Frequently Asked Questions (FAQ) | |
---|---|
|
WikiProject Manual of Style | |||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|
|
the Wikipedia Help Project | (Rated Project-class, High-importance) | ||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
Archives (index) |
---|
Threads older than 14 days may be archived by lowercase sigmabot III. |
Contents
- 1 What are the benefits of preferring full citations over short citations?
- 2 Citing a blast email
- 3 RFC: Is a change in citation markup method a change in citation style?
- 4 Extracting referencing and citation information from articles for evaluation
- 5 Wording at WP:Citation needed
- 6 Use of "Locations" in citing Kindle e-books?
- 7 "Request admin assistance" in the "Dealing with unsourced material" section
- 8 Citing sources with spelling mistakes in their titles
What are the benefits of preferring full citations over short citations?
Nowhere had a guideline or even an essay mentioned benefits of adding full citations and not just short ones, so what are they? Gamingforfun365 (talk) 05:06, 12 March 2016 (UTC)
- Not sure why you ask Gff365, but if you mean the case where When an article cites many different pages from the same source, the use of short citations in footnotes is recommended to avoid the redundancy of many big, nearly identical full citations. The only reason I can think of not to do that would be that it entails a more indirect process of tracking down the full reference: say you come across a ref marker [23] in the text of an article, so to find who said that and where, you jump to the numbered ref list and it only says "Bloggs, 2013, p. 343", then for full detail you have to scroll further down (or sometimes up) to another end section "Sources" or similar, hunting for something credited to a Bloggs and written in 2013. Not saying I would support a wholesale change of practice, mind...: Noyster (talk), 11:12, 12 March 2016 (UTC)
- Of course if the editor uses the harv templates for the short enties, it is just a single click to the full entry. @Gff365: If Gff365 is asking the question about instances where the full is not given, the answer is to minimize possible confusion, ambiguity and extraneous searching. --Bejnar (talk) 00:59, 13 March 2016 (UTC)
- And if you use {{rp}} to provide page numbers, even that step is unnecessary; all the [23] or whatever citation markers go directly to the full citation without any intermediary stages in a second section of shortened citations. — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼ 01:43, 13 March 2016 (UTC)
- Of course if the editor uses the harv templates for the short enties, it is just a single click to the full entry. @Gff365: If Gff365 is asking the question about instances where the full is not given, the answer is to minimize possible confusion, ambiguity and extraneous searching. --Bejnar (talk) 00:59, 13 March 2016 (UTC)
- A poorly formulated question, though the documentation on all this is indeed deficient. Each source in an article should have a full citation that includes all of the bibliographic information useful in identifying and finding a source. The issue that arises is when a source is used more than once, as it is not useful to repeat a full citation. There are two principal ways of handling this: a) "named refs" (the "<ref name=...>" construct), which allow a footnote (presumably containing the full citation) to appear in more than one place, or b) short cites ("Smith 2001"), usually implemented with Harv templates which provide a link to the full citation to where ever it is placed.
- Use of short cites also arises when editors want to get all the clutter of bibliographic detail out of the text by moving the full citations into their own section. This also makes it easier to find and edit the full citations (one doesn't have to search through a lot of text). ~ J. Johnson (JJ) (talk) 20:48, 13 March 2016 (UTC)
Each style of referencing has its own advantages and disadvantages. It's just a matter of taste which one each person prefers. However, for existing articles, you should simply follow the existing style, rather than trying to worry about it. The question of which style you prefer is really only relevant when you start new articles, or add references to articles that don't have them yet.
The use of short citations is neither required, recommended, or prohibited by the MOS; nor are "duplicated full" citations required, recommended, or prohibited. None of them is viewed as any better than the others from the point of view of the MOS. — Carl (CBM · talk) 21:09, 13 March 2016 (UTC)
-
-
- Sure, short citations are not required, no more than named refs. But duplication of full citations really should be avoided. So where the editors working on some article agree to, say, use of short cites with all of the sources listed alphabetically in a "Sourcss" section it is NOT "just a matter of taste" if,say, someone adds citations using {sfn}. It is NOT that "none of them is viewed as any better"; it is matter of consistent practice. ~ J. Johnson (JJ) (talk) 20:08, 14 March 2016 (UTC)
- Of course, if an article already uses a style without duplicate full references, it should be preserved. But, until/unless the MOS has a recommendation, the duplication of references, or lack thereof, is just one more optional matter of taste. There is no support in the existing MOS for "duplication of full citations really should be avoided". So if an article already uses some other system with duplicated references, that system should also be respected. — Carl (CBM · talk) 00:03, 15 March 2016 (UTC)
- WP:CITE is not part of the MOS, so why refer to the MOS? -- PBS (talk) 15:25, 15 March 2016 (UTC)
- CITE is part of MoS, just not all of it; it's half style guideline and half content guideline. Like WP:SAL (which a three way MoS/content/naming convention guideline), it was determined that it was more utilitarian to group it topically than functionally; we might need to do more of this. — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼ 12:32, 22 March 2016 (UTC)
- @SMcCandlish when where is the discussion showing that there was a consensus that this guideline is part of the MOS? -- PBS (talk) 23:37, 1 April 2016 (UTC)
- CITE is part of MoS, just not all of it; it's half style guideline and half content guideline. Like WP:SAL (which a three way MoS/content/naming convention guideline), it was determined that it was more utilitarian to group it topically than functionally; we might need to do more of this. — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼ 12:32, 22 March 2016 (UTC)
- The WP:IBID section says: "When an article cites many different pages from the same source, to avoid the redundancy of many big, nearly identical full citations, most Wikipedia editors use one of three options:..." which sounds like at least advice that duplicated or near-duplicated citations should be avoid, albeit not a mandate to do so. DES (talk) 23:10, 15 March 2016 (UTC)
- WP:CITE is not part of the MOS, so why refer to the MOS? -- PBS (talk) 15:25, 15 March 2016 (UTC)
- Of course, if an article already uses a style without duplicate full references, it should be preserved. But, until/unless the MOS has a recommendation, the duplication of references, or lack thereof, is just one more optional matter of taste. There is no support in the existing MOS for "duplication of full citations really should be avoided". So if an article already uses some other system with duplicated references, that system should also be respected. — Carl (CBM · talk) 00:03, 15 March 2016 (UTC)
- Sure, short citations are not required, no more than named refs. But duplication of full citations really should be avoided. So where the editors working on some article agree to, say, use of short cites with all of the sources listed alphabetically in a "Sourcss" section it is NOT "just a matter of taste" if,say, someone adds citations using {sfn}. It is NOT that "none of them is viewed as any better"; it is matter of consistent practice. ~ J. Johnson (JJ) (talk) 20:08, 14 March 2016 (UTC)
-
-
-
-
-
- Duplication of full citations is something most editors would naturally avoid, even without being told, so they do search for ways of handling that. Perhaps the MOS should say something stronger, as it really is not a good thing. Certainly not something we should find acceptable. ~ J. Johnson (JJ) (talk) 00:02, 16 March 2016 (UTC)
-
-
-
-
-
-
-
-
-
- Perhaps I should clarify that as "should say something stronger". The current wording (as DES has quoted) only suggests alternatives, and remains permissive in allowing duplicated citations. I think such duplication should be expressly deprecated, with an admonition that alternatives should be used. ~ J. Johnson (JJ) (talk) 20:37, 17 March 2016 (UTC)
-
-
-
-
-
-
-
-
-
-
-
-
- I don't think the overhead of using one of the methods to avoid duplicate full citations is justified if there are only a few duplicates. I do think duplicates should be avoided if there more than a few. I would count the total number of extra full citations when deciding; if five books were each cited in two places, that would be five redundant full citations. Jc3s5h (talk) 21:49, 17 March 2016 (UTC)
-
-
-
-
-
-
Duplicated citations can be confusing for the reader, and particularly in evaluating relatively new or challenged articles on a quick scan, cna make it appear that the article has more supporting sources than it actually does. I suggest that this guideline be amended to say that it is strongly encouraged to combine duplicated citations, and that it is always acceptable for an editor to use one of the described methods to avoid duplicated full citations, and this is not a change in citation style that needs consensus under CITEVAR. We don't want to prevent relatively inexperienced editors from adding proper cites just because they create duplicates, and the editor doesn't know how to combine them, or doesn't have time to do so. (cites are better than no cites.) But we want to strongly encourage the 'best practice' of combining them in a way appropriate to the overall citation style of the article. I would want to encourage combining even when there are only two occurrences of the same citation. That is my view. DES (talk) 22:27, 17 March 2016 (UTC)
- I don't believe the overhead is significant (and don't we have a statement some where not to worry about that?). Part of my antipathy to duplication of data comes from the database principle that every datum should have a single (and authoritative) instance, to avoid inconsistency. While this isn't too important for us, still, it is a sign of sloppiness. When citations are collected in a list (and sorted) duplications are obvious, and tend to be eliminated. But not so obvious in the text, where duplications just add to the clutter. If we really want to reduce processing overhead we should pull all of the citation templates out of the text (i.e., put them in their own section). ~ J. Johnson (JJ) (talk) 22:46, 17 March 2016 (UTC)
- P.S. Yes, I like DES' idea of encourage a "best practice". We might tolerate the temporary existence of duplication, but should encourage correction. ~ J. Johnson (JJ) (talk) 22:52, 17 March 2016 (UTC)
- I too agree with DES' idea of encouraging a "best practice". However, I strongly disagree with J. Johnson's suggestion of always pulling the full citations out. What JJ is talking about is only a problem in articles with a great deal of citation, and where individual sources are cited multiple times with variant pages numbers. Most of the time citations can be combined using the name function and one or two short cites. There is no need for draconian duplication in those cases. I.e., having both a short cite and a full cite for a bunch of single cites. --Bejnar (talk) 00:43, 21 March 2016 (UTC)
- Benjar, I believe you have misunderstood, and even over-interpreted, what I said. What I said was not to always pull out the full citations, but, if we really wanted to reduce processing overhead (DES' concern), to pull out all the templates. Note also that even where sources are cited only once there are benefits to moving the full citations (templated or not) to their own section: it greatly reduces clutter in the main text, makes the full citations easier to find, permits sorting (e.g,, alphabetization) of the sources, and makes clean-up (etc.) over the entire set of full citations easier because they're all in the same place.
- I too agree with DES' idea of encouraging a "best practice". However, I strongly disagree with J. Johnson's suggestion of always pulling the full citations out. What JJ is talking about is only a problem in articles with a great deal of citation, and where individual sources are cited multiple times with variant pages numbers. Most of the time citations can be combined using the name function and one or two short cites. There is no need for draconian duplication in those cases. I.e., having both a short cite and a full cite for a bunch of single cites. --Bejnar (talk) 00:43, 21 March 2016 (UTC)
-
-
- Now you say all this is "only a problem in articles with a great deal of citation". But consider this: if an article with a handful of full citations in the text gets expanded to the point that citations are getting duplicated, does CITEVAR preclude moving those citations out of the main text? To the extent that this (used with short cites) is a good practice, should it not also be a good practice to encourage this even before multiple cites make it necessary? ~ J. Johnson (JJ) (talk) 22:11, 21 March 2016 (UTC)
- J. Johnson, Bejnar, you have misunderstood my concern. I am not even slightly concerned with "processing overhead" here. I am concerned with reader confusion. It is my view that having duplicate full citations in the Notes list can lead to reader confusion, and to a misleading appearance of more sources being cited than there in fact are. This could be fixed by combining cites using named references, (and perhaps the {{rp}} template), or by using shortened cites, either with or without the harv templates (there are several ways to do short cites, all acceptable. If an article has a more or less stable citation style not using short cites, then i would advise using named refs to combine duplicates, and possibly using named refs and rp to combine different cites to the same source, or else to discuss on the talk page to establish a consensus to change the citation style. as for "pulling full cites out of the text" that can be done with List-defined references as well as with short cites -- personally I prefer LDR, but the guidelines properly do not prefer one of these methods over another. My only suggestion at this time is that this guideline advise that duplicate citations be combined by some method, and that much seems to have consensus. DES (talk) 00:38, 22 March 2016 (UTC)
- Also, when Jc3s5h wrote of "overhead" above in this thread, i think he or she meant the work of doing the editing to combine the cites, not processing overhead. There I disagree; if I am editing an article I would always combine even two duplicate cites using whatever method is consistent with the existing cites, most often named refs in my experience, as I tend to work on less-developed articles. DES (talk) 00:49, 22 March 2016 (UTC)
- (edit conflict) Re: J. Johnson's "does CITEVAR preclude moving those citations out of the main text?" – Of course it does, according to the interpretation of the few above convincinced that no one may touch any aspect of their citation coding. This is a nice proof that their interpretation is wrong (both in the sense of incorrect, and the sense of counter to WP's interests). — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼ 05:43, 22 March 2016 (UTC)
- Now you say all this is "only a problem in articles with a great deal of citation". But consider this: if an article with a handful of full citations in the text gets expanded to the point that citations are getting duplicated, does CITEVAR preclude moving those citations out of the main text? To the extent that this (used with short cites) is a good practice, should it not also be a good practice to encourage this even before multiple cites make it necessary? ~ J. Johnson (JJ) (talk) 22:11, 21 March 2016 (UTC)
-
-
-
-
-
- So when all of the slight misunderstandings are sorted out I think we are all pretty much on the same page as far as deprecating the duplication of full citations, for whatever reasons. Opinions vary as to the best ways of dealing with duplicated cites (uses of) a source. I think we could say something like each source should have a full citation, but it is not necessary nor (generally) useful to duplicate it. I say "generally" to allow for things like suggested reading lists. ~ J. Johnson (JJ) (talk) 22:12, 22 March 2016 (UTC)
-
-
-
-
- Always good to discuss, when individuals have questions, and I concur with foregoing discussion that reflects WP guidelines and policies. I would note, in closing to the above, that there are various means by which duplicated citations can be avoided, and various accepted forms of short versus long citation. In each field of academic study, there are usually are a small set of acceptable ways within each discipline (humanities, social sciences, business, natural sciences, etc.). Duplications of citations in the sciences often use the {{ rp | pg. }} markup; in the humanities and classics, it is often a bibliography followed at each point of citation by an acceptable abbreviated form of the source. The variability of accepted forms within and between disciplines, the need for consistency of presentation, and the demand for editor mutual respect are what underly the Wikipedia policy that the pattern of practice in place at an article (the precedent set by earlier editors) should be followed. That is, unless there is a clear policy- or guideline-based reason for taking the article in a new direction (or, unless the article defies all acceptable styles of its discipline, or has no consistency of style, in which case it is acceptable to establish order from the chaos). User:Leprof_7272 50.179.252.14 (talk) 14:59, 21 March 2016 (UTC)
- Actually, the method of citation used in a particular field is of almost no significance. It may be the reason why editors favor one style over another, but what counts is the consensus among editors. If the editors involved on a page cone to a consensus to more away from the standard format used in a particular field, and toward some other format, they may do so. Contrawise, changing an established format because "the article defies all acceptable styles of its discipline" should not be done without first obtaining consensus on the article talk page. A consistent citation style that satisfies the requirements of verifibility including text-source integrity, should be changed only after obtaining consensus. DES (talk) 01:13, 22 March 2016 (UTC)
- (edit conflict)The "the article defies all acceptable styles of its discipline" concept is totally inoperative here, per WP:NOT. WP is not written like, and does have its style of anyhting dictated to it, by the AMA, PLOS ONE, the Royal Historical Society, or Chicago and Turabian. WP is not even bound by the citaton or other style conentions of other encyclopedias, FFS. — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼ 05:52, 22 March 2016 (UTC)
- Actually, the method of citation used in a particular field is of almost no significance. It may be the reason why editors favor one style over another, but what counts is the consensus among editors. If the editors involved on a page cone to a consensus to more away from the standard format used in a particular field, and toward some other format, they may do so. Contrawise, changing an established format because "the article defies all acceptable styles of its discipline" should not be done without first obtaining consensus on the article talk page. A consistent citation style that satisfies the requirements of verifibility including text-source integrity, should be changed only after obtaining consensus. DES (talk) 01:13, 22 March 2016 (UTC)
- Always good to discuss, when individuals have questions, and I concur with foregoing discussion that reflects WP guidelines and policies. I would note, in closing to the above, that there are various means by which duplicated citations can be avoided, and various accepted forms of short versus long citation. In each field of academic study, there are usually are a small set of acceptable ways within each discipline (humanities, social sciences, business, natural sciences, etc.). Duplications of citations in the sciences often use the {{ rp | pg. }} markup; in the humanities and classics, it is often a bibliography followed at each point of citation by an acceptable abbreviated form of the source. The variability of accepted forms within and between disciplines, the need for consistency of presentation, and the demand for editor mutual respect are what underly the Wikipedia policy that the pattern of practice in place at an article (the precedent set by earlier editors) should be followed. That is, unless there is a clear policy- or guideline-based reason for taking the article in a new direction (or, unless the article defies all acceptable styles of its discipline, or has no consistency of style, in which case it is acceptable to establish order from the chaos). User:Leprof_7272 50.179.252.14 (talk) 14:59, 21 March 2016 (UTC)
Citing a blast email
If a reputable source (the creator of the content in question) sends a blast email to thousands of people, can that email be cited as a source? I'd like to add some detail to facts on a page, but the only source of this information is a blast email that the well-known creator of the content sent out to announce a new episode of the web series. No reputable online source has commented on this particular fact directly, but the current Wikipedia article states "It seems that ...", when in fact the content creator confirmed the assumption made in the Wikipedia article completely. — Preceding unsigned comment added by 216.80.102.52 (talk) 04:50, 25 March 2016 (UTC)
- If there isn't a place Wikipedia readers can go to view the email, even if the place is behind a paywall, it can't be cited. Jc3s5h (talk) 11:44, 25 March 2016 (UTC)
-
- Actually that's not true. In the age of the Internet people sometimes assume that a source isn't a source unless anyone ("I") have access to it somehow, but that is just not the case. If you check WP:SOURCE it says: "Source material must have been published, the definition of which for our purposes is "made available to the public in some form"." So something that was published, but is no longer available to Wikipedia readers to go check, still counts as a source. An obvious example of this is a book that was published, but that is now out of print and not held by any library. The book was published and copies of the book do exist, so it counts as a source. Another example would be radio broadcasts that are not recorded or archived anywhere the public has access to. The broadcast itself is publication.
-
- To the specific question about the blast email, if the email is sent to thousands of people and anyone could have signed up to receive the email, then that is a publication. The email was made available to the public. The fact that many people in the public did not sign up for the email does not mean it was not available to them. Some primary source documents only exist in one location in the world (like an archives for historical print documents), so technically there is a place people can go to check it if they want to (and, more importantly, have some way of getting to that archive to check). With an email there might be thousands of people who have a copy of it in their in boxes who could show it to you if you go visit them (the logistics of which might be a lot simpler than visiting an archives), so even in this case there still might technically be "a place Wikipedia readers can go to view the email". But like I said, Wikipedia does not even require that. It just requires that something must have been made available to the public in some form, and allowing just anyone to sign up for an email blast fits that definition. 99.192.55.195 (talk) 13:09, 25 March 2016 (UTC)
- It qualifies as a "source"... But not as a Reliable source. A reliable source must (at a minimum) be located in an archive that is open to the general public ... So anyone could gain access to it (As long as they are willing to spend the time and money to do so.) A private archive does not meet that standard, because it is not open to all.
- About the only way we might allow it if someone said "yes, I have a copy... I live at such and such address, and will show it to any random stranger who shows up at my door and asks to see it." Not likely. Blueboar (talk) 13:33, 25 March 2016 (UTC)
- To the specific question about the blast email, if the email is sent to thousands of people and anyone could have signed up to receive the email, then that is a publication. The email was made available to the public. The fact that many people in the public did not sign up for the email does not mean it was not available to them. Some primary source documents only exist in one location in the world (like an archives for historical print documents), so technically there is a place people can go to check it if they want to (and, more importantly, have some way of getting to that archive to check). With an email there might be thousands of people who have a copy of it in their in boxes who could show it to you if you go visit them (the logistics of which might be a lot simpler than visiting an archives), so even in this case there still might technically be "a place Wikipedia readers can go to view the email". But like I said, Wikipedia does not even require that. It just requires that something must have been made available to the public in some form, and allowing just anyone to sign up for an email blast fits that definition. 99.192.55.195 (talk) 13:09, 25 March 2016 (UTC)
-
-
-
- But that's not the policy. A lot of people don't seem to understand the policy or how citing sources works in general, so that's understandable. Like I said before, a lot of people think a source isn't a source unless they personally have some access to it. That is not the policy and has never been how citing sources has worked The policy is, as I quoted above: Source material must have been made available to the public in some form. An out of print book is something that was made available to the public in some form. The policy does not say the source must currently be available to the public, just that it once was. If you don't like the policy, you are free to lobby for a change, but that is the policy and it is how citing sources has pretty much always worked. 99.192.78.176 (talk) 15:57, 25 March 2016 (UTC)
-
-
-
-
-
- ADDENDUM: Jut to be doubly clear, the passage I have quoted comes from a section of the policy entitled Reliable sources and the subsection What counts as a reliable source, so your claim that it is a source but not a reliable source is just plain false. Reliability has to do with whether or not the source can be counted on to be accurate. Whether or not you, I, or people in general have access to the source has nothing at all to do with how accurate it might be. 99.192.78.176 (talk) 16:01, 25 March 2016 (UTC)
- Not quite, 78.176.
WP:RS also says, in the "definition of Published" sub section: "Additionally, an archived copy of the media must exist. It is convenient, but by no means necessary, for the archived copy to be accessible via the Internet." (emphasis added).So an out of print book not held by any library is probably not a reliable source any more, but if OCLC or a similar resource shows even a few libraries with copies, that is sufficient. If the email blast was in fact not archived anywhere, it is not a reliable source and should not be cited. DES (talk) 16:07, 25 March 2016 (UTC) - Blueboar, not quite correct either. Three must be an archive, but it can be restricted. For example, archives at some university research libraries are only accessible to "qualified researchers" or "Legitimate scholars". Such restrictiosn would not prevent citing a source, if an editor has verified the source and another could do so by obtaining the needed qualifications. This is similar to the rule on sources behind a paywall. DES (talk) 16:14, 25 March 2016 (UTC)
- Not quite, 78.176.
- ADDENDUM: Jut to be doubly clear, the passage I have quoted comes from a section of the policy entitled Reliable sources and the subsection What counts as a reliable source, so your claim that it is a source but not a reliable source is just plain false. Reliability has to do with whether or not the source can be counted on to be accurate. Whether or not you, I, or people in general have access to the source has nothing at all to do with how accurate it might be. 99.192.78.176 (talk) 16:01, 25 March 2016 (UTC)
-
-
-
-
-
-
-
- DES, (and call me 99.192. I have a dynamic IP address so the last two numbers change a lot, but the first two are always the same) You have quoted the definition out of context. The statement that an archived copy must exist is specifically a statement made in reference to "audio, video, and multimedia materials", not text. The contrast to published text is made clear in the first two sentences of the section. So if I own the only remaining copy of a book that was published and made available to the public in book stores years ago, it counts according to the policy. Text need not be archived. 99.192.78.176 (talk) 16:24, 25 March 2016 (UTC)
-
-
- 99 - we can not be sure you actually own the book/email/document you claim to own (Assume Good Faith only goes so far, and you could be making it up). Someone else has to be able to verify that a) you own an authentic copy of the source, and b) it says what you say it does. Blueboar (talk) 17:29, 25 March 2016 (UTC)
- 99.192 you are correct that I cited the guideline out of context, and that it doesn't now say what I thought it did. But I think that perhaps it should. I am very uneasy about citing a source that no one (or no one but the citing editor) can verify. To take an extreme example Aristotle is a very reliable source for what Aristotialian philosophy held. But if an editor were to cite Book 2 of his Poetics, which as far as anyone knows has not existed since the destruction of the Library at Alexandria, it would not be accepted, i think. An out of print book can be cited, but i think it is at best ill advised to cite an out of print book not held by any library at all, nor otherwise available to a diligent searcher (copies not routinely available on the used book market, say). As to the email blast, if it was not archived anywhere, and no one who didn't originally receive it can verify it, I don't think it should be given much if any weight as a source. Perhaps we need an RfC to cover this situation. DES (talk) 18:13, 25 March 2016 (UTC)
-
-
-
-
-
-
- Blueboar, you might have reason to think someone is just making it up if no one else has access to the book, but (1) that does not change the fact that the policy says what it does so (2) in a case where someone does not doubt that the book exists and says what it does it would be both against policy and kinda shitty to remove a source or piece of information just because you don't like the policy.
-
-
-
-
-
-
-
- DES, if someone said they know what Book 2 of Aristotle's Poetics says (or that he has a copy of it) despite it being believed to no longer exist, that might in itself be reason to think the person is lying and so to contest the claim that the source exists. But suppose that it were determined by whatever experts you like that exactly one copy of the book did exist and was privately owned by Bill Gates. Then the fact it is not available to anyone is not a reason to think it cannot be valid as a source.
-
-
-
-
-
-
-
- It is also worth considering just how much protection adding an "availability" criterion to books would actually have. If being available just means that one library somewhere in the world has one copy or that an exclusive archive somewhere in the world has one copy or even just one used bookstore somewhere in the world has one copy for sale then changing the policy to add an "availability" criterion for books would have no effect. All I would have to do as the owner of the only copy of the book is to list it as for sale online and make my asking price $1 million. Then PRESTO! the book is now available to anyone in the world who wants to check it!
-
-
-
-
-
-
-
- As for the email blast, if it was sent to anyone in the world who signed up to receive emails and thousands of people got it and hundreds of them still have their copy of that email, then it clearly exists. And for those worrying about availability, if any one person who has the email says, "I'm willing to meet up with anyone with my laptop and I can show them the email. Just pay me $50 and buy me a coffee" then the email is publicly available. If there is some reason to think the person is just making it up, then there can be a discussion about whether or not the source actually does exist, but once we agree it does exist then according to current policy, it is a reliable source. 99.192.53.220 (talk) 19:29, 25 March 2016 (UTC)
-
-
-
-
-
-
-
-
-
- Blueboar, you have misread. I was offering a hypothetical about how the "availability" criterion could be satisfied if someone needed to do so. But as I have pointed out (and DES now agrees I am correct about) text sources do not have an "availability" criterion for use. That criterion only applies to "audio, video, and multimedia materials". 99.192.55.221 (talk) 00:38, 26 March 2016 (UTC)
-
-
-
-
-
-
-
-
-
-
-
-
-
- You still don't get it, but whatever. The fact still remains that Wikipedia policy says the email is ok. If hypotheticals to explain why trying to change that policy would be pointless confuse you, then just ignore them. Because as it stands, policy says the email is a valid source. 99.192.55.221 (talk) 02:46, 26 March 2016 (UTC)
-
-
-
-
-
-
-
While I now agree that there isn't a specific availability/archive requirement in the guideline for text sources, i think that probably there should be. In any case, if this email blast were actually to be cited in an article, i would be inclined to revert and take the matter to the reliable source notice board arguing for removal of such a citation as unreliable. DES (talk) 09:57, 26 March 2016 (UTC)
- Editing based on what you think the policy should be when you agree it is not the policy is bad faith editing. If you think something is accepted by the policy, but you disagree with the policy, the appropriate step is to try to change the policy. I'm sure you can think of a number of times encountering editors who, when informed they are editing contrary to the rules, will try to argue what the rules "should be" and insist on their edits standing. People who edit based on what they want the rules to be just make Wikipedia editing more difficult for everyone. 99.192.75.131 (talk) 11:42, 26 March 2016 (UTC)
- 99.192, while it may be that under the current guideline there is no archive requirement for text, that does not eman thqat all published text is reliable, nor is there a precise definition of what is reliable in a given context. it is not at all bad faith editing to revert the use of a so7urce that one considers dubious per WP:BRD, and then to take the matter to an appropriate noticeboard for additional input and decision. DES (talk) 15:21, 26 March 2016 (UTC)
-
- The discussion has got a bit bogged down in detail. WP:RS says: Articles should be based on reliable, third-party, published sources with a reputation for fact-checking and accuracy. In what way does Wikipedia:Identifying reliable sources#Definition of published say that an e-mail is a published source, let alone one by a source with a reputation for fact-checking and accuracy? Peter coxhead (talk) 12:14, 26 March 2016 (UTC)
-
-
- Peter, I think you misunderstand the context of the question the person who started the thread was asking about. The question is not about any old email that goes from one person to another, but an email that is sent as a part of an official mailing list to thousands of people all at the same time. This is an email that essentially functions as a press release. If you check the only other edit that was made by the person with the IP address who started the thread, you will see that the question is likely in the context of emails Louis CK sends to people who subscribe to his email list and the content of the email in question is something about his web series Horace And Pete. So if, for example (this is just a hypothetical) he sends out an email to his list saying: "Episode 4 of Horace And Pete is now available for download. People have been asking me how many episodes of the series there will be. I will be making X number of them only. I hope you enjoy them" - then that looks like a pretty reliable source that there will be X episodes. Remember that anyone can sign up to get the emails and people who do get them include professional writers who have reviewed his work and sometimes quote from them, so they are clearly public documents. If anyone knows how many episodes there will be, It's Louis CK. So if he says there will be X of them then that is as reliable a source as you could ask for. So yes, I agree that there is not general rule that emails are reliable, but when it is an email from a public figure sent on to a list of thousands of people and anyone in the world can sign up for the emails and the content is about his own work, reliability and fact-checking is not an issue. 99.192.75.131 (talk) 13:55, 26 March 2016 (UTC)
-
-
-
-
- I do understand the context. I can only repeat that an e-mail is not a "published source" as per the definition at Wikipedia:Identifying reliable sources#Definition of published, as well as any normal use of the term. I don't dispute the accuracy of the e-mailer, but it's not the point: the point is that it wasn't published, however many people it was sent to, and the accuracy that matters is the accuracy of anyone that publishes the e-mail, e.g. by putting it on a blog. Peter coxhead (talk) 14:06, 26 March 2016 (UTC)
-
-
-
-
-
-
- Peter, we went through this in the discussion above. The page you link to is a guideline that partially quotes the policy, but is not policy itself. The policy, which I quoted above, is here: WP:SOURCE. The part of the policy that the guideline quotes is (as DES and I discussed above) about what counts as "published" for "audio, video, and multimedia materials", not text. And the part of the policy that the guideline page leaves out is this: Source material must have been published, the definition of which for our purposes is "made available to the public in some form." If you substitute the definition given for the word "published" you get a simpler reading that says: Source material must have been made available to the public in some form. An email to a list of thousands of people that anyone in the world can sign up for is as much of a publication by that definition as a print newspaper that anyone can subscribe to. In both cases only people who sign up for it get it, but anyone can. The email is arguably more of a publication, since there is no fee barrier or geographical location barrier. 99.192.75.131 (talk) 14:35, 26 March 2016 (UTC)
-
-
-
- 99 may have a point here... I think it may help if we use a different term to describe what we are talking about. The documents that 99 refers to are really "news letters"... Not email communications.
- A news letter that was distributed the old fashioned way (printed on paper, and sent to subscribers via the post office) can be considered reliable (in limited situations)... It would qualify as a self-published, primary source). The question is... Does that limited reliability change if the news letter is distributed via email? Blueboar (talk) 14:54, 26 March 2016 (UTC)
- Yes, 99.192 does have a point, Peter coxhead. Such an "email blast" or "electronic newsletter' probably should be considered as having been "published". Whether it was "reliably published" is another matter, of course, and whether it is suitable for use as a source may be yet another matter. But the definition of "published" is probably not the weak point in 99's argument. DES (talk) 15:21, 26 March 2016 (UTC)
- Note that Wikipedia:Published includes "A broadcast email, including email-lists if they are archived and public..." as one of its examples of things that do count as published. But note that there is some attention in that statement to the concept of archiving. DES (talk) 16:28, 26 March 2016 (UTC)
-
-
- And the qualifications are important: "archived and public". Can I now access this e-mail? Where is it archived? Peter coxhead (talk) 17:29, 26 March 2016 (UTC)
- Of course if you read that closely, it appears that "archived and public" is being applied to email-lists, but perhaps not to other sorts of "broadcast email"s. And of course Wikipedia:Published is an essay, not a guideline. Mind you, i think that the "archived and public" requirement should apply, but i'm not sure that it does. however, Wikipedia:Published also says "Sources that are ... not accessible (e.g., the only remaining copy of the book is locked in a vault, with no one allowed to read it) are never acceptable as sources on Wikipedia.". This is relevant to some commetns about out-of-print so8urces above, and to the new discussion linked below. DES (talk) 17:49, 26 March 2016 (UTC)
- And the qualifications are important: "archived and public". Can I now access this e-mail? Where is it archived? Peter coxhead (talk) 17:29, 26 March 2016 (UTC)
-
-
-
-
-
- Yes, the fact that that page is just an essay is a limitation on its applicability, but even then the essay does not seem as helpful as you might think. On the issue of vaults, the point being made is that if no one has access to a source it cannot be used. In the previous paragraph it says a source needs to be "accessible to at least some people." An out of print book that no library holds is not unavailable to those who own a copy, so it is available to at least some people. Same for an email that thousands received and many did not delete. On the issue of emails, the full sentence reads as follows: "A broadcast email, including email-lists if they are archived and public—but not email messages or other forms of personal communication sent only to you or a small number of people." Ok, but an email sent to subscribers of an email list that includes thousands of people is not a "personal communication sent only to you or a small number of people" and if there is no public archive of the email then it covered by neither part of the sentence. In short, the essay does not seem to say anything about whether or not to count non-archived non-personal emails. But that is where the word "including" is important. The sentence is saying broadcast emails count, including archived ones but not including personal ones. So if an email is neither archived nor personal, it counts as acceptable. 99.192.48.16 (talk) 18:10, 26 March 2016 (UTC)
-
-
-
- I have started a discussion on the broader point at WT:RS#Archive or availability requirement for text sources. DES (talk) 17:08, 26 March 2016 (UTC)
It's tautological (in the proof sense, not the criticism-of-wording sense) that the source must be available at some effort or other cost. No amount of lawyering about "private archives" or "published but no longer available to anyone anywhere" will get past the fact that it's the verifiability policy. If the source does not make it possible to verify the content, then it does not qualify. A possible solution in a case like this would be to post the blast e-mail to some site for discussion of spam or whatever, archive that page via archive.org, then cite the e-mail as a primary source and the forum as a |via=
, and archive.org as the |archive-url=
. Forums and other WP:UGC are not themselves reliable sources, but if you're using one as a carrier for the original content, it's hard to see what the issue would be (archive.org itself is a user-generated carrier of original non-user-generated sources; all this would do it add another carrier so that archive.org has something to archive). — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼ 18:35, 26 March 2016 (UTC)
- Just consider where this leads. Editors are discussing the interpretation of a source. One e-mails the author of the source, receives a reply, and posts this on a page on their own web site, which they then archive. (This has actually been suggested as a way of resolving a dispute; it's not just an idea I made up.) Would this be acceptable? The counter-argument is that it's not a broadcast e-mail, but the distinction seems a fine one to me. The problem is that in neither case would we know whether the e-mail had been accurately copied to the web site, only that whatever was there had been archived accurately. E-mails are not reliable sources, period. Peter coxhead (talk) 20:38, 26 March 2016 (UTC)
- An email quoted by a reliable source, such as a newspaper, is surely at least as reliable as id the sender had posted the same text on his or her own web site. (And when the fact being supported is "X made Y statement on date Z", that is reliable enough.) Many email lists are automatically archived at the time of sending, and such archives are often publicly accessible. Would that be any less reliable than a post by the sender on the sender's own blog? Or an interview with the sender quoted by a reliable news source? If we are depending on manual archiving there would be room for doubt, although little room if multiple receivers confirm the accuracy of the archive. I think you are being too quick to dismiss all emails, and with them Wikipedia:Published, Peter coxhead. DES (talk) 21:19, 26 March 2016 (UTC)
- An email reported by a source is as reliable or unreliable as the reporting source, just as anything else it reported would be. If the e-mail is automatically archived in some way that is publicly accessible, then it surely counts as "published". If not automatically archived, I don't see how a third party can verify that other people have received the same e-mail, whereas it's easy to verify a blog posting. Peter coxhead (talk) 22:20, 26 March 2016 (UTC)
- We do regularly cite e-mail archives; it's just a website like any other, and the fact that the content on it originated in a different medium (e-mail) doesn't seem to be material. We cite newsgroup archives, etc., the same way. — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼ 05:06, 2 April 2016 (UTC)
- An email reported by a source is as reliable or unreliable as the reporting source, just as anything else it reported would be. If the e-mail is automatically archived in some way that is publicly accessible, then it surely counts as "published". If not automatically archived, I don't see how a third party can verify that other people have received the same e-mail, whereas it's easy to verify a blog posting. Peter coxhead (talk) 22:20, 26 March 2016 (UTC)
- An email quoted by a reliable source, such as a newspaper, is surely at least as reliable as id the sender had posted the same text on his or her own web site. (And when the fact being supported is "X made Y statement on date Z", that is reliable enough.) Many email lists are automatically archived at the time of sending, and such archives are often publicly accessible. Would that be any less reliable than a post by the sender on the sender's own blog? Or an interview with the sender quoted by a reliable news source? If we are depending on manual archiving there would be room for doubt, although little room if multiple receivers confirm the accuracy of the archive. I think you are being too quick to dismiss all emails, and with them Wikipedia:Published, Peter coxhead. DES (talk) 21:19, 26 March 2016 (UTC)
RFC: Is a change in citation markup method a change in citation style?
Some editors believe that a change in the wikisource formatting of citations, such as a change to or from List-defined References, or a change to or from the use of citation templates, is a change in citation style, and under WP:CITEVAR should not be made without obtaining consensus first, if there is an existing established citation style. Others believe that as long as the visible citations are unchanged, changing the wikicode does not amount to a change in style. Various arguments have been made in recent threads here on WT:CITE, as well as on WT:MOS. I ask for comments to resolve this issue with clear consensus, if possible. DES (talk) 21:55, 26 March 2016 (UTC)
Background
This issue was being discussed at least as far back as 2010, in Wikipedia talk:Citing sources/Archive 28#Consistent style. Changes to WP:CITE in October 2015 favored the "coding is style" view, but were disputed at the time and since. Recently the matter has become heated once more. Extended dispute over he meaning of an important guideline is not good, and should be settled by wider discussion. DES (talk) 22:02, 26 March 2016 (UTC)
- Comment: Here's a link to a possibly relevant previous discussion from 2010. There may be others between then and now. – Jonesey95 (talk) 22:37, 26 March 2016 (UTC)
Yes, coding is part of style
- Yes, coding is part of style, because it causes the same kind of mental adjustments and tool selection adjustments that editors would have to make when changing among completely different kinds of styles (APA Style vs Chicago Manual of Style, for example). Jc3s5h (talk) 22:43, 26 March 2016 (UTC)
- The formatting of the wiki-code is at least as important to editors, perhaps more important, than is the rendered appearance. Moreover, even if it makes no difference in appearance at a given moment, the use or non-use of templates may make a significant difference in future, as the templates might well be changed to emit different output. Moreover even now it is not really possible for the output to be identical: most of the citation templates emit machine-readable microformats, which manually formatted citations that are otherwise identical will not. Thus a change to or from the use of citation templates is a change in rendered output, even if that change is not readily visible to the non-automated reader. Similarly the use of WP:LDRs facilitates particular editing styles, and not others. Changing existing citations to use or not use LDRs without consensus is undesirable, as neither is favored or disfavored by any guideline. DES (talk) 22:41, 26 March 2016 (UTC)
- Yes. How citations are rendered is a stylistic choice, as much as is choosing to use APA over MLA. We don't mandate that users format citations using {{cite}} over {{citation}}, or use LDR over inline refs; it's a choice that someone has to make at some point in the article's development. Changing that is not "coding cleanup" unless we have a community consensus on a single approach that all articles must use - and we aren't there now. Until that happens, this is the best way to avoid endless edit-warring over optional preferences. Nikkimaria (talk) 01:40, 27 March 2016 (UTC)
- Perhaps you mean
{{cite ...}}
(cs1) v.{{citation}}
(cs2) since{{cite}}
redirects to{{citation}}
. We should make the distinction between the cs1 templates and{{citation}}
because they render differently:- Author (2016). Title. Publisher. –
{{cite book}}
- Author (2016), Title, Publisher –
{{citation}}
- Author (2016). Title. Publisher. –
- The cs1|2 templates have
|mode=
which allows cs1 to render in the same fashion as cs2 and vice versa so that the reader does not know that there are differences in the coding. - —Trappist the monk (talk) 11:06, 27 March 2016 (UTC)
- Perhaps you mean
-
- it's a choice that someone has to make at some point in the article's development There are two problems with this: (1) allowing "someone" to make a choice is pure WP:OWN (2) as noted below, coding style often needs to change with article length, so a choice made at one point should not be privileged over choices made later. Peter coxhead (talk) 11:12, 27 March 2016 (UTC)
- "it's a choice that someone has to make" is a simple statement of fact: if you are the first person to add a reference to an article, you obviously must make some sort of decision about how you are going to format that reference, since there is no mandatory style. As for your point (2), no, it doesn't need to change. You might prefer using different formatting for longer articles vs shorter, but it's perfectly possible to keep it the same. Nikkimaria (talk) 15:33, 27 March 2016 (UTC)
- it's a choice that someone has to make at some point in the article's development There are two problems with this: (1) allowing "someone" to make a choice is pure WP:OWN (2) as noted below, coding style often needs to change with article length, so a choice made at one point should not be privileged over choices made later. Peter coxhead (talk) 11:12, 27 March 2016 (UTC)
- Yes, coding is an important aspect of style for editors and I agree with Nikkimaria — until we have templates that can invoke styles through a
style=
-parameter — it is important that we don't allow arbitrary changes by saying "coding is not style". ~99% of so called "code-cleanup" consists of wasteful edits which only intend to inflate edit-counts of the editors who perform them. CFCF 💌 📧 14:24, 27 March 2016 (UTC)
- Yes, coding is part of style. The reason we have WP:CITEVAR is to avoid situations where people are discouraged from contributing by being forced to use unfamiliar citation styles. Coding differences is probably the most significant deterrent from contributing. Changing for example simple inline short references (Johnson 2001:59), to linked ones using the harv template creates the same visual output style, but the second forces the editor to learn much more wikimarkup. WP:CITEVAR just needs to be respected. This of course does not mean that you cant change <ref>M</ref> to <ref name=M> (which I think is basically the same style and requires the same level of expertise), but it does mean that if someone takes exception to that choice you need to go to the talkpage and argue your case.·maunus · snunɐɯ· 18:59, 27 March 2016 (UTC)
- @Maunus: Funnily enough if you make coding part of style sensu CITEVAR, then no-one can add "(Johnson 2001:59)" to a
{{Harv}}
based article, causing precisely the deterrent you want to avoid. All the best: Rich Farmbrough, 21:27, 31 March 2016 (UTC).
- @Maunus: Funnily enough if you make coding part of style sensu CITEVAR, then no-one can add "(Johnson 2001:59)" to a
- Yes. Coding is part of how citations are written. In fact, CITEVAR's origins go back to preventing editors from forcing citation templates onto manual cites. I don't use citation templates, so when I edit articles where templates are used, I add manual refs in a style that copies the style the templates use. But I fully expect someone to convert those refs to templates, because the preference of the first major contributor, or whoever is most active on the page, is for templates. Similarly, when manual cites are dominant, I expect those who prefer templates not to impose them. SarahSV (talk) 02:03, 28 March 2016 (UTC)
No, coding is not part of style
- Style is what readers see. Code is how we make that happen, and the exact methods for doing this change over time. Furthermore, the attempts by certain editors to prevent other editors from improving citation markup, without changing the rendered style of the citations, violates WP:EDITING, WP:OWN, and WP:NOTBUREAUCRACY policies; no one needs permission to do coding cleanup. It would also interfere with various bot and AWB work; is an unreasonable burden on other editors (to remember which tiny handful of editors out of our many thousands believe that no one is allowed to touch "their" cite coding); is not practical (citation coding requirements change over time, and thus must be updated whether their "owner" understands or not). And it's against the WP:5P ("mercilessly edited") and WP:COMMONSENSE principles. Among many other reasons covered in more detail in #Instruction creep in WP:CITEVAR, above, the most obvious being WP:CREEP. — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼ 22:03, 26 March 2016 (UTC)
- Clearly, otherwise every change to a Cite template would result in mass bannings and nuclear war. Moreover many improvements have made parameters obsolete. Many editors do not know how to use Cite templates, and welcome others improving their referencing. Some of us are not allowed to use things like "refill" and rely on others applying it.
- There are enough restrictions on who can edit and what they can do already, without making minor changes in references yet another prohibition.
- All the best: Rich Farmbrough, 18:55, 27 March 2016 (UTC).
- The best solution to problems of this kind is to gain community consensuses on the largely inconsequential details that we spend so much time arguing about, and codify those consensuses in the guidelines so that we are all on the same page. Except for the few who refuse to follow a guideline, even on such trivial details, because they disagree with it and "it's only a guideline", that solution would put an end to the arguing.
For example, we should be able to reach a consensus that a refname should be meaningful. And we should be able to reach a consensus on preferred parameter formatting within a citation template. There is no merit in an argument that such decisions should be left to local consensus because, for example, some articles might need a space before the separator character while others might be better served by no space there; these are cases where one size could in fact fit all.
The common argument against such guidelines, CREEP, is easy to defeat. Guidelines are justified when they discourage editors from spending a vast amount of time arguing about relatively unimportant things. A legitimate function of guidelines is (or should be) the elimination of bikeshed opportunities. The guiding principle should be sensible use of editor resources—not individual freedom of expression.
Another opposing argument, that it would make it more difficult to learn editing, is also fallacious. No editor would be forced to conform, and no editor would be chastised if they did something "wrong", but they should not object when "their" coding is modified to conform by someone else (preferably with a link to the guideline in the editsum). Again, no time wasted arguing about it.
I see no other good solution to the problems addressed in this RfC, but to corrupt the intent of CITEVAR would be a step in the wrong direction. ―Mandruss ☎ 09:23, 28 March 2016 (UTC)- In that example I like meaningful ref names. But I would not object to someone improving an article by naming references with "ref1" etc.. Someone else can add meaning later (if they want to). However someone changing "BBC" to "Ref1" would get a "stern look".
-
- I think that means I agree with you.
- All the best: Rich Farmbrough, 23:02, 29 March 2016 (UTC).
- Seems a bit superfluous at this point. If I edit a page with the VisualEditor, I end up inserting citations which are "styled differently" from those on the page currently without thinking about the current style or knowing that I'm inserting a variant. Likewise I am very skeptical of proposals which would hem in our ability to update references to templates (or some future tool) by invoking a longstanding policy that was developed for (largely) different purposes. Protonk (talk) 16:27, 10 April 2016 (UTC)
Partly, some aspects of coding are relevant, others not
- I am a strong advocate of using templates for citations. They are simpler to use, consistent in rendered appearance and facilitate the generation of metadata. I would favour a change in policy that favoured the widespread and general conversion to templates, on the grounds that they are objectively better. That said, I know that some editors oppose their use. I see, at minimum, a need to separate the question of whether to use templates from some other questions about their more subtle use.
- There are (at least) two other issues of whitespace, formatting and layout that have no effect on the rendered page and for one of them is only of strong relevance within the templated context anyway. Despite that, both have been contentious of late. These are the use of
{{Reflist|refs= ...}}
and also the layout of parameters to a citation template, either in-line or per-line, and with or without whitespace. Whatever is (if anything) decided about conversions to and from templates overall, we should separate out these changes as clearly being, "minor changes that affect neither rendered behaviour, nor the overall use or not of templates". - As to our behaviour over these changes, then I see that there is very little reason to prohibit changes in either direction, other than a general reluctance against churn or edit-warring. As they are so inconsequential to the finished result, there can be little strong reason to favour one over the other. For the use of lists of refs in particular, this is likely to become more attractive as an article grows in size. The current view that "preserving the status quo is all that matters" is deeply unhelpful here. Andy Dingley (talk) 02:12, 27 March 2016 (UTC)
- Agreed. I would modify my own comment above with the suggestion that these matters might be something we want to have guidance on, but it/they must be separated out from the style issue. The problem is that we have a guideline on not changing from AMA to Harvard style (or whatever). Some people separately want to also not change from LDR to entirely-inline references, or (when inline) from horizontal to vertical or (when LDR) from vertical to horizontal. These are not style matters, they're coding standards. But by shoe-horning them into the sentence on style (what the reader sees) it opens the unintentional back door of people editwarring to the death other whether you can change their
<ref name=Washin_08/>
[or worse – I see crap like<ref name=_39xps#p8/>
with increasing frequency] to<ref name="Washington 2008" />
. The present situation is untenable, robbing Peter to pay Paul by defending one kind editwarring to prevent another. — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼ 03:02, 27 March 2016 (UTC) - Agreed. We need to separate out the different issues. In particular, a coding style for a short article, however long-standing, is often not appropriate for a longer article, particularly when the same sources are used repeatedly. It's unacceptable that a coding style adopted when an article first ceases to become a stub is then imposed on editors when the article undergoes a major expansion; it can discourage them from expanding the article. (I've a short list of articles I won't work on because of encountering this issue.) On the other hand, there are different ways of dealing with referencing in long articles (e.g. using the sfn family of templates or not), and flipping between them isn't helpful, which is why I say the issues need to be separated out. Peter coxhead (talk) 11:06, 27 March 2016 (UTC)
- If editors working on an article think that, as it grows, a different coding style would be an improvement, all that they need do is hold a brief talk page discussion to establish that that is the new consensus for the article. What should not happen is for one editor to decided that he knows better than all other editors and make a wholesale change without consulting other editors who work on the article. What is even less desirable is edit wars over coding style. The point of CITEVAR is to prevent such edit wars, and prevent such unilateral undiscussed changes. But if a change is clearly an improvement, then obtaining consensus to change should be quick and easy, If it isn't so easy, maybe its status as an improvement isn't so clear. DES (talk) 16:09, 27 March 2016 (UTC)
- Or maybe there's a small group of editors with fixed views and a sense of ownership. The point is that most articles have few editors working on them so a local consensus builds up. That's why we need project-wide guidance. Peter coxhead (talk) 18:30, 27 March 2016 (UTC)
- Yes. The key problem here is the "other editors who work on the article" sentiment, the WP:VESTED / WP:OWN belief that whoever has put in more time at a particular article has "tenure" at it and must be obeyed or at least deferred to. This is an un-wiki and anti-policy viewpoint that needs to be scrubbed away vigorously and with strong solvents until no trace remains. — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼ 21:16, 28 March 2016 (UTC)
- Whereas the idea that who ever manages to get their preference encrusted in the MOS has the moral right to make everyone else conform to their wishes on any detail of writing and presentation is profoundly wikipedian? People who are WP:VESTED in their WP:OWN citation preferences are no less WP:VESTED than those who happen to WP:kNOW about a topic and therefore take a special interest in writing about it and like to have some say about how that knowledge is represented.·maunus · snunɐɯ· 23:59, 31 March 2016 (UTC)
- @Maunus: No-one is suggesting imposing a single visible uniform citation style, and I would be strongly opposed if they did. The main issue is whether WP:CITEVAR should give the first main contributor even more control in even more detail over the internal coding of referencing, which is quite a different matter. Peter coxhead (talk) 08:22, 1 April 2016 (UTC)
-
- In fact I think, lots of people are waiting for any sign that citevar is loosing consensus to do just that. It is not about "more" control, since the difference between hand formatted and coded referencing is in fact the original context for wp:citevar. Of course the same rules should apply to the case with the difference between <ref> tags and {{sfn|}} cites for example. There just is no reason why anyone's preference for one or the other should overrule the first style that is systematically used in an article. Changes to citation coding can always be changed through normal talkpage consensus. But as long as it is one persons preference vs. the preference of another, then of course it makes sense to use WP:CITEVAR as a grandfather clause.·maunus · snunɐɯ· 18:07, 1 April 2016 (UTC)
-
- @Maunus: No-one is suggesting imposing a single visible uniform citation style, and I would be strongly opposed if they did. The main issue is whether WP:CITEVAR should give the first main contributor even more control in even more detail over the internal coding of referencing, which is quite a different matter. Peter coxhead (talk) 08:22, 1 April 2016 (UTC)
- Whereas the idea that who ever manages to get their preference encrusted in the MOS has the moral right to make everyone else conform to their wishes on any detail of writing and presentation is profoundly wikipedian? People who are WP:VESTED in their WP:OWN citation preferences are no less WP:VESTED than those who happen to WP:kNOW about a topic and therefore take a special interest in writing about it and like to have some say about how that knowledge is represented.·maunus · snunɐɯ· 23:59, 31 March 2016 (UTC)
- Yes. The key problem here is the "other editors who work on the article" sentiment, the WP:VESTED / WP:OWN belief that whoever has put in more time at a particular article has "tenure" at it and must be obeyed or at least deferred to. This is an un-wiki and anti-policy viewpoint that needs to be scrubbed away vigorously and with strong solvents until no trace remains. — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼ 21:16, 28 March 2016 (UTC)
- Or maybe there's a small group of editors with fixed views and a sense of ownership. The point is that most articles have few editors working on them so a local consensus builds up. That's why we need project-wide guidance. Peter coxhead (talk) 18:30, 27 March 2016 (UTC)
- If editors working on an article think that, as it grows, a different coding style would be an improvement, all that they need do is hold a brief talk page discussion to establish that that is the new consensus for the article. What should not happen is for one editor to decided that he knows better than all other editors and make a wholesale change without consulting other editors who work on the article. What is even less desirable is edit wars over coding style. The point of CITEVAR is to prevent such edit wars, and prevent such unilateral undiscussed changes. But if a change is clearly an improvement, then obtaining consensus to change should be quick and easy, If it isn't so easy, maybe its status as an improvement isn't so clear. DES (talk) 16:09, 27 March 2016 (UTC)
- Agreed. I would modify my own comment above with the suggestion that these matters might be something we want to have guidance on, but it/they must be separated out from the style issue. The problem is that we have a guideline on not changing from AMA to Harvard style (or whatever). Some people separately want to also not change from LDR to entirely-inline references, or (when inline) from horizontal to vertical or (when LDR) from vertical to horizontal. These are not style matters, they're coding standards. But by shoe-horning them into the sentence on style (what the reader sees) it opens the unintentional back door of people editwarring to the death other whether you can change their
Extended discussion of the Style RFC
- It doesn't really matter whether the use of citation templates is "part of style" or not. The key point is that there has never been consensus for either of these two options: (1) all citations should be converted, over time, to use citation templates; (2) all citations should be converted, over time, to not use citation templates. Because there is no general agreement over whether to use citation templates, the guideline here is neutral about the use or non-use of templates, and at the same time the guideline here discourages users from converting articles that are well-established with or without templates from being randomly "switched" to the other method. Even if citation templates are "not part of style", the guideline would still explicitly say not to change articles from one method to another, at least not without gaining consensus that the current method is inappropriate for the needs of the article. That fact is really what should be discussed at the RFC; the question about what is "really" a "part of style" is mostly irrelevant. — Carl (CBM · talk) 22:18, 26 March 2016 (UTC)
- @CBM: the October 2015 addition is concerned with the placement of citation templates, not whether they should be used or not, and so goes well beyond "templates or not". The current wording is used to defend all kinds of "coding style", such as whether the parameters to templates are placed on separate lines or not. What absolutely needs to be discussed is what counts as "style", and how to prevent continual creep of the guidelines. Peter coxhead (talk) 22:42, 26 March 2016 (UTC)
- As WP:CITEVAR says that what should not be changed without consensus is "style" the question of what requires a consensus discussion to change is the question of what constitutes "style" unless we opt to change how CITEVAR is worded. The real question here is what things require a local consensus in advance to change. DES (talk) 22:46, 26 March 2016 (UTC)
- I think that SMcCandlish is badly mistaken is several points made in the "no" section above. An insistence that coding style as well as rendered style be RETAINED, and not changed without consensus, is not a violation of WP:OWN It does not privilege any one editor, what it does privilege is the established stable style, if one exists. Anyone may change it after obtaining consensus, no one may change it without doing so, not even the "first major contributor". Nor does it require remembering "which tiny handful of editors out of our many thousands believe that no one is allowed to touch "their" cite coding" because it applies equally to every editor's cite coding, or more exactly, to every article's existing consistent coding style, where one does exist. Neither bots nor AWB users should be making automated or semi-automated changes that change such a style -- they might well make changes to move inconsistent, more recently added cites to the pre-existing style of the article. That will require a good algorithm to determine that style, or else use of human judgement to filter the list of articles to be worked on. But that is probably a good idea anyway. As this has been understood by many editors (and I argue has been the accepted albeit not unopposed consensus) for years, this is not CREEP, it is merely clarifying an existing guideline. DES (talk) 22:57, 26 March 2016 (UTC)
- It absolutely privileges any one editor, namely a) the first major contributor who believes the have WP:VESTED editorial rights (a falsehood perpetuated by the wording in this page, and often vigoriously defended by wikilawyering by various people defend that anti-policy wording), or b) any random editwarrior who wants to over-control every aspect of a squatted-on article which pretty much no one else cares about, but which some gnome tried to improve before stepping on the hidden land mine that the page has a would-be owner. These are the two most common cases I encounter. I believe they're also among the cases Peter coxhead runs into, though I think his major concern is orthogonal to my main one, being only one of my secondary ones: people resisting LDR formatting without any actual rationale for their stonewalling, just an "I don't like it" complaint, using the incorrect and anti-policy interpretation of CITEVAR as a system-gaming platform from which to filibuster changes they don't understand or want to resist for the same of opposing change generally, but cannot formulate an actually rational objection to.
If you have an issue with what a bot or AWB is doing, take it up at the bot and AWB talk pages. There may be a potential problem of an AWB user mistaking a lone case of a citation done one way as the norm in an article actually otherwise done a different way, and mis-normalize to the non-majority style, but this doesn't seem to come up often enough to care about. It seems to cause fewer problems than editor-vs.-editor filibustering of code improvements, which is the reason we're here disputing your view on this.
A more sensible solution to the whole matter would be to normalize to the majority style in the article (aside from recent changes that altered existing citations). If I come to "your" C-class article you just improved from a stub, and I triple the size of it and the number of citations in it, why should you have any more say over what the citation style is to be now, other than as one editorial voice among all others? I know Curly Turkey takes this view (it's his #1 concern raised about CITEVAR, at any rate), but perhaps it's a different proposal for another time.
The real point is that whatever the best solution may be, there's definitely not a consensus in favor of the one you prefer, or there would not be so much dispute about this. "I get my way pretty often in bitter fights, and people give up because I fight to the end" is the approach too many have taken to this. It is not evidence of consensus, but of tendentious possessiveness that everyone rolls they eyes about and moves on because its petty and a waste of time. Ask yourself: How does preventing Peter Coxhead from using LDR in this article, or even preventing Jane Doe from using APA references in this article instead of Vancouver ones, for that matter, help the reader? Ask yourself also this: If this provision was enacted to prevent editorial disputes but is just leading to a long-term and increasingly dug-in editorial dispute of a broader nature, how well is it working? We tried it for a while, for lack of anything better to try, and it has not worked well. There's an old saying that the definition of insanity is to keep doing the same thing but expect different results. So, let's not be crazy. Is broke, do fix it. — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼ 00:16, 27 March 2016 (UTC)
- It absolutely privileges any one editor, namely a) the first major contributor who believes the have WP:VESTED editorial rights (a falsehood perpetuated by the wording in this page, and often vigoriously defended by wikilawyering by various people defend that anti-policy wording), or b) any random editwarrior who wants to over-control every aspect of a squatted-on article which pretty much no one else cares about, but which some gnome tried to improve before stepping on the hidden land mine that the page has a would-be owner. These are the two most common cases I encounter. I believe they're also among the cases Peter coxhead runs into, though I think his major concern is orthogonal to my main one, being only one of my secondary ones: people resisting LDR formatting without any actual rationale for their stonewalling, just an "I don't like it" complaint, using the incorrect and anti-policy interpretation of CITEVAR as a system-gaming platform from which to filibuster changes they don't understand or want to resist for the same of opposing change generally, but cannot formulate an actually rational objection to.
-
-
-
- Just to clarify: my particular examples were ukiyo-e and comics, wher I completely scrapped what was there are rewrote the articles from scratch. This is not SMcCandlish's example of "If I come to 'your' C-class article you just improved from a stub, and I triple the size of it and the number of citations in it". I don't necessarily disagree with SMcCandlish, but it's not really the same as my situation. My point was that it would be merely disruptive and WP:LAWYER-ly to force the previous ENGVAR (assuming there was one) on an article that was completely rewritten from scratch. Curly Turkey 🍁 ¡gobble! 11:15, 27 March 2016 (UTC)
- Thanks for the clarification; I wasn't meaning to mis-analogize your cases. We are arguing the same point, I think. The wikilawyering about forcing the previous CITEVAR (or, rather, against introducing a consistent one or a better one) after major article improvement is what brings me to this discussion in the first place. I never have trouble with people over citation formatting except when expanding a poor, barely-above-stub article, and thereby rousing some page "owner" from a year or three ago who starts using every excuse they can think of to put the article back to their perfect version. I've also seen (and experienced) it used in a hounding manner by editor A who has a personal beef with editor B, following B around article after article in a particular topic area in which A desires control and desires B's absence, using CITEVAR as a bogus revert excuse, just another way to may life difficult for B at "A's own" articles. This can be particular destructive, given the habit of some editors of using a revert-all-work-punitively-to-object-to-one-particular-change tactic of editorial participation discouragement. This kind of nonsense is only possible under the "CITEVAR means code not style, despite being written to address style" interpretation. — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼ 21:12, 28 March 2016 (UTC)
- Just to clarify: my particular examples were ukiyo-e and comics, wher I completely scrapped what was there are rewrote the articles from scratch. This is not SMcCandlish's example of "If I come to 'your' C-class article you just improved from a stub, and I triple the size of it and the number of citations in it". I don't necessarily disagree with SMcCandlish, but it's not really the same as my situation. My point was that it would be merely disruptive and WP:LAWYER-ly to force the previous ENGVAR (assuming there was one) on an article that was completely rewritten from scratch. Curly Turkey 🍁 ¡gobble! 11:15, 27 March 2016 (UTC)
- So, in lieu of the current policy, we would have some number of completely avoidable edit wars, and a larger number of completely avoidable discussions on talk pages where people waste their time arguing whether LDR is better or worse, rather than working on actual article content. This is the WP:BIKESHED effect spread over thousands of articles. The current guideline has an easy fix: just leave the current style alone. That fix, like ENGVAR, *has* worked well. Very rarely do I see people come around my large watchlist randomly changing citation styles. Almost everyone does what the guideline says. The small number of self-appointed "gnomes" who insist that their preferences should override the preferences of everyone else are usually reasonable enough to cut it out once the fact their their preferences are not actually guidelines is noted. — Carl (CBM · talk) 00:29, 27 March 2016 (UTC)
- This isn't a policy. WP:EDITING is the policy. The current guideline is not up for deletion, so I'm not sure what you mean by "in lieu of". What's under discussion here is whether editor A can prevent editor B from improving underlying citation code, in ways that are invisible to the reader, and can filibuster in this way with no rationale at all for the objection other that bible-thumping CITEVAR as if it applied to citation code formatting, not style as it actually says. Editwarring would not magically become permissible all of sudden, under any interpretation. The very fact that editor A in this scenario is gaming a guideline, against policy, to engage in a form of editwarring, however, is a good summary of the problem that brought us here. I'm not sure why you think the bicycle shed and debate problem is in any way improved by the interpretation you seem to favor. The exact opposite is the case. If editor B is making improvements to the citation code that make the article easier to work with, this increases the likelihood that people will work on it, and the efficiency with which they can do so. No one without a real argument is likely to get in the way. However, if there's a perception that CITEVAR allows someone to object to any citation-related alteration of any kind at all, and that they'll get their way if they can stonewall again consensus-forming without a real reason, just CITEVAR by itself – which is precisely the interpretation being advanced – this gives a very strong incentive to WP:OWN / WP:VESTED types to raise as much hell as possible on the talk page, hand-waving at CITEVAR until they are blue in the face, without ever raising a single substantive objection to the changes. If another couple of fans of this interpretation of CITEVAR show up (and they always seem to as if out of the blue ... I wonder why that is, since so few editors actually hold this view?), the over-controller of the citation code will probably WP:WIN, and the article will continue to be their personal playground, and will suffer in quality as a result. — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼ 02:29, 27 March 2016 (UTC)
- "If editor B is making improvements to the citation code that make the article easier to work with," then there will be consensus to document those improvements directly into the WP:CITE or WP:MOS guideline, and CITEVAR won't apply. Invariably, the actual issues are with things for which there is *not* a general consensus that they are improvements, such as list-defined references, citation templates, etc. The argument by editor B that they are "improving" the article falls flat in these cases, where editor B is only imposing their own optional preferences over someone else's optional preferences, and there is no consensus that either is better. If the guideline says that either of two options is acceptable, there is no reason to think that changing from one to another is an "improvement". — Carl (CBM · talk) 02:51, 27 March 2016 (UTC)
- We're talking past each other, since our concerns about CITEVAR are largely unrelated. I can tell you from a mostly observer perspective (LDR isn't a big issue to me) that most of us have figured out by now that LDR is very helpful in long articles, and not so much in short ones. The exact same thing is true of SFN formatting and splitting references up into two refs sections, short and long. The disputes really are about a) which of these is a better approach, and b) at what point an article needs one of them. The underlying brainfart in CITEVAR, when wrongly applied to cite formatting not cite style, is that it is massive roadblock to ever coming to consensus on these matters at any article at all, because it ossifies the status quo. Secondly, the idea that there is not a broad consensus for templated references is totally untenable, sorry. There are a handful of people who hate them, and that's it. Everyone else uses them, and there is no real controversy in imposing them. I do it very frequently, and have run into people reverting me on that, in over ten years, fewer times than I can count on a single hand. Maybe reversing this situation is important to you in this RfC. Good luck. I have my eye on a more likely outcome: no one has any business reverting me on CITEVAR grounds if they have put in
<ref name=7>
and I change this to<ref name="Hendricks 2015">
so it's actually helpful and isn't meaningless noise. — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼ 03:26, 27 March 2016 (UTC)
- We're talking past each other, since our concerns about CITEVAR are largely unrelated. I can tell you from a mostly observer perspective (LDR isn't a big issue to me) that most of us have figured out by now that LDR is very helpful in long articles, and not so much in short ones. The exact same thing is true of SFN formatting and splitting references up into two refs sections, short and long. The disputes really are about a) which of these is a better approach, and b) at what point an article needs one of them. The underlying brainfart in CITEVAR, when wrongly applied to cite formatting not cite style, is that it is massive roadblock to ever coming to consensus on these matters at any article at all, because it ossifies the status quo. Secondly, the idea that there is not a broad consensus for templated references is totally untenable, sorry. There are a handful of people who hate them, and that's it. Everyone else uses them, and there is no real controversy in imposing them. I do it very frequently, and have run into people reverting me on that, in over ten years, fewer times than I can count on a single hand. Maybe reversing this situation is important to you in this RfC. Good luck. I have my eye on a more likely outcome: no one has any business reverting me on CITEVAR grounds if they have put in
- "If editor B is making improvements to the citation code that make the article easier to work with," then there will be consensus to document those improvements directly into the WP:CITE or WP:MOS guideline, and CITEVAR won't apply. Invariably, the actual issues are with things for which there is *not* a general consensus that they are improvements, such as list-defined references, citation templates, etc. The argument by editor B that they are "improving" the article falls flat in these cases, where editor B is only imposing their own optional preferences over someone else's optional preferences, and there is no consensus that either is better. If the guideline says that either of two options is acceptable, there is no reason to think that changing from one to another is an "improvement". — Carl (CBM · talk) 02:51, 27 March 2016 (UTC)
- This isn't a policy. WP:EDITING is the policy. The current guideline is not up for deletion, so I'm not sure what you mean by "in lieu of". What's under discussion here is whether editor A can prevent editor B from improving underlying citation code, in ways that are invisible to the reader, and can filibuster in this way with no rationale at all for the objection other that bible-thumping CITEVAR as if it applied to citation code formatting, not style as it actually says. Editwarring would not magically become permissible all of sudden, under any interpretation. The very fact that editor A in this scenario is gaming a guideline, against policy, to engage in a form of editwarring, however, is a good summary of the problem that brought us here. I'm not sure why you think the bicycle shed and debate problem is in any way improved by the interpretation you seem to favor. The exact opposite is the case. If editor B is making improvements to the citation code that make the article easier to work with, this increases the likelihood that people will work on it, and the efficiency with which they can do so. No one without a real argument is likely to get in the way. However, if there's a perception that CITEVAR allows someone to object to any citation-related alteration of any kind at all, and that they'll get their way if they can stonewall again consensus-forming without a real reason, just CITEVAR by itself – which is precisely the interpretation being advanced – this gives a very strong incentive to WP:OWN / WP:VESTED types to raise as much hell as possible on the talk page, hand-waving at CITEVAR until they are blue in the face, without ever raising a single substantive objection to the changes. If another couple of fans of this interpretation of CITEVAR show up (and they always seem to as if out of the blue ... I wonder why that is, since so few editors actually hold this view?), the over-controller of the citation code will probably WP:WIN, and the article will continue to be their personal playground, and will suffer in quality as a result. — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼ 02:29, 27 March 2016 (UTC)
-
-
- SMcCandlish above wrote that CITEVAR, when applied to coding style rather than visible style, ossifies the status quo. If taken to its extremes (which some of those posting here seem determined to do) it does, and so would have prevented some of the recent developments in editing tools and the re-working and improvement of the cite/citation templates. Fortunately, those involved have rightly ignored this extreme interpretation. Peter coxhead (talk) 07:47, 29 March 2016 (UTC)
Identifying the actual concerns and their validity
This RfC is basically a false dichotomy and we're talking past each other a lot because many of our concerns are tangential to those of others. I can identify at least all of the following issues (not all of which are valid) in the above discussions, and feel free to add more. Some are clearly much more legitimate than others, and it's pretty obvious that there's a culture clash here between GA/FA-focused editors and stub-expansion editors, both of which camps feel they're doing more important work than the other.
- There are concerns that undiscussed changes of style (i.e. visual display) of citations in an article may be disruptive of article stability. (This is alleged to affect both readers and editors.)
- There are concerns that undiscussed changes of certain kinds of coding standards (article-wide formatting) of citations in an article may be disruptive of article stability. Examples include WP:LDR, WP:SFN, and the conversion to or stripping of templated citation formatting. (This is a different analysis, and largely on affects particular editors who like or dislike particular code approaches, though removal of template citations also destroys metadata and introduces maintenance problems.)
- There are concerns about WP:GAMING by using the pretense that undiscussed changes of trivial code formatting (a third category of change, affecting a particular cite, set of cites, or use of particular templates) may supposedly be disruptive of article stability, without any plausible rationale to support such a notion.
- There is a dispute (thus the RfC above) about whether these concerns should be commingled as if identical despite raising different potential problems in different circumstances, or treated separately for clarity (or even at all, in the case of code).
- There are concerns about misuse of CITEVAR to prevent coding-standards changes (the article-wide sort) that are needed for editorial efficiency, on the basis that it's "change in style", a disputed claim.
- There are concerns that over-extension of CITEVAR's scope enables furtherance of interpersonal disputes between editors by providing a "weapon" for revertwarring over trivia, to the detriment of community peace, article quality, and maintainability.
- There are disputes about which citation style (i.e. visual display) is most appropriate for a particular article, often tied to what is done in journals in the field to which the article's scope most pertains (though there can often be more than one, thus many of these disputes).
- There are disputes about whether WP should be importing external styles, and they feel we should stick with WP:CS1. (There does appear to be consensus to permit these styles.)
- There are disputes about whether WP should be permitting made-up styles that are neither CS1 [including CS2, now handled by CS1 templates, and increasingly merging], nor a recognized, documented style found in professional literature. (There is a local consensus at this page to permit this, but it has not actually be subjected to a site-wide consensus discussion. Given WP's hostility to "made up nonsense" it is actually unlikely that the community really supports the idea that fake citation styles cannot be changed without a big discussion, even if it would support the idea that citations may be added in any form as long as they're usable.)
- There are disputes about which citation-grouping coding standards (invisible to the user) are best for editors, at a particular level of article development, or when the use of WP:LDR or WP:SFN is warranted.
- There are disputes about which citation template-layout coding standards (horizontal vs. vertical) are best for editors, depending on citation placement. (In reality, we know from experience that horizontal works best for full citations when used inline, since it does not interfere with the ability to understand that paragraph structure of he document, and that vertical works best when full citations are grouped at page bottom, via either SFN or LDR, since it makes the citation details easier to parse and each cite easier to distinguish from the next.)
- There are disputes about which citation template parameter-spacing coding standards are best for editors. (In reality we know that: 1) in horizontal cites,
{{cite foo |para1=value1 |para2=value2 |...}}
is the vast-majority usage and that weird formatting like{{cite foo | para1 = value1 | para2 = value2 | ...}}
, or{{cite foo| para1= value1| para2= value2| ...}}
is cleaned up on sight as a parseability problem, with virtually no one ever reverting it; and 2) aligning=
characters in vertical templates is common but not enforced, and also virtually never reverted.) - There are concerns about misuse of the "consistent" provision to game the system by adding a few cites in a different style then claiming the style is inconsistent and imposing a new one. (See below for a tool that could short-circuit such behavior, if it's actually even happening.)
- There are disputes about expectations that an editor who did more work at an article – a) to get it past stub stage, b) in total, or c) recently – has more say about what citation style is to be used, what citation coding standards are employed, and what citation cleanup tweaks are permissible. (Policy does not support either of the first two cases – see WP:EDITING, WP:OWN and associated pages like WP:5P and WP:VESTED – and seems to favor the latter, since all of WP is built and maintained by the WP:CONSENSUS of who is doing the work now, not who was around in 2007, and consensus can change at any time. Further, the "first major contributor" rule was determined to be problematic at MOS and removed, further evidence of consensus against the first of these "control criteria". Nevertheless, we have a long-standing proviso site-wide that in the event of a dispute and its failure to reach resolution, preserve the status quo ante.)
- There are concerns that even the "first post-stub revision to introduce a style" rule, which replaced "first major contributor", may interfere with our normal status quo ante rule.
- There are concerns that despite attempts to resolve this WP:CONLEVEL conflict that people are still incorrectly interpreting CITEVAR (and WP:ENGVAR, WP:DATEVAR, WP:TITLEVAR) as "I did it this way, so you can't change it" WP:OWNership rules, when in fact they were intended as last-resort default when no other solution can be reached.
- There is a dispute about CITEVAR and other style-related material in WP:CITE being a WP:POVFORK from WP:MOS, and how to reintegrate them to prevent further conflict.
I think the causes of this fracture are: Firstly, GA and FA tend to give a lot of deference to whoever worked the most on an article (this is a choice by the participants at those processes, and is not a WP policy matter when it comes to who may edit and how). Secondly, some WP:MOS-focused editors, early on, resisted the use of external citation styles that do stylistically unusual things like using smallcaps extensively; but none of the current MoS regulars seems to care any longer, so this rift is illusory and only being maintained by WP:CITE regulars.
- I think per point 4, the point-by-point analysis suggests that there should really be multiple RFCs. The RFC above doesn't say "making refnames understandable and converting between templated/untemplated citations are the same", it says "CITEVAR covers coding yes/no". We're getting more division on that because of different interpretations of "coding"; if we ran more limited RFCs on some of the individual points above (particularly 1-3), we would get a clearer consensus. Points 7-12 really shouldn't be part of a CITEVAR RFC unless/until a different RFC is run to determine which style is better/preferred/deprecated, because until that point all styles are available on the table. (And I would like to see some citations on some of your interpretation above). Nikkimaria (talk) 19:54, 29 March 2016 (UTC)
- Right. The point of this exercise is to see what the actual sticking points are, and RfC those separately. — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼ 05:07, 2 April 2016 (UTC)
Extracting referencing and citation information from articles for evaluation
Recent discussions, especially Instruction creep in WP:CITEVAR and RFC: Is a change in citation markup method a change in citation style?, got me wondering if there could be a tool that might aid editors in evaluating the coding part of citations. To that end I've hacked up a proof of concept template that takes a stab at doing that.
The template is {{ref info}}
. It can be saved in an article so that the referencing information is automatically available to any editor who clicks the Show preview button or as the examples here show, can take a single parameter that is the title of an article of interest:
{{ref info}} |
{{ref info|alzheimer's disease}} |
{{ref info|Barack Obama}} |
{{ref info|aristotle}} |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
|
|
This is just a hack that requires improvement if it is to be useful. I would like your opinions. Does the concept have merit? Should it be pursued? What information should the template provide? How should the results be displayed? What else?
Please comment at the template's talk page.
—Trappist the monk (talk) 19:52, 27 March 2016 (UTC)
- That is really cool, and useful! Only things I can think of right off-hand to add to it are: noting if the page has citations divided into two short and long sections; the article class(es) assigned on talk page; a
|sidebar=y
to wrap it in something like{{archive box}}
; count of citation errors (of the red alert kind); detect citation dispute/cleanup templates (dead URL, unreliable source, failed verification, etc.); maybe detection of some variant templates in Category:Footnote templates and Category:Citation templates and subcats thereof (but not Category:Specific-source templates – the general consensus at TfD is converging on the idea that all of these need to substitute cleanly and be subst-only templates so that they generate standard CS1 (or 2) citations, or they're liable to be deleted; I'm in process of working this out, but the amount of actual work to implement that solution is daunting, since it's over 600 templates to convert in very fiddly ways).
- I agree, Trappist, this is cool and potentially very useful. One thing I would like to suggest, an additional parameter taking a revision ID. Then when there is a talk page discussion about how citation markup has changed in a particular article, the output for two specific revisions could be compared side by side. This might even eventually be a tool used at GA or FAC reviews where consistency is valued. Thanks. DES (talk) 17:18, 29 March 2016 (UTC)
-
- Indeed. It could also obviate the "someone could add a few citations in a different style, then claim inconsistency, and impose a new style" WP:GAMING concerns that some have had, since it could easily show that a particular style was largely consistent before and after the change. We could even have a specific cut-off, e.g. 75% consistent = "consistent" or whatever. This could also obviate countervailing concerns I and Curly Turkey and Peter Coxhead have about ossification of style (i.e., if the article expands by more than 50% or whatever in number of valid citations due to someone's recent work, old citation bets are off, and normal WP:BOLD would apply, with a necessity to justify keeping the old one on its actual merits not on "local tradition". Putting actual numbers to this should forestall a lot of disputes, and bring CITEVAR back into common sense territory and resolve many of its conflicts with WP:EDITING, WP:BOLD, WP:OWN, etc.
It still does not address the clear distinction between code and content, and I think we're arriving at a tripartite distinction between: 1) actual citation style (CS1, Vancouver, Jim-Bob's made up one, whatever) seen by the reader; 2) coding standards like LDR and SFN, or vertical vs. horizontal templating, or mass-conversion to/from using templates at all; and 3) trivial coding tweaks like using canonical not alias parameter names of templates, using
<ref name=
values that are human-readable, or using a particular footnote template that provides an additional feature. Regardless of those questions,{{Ref info}}
is a major step in the right direction. — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼ 17:36, 29 March 2016 (UTC)
- Indeed. It could also obviate the "someone could add a few citations in a different style, then claim inconsistency, and impose a new style" WP:GAMING concerns that some have had, since it could easily show that a particular style was largely consistent before and after the change. We could even have a specific cut-off, e.g. 75% consistent = "consistent" or whatever. This could also obviate countervailing concerns I and Curly Turkey and Peter Coxhead have about ossification of style (i.e., if the article expands by more than 50% or whatever in number of valid citations due to someone's recent work, old citation bets are off, and normal WP:BOLD would apply, with a necessity to justify keeping the old one on its actual merits not on "local tradition". Putting actual numbers to this should forestall a lot of disputes, and bring CITEVAR back into common sense territory and resolve many of its conflicts with WP:EDITING, WP:BOLD, WP:OWN, etc.
I had thought about using an id to look at older versions of an article. I've just spent a bit of time trying to get that to work, but alas, no success. I've posted at WP:LUA to see if there is a solution.
I'll take the suggestion portions of the above comments and make a list of them at Template talk:Ref info.
—Trappist the monk (talk) 19:25, 29 March 2016 (UTC)
- I agree it's cool. One bit of info for articles with templates that's nice to know is the date format of publication dates and the set {accessdate, archivedate}.
- A tangential matter: it would be nice to have an agreed-upon way to record which (if any) external style guide an article follows, such as MLA or Chicago. But that would be separate from this tool.
- It's also worth noting that the tool does not distinguish between an article with no citations at all versus an article that has parenthetical citations but no templates. Jc3s5h (talk) 20:11, 29 March 2016 (UTC)
- Example of that last case?
- —Trappist the monk (talk) 22:35, 29 March 2016 (UTC)
- Depends on what Jc3s5h meant by "parenthetical citations but no templates". Does "templates" refer to properly formatted footnotes? Does "parenthetical citations" refer to the kind one might see in print books where "(Keene 1999 : 165)" appears inline but one would need to look closely to find them? If the answer to both is "yes", then I have seen such articles around, but I can't name them off the top of my head, and they would be difficult to locate on request, except by memory. These articles have their own problems that are largely unrelated to verifiability (they were probably written by new users unfamiliar with Wikipedia policies and guidelines, and, worse, may have been copy-paste jobs). Hijiri 88 (聖やや) 06:52, 3 April 2016 (UTC)
- Tropical year. Jc3s5h (talk) 09:55, 3 April 2016 (UTC)
- Yeah, that's exactly what I thought. I actually don't see a problem with not making the distinction, though, since the citation style in that article is much less practical for an online encyclopedia article than the more standard footnote style, and if I had unlimited time and resources and get rid of it across the board. Once the articles are improved thusly, the shortcoming in the tool would no longer be relevant. Hijiri 88 (聖やや) 10:29, 3 April 2016 (UTC)
Wording at WP:Citation needed
Opinions are needed on the following matter: Wikipedia talk:Citation needed#Apologies, but object strongly, to content in article here based on decades of experience. A WP:Permalink for it is here. Flyer22 Reborn (talk) 23:18, 29 March 2016 (UTC)
Use of "Locations" in citing Kindle e-books?
Hi!
Background on why I'm asking this |
---|
I was recently TBANned from the topic in which I am most interested and on which I have access to the best print sources. I don't want to go around English Wikipedia adding a bunch of citations to Japanese-language print sources to which most other editors don't have access to articles on, say, Chinese literature and the Bible, but I also don't want to pay and wait for international shipping on print books in English from American and European publishers. So I've been using my Kindle a whole lot more. But I'm not sure about something. |
I read Wikipedia talk:Citing sources/Archive 31#How do you cite E-Books from Kindle?, but it might be out of date. User:Kmhkmh said use whatever description works to narrow down the content in question for that "Chapter 5, Section 2, Paragraph 8" ais fine, but now (perhaps not true back in 2011?) I can provide much more precise "Loc"s to give almost the exact line I'm citing, but this is only useful if one has the Kindle version, and I'm worried that providing a unique and original citation format that includes something like "Kindle edition. Locs 22405-22417." would be especially unattractive. It's also tedious trying to locate these exact Locs, as I don't know how to do it beyond zooming in to the max and scrolling around...
Thoughts?
Hijiri 88 (聖やや) 06:43, 3 April 2016 (UTC)
See also my recent hack job on Xiaopin (literary genre). Hijiri 88 (聖やや) 08:50, 3 April 2016 (UTC)
-
- I think I can answer why I don't believe this solution is feasible, as least for now.
- As I see it there are couple problems/questions I'd want sorted before anyone tries to create anything like this:
- Do kindle-locations work on other formats such as
.epub
or.mobi
as well or are they different? - Is there any way to convert these locations to something applicable to a .pdf version of the book?
- Does each version i.e. bought from Amazon, Barnes & Noble; borrowed from Overdrive, Proquest, Adobe Digital editions, Elib, etc. have different ISBNs? Are they identical with identical location info?
- Do similar locations work for content that is in audiobook format?
- Should we ever allow only a kindle location without page, chapter, section, paragraph?
My opinion is no, this would create a dependence on a single distributor and promote monopoly/require having to buy ebooks to verify content as opposed to just popping by your nearest library.
- Do kindle-locations work on other formats such as
- In addition to these questions there is a different way to handle locations — namely with hashes of the content, which can direct right to the content in question in a way that can be converted across formats. This is for example applied by the Hathi Research Center https://sharc.hathitrust.org.
P.S. Can't find any source for how to cite with hashes right now, will get back to you if I find anything
CFCF 💌 📧 10:32, 3 April 2016 (UTC)
- There's nothing wrong with simply citing "|loc= chapter X", though it's less than ideal. Commercial .epubs also tend to have Adobe pagination (though not all), but it has to be turned on (on my Kobo the hard-coded page numbers then show up in the margin). I don't know whether there's a Kindle equivalent. Electronic versions should have different ISBNs from print versions. I wouldn't cite line numbers or paragraphs, as it's not a standard many people would recognize. Curly Turkey 🍁 ¡gobble! 12:36, 3 April 2016 (UTC)
-
- Electronic versions should have different ISBNs from print versions. They should, but seemingly they don't... I learned in college to get what I need from the title page and bibliographical information just after that. In the article I linked above I was consulting the Kindle edition, but when I checked the third "page" it said "ISBN 0-231-10984-9 (alk. paper)"... Hijiri 88 (聖やや) 13:00, 3 April 2016 (UTC)
- And even if they have separate ISBNs to physical books, it's also relevant to know if they have separate ISBNs between different ebook versions. CFCF 💌 📧 13:08, 3 April 2016 (UTC)
- Well, all the more reason to stick to "|loc= chapter X" if you don't have page numbers. It's no worse than citing a long unpaginated web article. Curly Turkey 🍁 ¡gobble! 21:39, 3 April 2016 (UTC)
- Well, yeah, except that if someone else has a paper copy of the book and wants to check up on my cites, they can't Ctrl+F their paper copy like they can a long unpaginated web article. ;-) Hijiri 88 (聖やや) 03:59, 4 April 2016 (UTC)
- Well, if you think they'll count the paragraphs, go right ahead—there's no rule against it. I'm just letting you know that "|loc= chapter X" is just fine, and will even pass FAC. Curly Turkey 🍁 ¡gobble! 04:05, 4 April 2016 (UTC)
- Meh. I've seen people claim it was okay to just name the book they read it in, and you and I both called CurtisNaito out on his massive page-range citations... Hijiri 88 (聖やや) 04:10, 4 April 2016 (UTC)
- He had access to the actual pages, and the ranges were sometimes absurd. But if that was the only problem with his editing, it wouldn't have been a real problem, just irritating and lazy. Curly Turkey 🍁 ¡gobble! 06:30, 4 April 2016 (UTC)
- Meh. I've seen people claim it was okay to just name the book they read it in, and you and I both called CurtisNaito out on his massive page-range citations... Hijiri 88 (聖やや) 04:10, 4 April 2016 (UTC)
- Well, if you think they'll count the paragraphs, go right ahead—there's no rule against it. I'm just letting you know that "|loc= chapter X" is just fine, and will even pass FAC. Curly Turkey 🍁 ¡gobble! 04:05, 4 April 2016 (UTC)
- Well, yeah, except that if someone else has a paper copy of the book and wants to check up on my cites, they can't Ctrl+F their paper copy like they can a long unpaginated web article. ;-) Hijiri 88 (聖やや) 03:59, 4 April 2016 (UTC)
- Well, all the more reason to stick to "|loc= chapter X" if you don't have page numbers. It's no worse than citing a long unpaginated web article. Curly Turkey 🍁 ¡gobble! 21:39, 3 April 2016 (UTC)
- And even if they have separate ISBNs to physical books, it's also relevant to know if they have separate ISBNs between different ebook versions. CFCF 💌 📧 13:08, 3 April 2016 (UTC)
- Electronic versions should have different ISBNs from print versions. They should, but seemingly they don't... I learned in college to get what I need from the title page and bibliographical information just after that. In the article I linked above I was consulting the Kindle edition, but when I checked the third "page" it said "ISBN 0-231-10984-9 (alk. paper)"... Hijiri 88 (聖やや) 13:00, 3 April 2016 (UTC)
"Request admin assistance" in the "Dealing with unsourced material" section
Why does Dealing with unsourced material say "If an article is unreferenced, you can tag it with the {{unreferenced}} template, so long as it is not nonsensical or a biography of a living person, in which case request admin assistance"? What is it that an admin would do that we would not expect any editor to do? I can see that an admin's assistance might be needed in a dispute over biographical material, but if that's the intention it needs to be a little clearer. Mike Christie (talk - contribs - library) 10:31, 7 April 2016 (UTC)
- If the article is nonsensical, it is eligible for quick deletion, which can be done by an admin. If a particular version of an article has a passage about a living person which is seriously inappropriate, an admin can hide that revision so only other admins can see it. Jc3s5h (talk) 10:53, 7 April 2016 (UTC)
- Good points, but neither is clear from the current text. How about changing the text to suggest a CSD tag for the first point, and to make the second point clearer? Perhaps "If an article is unreferenced, you can tag it with the {{unreferenced}} template. If the article is nonsensical, it may make more sense to tag it for speedy deletion. If the unreferenced material relates to the biography of a living person, and is seriously inappropriate, it may need to be hidden from general view, in which case request admin assistance"? With appropriate links, of course. Mike Christie (talk - contribs - library) 11:06, 7 April 2016 (UTC)
- Unreferenced article: If it's nonsense tag it for G1 speedy deletion. If it's a BLP and gives no sources of any kind it qualifies for a BLPPROD tag. If, in addition, it amounts to an attack page, it qualifies for G10 speedy deletion (placing the tag will automatically blank the article). In any other case try to talk to the page creator, find references yourself, or, failing those, place the
{{unreferenced}}
template and consider sending it for AfD. I see no place for vague advice to "seek admin assistance". Can we agree to change the text in this way?: Noyster (talk), 12:24, 7 April 2016 (UTC)- Looks good to me; perhaps add mention of the need for hiding revisions in some cases, and how to request that? Mike Christie (talk - contribs - library) 12:38, 7 April 2016 (UTC)
- Unreferenced article: If it's nonsense tag it for G1 speedy deletion. If it's a BLP and gives no sources of any kind it qualifies for a BLPPROD tag. If, in addition, it amounts to an attack page, it qualifies for G10 speedy deletion (placing the tag will automatically blank the article). In any other case try to talk to the page creator, find references yourself, or, failing those, place the
- Good points, but neither is clear from the current text. How about changing the text to suggest a CSD tag for the first point, and to make the second point clearer? Perhaps "If an article is unreferenced, you can tag it with the {{unreferenced}} template. If the article is nonsensical, it may make more sense to tag it for speedy deletion. If the unreferenced material relates to the biography of a living person, and is seriously inappropriate, it may need to be hidden from general view, in which case request admin assistance"? With appropriate links, of course. Mike Christie (talk - contribs - library) 11:06, 7 April 2016 (UTC)
OK, here's a proposed rewording of the section, based on Noyster's comments above.
If an article has no references at all, then:
- If the entire article is nonsense, tag it for speedy deletion using criterion G1.
- If the article is a biography of a living person, it can be tagged with {{subst:prod blp} to propose deletion. If it's a biography of a living person and is an attack page, then it should be tagged for speedy deletion using criterion G10, which will blank the page.
- If the article doesn't fit into the above two categories, then consider finding references yourself, or commenting on the article talk page or the talk page of the article creator. You may also tag the article with the
{{unreferenced}}
template and consider nominating it for deletion.For individual unreferenced claims in an article:
- If an unreferenced claim is doubtful but not harmful, you may remove it from the article or, alternatively, use the {{citation needed}} template, which will add an inline tag. If you choose to use the tag, please go back and remove the claim if no source is produced within a reasonable time.
- If a claim is doubtful and harmful, remove it from the article. You may want to move it to the talk page and ask for a source, unless it is very harmful or absurd, in which case it should not be posted to the talk page either. Use common sense.
Comments? Mike Christie (talk - contribs - library) 18:44, 10 April 2016 (UTC)
Citing sources with spelling mistakes in their titles
The article Sabah Chinese Association is referenced using this source, the records of the Sabah state government. The title of the page in question is "Govermental Records", i.e. with a spelling mistake.
Parkywiki corrected the spelling of the reference, but I reverted on the basis that the title of the source is effectively a quotation, and we should leave it as it is. Parkywiki then reverted to the correct spelling on the basis that it's clearly an error (which I agree with) and that correcting it will avoid spell-check issues in future.
Is there an agreed way of dealing with this? Cheers, Number 57 23:32, 9 April 2016 (UTC)
- @Number 57: It depends how "significant" you consider the error to be. The relevant guideline is MOS:QUOTE. I would argue that this qualifies as "trivial spelling and typographic errors" and so should be corrected, but alternatively you could denote the error with {{sic}}. Nikkimaria (talk) 00:44, 10 April 2016 (UTC)