware of in constructing the theme of the list, but it is difficult to nail down to a line since some novelty in organization may actually be nature and appropriate. --MASEM (t) 22:47, 26 June 2012 (UTC)

Lists need notability like any other article, one needs to show someone is interested in such a list by having something similar in a reliable source. The only special exception for lists is that one is allowed to set up a article on a section of a list, e.g. composers from A to E. Dmcq (talk) 23:49, 26 June 2012 (UTC)

The problem is that WP:SALAT does not tell about that. Per WP:SALAT, it is quite possible to create, for example, a List of Jewish mass-murderers, List of black homosexuals, etc. Not only such list will create non-existent topics, they would violate WP:NPOV also. We need to link list-related MOS to the NOR policy and to explain clearly that notability criteria for lists and for articles should be essentially the same.--Paul Siebert (talk) 01:03, 27 June 2012 (UTC)

Dmcq, when we developed WP:LISTN, we never came to that conclusion because there's just too many exceptions. We strongly recommend lists be notable themselves, or be a grouping that is notable, but we could not say lists that didn't fall into either case were or were not notable. And yes, there's definitely a need to link NOR/NPOV in discussion of list topics, but we just need to be aware that we can create novel groupings of elements that have never been published before as long as its not POV pushing or an unique way that doesn't fall naturally out of regular coverage. --MASEM (t) 01:37, 27 June 2012 (UTC)

• Disagree. What Paul Siebert and TFD are essentially proposing is that it should only be possible to create a List of fictional dogs if there exists a single source that lists all fictional dogs contained in that list. He is proposing that the use of multiple sources to create such a list is deemed to be synthesis and thus such a list be prohibited. Clearly that is too restrictive and is contrary to the whole purpose of a list which is to aid navigation through the organisation of articles. --Nug (talk) 02:18, 27 June 2012 (UTC)
• I'm with Nug. This comes up every now and again. NOR does not prohibit you from making a list merely because your sources don't include every item, or because your sources deal with it in prose. If you've got an entire book on The Top 100 Causes of Fever, and it uses paragraphs throughout, you may (1) present that information in a list format if you so choose and (2) expand the information citing other sources) to include items not in that one source. WhatamIdoing (talk) 03:10, 27 June 2012 (UTC)

Yes, NOR do not prohibit me from making a list if the sources don't include every item. However, and I explained that to Nug elsewhere, the topic per se must exist, i.e., some source must exists that groups majority of the list's items according some non-trivial criterion. Thus, you may create a list of WWII battles based on some WWII handbook, and add to that list some battles not mentioned in this concrete source. However, you cannot create a list of black rapists based on the book about mass rapes during the WWII if no non-fringe source exists that defines such a topic (rapes by Afro-american, or Caucasian, or Jewish, etc military personnel during the WWII).

In actuality, the WP:SAL lede says: "Being articles, stand-alone lists are subject to Misplaced Pages's content policies, such as verifiability, no original research,neutral point of view, and what Misplaced Pages is not, as well as notability guidelines." The problem is that that thesis has not been properly explained in the guidelines' body. In addition, no strict dividing line exists between the lists and the articles (for example, compare this ans that), so it is quite possible to create a piece of original research by gradual expansion of the list created based on the arbitrary chosen criterion.

In my opinion, the problem may be resolved if we add to the policy that strictness of the notability, NOR and NPOV criteria directly correlate with the amount of prose in the list: if the list consists of a very brief and descriptive lede followed by short descriptive single sentence items, the NOR criteria should be much milder.--Paul Siebert (talk) 03:23, 27 June 2012 (UTC)

I don't think this restriction would solve any problems. Also, WP:N plainly states that notability does not affect what or how much you can put on the page, just whether we should have that page at all. WhatamIdoing (talk) 18:09, 28 June 2012 (UTC)

In any event, WP:SALAT currently says: "The potential for creating lists is infinite. The number of possible lists is limited only by our collective imagination. To keep the system of lists useful, we must limit the size and topic of lists." Some users may interpret that as a carte blanche for creation of any list they want. In my opinion, the sentence

"The number of possible lists is limited only by our collective imagination."

should be modified as follows:

"The number of possible lists is limited only by our collective imagination, and by our content policy WP:NOR, WP:NPOV and WP:V."

--Paul Siebert (talk) 19:52, 28 June 2012 (UTC)

Lists are properly seen as navigation aids, that is why they are discussed in Misplaced Pages:Categories, lists, and navigation templates. Your proposal is still too restrictive. --Nug (talk) 20:06, 28 June 2012 (UTC)

My proposal is by no means restrictive. Without this reservation, the WP:SALAT section simply contradicts to what the lede of the same page says ("Being articles, stand-alone lists are subject to Misplaced Pages's content policies"). As this discussion demonstrates NOR NPOV and V are applicable to stand alone lists, no matter if they are seen as navigation aids or not.--Paul Siebert (talk) 20:23, 28 June 2012 (UTC)

• Support change to WP:SALAT as detailed by Paul Siebert, i.e. to say WP:NOR, WP:NPOV andWP:V apply. People seem to have got the idea that the manual of style overrides the content policies and guidelines. Dmcq (talk) 22:16, 28 June 2012 (UTC)

WP:NOR applies in what way with respect to lists? That all items in a list should be sourced to a single publication? This is in essence what Paul Siebert is proposing. --Nug (talk) 22:27, 28 June 2012 (UTC)

Nug, you again misinterpreted my words: I didn't say all items need to be listed in the same source. My point is that some reliable non-fringe source is necessary to define the topic. Do you see a difference? I believe my explanation is crystal clear. Please, do not misinterpret my words any more.--Paul Siebert (talk) 05:04, 29 June 2012 (UTC)

• Comment. The background to Paul Siebert's request here is this discussion at Talk:Soviet_occupations#List_Article, his contention being that there is no one single source that mentions all the Soviet occupations mentioned in the list. Although all individual occupations mentioned in the list are cited to reliable sources, he contents that aggregating them into a list is synthesis and thus should be prohibted. --Nug (talk) 22:22, 28 June 2012 (UTC)
  • Yea, Paul's view is completely counter to how list articles can be constructed. No single source has to discuss all the occupations in whole, but the grouping of the occupations (presumably each well sourced) is completely fine particular since its a neutral, non-POV state ("here are the occupations of the Soviet Union"). And because the individual occupations are notable themselves, there's no issue there as well. --MASEM (t) 22:27, 28 June 2012 (UTC)

First of all, you Masem seems to argue not with what I propose, but with Nug's interpretation of my words: " What Paul Siebert and TFD are essentially proposing is that it should only be possible to create a List of fictional dogs if there exists a single source that lists all fictional dogs contained in that list." In actuality, what I say is the following: since per WP:SAL, list articles, similar to other articles, are subject to our content policy, we can add new items in the list that belongs to some existent topic, but the list cannot create a non-existent topic. If the Nobel prize winners A, B, C, and D are Caucasian males, and each of them is notable enough to have his own article, it would be hardly correct to create a list "List of Caucasian males who got a Nobel prize". We can create such a list only if some mainstream reliable source exists that discusses a category "Caucasian male Nobel prize winners. Otherwise, we open a Pandora box: we create a topic, it may grow to a full article and eventually we get a piece of non-neutral original research.

With regard to "occupation of the Soviet Union", have you noticed that that article (which, according to Nug is a list, but it is meets all criteria of the full article) is not about the occupations of the Soviet Union, but the article about occupations by the Soviet Union. I asked, for several times, to provide a source that links such different events as annexation of the disputed province Bessarabia (a former Russian province, which was annexed by Romania, the act, which had never been recognized by the USSR and many other states), Soviet occupation of Nazi Germany after the Allied victory in the WWII, and the war in Afghanistan (which is not described as Soviet occupation in the main article, so it is simply a POV-fork). I have been waiting for almost three months, but no sources have been provided to link all those events. Meanwhile, when I type "Soviet occupations" google gives mainly the links to Misplaced Pages itself, and to its mirrors, thereby confirming that Misplaced Pages has created a new topic. --Paul Siebert (talk) 05:01, 29 June 2012 (UTC)

PS. By writing that, I by no means argue against a disambiguation page, similar to the Japanese occupation page. --Paul Siebert (talk) 05:10, 29 June 2012 (UTC)

Paul, if you first find a source that links Akamaru, Kiba Inuzuka's dog in the television series Naruto with Chief from The Fox and the Hound, two dogs listed in List of fictional dogs, then I will find a source that links the Soviet military occupation of Afghanistan with the Soviet military occupation of Germany. According to Paul "List of fictional dogs" can only exist only if some mainstream reliable source exists that discusses a category "fictional dogs". --Nug (talk) 05:40, 29 June 2012 (UTC)

There are plenty of lists that are "Lists of X" where "X" is not a notable topic in of itself, but is one that derives from natural or logical extension of other notable topics. Yes, care must be taken that that grouping is not biased or created to push a point; NOR and NPOV apply to lists in as much as they apply to articles. But you do not need to have one single source that link all those events to X, as long as individual sources for each event declare it part of that arena. The Soviet Occupation article would likely be considered a list because it is simply grouping several already-existing articles as well as some non-notable events into a navigation aid. --MASEM (t) 05:50, 29 June 2012 (UTC)

The Soviet occupations article is definitely not a single topic but a list type article and should be treated as such. It looks like original research to me trying to make an integrated article out of them, that would require a source talking about the topic as a subject rather than just a list. Dmcq (talk) 10:49, 29 June 2012 (UTC)

Well obviously it is a list of military occupations by the Soviet Union. The lede describes a clear criteria for inclusion and each section has a link to the respective article and a summary of that article. --Nug (talk) 13:24, 29 June 2012 (UTC)

I agree it is a list article in its approach, but again, that doesn't require that we have one single source that links them all together. Is there any question that the former Soviet Union has occupied other nations? If not, this is a natural grouping of such occupations as a list, it doesn't matter that no source explicitly lists each one in a single shot. --MASEM (t) 13:35, 29 June 2012 (UTC)

Again, we do not need a single source that links all of them together. However, we do need a source that sets the topic: is there any question that some white men got Nobel prize? However, we can create a list of all Nobel prize winners, a list of American Nobel prize winners, but I don't think a list or white male Nobel prize winners would be a legitimate subject of the list. Similarly, it is possible be to create a list of countries under Soviet occupation as a result of the WWII (that is a part of a single story), however, a list of countries that are believed to be occupied by the USSR during different periods of its history is hardly a correct subject of the list: I asked, for several times, for a source that links WWII, post-war and Cold War Soviet occupation within the frames of the same concept, however, no sources have been provided.--Paul Siebert (talk) 14:31, 29 June 2012 (UTC)

--Paul Siebert (talk) 14:05, 29 June 2012 (UTC)

"believed to be occupied by the USSR during different periods of its history" - as long as 1) they are being called "soviet occupation" or a near-equivalent phrase, and 2) backed by reliable sources for each entry, there's no NOR/NPOV going on here in combining all Soviet occupations. I do recognize the concern when you say "believe", as I could see the possibly of someone in a politically/ethnically-charged manner assert that some action by the Soviet Union was an "occupation" but no RS ever back that up; I have not reviewed every statement on that list to confirm this, but AGF that this is not the case. We use a similar logic on List of commercial video game failures in that we require sources to call out the entries as commercial failures instead of just making assumptions without sources. --MASEM (t) 14:22, 29 June 2012 (UTC)

Using the same logic, as long as some white males are believed to get a Nobel prize, and that is confirmed by reliable sources, there is no NOR/NPOV issues with combining them together in the List of white male Nobel prize winners.

Well, I'll ask the same question is a different way: is it correct to create a full article that discusses all Soviet (American, British, etc) occupations (or white male Nobel prize winners) in a single article devoted specifically to this subject? I believe, in the absence of RSs that confirms that such a topic does exist in literature the answer would be "no". However, for the lists, which must comply with the same content policies, the answer is, according to you, "yes". Don't you see any inconsistencies here?--Paul Siebert (talk) 14:39, 29 June 2012 (UTC)

The reasoning on the white male Nobel prize list being a problem is that there is implicit gender and racial biases when we talk about a large grouping of people; to classify any list out by gender, race, religion, sexual preference, etc. is going to be a non-neutral action. To demonstrate that we (the Wikipedian editors) are not introducing that bias ourselves, we'd need reliable sources that calls out this grouping as novel, at which point its not our OR or POV that we're putting into text. You'd very likely never see any type of "white male Nobel prize" (doing a quick glance through Google) but there's plenty of reasons to have "female Nobel prize winners" (as this group has been called out) or similar "minority" classifications. But that just doesn't seem to exist for white males, so while assembling the list is not violating NOR, it is a problem with NPOV unless sources defer that claim for us.

And no, "List of X" does not depend on "X" being a valid topic for an article. I doubt there is any specific source that describes, in generic terms, the idea of a "Soviet occupation" without ever referring to any specific occupation. But there have been several Soviet occupations stated by reliable sources over the years, so there's no doubt they happened - before, during, and after WWII It's clear that the concept is related to the notable topic of the USSR, and more likely under History of the Soviet Union; its even better that its limited to the lifetime of the Soviet Union and not of the previous Russian empire or the post-dissolution. There is thus no NPOV connecting these together, and as long as each occupation is clearly identified as such by RSs, no OR in putting them into a single list. --MASEM (t) 15:17, 29 June 2012 (UTC)

And to add (after seeing a response below), a list that describes all occupations of other countries by another country (extending outside of the Soviet Union), would not be OR or NPOV as well, as long as each occupation is sourced to be described that way in RSes. There may be organization and size issues (I figure there's a LOT of those in human history) but the discussion of all known country occupations is a completely fair list topic. --MASEM (t) 15:20, 29 June 2012 (UTC)

You are talking about "discussion", but the List of fictional dogs does not discuss anything, and the List of military occupations does not discuss either. By allowing stand-alone lists to expand we allow users to create the topics that do not fit notability criteria applied to normal articles. In connection to that, see my comment below.--Paul Siebert (talk) 15:46, 29 June 2012 (UTC)

• Oppose to change. There is nothing wrong with creating pages like List of Japanese occupations, List of military occupations by the United States, etc. This does not contradict NPOV or anything. The objects included in these lists are notable and obviously belong to the same category. Such lists can have such format and include such amount of information about every item as decided by WP:Consensus at talk page of the corresponding list article, not here. The items of the list can be taken from multiple sources that are not necessarily consistent with each other. They are even required to be taken from multiple sources per WP:NPOV. My very best wishes (talk) 14:00, 29 June 2012 (UTC)

Firstly, Japanese occupation dab page already exists. However, it combines the events that have been combined together (by many sources) according to a very concrete criterion: all of them occurred during the WWII (and occupation of Japan was an immediate result of the WWII). With regard to List of military occupations by the United States, I have a double feeling. Initially, I had no objections against the List of military occupations by the USA, USSR, or similar articles. However, upon realizing that stand alone lists must comply with the same NOR and NPOV criteria as other articles do I started to doubt. Our NOR policy says: "Do not combine material from multiple sources to reach or imply a conclusion not explicitly stated by any of the sources." By combining all white male Nobel prize winners in the same list we imply the idea of white male supremacy. By combining all military occupations by some concrete big country we inevitably imply it was conducting aggressive and expansionist policy during its whole history. If such a conclusion does not explicitly exists in some mainstream reliable source, it is synthesis. If lists comply with the same content policy, such an approach is hardly legitimate.--Paul Siebert (talk) 14:17, 29 June 2012 (UTC)

No, I do not think that such lists actually imply anything beyond the existence of many occupations. For example, a list of software would hardly imply anything except the existence of numerous software. Of course if a list of occupations by nation X includes a direct statement that nation X is a "nation of occupiers" (in fact such statement can frequently be well sourced), that might be a matter of concern related to specific list, but not to the general policy. My very best wishes (talk) 14:45, 29 June 2012 (UTC)

By combining all US (Soviet, British, etc) occupations together we imply a conclusion that those countries conducted expansionist policy during whole their history. By combining all white male Nobel prize winners we imply the idea of white male supremacy. All of that may be right or wrong, however, we need a source that tell that explicitly.

And, by the way, you seem not to understand the main idea of this thread: I propose no changes to policy, I propose to bring the guidelines in accordance to our policy.--Paul Siebert (talk) 14:48, 29 June 2012 (UTC)

A reader (like you or me) has every right to personally interpret the list of Microsoft software as a proof that Microsoft was conducting aggressive and expansionist policy during its whole history, but it does not make such list problematic. And if the claim about "aggressive and expansionist policy" can be reliably sourced, there is nothing wrong with including such claims in the introduction.My very best wishes (talk) 17:03, 29 June 2012 (UTC)

• Oppose change - a cure in search of a disease, a solution in search of a problem, and a great example of instruction creep. Cheers. Collect (talk) 14:43, 29 June 2012 (UTC)

Could you please elaborate on that? Your post contains no concrete statement, so it is just a universal "IDONTLIKEIT".--Paul Siebert (talk) 14:50, 29 June 2012 (UTC)

1. You have not demonstrated that a problem exists.

2. You have not demonstrated that this would solve an existing problem.

3. This is a "new instruction" hence clearly is "instruction creep."

To wit: Instruction creep occurs when Misplaced Pages guideline or policy pages contain statements that expound upon unimportant details and minutiae.

Thus my post is not an IDONTLIKEIT post - it is grounded in the facts and substance of what has been presented. Collect (talk) 18:52, 29 June 2012 (UTC)

Upon looking at List of fictional dogs I partially understood Masem's point. Yes, such a list has a right to exist, because it implies no conclusions and is used for navigational purposes only. The close analog of such list would be List of Nobel laureates and List of military occupations. These lists contain no original research, imply no conclusions and are in full accordance with out core content policies. However, a List of white male Nobel laureates would violate our policies if no mainstream sources exist that define such a topic. Similarly, list of military occupations by country, during whole history of this concrete country implies some conclusion, and if no sources exist that supports such a conclusion, then we can speak about a violation of our policy.
In addition, in some lists, the prose has a tendency to expand, and such a list may become a full scale article. In connection to that, I am asking again, what our policy tells about that, and, if different criteria are applied to the lists and full articles, what are the criteria that allow us to discriminate the former from the latter?--Paul Siebert (talk) 15:33, 29 June 2012 (UTC)

You're missing the NPOV aspect. "White male Nobel laureates" is immediately a biased statement, simply because of the weight of gender/racial equality concerns. "Occupations by the Soviet Union" is not biased at all; the Union did occupy countries, they were called "occupations", etc. There's no conclusion to be made from that beyond listing it out. If it was more the case of said occupations of being of what individual editors made of it (hypothetical example: a Czech calling the 1968 action as an "occupation" why a Russian calling the same thing "a nice friendly visit") so that there was a large disagreement in RSs of what really was an occupation or not, ok, there may be some question, but this doesn't appear to be the case; the majority of RS agree each case can be defined as "occupation". So there's zero POV pushing happening here as best I can tell. --MASEM (t) 15:49, 29 June 2012 (UTC)

I am not missing anything. If something is not biased, that is easy to demonstrate. We do not outline a limited set of statements that are intrinsically biased, and we do not provide a carte blanche for other topics.

Regarding your "There's no conclusion to be made from that beyond listing it out", that depends on the form of the list. If the list has the same style as the List of fictional dogs, yes, there is probably no problems with such a list, for example, a List of countries occupied by the USSR from 1939 to 1991. However, when we are talking about an expanded list, I see almost no difference between such a "list" and a full article. In connection to that, I repeat my question: what is the difference between the "expanded list" and full article, and why the NOR/NPOV criteria should be milder to the former than to the latter?

Regarding your second point, yes, if different terminology can be applied to some item of the list, NPOV should be taken into account. In your example, "a nice friendly visit" to Czechoslovakia would be a totally fringe views, however, "occupation of Afghanistan" is described as a "proxy war" in the main article, not an occupation. Therefore, a list should not give undue weight to one existing viewpoint. However, that is a different story.--Paul Siebert (talk) 16:14, 29 June 2012 (UTC)

There is no difference in the application of NOR/NPOV between a list and an article, and certainly not in this case. All pages in mainspace, short of disambiguation pages, are expected to meet all content policy and guidelines, and the only place where lists and articles really come up is if you're trying to go to Featured content, where we decide if its more a list or more an article. You're trying to argue that some exception is being made on the Soviet occupation page but there isn't. It complies fully with NOR and NPOV. And as for the Afghanistan piece, there's plenty of sources search on "Soviet occupation of Afghanistan", so just because WP calls it a proxy war doesn't mean that RS never called it an occupation. --MASEM (t) 17:19, 29 June 2012 (UTC)

Well, is there is no difference, then explain me why do you oppose to addition of that explanation to WP:SALAT? I see the need in such an explanation, because during some discussion I saw the argument that, since the article is in actuality a list, then we are not restricted with NOR/NORN/V policies here. Since that argument was put forward by an experienced user, I think the WP:SALAT is ambiguous.

With regard to "Soviet occupations", I am not trying to argue about that here. You probably noticed that I didn't bring this argument here. I am perfectly comfortable to discuss any other example.--Paul Siebert (talk) 17:59, 29 June 2012 (UTC)

I personally never said that NOR/NPOV shouldn't be added here. But what does seem to be in play is that you're trying to argue for their addition here to nullify the Soviet occupation (and other similar articles) article, starting with the false premise that a topic needs to be discussed in full by a source to be an appropriate topic for WP, which is a fallacy. Again we should be applying NOR and NPOV to list contents, but remember that NOR does allow for non-biased synthesis of information which is what we're talking about. --MASEM (t) 18:06, 29 June 2012 (UTC)

If you never said that NOR/NPOV shouldn't be added to SALAT, then I see no reason to argue with you about that. We need just to implement these changes in SALAT and forget about that.

With regard to Soviet occupation, are you talking about my real intentions, or about Nug's interpretation of them? My views of the "Soviet occupations" page evolved with time. Initially, I asked the users who worked on this article to provide sources that demonstrate that such topic exists. In response, they agreed to change the lead and declated that the article is a list now (although some of them still maintain that this is an article about some single subject). Upon changing the lede they declared that the issue had been resolved, because, since the article is a list, no sources that prove the existence of the article's topic in general are needed.

I didn't mind to keep this list, although, I insisted on its renaiming into the "List of Soviet occupations and on bringing it into accordance with List of military occupation style, because the article in the current form is the article, or summary style article (similar to the WWII article), but by no means it is a list. However, upon meditation, I realized the we have a general issue, and I decided to discuss it here. I believe this my explanation should resolve the issue. In future, please, do not accuse me in attempts to modify policy to push some local POV.--Paul Siebert (talk) 18:19, 29 June 2012 (UTC)

The problem is that you started with a statement like "By arbitrarily choosing the list's subject, a user may create new topics. By expanding the annotations and the lede of the list article users can create de facto full articles about the topics that do not exist in literature, because the difference between the article and extended lists is not obvious." which has been pointed out is not true. It didn't matter if it was specifically the Soviet occupation article, just that that statement was a fallacy. NOR/NPOV still apply to lists and list content (so I do support their addition here), but neither of them prevent the type of list your original statements suggested.

And no, renaming doesn't have to be done either. List articles, expanded or not, do not require them to start "List of". I've found its generally more the case as the list becomes more a navigational aid that "List of" is used more often but certainly it is not required. As long as it is clear what the inclusion rules are (in the case of Soviet occupation, its right there, first sentence), the title doesn't have to indicate that. --MASEM (t) 18:32, 29 June 2012 (UTC)

All of that is of secondary importance. I see you have no objections to updating of WP:SALAT, and that is good. With regard to the rest, see my questions below. If you believe that the same criteria are applied to SALs, to summary style articles and to articles, please, explain, can I take several items, each of them are notable per se, combine them into a list based on some nominal linkage between them, expand the list, and eventually get a summary style article about the topic that does not exist in literature?--Paul Siebert (talk) 18:38, 29 June 2012 (UTC)

To your question: yes, but only if that linkage is not something that is POV-loaded or biased, or that would require that POV/biasing would be necessary to shoehorn in topics. "Soviet occupations" is fine as the country did perform several - albeit different reasons, but it was still the same country, so that's a non-biased link. "White male Nobel prize winners" is a problem since you're introducing the bias of race and gender. The same problem would apply to "Female Nobel prize winners" but here while there is the bias, we can find sources that describe it out that way, thus removing the bias from WP's editors and to the RS, which apparently see it this way. --MASEM (t) 18:42, 29 June 2012 (UTC)

Do you mean NPOV are applicable, but NOR is not?--Paul Siebert (talk) 18:45, 29 June 2012 (UTC)

NOR applies too, but keeping in mind that NOR allows for reasonably simple, obvious synthesis, of which, say, gathering together the Soviet occupations. --MASEM (t) 18:48, 29 June 2012 (UTC)

Then it would be useful to define the limits of simple and obvious synthesis that are allowed per NOR. For example, let's consider the following case. Some reliable sources describe behaviour of some Western powers towards some small states or nations as "betrayal" ("western betrayal at Munich", "western betrayal in Yalta", "western betrayal of Arabs" etc.), so each of those topics is notable per se. In connection to that, I have following questions. What of the following can be considered as "simple and obvious synthesis that is allowed per NOR":

1. Creation of the "List of western betrayals";
2. Development of such a list into an extended list;
3. Creation (based on this extended list) of a summary style article "Western betrayal" that starts with "Western betrayal is the term which refers to some aspects of the foreign policy of several Western countries during the period from the Treaty of Versailles in 1919 through World War II and to the Cold War".

As this search demonstrates, each separate topic is notable and verifiable.--Paul Siebert (talk) 20:37, 29 June 2012 (UTC)

The only thing that would prevent this is the quality of the RS's in how they define "betrayal", as it is a potentially more opinionated word than "occupation". If the only sources for these are, say, for the Yalta pieces are coming from Yalta sources, and so on, their reliability is in question, and there's like NPOV going on. But if dozens of historians all agree they're all individually betrayals, that's fine. --MASEM (t) 22:46, 29 June 2012 (UTC)

I don't think this interpretation of NOR policy is correct. Many instances of different "betrayals" by one or another Western power do not form a single concept "Western betrayal". Just compare this discussion with the discussion in the previous section: much more innocent things are not allowed per NOR.--Paul Siebert (talk) 23:44, 29 June 2012 (UTC)

The problem is that you have a team that is "loaded", that being the word "betrayal". Yes, everyone knows what it means, but how it is used in the geopolitical stage? I have a feeling if you looked through highly reliable sources of a purely academic nature, you'd find several different definitions. On the other hand, the geopolitical meaning of "occupation" is pretty dang clear. Again, there are fringe takes on the word, but in academic circles, it seems to be a pretty sound, concrete topic. It would be very difficult to contest any of the Soviet occupation as "occupations" but it would rather easy to contest "betrayals". But you can always be very explicit in what "betrayal" means - you must have such-and-such a source confirm this, or it must be X doing Y to Z, or whatever, and avoid the fringe interpretations.

The end result: there is absolutely nothing wrong with grouping numerous elements that share common aspect X as long as the aspect of X can be reliably sourced (WP:V), does not require "creative" interpretation or correlation (NOR) and does not create a bias (NPOV). --MASEM (t) 00:06, 30 June 2012 (UTC)

Masem, you will be surprised, but the term "occupation" is also not as clear as you think, and some events that are described as "occupation" by one author are described otherwise by another one. However, I suggest you to forget about occupation. I came to this page for quite a different reason. (Regarding "Soviet occupations" I initially was pretty comfortable with the idea to convert it to the classical list, similar to List of fictional dogs). Therefore, again, please stop your allusions: I am here not because I want to change policy to push some POV.

With regard to betrayal, that is a part of a real dispute (I just changed some arguments for simplicity): if one author speak about "western betrayal" of Czechoslovakia by Britain and France in 1937, and another author speaks about "Western betrayal" of Poland by the US and the UK in 1945, can we speak about some general "western betrayal" as a general concept that is used to describe policy of western powers in XX century?

This is a very interesting question, and I am even contemplating to start an RfC on that account: can Wikipedians create new topic by themselves based on well sourced and notable examples that may be (or may be not) combined within some more general concept, but which have not been combined so far in reliable mainstream sources? I cannot predict the result of such a RfC, but, in my opinion, such a question deserves to be asked.--Paul Siebert (talk) 00:33, 30 June 2012 (UTC)

The selection metric you're using is actually very important here. "Occupation" might be a loaded term, but in reference to everything on the Soviet occupation page, those examples are far from it. "Betrayal" is a loaded term - is it as simple as a broken verbal promise? - and thus if we group betrayal together we've got to make sure that the reliable sources are talking the same thing. I can't recollect the exact cases above, but if one case was where country A was betrayed by B causing significant economic damage to A, and another case of country C being shunned because country D didn't hold their promise to hold an event these but otherwise causing no other issues, there's no way those two would be connected. But this is because "betrayal" has a wide breadth of meaning, particularly into the fringe. A list would be possible, but its got to have a strong definition of what betrayal is for each item added to it to prevent trivial/fringe considerations. Again, I bring up List of commercial failures in video gaming in which we don't include any "flop" but ones specifically called out by the industry press as commercials failures; Similarly, List of Internet phenomena requires a highly-reliable source to claim the material is viral or the like, despite lots of ways in the fringe you could says something's popular on the Internet. Making that type of definition (explicitly in the article text, or in the talk page) to limit what elements are included or the like is ok, as long as we're still following NPOV and WP:V.

There's also a flip way of looking at this. Imaging you could write an article that would encapsulate the list, talking about the subject that's at the next highest level; in the case of Soviet occupation, it may be "Military actions of the Soviet Union". (and lets say that's a true article). Does it make sense that there would be a section "Soviet occupations" in that article? Or in the case of betrayals, lets say its an article on Intra-country politics, and would a section on "betrayals" make sense? If that section makes sense to have because its one way to organize the larger information, then its probably okay as a standalone list. This is not a test that works well in all cases but works quite often. --MASEM (t) 00:50, 30 June 2012 (UTC)

Re your "A list would be possible". Whereas I tend to intuitively agree that for some topics lists are acceptable, and articles are not, I still cannot understand that, taking into account that, according to WP:SAL, the same criteria are applied to lists and to articles.

Re your " Does it make sense that there would be a section "Soviet occupations" in that article? ", no. Different events currently described in the "Soviet occupations" article are seen quite differently in literature: Bessarabia (a former Russian province annexed by Romania) had been annexed by the USSR, and this fact is not a subject of debates in literature (except, probably, Romania). The events in the Baltic states are also described differently, from "annexation/incorporation" through "occupation quasi-bellica' to "military occupation" (depending on the views of one or another author). Occupation of Czecholovakia was a typical military occupation. Occupation of Germany was a very strange case (some authors even use the occupatio sui generis formula). Occupation of Afghanistan was a proxy war similar to Vietnam, and so on. As a result, those events will never be analyzed together in the hypothetical "Military actions of the Soviet Union" article; each of them will be discussed in the proper historical context.

However, all of that is just details. I am more interested to analyze a general question: can we create new topics by themselves?--Paul Siebert (talk) 01:42, 30 June 2012 (UTC)

Within limits/reason we can, we not just have list but articles as well the exact naming of which exact naming does not exist in sources but the subject as such is still considered notable. In these cases usually a purely descriptive term is used. But note it may only be "new" in the sense that there is no established name or definition in the sources, it's actual content however is not new at all.--Kmhkmh (talk) 09:18, 30 June 2012 (UTC)

Oppose I have a problem with how the suggested change might be used. I see no problem with example of the "list" of soviet occupations. I think we need to keep in mind the goal of our guidelines is to ban "real" (novel) OR and "real" (novel) synthesis, but not to ban informative and useful articles/lists which could be seen as possible OR or SYNTH violation if you try really hard. I have hard time seeing anything novel in that example. Summarizing/categorizing/listing/combining material from different sources is standard task in any encyclopedic writing.--Kmhkmh (talk) 16:17, 29 June 2012 (UTC)

I am not sure you understand what we are talking about. The discussion is primarily about bringing WP:SALAT (a WP:SAL's section) in accordance with the WP:SAL itself. If some topic contains no original research or it does not violate the neutrality policy, it can be easily demonstrated. However, to say that "The potential for creating lists is infinite. The number of possible lists is limited only by our collective imagination. To keep the system of lists useful, we must limit the size and topic of lists." (as WP:SALAT currently says) would mean to give a green light to various violation of NOR, V and NPOV policy.--Paul Siebert (talk) 16:48, 29 June 2012 (UTC)

I can't the see the (alleged) violation you're talking about, in particular not in that line yu've cited. Also if you're primary aim is to change the text of WP:SALAT you should discuss that there rather than here.--Kmhkmh (talk) 17:58, 29 June 2012 (UTC)

See my responce to Masem. It was not me who started a discussion about "Soviet occupations". I am discussing a policy in general. If we assume that the same criteria are applied to the articles and to the lists, then the opposite is equally correct: we can apply the same criteria to the articles as to the lists. If we assume that the lists can combine several items that are notable per se according to some arbitrarily chosen criteria, then the articles can do that too. In other words, we are allowed to write WP articles about non-existent topics by combining several existing notable topics, for which needed reliable sources exist. That is not how I saw our policy.--Paul Siebert (talk) 18:04, 29 June 2012 (UTC)

Why would you assume an identical treatment of articles and lists to begin with?--Kmhkmh (talk) 18:23, 29 June 2012 (UTC)

I do not assume that. It is what WP:SAL (its lede) says. In any event, my questions here are general: (i) are the same criteria applied to the articles and to the lists (as you can see, some participants of this discussion believe that they are); (ii) if the criteria are different, then what is the concrete difference; (iii) if different criteria are applied to the lists and to the articles, what is a borderline between SALs, summary style articles and full articles, and which parts of NOR policy are applied to each of them? I see some gray zone here, which may serve as a serious source of original research.--Paul Siebert (talk) 18:28, 29 June 2012 (UTC)

The concrete difference as far as I'm concerned is that a list does not have to cover a topic that has been covered in detail, it just has to be based on lists that people have thought worth making. For instance one would have trouble with an article on universities in the US as there isn't really a topic that people go into in any detail like Europe even though Europe is composed of countries. However people do make lists of things like American universities. Dmcq (talk) 18:37, 29 June 2012 (UTC)

American universities is a legitimate subtopic of the Education in the United States article. I see no problem with V, NOR and NPOV here.--Paul Siebert (talk) 18:40, 29 June 2012 (UTC)

List of London railway stations then. Dmcq (talk) 23:55, 29 June 2012 (UTC)

That is obviously a quite legitimate subject for the list (as a supplement to the London metro article). I am not sure I understand the idea you are trying to convey.--Paul Siebert (talk) 00:37, 30 June 2012 (UTC)

What about List of fictional canines? What article does that supplement? It is subsidiary to the list of fictional animals. In just the same way Soviet occupations is subsidiary to the List of military occupations. --Nug (talk) 01:02, 30 June 2012 (UTC)

That is an interesting question. But that is a part of a bigger question. See our discussion with Masem.--Paul Siebert (talk) 01:42, 30 June 2012 (UTC)

Re what I was trying to convey, you asked what was the concrete difference between an article topic and a list topic. I answered that lists do not have to be dealt with in depth as a subject, they just have to be things that people have thought worthwhile to list, i.e. they have list notability not topic notability. Dmcq (talk) 12:54, 30 June 2012 (UTC)

The problem I have with that is it means I could add my niece to the list of notable people from Denver. Or a friend's film company to a list of film companies. Dougweller (talk) 13:45, 30 June 2012 (UTC)

That is a question of what can be included in a list which is covered by what the list inclusion criteria and verifiability. Editors may or may not require that the entries are notable for instance after checking what the list is about. If all that is required is verifiability then it is possible your friend's film company may be included even if nobody is particularly interested in it. Dmcq (talk) 16:16, 30 June 2012 (UTC)

RfC on the verifiability policy lede

• Misplaced Pages:Verifiability/2012 RfC

Hello all. I'd like to draw your attention to an RfC about the lede of the verifiability policy. We have been drafting this RfC for some time as part of a MedCab mediation, and it is finally ready for comment. In the RfC we have included a few specific drafts of the policy lede for you to comment on, and we have twelve general questions to find editors' views about how the lede should look. All editors are warmly invited to join the discussion at the RfC page. Thanks! — Mr. Stradivarius on tour 02:04, 29 June 2012 (UTC)

Newspapers (again...)

I just want to make sure I've not missed something that may have changed in regards to the primary/secondary nature of newspapers.

There's been discussion about the appropriateness of "X On Twitter" articles (This is not to clear that issue up). A discussion I started is at WP:VPP#Appropriateness of "X on Twitter" (or similar) articles. One protracted issue that has come up is the nature of newspapers articles as primary or secondary sources. Last I checked, and this is consistent with the wording of the policy page and other places on WP that talk about primary/secondary sources, newspaper sources are generally considered primary if they are reporting on a topic and not providing any type of analysis or critical assessment; another user (Alanscottwalker (talk · contribs)) is insisting because the newspaper is not the original source and had editorial control to chose to publish it, they're secondary sources.

As the case in point, I found an example of an article that is talking about one of the celebs with a Twitter account, and the article quotes one of the celebs tweets - and says nothing else about the Twitter account . In considering the application of that article to a WP article specifically on the celeb's Twitter account (and not on the celeb himself), that, by all standards that I've read and checked is flat out a primary source. (It may be secondary for the celeb, but certainly not for the Twitter account). Alanscottwalker insists that its secondary because the news-reporting group used editorial control to select that quote and thus implicitly added an evaluative aspect to it, thus making is secondary.

I want to make sure I'm not mistaken in that Alan's view here specifically on this example is completely at odds with standard consideration of primary and secondary sources. --MASEM (t) 12:43, 4 July 2012 (UTC)

You're right. The term secondary was used for so long as a synonym for any positive quality (independent, fact-checked, peer-reviewed, non-self-published, etc.) that some editors are pretty confused. See WP:PRIMARYNEWS for a summary. WhatamIdoing (talk) 02:30, 16 July 2012 (UTC)

Synthesis example

The "more complex example of original synthesis" is not a good one. It claims that "the second paragraph is original research because it expresses a Misplaced Pages editor's opinion that, given the Harvard manual's definition of plagiarism, Jones did not commit it." But it doesn't: it expresses the simple logic that IF Jones did such-and-such (copying references without consulting them), then such-and-such would be the case (this would contravene the Harvard manual, but not fall under its definition of plagiarism)." Certainly this expresses the editor's understanding of what the Harvard manual says, but understanding is a part of writing about anything, is always open to correction, and is not by itself OR. A reference to the manual itself may be sufficient. If the manual is not so explicit, then reference to a reliable source that comments on the manual and its definition of plagiarism may indeed be desirable, but such a reference need not comment specifically on Smith and Jones as stated in the article. — Preceding unsigned comment added by 163.1.246.64 (talk) 13:54, 4 July 2012 (UTC)

Although Smith accused Jones of plagiarism, he did not accuse him of violating the practice recommended in the Harvard manual. If Jones were formally accused of plagiarism by Harvard, they may present another manual whose rules he violated. Or an author may sue Jones based on state or federal laws. Perhaps Smith meant that Jones' actions were plagiarism because he used another author's ideas without citing that author. These are determinations that we cannot make. TFD (talk) 16:32, 7 July 2012 (UTC)

Indeed, but I don't think they're relevant here. The supposed OR is not commenting on the intentions of Smith or on the actual actions of Jones. It discusses how a hypothetical action ("If Jones ...") relates to the actual manual. If interpretation of the manual is difficult, then a reference for that is desirable, but no specific reference to the actual Smith or Jones is required, which is the main claim being made. — Preceding unsigned comment added by 82.10.109.67 (talk) 14:14, 5 August 2012 (UTC)

WP:WAF conflict with WP:NOR?

I am confused. Recently I added this talk section (please read) for the TV series The Big Bang Theory. In it, I warned that I would delete a section in the article that contained about 200 lines. My rationale was that this section was either (a) almost entirely based on the episodes of The Big Bang Theory and that is regarded (see WP:TVPLOT) as a primary source, which allows no interpretation (see WP:PSTS), or that (b) the majority of statements in the section should have inline citations to secondary sources.

However, the WP:WAF guideline, which covers classic plays as well as comic books, talks about in-universe perspective, based on the work itself. It implies that we can use in-universe perspective for fictional works, with limitations. This is even more egregious on the talk page (for example, see this). WP:NOR says nothing about in-universe perspectives or their effect on primary sources.

The policy under WP:NOR seems to be:

Any interpretation of primary source material requires a reliable secondary source for that interpretation. A primary source may only be used on Misplaced Pages to make straightforward, descriptive statements of facts that can be verified by any educated person with access to the source but without further, specialized knowledge. For example, an article about a novel may cite passages to describe the plot, but any interpretation needs a secondary source. Do not analyze, synthesize, interpret, or evaluate material found in a primary source yourself; instead, refer to reliable secondary sources that do so.

Normally, I would say that policies trump guidelines but the history of WP:WAF gives me pause. Allowing some in-universe perspective is not a new idea. The majority of articles about fiction do use in-universe perspective for things other than the plot.

Does a "straightforward, descriptive statements of facts" include fictional "in-universe perspectives", which, of course, must be based on the literary text, a primary source? I would like a clear statement in WP:NOR about using the work itself for fiction.

--RoyGoldsmith (talk) 17:05, 13 July 2012 (UTC)

The plot guideline seems to be well accepted. According to that one can put in a short neutral description of a plot. The in-universe bit only extends to describing things and one shouldn't assume too much in-universe background. You can say 'he erected a psychic shield' without worrying about the real world existence of such things, but one should be careful to avoid interpretations like 'in a satire of Soviet bureaucracy' unless they have been made very obvious or there is a secondary source. This is interpreted as being in line with only using primary source for facts. Dmcq (talk) 18:53, 13 July 2012 (UTC)

First, where would I find the "plot guideline" you refer to? Second, I would like to get your opinion. Which of the following article sections would need secondary sources:

• Characters, see Hawaii_Five-0#Main_cast or Alphas#Main_characters;
• Settings, see General_Hospital#Setting;
• Themes, see The_Big_Bang_Theory#Elements_of_the_show;
• Season summaries (in addition to episode summaries), see Private_Practice_(TV_series)#Series_Overview

I've taken all the examples from TV series; believe me, comic books would be far worse.

As a comparison, look at The Simpsons featured article. It has secondary source inline citations for Characters, Setting and Themes, as well as Hallmarks and Culture. --RoyGoldsmith (talk) 03:31, 14 July 2012 (UTC)

Non-interpretive summaries of primary works are completely acceptable. If you can find secondary sources to back it up, great, but it's generally not required if you're summarizing a single work. But as you move towards series-wide themes for things like TV shows or comics, broader claims may become a bit more interpretative depending on how they are taken, and secondary sources should start being used more. But again, if the claim of a broad scope is obvious to any reasonable person that would have access to the primary source (eg, the BBTheory has a lot of science-related humor) it doesn't need sourcing. --MASEM (t) 03:54, 14 July 2012 (UTC)

See WP:PLOTSUM for the plot summary guidelines. Dmcq (talk) 04:27, 14 July 2012 (UTC)

As I said before, I only want to bring NOR policy in line with current consensus. In my opinion, all statements within Misplaced Pages guidelines and essays should be based on some policy (or, failing that, the lack of any applicable policy). How about the following change (in boldface) to one sentence of WP:NOR that I quoted above:

Any interpretation of primary source material requires a reliable secondary source for that interpretation. A primary source may only be used on Misplaced Pages to make straightforward, descriptive statements of facts that can be verified by any educated person with access to the source but without further, specialized knowledge. For example, an article about a novel may use plot elements to describe the characters, themes, etc., but any interpretation (such as literary criticism) needs a secondary source. Do not analyze, synthesize, interpret, or evaluate material found in a primary source yourself; instead, refer to reliable secondary sources that do so.

--RoyGoldsmith (talk) 13:33, 15 July 2012 (UTC)

That's just unnecessary cruft. More is less with cruft. Dmcq (talk) 01:22, 16 July 2012 (UTC)

Actually about 95% of Misplaced Pages (everything except direct quotes) violates a rigorous literal interpretation of wp:nor/ wp:synth. But the de-facto policy is to bend it a little, allowing straightforward neutral summarization. North8000 (talk) 17:36, 15 July 2012 (UTC)

You've said that for years, but nobody agrees with you that everything eccept quotations and copyvios is a NOR violation. In fact, the third paragraph of the lead to this policy directly tells you that you are supposed to summarize sources in your own words, while retaining the meaning of the sources. WhatamIdoing (talk) 02:43, 16 July 2012 (UTC)

I understand what North8000 is saying. He is talking about an extreme interpretation where every editorial action should be based on secondary sources. The answer is that the level of synthesis that is forbidden is that which would be considered creative. In other words, WP:NOR says that editors are not allowed to do themselves the sort of transformation that converts primary source material to secondary source material. --SmokeyJoe (talk) 03:47, 16 July 2012 (UTC)

Actually, it is my impression that North believes that "everything except direct quotes" is a violation of a strict reading of NOR. Like I said, nobody else agrees with him, but he does seem to sincerely believe that summarizing a multi-page source into a single sentence or paragraph, even if your summary is obviously a plain, neutral summary of the source, "violates a rigorous literal interpretation of wp:nor", to use his words. WhatamIdoing (talk) 00:26, 18 July 2012 (UTC)

Given that WP:NOR, despite being "core policy", is flexible and not prescriptively written, in the phrase "is a violation of a strict reading of NOR", I'd equate "strict" to "mis-". WP:NOR is not intended to be read that way. I'd ask North8000 if many editors on the ground are reading it that way. --SmokeyJoe (talk) 00:55, 18 July 2012 (UTC)

• I think that Roy has merely rediscovered that applying the advice of WP:WAF causes violations of WP:NOR to become obvious. Go back to the previous in-universe form and look carefully; it would have contained the same OR only better disguised. --SmokeyJoe (talk) 03:53, 16 July 2012 (UTC)
  • It's not obvious; NOR doesn't depend on if its in- or out-of-universe. (that's a matter of writing style than the correctness of the information.) You need to be a more explicit to say what the problem is, as as give, WAF follows standard practice for using primary sources to summary primary works. --MASEM (t) 05:33, 16 July 2012 (UTC)
    • When something is written in-universe, it is very easy to include multiple references from the work of fiction itself to support an analysis. When writing out-of-universe, the work-of-fiction references no longer suit, and then the lack of references makes it obvious that some synthesis is being presented. --SmokeyJoe (talk) 06:06, 16 July 2012 (UTC)
      • That's not what "in-universe"/"out-of-universe" distinction is. OOU basically means to discuss the plot/characters as the viewer sees them, not as the characters see them. Say a character holds a secret that the viewer is obviously aware of, but no other character in the work knows this until the end of the work. In our summary of it, we can expose this secret, describing it without being coy about it. OOU writing most commonly manifests itself in being reasonably creating in untangling simultaneous-running plot lines into a single one, putting events in a order that is easiest and more concise to explain than the scene-by-scene of the work itself. But in-universe/OOU has zero to do with anything about sourcing. --MASEM (t) 12:59, 16 July 2012 (UTC)
        • You seem to be disagreeing with me, but I'm not sure with what. What you say is correct. I didn't mean to repeat any of the "in-universe"/"out-of-universe" distinction, WP:WAF explains it well. I think what I was saying is that in-universe writing makes pseudo-sourcing easier. --SmokeyJoe (talk) 03:00, 17 July 2012 (UTC)
          • Yes, I am , as you're implying that to write out-of-universe requires extra sourcing beyond the work itself. This is not true. A properly written out-of-universe summary is sufficiently sourced to the primary work, period. We only encourage this because when writing in-universe, it encourage original research from that perspective, not because we are requiring more sourcing. --MASEM (t) 03:51, 17 July 2012 (UTC)
            • Every fiction article should have sourcing beyond the work itself. The plot summary is an obvious exception. I didn't think we were talking about plot summaries. --SmokeyJoe (talk) 00:50, 18 July 2012 (UTC)
              • There is no question, per NOT#PLOT, that we need sourcing beyond just a plot summary. But that is not what we talk about when we talk out-of-universe; out-of-universe describes the type of tone one takes when writing plot summaries, or more particularly, when writing on characters and larger meta-plots. We write as a distanced all-knowing viewer, not as invested character. Both tones still can rely on the primary work and require no additional secondary sourcing for supporting the plot summary. Everything else in a fiction article, heck yes, we need better sourcing but that's not the original question poised here. --MASEM (t) 01:08, 18 July 2012 (UTC)

It says 'but any interpretation needs a secondary source.' Nothing about in universe there. Dmcq (talk) 08:40, 16 July 2012 (UTC)

"Interpretation with a secondary source" will get you to the right place nicely. Few reliable independent secondary sources are in-universe. The in-universe type advice can be helpful in parallel. If in doubt, note that one is policy. --SmokeyJoe (talk) 03:00, 17 July 2012 (UTC)

Responding to WhatamIdoing, I stand by what I said, at least with respect to the nuts-and-bolts operative clauses of wp:nor/wp:ver. It's also true that the "intention" clause that you cite says that summarization is OK. That, plus common sense is what makes this work....most of the time. But once it degenerates into e.g conducting a POV war via wiki-lawyering, the nuts and bolts operative clauses always win out over the "intention" ones, and the warrior can use the fact that summarization in some respect always violates those clauses to knock out any material that they wish. Think about what summarization actually is. It is synthesis combined with selectively leaving out material. (The latter is fully legal despite some rumors otherwise) Like most policies it work most of the time but is prone to misuse, also a frequent occurrence. Tweaks would reduce the latter. Sincerely, North8000 (talk) 12:12, 16 July 2012 (UTC)

It's the cases where people say 'Oh I see now' what it means where one should be thinking about rephrasing to cope with misunderstanding. If they argue after having it explained a number of times then explaining better is probably not a solution. Dmcq (talk) 13:29, 16 July 2012 (UTC)

Not sure that I understand. But the problematic cases for most policies aren't cases where someone doesn't understand the overall policy. These are people who misuse operative clauses to suit their own agenda. Violations of policies are quickly corrected. Mis-uses are more common because they don't get corrected and just keep going. North8000 (talk) 14:10, 16 July 2012 (UTC)

Dmcq, I agree with you: Misplaced Pages has gotten far too bureaucratic for my taste. The primary lesson I've gotten from all this is that you have to judge truth along with verifiability before you challenge anything. If you believe something is true, it probably doesn't need for you to insert a CN-tag.

This is different than BLP and most articles on current politics (see US debt-ceiling crisis, Tea Party movement or Abortion). Unlike fiction, in these articles practically every statement requires a source. Once you have a source (or sources), you may spend endless hours discussing reliablity and weight before you get consensus of the active contributers to add one sentence. In fiction, my guess is that we have far fewer editor eyes to review and much less contentious opinions to referee.

I'm mostly from this world of writing about politics in Misplaced Pages. I'm still wondering whether we can do something in WP:NOR that makes it less likely that someone like me may interpret the definition of primary sources too legalistically. How about adding one word:

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

The rest would remain the same as it is now. Other words we could use in place of creative might be: imaginative interpretation, original interpretation, inventive interpretation, novel interpretation or something else. What's your idea? If I had seen something like this in the NOR definition of primary sources, I'd have been a lot less prone to starting this whole rigmarole in the first place. --RoyGoldsmith (talk) 00:37, 17 July 2012 (UTC)

I just had a good look through the guideline writing about fiction and it seemed okay to me. However when I looked at one of its exemplars Gulliver's Travels it did seem to be badly lacking in citations in some important places. The section Gulliver's Travels#Major themes has no citations at all even though it is pure interpretation and there must be loads of people who have analyzed the work. My guess is the problem is with getting editors to implement the guideline. Dmcq (talk) 09:35, 17 July 2012 (UTC)

I don't think that adding the word "creative" is going to do anything except cause problems from people who are obviously going too far, but who say that their interpretations are so obvious that they're not creative and are therefore okay.

What we really need is for a few people to understand that "interpretation" is about assigning a meaning to something, not simply describing it. For example: Georgia O'Keeffe made a famous painting of a cow's skull on a red, white, and blue background. You may describe the painting, using the painting itself as your source: rectangular painting containing one cow's skull, centered, on a tricolor background. You may not "interpret" the painting: you may not talk about even the most "obvious" (to you) and "uncreative" meanings of the painting's contents. There might well be patriotic symbolism the choice of the colors, but you need a secondary source to write that. The skull may make you think of your own mortality, but you need a secondary source to write about that.

Description is not interpretation and is not banned. WhatamIdoing (talk) 00:35, 18 July 2012 (UTC)

I agree with WhatamIdoing. Non-creative interpretation is so narrow as to not be worth mentioning, and the problems introduced outweighs benefit. "Interpretation" already implies a degree of copyrightable creativity. Is it OK me for to copy your interpretation and present it elsewhere, without attribution, as if it is my interpretation? I think not. While there is a low-end to interpretation, which is not creative, it is probably better described as description. Someone might interpret the cental thing in Cow's Skull: Red, White, and Blue as looking like an animal skull. This would not be creative, but I think that in this case "interpret" is an exaggeration of "describe". --SmokeyJoe (talk) 01:12, 18 July 2012 (UTC)

I think that a good tweak be to more deeply embed "summarization is OK" into the operative parts of wp:nor & wp:ver, especially when the content of the summarization is not questioned. North8000 (talk) 01:14, 18 July 2012 (UTC)

I think Potter Stewart's comment regarding pornography applies to "interpretation" as well. It's very difficult to codify precisely where the line is overstepped between "describing what happens in the plot" and "analysing what happens in the plot" sometimes, but it's rarely difficult to tell the difference in a given bit of prose. (Or at least it is to me. YMMV, and obviously this is a major problem in almost all non-stub articles on fiction on en-WP and always has been.) It may be best tweaking the guideline so as to note this effect and to advise editors who disagree on whether a given summary strays too far into personal interpretation to take it to talk. Chris Cunningham (user:thumperward) (talk) 12:54, 18 July 2012 (UTC)

I tend to think that when people are sincerely questioning the content itself as overstepping summarization, there usually is really an overstepping problem. When they don't really question the material itself but just quote rules to knock out the material, now we've gotten into where the inconsistency that I pointed out above becomes problematic. That's why I think that this rule (and others) would work better if a part of the process of invoking the rule is to question the material. This is just procedural, not a criteria for the outcome. So, the person raising the question does not need to debate the question nor win a debate on it, they just need to just raise it. North8000 (talk) 20:21, 19 July 2012 (UTC)

Getting back to my original problem (what should we do about The_Big_Bang_Theory#Elements_of_the_show), I still have trouble with NOR policy being understood one way for fiction and a totally different way for, say, current politics. I think the difference is that a work of fiction is the primary source itself while, in other articles, you're talking about primary sources that describe the article's subject but are not the article's subject itself.

I examined various other, non-fictional works that were themselves the subject of their article; for example, the Bible and Common Sense (pamphlet), among others. Almost all described their subject based on the work itself. (For example, see this section in Common Sense.)

How do we allow this difference to be expressed in the NOR section that I quoted up above? If for no other reason than preventing anal-retentives like me from starting this discussion all over again. --RoyGoldsmith (talk) 01:57, 20 July 2012 (UTC)

On regards to the specific example of that section of the Big Bang Theory, you're absolute right that most of its problems in NOR. One station "The show centers around physics" (paraphrasing) is actually ok - you only have to watch a few episodes of the show to come to this obvious conclusion. But most everything else in that section is saying "The show has theme X because of these examples." without sourcing. As we now talking about the show and not the narrative itself, that steps out of the primary sources and requires secondary sourcing. Contrast this to the Themes section of Inception, which does supply appropriate sourcing to discuss the meta-details. So to your point, when talk about the broader scene of a work and not just the events as they happen on screen, we do expect better sourcing than just the primary work itself. But when you're just talking the narrow aspect of the events of one episode or a faith summary of the plot of a movie, the primary work is sufficient. --MASEM (t) 02:23, 20 July 2012 (UTC)

I'd agree that the material which you are discussing falls within the 95% of Misplaced Pages that is in violation of a strict/literal interpretation of some operative clauses of wp:nor/wp:ver. Assuming we aren't going to start deleting that 95% the question becomes a reasonable interpretation per the norms of Misplaced Pages and the "intent" and other portions of those policies. This one is a bit more out there than the norm because much of it is is a (IMHO excellent) summary from primary sources. My opinion/advice is that if you dispute the summary (or parts of it) itself, then do so and bring out the rule book; if not don't. North8000 (talk) 11:28, 20 July 2012 (UTC)

What I'm looking for is a (hopefully small) change to WP:NOR's policy on primary sources. That is the paragraph under Primary sources that begins with Policy.

I think we've all agreed that this policy doesn't state explicitly the use of primary sources in fiction and other, written works. (If you think it does, please, show me where.) Talking about common sense doesn't work because my common sense may be different than yours. (For example, I would still basically delete the entire The_Big_Bang_Theory#Elements_of_the_show section as well as Common Sense (pamphlet)#Paine's arguments against British rule, given the current NOR policy.) I believe that the policy should be changed but my belief doesn't make it policy. --RoyGoldsmith (talk) 13:22, 20 July 2012 (UTC)

Arguably what you are looking for is already under WP:SYNTH - we can't come to conclusions from facts from primary sources that are otherwise not obvious. We can add the entirity of the Big Bang Theory episodes to conclude "hey, they make a lot of jokes about physics", that's a non-SYNTH, obvious conclusion - but also one I suspect that it would take a trivial amount of time to search and verify with secondary sources. However, there are plenty of other statements that add 2 or three episodes together and claim long-running themes and that may be more difficult to source; that would be SYNTH. SYNTH as it is is fine, what probably is needed is clarification in WAF that points back to SYNTH to say that whenever you start to generalize a work of fiction (moving beyond the plot summary), secondary sources start to become required to do that.

Also, as a point: we prefer not to delete material if it doesn't immediately fail core problems like BLP or NPOV, particularly if the material is likely true and just lacks sourcing. The section in Big Bang Theory screams for the need to sourcing, and if I were taking it to Good Article or better, I'd either find sources or remove that section before GA, but as it is developing, it makes no sense yet to remove; the claims made all appear to be true, but just presently are technical NOR violations. --MASEM (t) 13:51, 20 July 2012 (UTC)

The policy on primary sources states (emphasis mine) "Do not analyze, synthesize, interpret, or evaluate material found in a primary source yourself; instead, refer to reliable secondary sources that do so." I therefore conclude that the examples in WP:SYNTH refer to reliable secondary sources.

My problem is that primary sources are used differently in articles that have written works (including fiction) as their subject. In that case, the written work is treated (and, IMO, deserves to be treated) as the most important source -- the canon, if you will. But nowhere in NOR or the other core policies does it say so, either explicitly or implicitly.

Let's say that you found an article in a reliable secondary source (bear with me) interviewing a writer of The Big Bang Theory and he said "I know many people think that the show focuses on science, especially physics, but they are wrong. We could have used accountants or computer programmers or even lawyers...". Would that source, simply by being secondary, make the current Misplaced Pages material (based on a primary source) wrong? If anyone thinks no or thinks they could weasel around it (for example, by saying that the writer really meant to say that "the show must focus") then I think they're engaged in the original synthesis: that they think that their opinion is as good as the writer's.

And I would agree with them. That's if we were just having a conversation. But I don't think anyone should have a discernible opinion when acting as a Misplaced Pages editor. Even if just "summarizing".

On your second point, I agree: we ought to move slowly, especially about deleting material. I am a firm inclusionist- I hate removing anything. Why else would I be having this involved discussion? But do you believe that the section in the BBT we've been discussing is based on anything other than its primary source or that, given time, the vast majority of the material can be sourced? The warnings about have been posted for a year and a half. If they were going to find secondary sources, wouldn't they have done so already? --RoyGoldsmith (talk) 22:25, 23 July 2012 (UTC)

The short answer is that we already have plenty of advice in NOR, WAF, NOT#PLOT, and elsewhere on putting too much reliance on primary sources, and using them to make claims outside of summarizing a work and for simple, obvious WP:SYNTH, in the absence of more authoritative sources to say otherwise. As in your example, if a writer confirms something that is completely against the obvious interpretation, we go with the authority, not the interpretation.

As to the BBT section in question, I don't know enough about the show to know if those are accurate but they do seem to be reasonably truthful statements that otherwise aren't being put there to create bias or falsify info. I wouldn't be making an article a Good Article with that section as is, and would ultimately require better secondary or third-party sourcing to affirm that or otherwise remove the claims; but as a currently airing/produced TV show, I would expect one shouldn't have a problem to find sources, so I wouldn't be quick to remove it. A better example I can vouch for is Mythology of Fringe which I created, and does have blocks of text without direct sourcing and relies on the primary. I've not spent a lot of time with it, but I do know that if someone came and begged for improvements from it from, I know I can source every element give from secondary third-party sources that talk about the show, as well as information from the producers/directors. Now, if we were suddenly talking about an old TV show, say, "I Love Lucy" and someone started throwing up similar types of thematic elements based only on the primary source, I'd beg to question if that is synthesis and if they would ever be able to find sources for it since the show's long off the air and production and many of the original people working on it have passed away. It's basically a matter of giving editors the benefit of the doubt. Basically, if it doesn't look like the material is a total fabrication or otherwise harmful to WP, but only begs for better sourcing, it is better to leave it in place and nudge those editing it to get the improvements in place. --MASEM (t) 23:50, 23 July 2012 (UTC)

OK, you've convinced me. Temporarily and tentatively but I'm convinced. For now. If you can get Mythology of Fringe to survive for more than a year without being shortened by 75% I guess I have to agree. For the time being, I'll only add or replace the episode templates in the Elements of the show section with non-primary source needed tags.

In "A primary source may only be used on Misplaced Pages to make straightforward, descriptive statements of facts...", my interpretation of "facts" was that fiction is not and could not ever be, by definition, facts. So that any synthesis beyond the sequential plot elements would have to be "interpretation". This was such an ingrained aspect of my psyche that I didn't realize it before now. I guess what the policy means is that "facts" can include fictional, in-universe facts as well.

This was acerbated by having the bulk of statements in the Elements of the show section rely on too many episodes. You can't specify episode numbers for "Much of the show focuses on science, particularly physics" because (IMO) two-thirds of the episodes support that (in-universe) "fact".

There is another, related point I want to make but, since it's another point, I'll make it in another talk section below. Thanks again. --RoyGoldsmith (talk) 15:15, 29 July 2012 (UTC)

IUPAC names for chemicals, especially for drugs

The International Union of Pure and Applied Chemistry (IUPAC) has a nomenclature for chemical substances, so that one can derive a name if the chemical structure is given. (Resulting in names such as "2-Amino-4-(ethylcarbamoyl)butyric acid" for theanine.) For the sake of completeness: sometimes there is more than one way to derive a name for a structure. Both {{chembox}} and {{drugbox}} have fields for IUPAC names, and one editor has started fact-tagging these a while ago. Now the problem is that the obvious sources (PubChem, Chemspider) aren't reliable, and articles in medical journals generally don't bother with chemical nomenclature. Even articles in chemical journals often have the structures but not the names. Deriving IUPAC names by ourselves is OR. Any way to solve this? --ἀνυπόδητος (talk) 11:19, 16 July 2012 (UTC)

It should be noted here, that in principle any creation of those names is 'original research' in a way (also outside Misplaced Pages) - the IUPAC has schemes (an algorithm) with which you can generate the name of a compound based on structure - they do not check or validate the names that one is giving to a compound - anyone can generate a name, publish that in a source and it is to the discretion of the peer review systems whether the names are actually checked (I have never seen that happen, and the reviewers would follow the same schemes anyway). --Dirk Beetstra 11:41, 16 July 2012 (UTC)

In violation of WP:BEANS, I'd like to mention here that SMILES and InChI (both having fields in chembox and drugbox) are also text representations of chemical formulae that can (and generally have to be) computed from the structures, so the mentioned problem applies to these as well. (Thanks to Dirk Beetstra for the tip.) --ἀνυπόδητος (talk) 13:49, 16 July 2012 (UTC)

The core of the problem is that generating systematic names from anything but simple chemical structures is complex. The set of rules (or more accurately, guidelines, I believe) described by IUPAC are not sufficiently comprehensive to make unambiguous IUPAC name assignments for complex multi-functional chemical compounds. There are plenty of computer algorithms out there that do a decent job, but they do not produce the same results in all cases. And since IUPAC does not produce or endorse any official algorithm (or other method), as far as I know, Misplaced Pages and other databases must work with what we have. If there is a way to determine a database that has the most reliable IUPAC names, we should go with that; but I'm not confident that could determined. In the absence of any "official" IUPAC name authority, perhaps we should just dispense with IUPAC per se, and simply go with a more general "systematic name", meaning any name from which one can unambiguously derive the correct chemical structure. In the mean time, we shouldn't be putting {{citation needed}} on any IUPAC name if we do not recognize any sufficiently reliable source for such a name. -- Ed (Edgar181) 17:31, 16 July 2012 (UTC)

But even "non-IUPAC systematic names" will be OR unless we find a reliable database, which is unlikely given the existence of computer algorithms. --ἀνυπόδητος (talk) 17:34, 16 July 2012 (UTC)

Yes, but I'm not too concerned about the OR aspect. OR concerns are important where subjectivity is an issue, but systematic chemical names can be objectively and readily verified by widely available chemistry drawing software or at various websites. -- Ed (Edgar181) 17:41, 16 July 2012 (UTC)

I'm concerned about OR because dozens (or more) systematic names have been fact-tagged, and neither PubChem nor ChemSpider are reliable sources. So maybe the question is whether using chemistry drawing software for verification counts as OR. --ἀνυπόδητος (talk) 17:46, 16 July 2012 (UTC)

Let's put a real example into your question: If I take the IUPAC name "2-Amino-4-(ethylcarbamoyl)butyric acid" which is currently in theanine and tagged with {{citation needed}} and I plug it into this webpage from Univ Cambridge which parses IUPAC names, it returns a chemical structure that matches what is in our article. Is that original research by Misplaced Pages standards, or not? And is that sufficient verification to remove the "citation needed" tag? -- Ed (Edgar181) 19:11, 16 July 2012 (UTC)

• I came here from a note at WT:WikiProject Pharmacology. I don't really see a problem with OR in this case. It's not any more an OR violation than would be computing the molecular weight of a compound from the periodic table masses of the individual atoms, or converting a measurement to or from metric. The principle of OR is to prevent editors from passing off opinions and pet theories as facts. To reply to Edgar's specific example, I think it would be reasonable to remove the cn tag, while best practice would be, instead, to provide an inline cite to a book that contains the structure, which I bet exists somewhere. --Tryptofish (talk) 21:57, 16 July 2012 (UTC)

If no reliable publications on a chemical use its IUPAC name, why do we need to give one? Someguy1221 (talk) 22:01, 16 July 2012 (UTC)

It's not really the case that no reliable publications use IUPAC names - most chemical databases will list IUPAC names. The problem is that reliable sources can differ in the ways they determine the IUPAC name and Misplaced Pages chemists have not determined which otherwise-reliable sources are sufficiently reliable for IUPAC names. -- Ed (Edgar181) 23:27, 16 July 2012 (UTC)

I hope you all remember how the whole situation started. After this we are back full circle. --Dirk Beetstra 04:18, 17 July 2012 (UTC)

I was contributing to the section above and I lost my place and started reading your section. As a complete ignoramus of chemistry and pharmacology, may I give two-cents. First, NOR only applies when you think something is wrong and the first step is to request reference(s). Do you think that "2-Amino-4-(ethylcarbamoyl)butyric acid" is wrong for theanine or that it's one of many IUPAC names? If the latter then why not add all valid names in a list, with an inline citation, if anyone requests it? Isn't this what WP:WEIGHT is all about?

This is just an idea and I freely admit that my lack of knowledge about the subject may make my opinion moot. --RoyGoldsmith (talk) 15:27, 17 July 2012 (UTC)

We've no idea whether Plasmic Physics thinks this name is wrong, he just goes around and adds fact tags to chemical names. Everything else (there being several possible names and all) isn't a problem in my opinion. --ἀνυπόδητος (talk) 17:14, 17 July 2012 (UTC)

Actually, I think that RoyGoldsmith has it right. If, hypothetically, someone has a real factual doubt about a particular chemical name (apparently not the case here), and they discuss it on the article talk page, then it should be sorted out. But if someone is just going around adding cn tags, it's probably worth asking them why, but if they don't have a specific reason, it's reasonable to revert them and request that they stop.

(Another example occurred to me. If you look at the end of Theanine, there are a bunch of interwiki links to Wikipedias in languages other than English. In all likelihood, those links got there from editors translating the title into/from other languages, but it would be a stretch to call that translation original research as it's defined here. We don't expect someone to cite a dictionary that such-and-such is the (language) word for theanine, and it would be silly to put a cn tag on such a link. If someone realizes that the link is incorrect, then correct it, but there's no need to worry about OR.) --Tryptofish (talk) 20:39, 17 July 2012 (UTC)

As a technical matter, if any reliable source in the world has ever claimed that this is the (or "an") IUPAC name for a given chemical, then there is no NOR violation. Providing a citation proves there is no violation, but if such a source exists, then there is no NOR violation even if there is no citation.

I would like to know why this person believes that his tagging is helpful. WhatamIdoing (talk) 00:42, 18 July 2012 (UTC)

I agree that there is no NOR violation here. For any given chemical structure, there are a number of valid IUPAC names that can be unambiguously used to re-derive the structure (many to one relationship). There is also something called the Preferred IUPAC name, which is supposed to be a unique, canonical name for the structure (one to one relationship). The fields in the chembox (IUPACName) and drugbox (IUPAC_name) are for any valid IUPAC name and does not mandate a preferred name. As long as an chemical software package can regenerate the structure from the supplied IUPAC name, that should be sufficient justification for including the name in an infobox and no citation is required. Boghog (talk) 05:49, 18 July 2012 (UTC)

Do note, that the rules for the Preferred IUPAC name did not finish yet (I am 'following' the discussions that should lead to them), the official guidelines/rules for that have not been published in a finished form. --Dirk Beetstra 05:53, 18 July 2012 (UTC)

As I said, we are going full circle. This situation started a long time ago (I'll try and find the discussions), where self-generated names were put on chemicals, where there were serious concerns that those names were not correct and many were reverted. IIRC, that self-generation of names was deemed original research, and hence, people were asking for references for those self-generated names ... --Dirk Beetstra 04:01, 18 July 2012 (UTC)

Oh, it is too much to read - many threads in Wikipedia_talk:WikiProject_Chemicals/Archive_2010

• -> Wikipedia_talk:WikiProject_Chemicals/Archive_2010#user:Plasmic_Physics
• Wikipedia_talk:WikiProject_Chemicals/Archive_2010#user:Plasmic_Physics_.28continued_from_above.29
• Wikipedia_talk:WikiProject_Chemicals/Archive_2010#Controversial_use_of_the_term_.E2.80.98Oxoazinic_acid.E2.80.99_in_the_chembox_of_Nitric_acid
• Wikipedia_talk:WikiProject_Chemicals/Archive_2010#Plasmic_Physics_edits
• Wikipedia_talk:WikiProject_Chemicals/Archive_2010#Block_for_Plasmic_Physics
• Wikipedia_talk:WikiProject_Chemicals/Archive_2010#Problematic_use_of_SystematicName_for_Beryllium_oxide
• Wikipedia_talk:WikiProject_Chemicals/Archive_2010#Substitutive_IUPAC_names_are_most_often_not_appropriate_for_inorganic_compounds
• -> Wikipedia_talk:WikiProject_Chemicals/Archive_2010#OR_introduction_of_systematic_names
• Wikipedia_talk:WikiProject_Chemicals/Archive_2010#Cleanup_of_overspecialized_silanes_categories_created_by_User:Plasmic_Physics

moving into 2011

• Wikipedia_talk:WikiProject_Chemicals/Archive_2011#More_overspecialized_categories
• Wikipedia_talk:WikiProject_Chemicals/Archive_2011#List_of_hidden_systematic_names
• -> Wikipedia_talk:WikiProject_Chemicals/Archive_2011#Systematic_Name

(I may finish this list later, but there are more in the archives of WT:CHEM). I've marked some with a <-, which are the ones which are of interest. --Dirk Beetstra 04:16, 18 July 2012 (UTC)

Responding to Tryptofish 20:39, 17 July: A minor point but according to WP:V, it's not legit to revert CN-tags (see WP:BURDEN):

The burden of evidence lies with the editor who adds or restores material. You may remove any material lacking an inline citation to a reliable source.

In other words, the person who objects doesn't really have to insert CN's -- that's just a courtesy. They can delete any statement without an in-cite immediately, just like BLPs. Then it's up to the editors who believe in that IUPAC name to find a reliable reference.

Reponding to Dirk Beetstra 04:16, 18 July: OK. Then what would you like as a resolution to this section? Has ἀνυπόδητος gotten his answer? --RoyGoldsmith (talk) 00:43, 19 July 2012 (UTC)

Yes, it's a minor point, but sometimes BURDEN becomes too much of a burden, the way it gets interpreted. I agree that it's bad form simply to revert the tag when a cite is legitimately needed, but it seems to me to be perfectly reasonable to revert one that was placed either in error or stupidly. Please note that I did specify that one should attempt to consult with editor placing the tag, before considering reverting it. --Tryptofish (talk) 20:16, 19 July 2012 (UTC)

Sometimes you really should revert fact tags. The rules about not rewarding vandals and pointy behavior still apply. WhatamIdoing (talk) 20:33, 19 July 2012 (UTC)

I love that diff! It's on my list of the best one, two, three, four.... --Tryptofish (talk) 21:19, 19 July 2012 (UTC)

Eyes

I'd appreciate it if some more people kept an eye on Template:Content policies for a little while. It seems to be attracting links to pages that aren't content policies. WhatamIdoing (talk) 16:33, 23 July 2012 (UTC)

Too many primary sources?

The sections below are all comparisons of films to their sources (the novels or plays that the movies were based on). These sections, which present significantly different styles, have no, or very little, inline citations.

• The Hobbit (1977 film)#Comparison to the source material
• Annie (1999 film)#Comparison to 1977 musical and 1982 film
• The Lord of the Rings: The Two Towers#Comparison to the source material

The theory has been advanced that these sections (and other ones like them) should be removed in their entirety because, since they were based on at least two primary sources (the work itself and its antecedents), they are obviously interpretations of those two sources and thus, original research and/or synthesis.

Do you think that these sections should be totally removed because of WP:NOR? Why? Would you change them? How? Or would you indicate by tags that they should be changed? If so, for how long? --RoyGoldsmith (talk) 16:51, 29 July 2012 (UTC)

This is probably covered by WP:PLOT... briefly noting differences in plot between two versions of the same story (purely descriptive statements, involving no analysis, interpretation or conclusion) is not OR. Blueboar (talk) 13:32, 1 August 2012 (UTC)

Technically true, but I do find that it is very easy for readers to either slip in every tiny change, or start engaging in subtle OR here. My opinion here is that such sections border on trivia, and what differences to include should be guided by third-party sources. --MASEM (t) 13:42, 1 August 2012 (UTC)

Replaced with the correct text from the consensus version of WP:V?

I still think this is premature. Certainly the closing statement did not say that "VnotT" would be removed. It says "there remain open questions about the exact status and place of the phrase "Verifiability, not truth", and recommend that the community continue discussion on these points. In particular, jc37 considers that there is consensus for option D only insofar as it forms a basis for such continued community discussion." Having said that, as this is just one small paragraph, perhaps there is no problem. Dougweller (talk) 15:43, 31 July 2012 (UTC)

We should definitely think about this carefully and go slowly. I don't object to rewording in light of the recent RFC at WP:V, but we do need to remember that the koan "The threshold for inclusion in Misplaced Pages is Verifiability, not Truth" directly relates to this policy... Indeed that specific wording originated here in WP:NOR. It was stated here before it was added to WP:V.

If we think of VNT as a statement about Original Research (as opposed to it being a statement about Verifiability), I think the koan is still conceptually valid (an unverifiable truth is excluded - on the grounds that it is OR.) Blueboar (talk) 12:34, 1 August 2012 (UTC)

The closing statement was to use option "D". Wp:ver and wp:nor are a 90% overlap of each other and wp:nor certainly needs to be updated. Further, the item updated here is a statement of what is IN wp:ver. It certainly must be a correct (and thus updated) statement of what is in there. Sincerely, North8000 (talk) 12:51, 1 August 2012 (UTC)

Which is why I am not opposed to discussing a rewording... what I am suggesting is that the koan (or something very like it) might still have a place in this policy... not as a restatement of what is stated at wp:ver, but as it originally was - a statement to clarify an important aspect of NOR. Blueboar (talk) 13:14, 1 August 2012 (UTC)

I have to admit to being a little surprised at my own reaction to this issue. After all, I'm a long-time supporter of VnT, and I'm a big believer in taking things slowly and discussing them fully. But nonetheless, I agree with S Marshall's removal of the sentence! The purpose of the "other policies" section is to point the reader to what those other policies currently say. And that phrase has now been deprecated to a footnote, so it shouldn't be where things start off here. But taking this discussion a step further, I wonder why we even have a lengthy section about V and NPOV here at all. Why not state briefly, as in the V lead section, that the policies work together, and link to them, and leave it at that? --Tryptofish (talk) 13:40, 1 August 2012 (UTC)

Truth be told, these policies are a 90% duplicate of each other and should probably be merged. The "10%" is basically emphasizing the aspect of it's title, which then, as it inevitably should, goes to verifiability. The more realistic (and possibly better) alternative is to start slowly taking out the 90% of it that is a duplication of wp:ver. North8000 (talk) 14:01, 1 August 2012 (UTC)

VNT separate from the "Other Policies" section

I am going to start a new sub-thread here... because my comments above were not really discussing a specific edit, or even the language of the "other policies" section... what I was asking is a slightly different question: Is there a place for the VNT koan (or something like it) somewhere in the NOR policy? (Not in the "other policies" section).

Remember... VNT originated in this policy. It was originally crafted to explain an aspect of Original Research. It was subsequently added to WP:V (in the context of being a restatement of something that this policy said). My question is, regardless of what WP:V now says... does it make sense for this policy to include the VNT koan (or something like it)? If the answer to that question is "yes", then we can move on to discussing the specifics of exactly where and how to include it. Blueboar (talk) 14:31, 1 August 2012 (UTC)

The answer is no. Surely it's fairly obvious that the correct response to the resolution of such an epic battle as has just been had is not "so is it OK if I put it in a different policy, then?". Formerip (talk) 14:33, 1 August 2012 (UTC)

Agree. (except I'd say "decision" rather than "battle") Blueboar, unless I am misunderstanding, this seems very uncharacteristic of you. Sincerely, North8000 (talk) 14:54, 1 August 2012 (UTC)

It isn't a question of moving it to a different policy... as it has always been part of this policy (even before it was added to the WP:V policy). I suppose what I am asking now is whether its depreciation over at WP:V means we should remove it from this policy, or whether we should return it to its former status as a stand alone element of this policy - separate from "other sections"?

As to the personal comment... I am not sure why you find my asking about this uncharacteristic ... I have been arguing all along that VNT is important in terms of understanding NOR. That's why I was initially so reluctant to depreciate it at WP:V. Sure, I (eventually) came to understand why it was problematic at WP:V... but I still think it is important to state here at WP:NOR. I am very flexible as to how we state it (I am not locked into any specific wording)... but as a concept I think it is important to state in this policy and would strongly oppose removing it completely. Blueboar (talk) 15:31, 1 August 2012 (UTC)

(added later) I didn't mean that the viewpoint was uncharacteristic, I meant that the angle you are taking on this is, respectfully, pretty weak and far-fetched, and that is uncharacteristic of you. North8000 (talk) 16:11, 1 August 2012 (UTC)

• Please reassure me that you're joking.—S Marshall T/C 15:45, 1 August 2012 (UTC)

No Joke. Blueboar (talk) 15:52, 1 August 2012 (UTC)

Okay, I'm not sure if I have the heart for this discussion but let's just try it out.

The consensus version that appears at WP:V and now in WP:NOR reads "Even if you're sure something is true, it must be verifiable before you can add it." What's missing from this explanation?—S Marshall T/C 15:58, 1 August 2012 (UTC)

It's even more no-brainer than that. The section that it was in was the "other policies" section, and the wp:verifiability subsection, and it was and is a statement of what is at wp:verifiability, as such. North8000 (talk) 16:08, 1 August 2012 (UTC)

Blueboar, when you say "it has always been part of this policy", do you mean to say that at some point in the dim past WP:V was a section in WP:NPOV, which was then spun out and became a page of its own? (I have no idea, this is just a hunch). Formerip (talk) 16:14, 1 August 2012 (UTC)

No... I am saying that both the concept behind VNT, and the specific phrasing "The threshold for inclusion in Misplaced Pages is Verifiability, not Truth" started out as part of the WP:No Original Research policy... and has remained part of this policy since it was created. The language was originally intended as an explanation of an important policy point (that we should not add unverifiable material, no matter how true it might be... because doing so is a form of OR). It was not created for WP:V and then added here... it was created here and subsequently added to WP:V (as a repeat of what this policy already said)... and it was only after it was repeated at WP:V that it took on a life of its own and became problematic.

To reply to S.Marshal's question... I suppose what is missing is a statement that ties that sentence, conceptually, into NOR... something to clearly explain the reason why we should not add unverifiable material - even if that material is true (the reason being that doing so is form of Original Research). This is the core concept that the VNT koan was written to express.

Thinking further... I suppose I am expressing my concern that we might be tossing the baby out with the bathwater... I am asking people to pause before we rush around deleting every instance of the "toxic triad"... and to question whether, in the context of NOR, the koan is really all that toxic (and to consider that perhaps it is significantly less toxic in this context than it was over at WP:V). While VNT has been appropriately depreciated at WP:V, does that necessarily mean that it should be depreciated here - retained in its original context. I am asking people to consider the possibility that VNT continues to be appropriate here, even though it was not appropriate at WP:V.

Now, if we can get the concept across without using the koan, I am quite willing to consider options. I am not insisting on retaining the language of the koan (what I insist on is retaining the original concept behind the koan)... I merely asking whether there remains a legitimate place for the koan in this policy. Blueboar (talk) 17:11, 1 August 2012 (UTC)

Your response doesn't seem very clear. What I'm suggesting in that VnT originated on the NOR page because there was at the time no V page. If that's the case, then it's even less of a big deal that it would otherwise be.Formerip (talk) 17:35, 1 August 2012 (UTC)

The earliest incarnation of "Verifiability, not truth" that I've been able to trace is an edit to a page that's now called Misplaced Pages:No original research (draft rewrite 5th December 2004 to 5th February 2005) (the edit being this one of 8th November 2004). The phrase appears fully-formed: "The threshold for inclusion is verifiability, not truth". (If anyone knows of an earlier version that explains more about where it actually came from or why, then I would be interested to see it.) WP:V existed then, having first been created on 2 August 2003, but there seems to have been quite a bit of reorganisation of the two policies with text moving between them. VnT got moved to WP:V in this edit by Uncle G.

Examining the history is instructive. The original version seems much less objectionable to me, because the phrase is completely surrounded by context and explanation. By the time I encountered it, all the thought and reasoning had been cut and the key words emphasized in bold, and it had been moved into the first sentence of the lede of Misplaced Pages's only positive content policy. (What I mean by "only positive content policy" is that all the other content policies are negatively phrased: they describe things you shouldn't do. WP:V is the only policy that describes what you should do. To my mind this is an important distinction, and it's why I think it's a good thing that V and NOR are separate.)

In any case, now that I understand Blueboar's objection better, it seems easily-overcome to me. Blueboar says: "I suppose what is missing is a statement that ties that sentence, conceptually, into NOR... something to clearly explain the reason why we should not add unverifiable material - even if that material is true". I agree that this would be a beneficial thing to explain. I would like to point out that the phrase I removed does not, in any way, explain it; the sentence "The threshold for inclusion on Misplaced Pages is verifiability, not truth" contains exactly zero reasoning. But I would welcome discussion about how we could explain it to Blueboar's satisfaction.—S Marshall T/C 21:15, 1 August 2012 (UTC)

First of all, I strongly believe that Blueboar's raising of the question is both good faith and thoughtful, and I hope that no one else is really calling that into question. Blueboar's question, whether the possibility that VnT arose here historically, instead of the policy page that was just discussed in the RfC, should mean that we evaluate it differently here, is an entirely reasonable question to ask.

So here is how I would answer it. Everything on Misplaced Pages is dynamic, such that it may always be changing (and maybe even improving). The purpose of the threshold-VnT sentence, no matter how it arose originally, was to explain the concept of verifiability. Not to explain the concept of original research! Therefore, the recent consensus to downgrade that sentence's status at WP:V means that it's entirely appropriate to do likewise here. The only reason not to would be one that I'm not seeing in this discussion: that VnT somehow clarifies one's understanding of original research in a way that it does not clarify one's understanding of verifiability. Obviously, that's not the case. (I recognize that it originally communicated the point that "truth" is a form of OR. But that's not the same thing. One can fully understand what OR is, without that sentence. As long as we continue to tell readers of the policy page that it works together with V (and NPOV), no one is going to think that NOR allows VnT's concept of "truth".) So I (who have long been a stick in the mud for VnT, let's not forget!) am all in favor of deleting it here. --Tryptofish (talk) 00:31, 2 August 2012 (UTC)

Blueboar wrote: "Now, if we can get the concept across without using the koan, I am quite willing to consider options." S.Marshall wrote: "I would welcome discussion about how we could explain it to Blueboar's satisfaction." So how about this? "Truth without verifiability is OR." Kalidasa 777 (talk) 08:42, 2 August 2012 (UTC)

Thanks to S. Marshall for the links to the early history of the VNT koan... I encourage everyone to read them. Understanding the original intent of the koan, and how it evolved over time, will better inform our discussions.

The type of evaluation and discussion that is now going on is exactly what I was asking us to do.

As for Kalidasa's suggestion... hmmm... that might do it. I'll think about it. Blueboar (talk) 14:04, 2 August 2012 (UTC)

OR, truth

When I think of my experiences looking at others' edit summaries and talk page discussions for articles, when someone tries to put OR into an article, they are simply told that it is OR, and usually the material is successfully deleted, without any discussion of truth. In some cases the claim that something is OR is disputed, but the discussion does not turn to whether or not it is true, but whether or not it is OR. --Bob K31416 (talk) 09:42, 2 August 2012 (UTC)

"OR" has become Misplaced Pages shorthand for "does not comply with wp"ver", and "unsourced". WP:ver/nor creates a (sourcability) requirement for inclusion of material. And say that there are no exceptions to meeting that requirement (e.g. by claiming that it is true.) Striving for accuracy of material is a related but different topic, and a good goal. North8000 (talk) 23:09, 2 August 2012 (UTC)

Re “ "OR" has become Misplaced Pages shorthand for "does not comply with wp"ver", and "unsourced".” — Not for me. OR means unpublished info that has originated with the editor that put it in, i.e. the editor’s original research. If one interprets OR to mean something more general, it would lose its usefulness of specifying a particular type of information that isn’t allowed in Misplaced Pages.

But that is a digression. I began this section to share my personal experience and implicitly asking other editors to consider their own experiences, not alleged experiences of others, regarding the situations that I mentioned in my message. --Bob K31416 (talk) 11:53, 3 August 2012 (UTC)

Sidebar/digress comments: Whether it's right or wrong, I think that it is useful to point out what the common use/meaning of the term "OR" is. And the range of unsourced/unsourcable material can range from new creative personal research (which is what the term brings to mind) to someone putting in a wrong fact (e.g. Babe Ruth's date of birth) from memory/personal knowledge. IMHO, from a structural/policy standpoint, they are both the same to Misplaced Pages.

On the main topic of your comment, I really didn't understand what you were getting at, but was pointing out a related distinction anyway. That the following are two different things:

1. WP:Ver/NOR establishes a requirement that must be met for inclusion. Meeting it is a requirement for inclusion, not a force for inclusion.
2. A quest for accuracy.

In some ways #1 helps #2, but they are two different things.

In other different ways, #1 also inadvertently hindered #2, but that has been fixed.

In cases where objective accuracy exists, the operative mechanics boils down to 4 cases:

1. Meets wp:ver/nor and is false: WP:ver/nor has no effect. Other Wikipedian processes are free to determine inclusion/exclusion
2. Meets wp:ver/nor and is true: WP:ver/nor has no effect. Other Wikipedian processes are free to determine inclusion/exclusion
3. Fails wp:ver/nor and is false: WP:ver/nor prevents the material from being in Misplaced Pages.
4. Fails wp:ver/nor and is true: WP:ver/nor prevents the material from being in Misplaced Pages.

Sincerely, North8000 (talk) 12:27, 3 August 2012 (UTC)

This actually raises some interesting side questions... are WP:V and WP:OR distinct concepts or simply different ways of expressing the same concept? Is it possible for something fail WP:V and yet still pass WP:OR? And is it possible for something pass WP:V and yet still fail WP:OR? Blueboar (talk) 13:05, 3 August 2012 (UTC)

IMHO they are the same concept. The only difference (the 10% that is not overlap) is wp:nor places extra emphasis on and provides extra discussion on one type of material that is excluded by wp:ver. North8000 (talk) 13:18, 3 August 2012 (UTC)

WP:NOR and WP:V are similar, but are distinct concepts. WP:V states that a reliable source must exist that supports the statement; WP:NOR adds that sources must not be improperly used to create original research. It is possible for something to meet WP:V but fail WP:NOR. The classic example is the one given in WP:Synthesis, reliable sources exist for the definition of plagiarism in the Harvard Writing with Sources manual, but inclusion of that definition would be OR unless a published source refers to it in the context of the dispute. LK (talk) 13:43, 3 August 2012 (UTC)

I think that puts a better point on it, addresses an abstract variant more directly, and explains it better. But I think that that is also a violation of wp:ver. North8000 (talk) 14:52, 5 August 2012 (UTC)

OR, Verifiability, Accuracy

North, you keep saying that verifiability is "not a force for inclusion". If you mean that verifiability is not a guarantee of inclusion, then I agree. However, if you mean that verifiability is not, in any circumstances, an argument for inclusion, then I can't agree.

A comparatively simply example from the WP:Inaccuracy essay -- report about road closures in Oslo after 2011 attacks, extent of road closures questioned by an Oslo resident, on the grounds that "I was there". My own thoughts about this...

• It is clear that we cannot simply add a previously unpublished eye-witness report into mainspace, as that would be OR.
• Is someone saying "I was there", sufficient grounds, by itself, to remove the information that was found in the verifiable published report? I don't think it is.
• On the other hand, I'd agree that in such a case it is a very good idea to check what other published sources have to say about how much of Oslo was closed. In fact WP editors did consult further sources in this instance.

I think it is a very important principle that, as the RfC-approved lede of WP:V says, "content is determined by previously published information rather than by the personal beliefs or experiences of its editors." Kalidasa 777 (talk) 23:45, 4 August 2012 (UTC)

Well.... OR can be discussed on the talk page, and while it should not influence what information we present in the article, the discussion can sometimes influence how we choose to present to information. For example, suppose we are writing about a road that closed (not talking about the ones in Oslo now), and based on a source we say: "The road was closed in August of 2011"... then a Wikipedian says "that's incorrect... I live there and drove on the road every day until it closed... it did not actually close until September", we can reach a consensus to add attribution: "According to <source> the road closed in August of 2011" ... or we could compromise and say: "The road closed during the late summer of 2011" (which effectively resolves the conflict between the OR assertion and the source). What we can not do is say the road actually closed in September, based purely on the OR assertion. Blueboar (talk) 02:30, 5 August 2012 (UTC)

I don't think you can call that an OR assertion, it rather a primary source. Writing WP content based on it, that's usually OR.--Kmhkmh (talk) 03:15, 5 August 2012 (UTC)

Kalidasa 777, one/curse/gift of mine is that I can see the logical/structural underpinnings of policies, conversations and think along those lines. I mentioned "curse" because, when I try to communicate it on a more complex ones it often comes off sounding like Greek. Your example is one of those so here goes. The short answer is that the fact that the material meets wp:ver means that wp:ver allows ("allows" means only "does not prevent") the material to be included, and that's the extent of wp:ver for your example. So it will be up to other Misplaced Pages policies, guidelines and processes to determine whether or not the material stays. Who knows what those could be on any particular example. A few examples of how it might flow for your given example could be for the editors at the talk page to look at it and say and decide:

• A "RS" source says that it is true, and a credible sounding editor/argument says that it is false. That means that we should look for more sources on this question.
• "That information was sort of off topic and not useful anyway. Before this question came up we were on the fence anyway on whether to keep/remove it. Now, with this additional new question, let's just leave it out."
• They make one of the decisions that Blueboar described.

Now on your:

"Is someone saying "I was there", sufficient grounds, by itself, to remove the information that was found in the verifiable published report? I don't think it is."

I would have to respectfully say that you have accidentally misheard what I said into a straw man version. While I agree with your conclusion "I don't think it is", I must note that nobody is arguing that a personal opinion is sufficient grounds to force removal of material, and so that is not relevant to what I said. Your conclusion does not conflict with or relate to my argument. Again, I say this respectfully; the bluntness is only for brevity& clarity.

Finally, when I said that meeting wp:ver is a requirement for inclusion, not a force for inclusion, I should clarify that I was talking about the policy, not the sourcing that made it meet the policy. The sourcing that enabled it to meet wp:ver would certainly matter in the discussion. Sincerely, North8000 (talk) 11:31, 5 August 2012 (UTC)

Original mathematical proofs are not interpretations

I would like to request that the policy be clarified concerning mathematical proofs. Almost all mathematics articles on Misplaced Pages are based on a specific standard axiomatic system that only a small minority reject. And with the exception of articles that deal specifically with axiomatic systems, the rest do not explicitly state this set of assumptions, but under these assumptions, all mathematical proofs are absolutely true regardless of whether they are original research or whether they have been published before. Therefore editors have always been freely writing new proofs or variants of them because they can be systematically and unquestionably checked to be correct. In short, there is no room for interpretation, so original research is perfectly fine. Of course, it might actually be a good thing if every single proof had a link to the axiomatic system within which it is established, but the typical reader would not care, so omitting it as has always been the case is fine too. Lim Wei Quan (talk) 05:26, 6 August 2012 (UTC)

New proofs that have not been reviewed by reliable sources may have errors. We as WP editors cannot rely on other editors finding and correcting them. Ergo, they are original research and cannot be included. --MASEM (t) 05:39, 6 August 2012 (UTC)

First of all most math articles do not contain proofs (see Misplaced Pages:Manual_of_Style_(mathematics)#Proofs). As far as describing proofs in WP are concerned "straight forward"/"obvious" or minor variations of known proofs are up to an editor discretion, but new proofs are considered WP:OR for the reason already stated above. WP is not a publication venue for new material, but it merely collects and describes what was already published elsewhere (in reliable, reputable sources).--Kmhkmh (talk) 06:59, 6 August 2012 (UTC)

Misplaced Pages is not a place for "proofs", but it is a place for "explanations". In the article IBAN, I wrote "It can be shown that ..." and in the reference, I stated "This is an undergraduate level problem". In other words, if you have a degree in maths, by all means comment, if you don't, please leave comments to somebody who can assess this statement. That article might well serve as a point of reference for this thread. Martinvl (talk) 10:29, 6 August 2012 (UTC)