Wikipedia talk:Citing sources/Archive 50

Archive 45

←

Archive 48

Archive 49

Archive 50

Archive 51

Archive 52

→

Archive 55

Foreign lang template to throw after a foreign-language reference? (references without citation templates)

Back in ye olde days of Wikipedia, there used to be an unobtrusive template like lang|en lang|fr lang|ru that would just put a smaller font (Ru) or (Fr) after a footnote reference that's in a foreign language, and automatically added the article to some hidden category for such works. I know {{in lang}} exists now, but it's bulky and plain-text and doesn't appear to have any real benefit that isn't to be had by simply writing "In Russian" myself. Any help finding something smaller? I'm working on an article that is almost entirely non-English sources and I'd rather not put "In German", "In French" after each citation. HaltlosePersonalityDisorder (talk) 03:15, 7 October 2020 (UTC)

There's {{lang}} and {{lang-x}} that might help, but I'm not sure they are exactly what you're after. Martin of Sheffield (talk) 15:30, 7 October 2020 (UTC)

If you’re using the CS1/2 citation templates, there’s a |language= parameter you can use that takes the ISO 639 code; is that what you’re looking for? Umimmak (talk) 16:00, 7 October 2020 (UTC)

In the CS1/2 citation templates, there are also parameters like |script-title= which can be prepended with an ISO language code for non-Latin languages, as in |script-title=zh-Hant:瘋狂亞洲富豪. Furthermore, if there's a Wikipedia article you want to link to where the article in another language is the only one that exists or is more complete than the English Wikipedia article, I use {{ill}}; so for example for the Chinese-language Hong Kong newspaper Ta Kung Pao the Chinese article is more extensive in discussing nuances of its coverage and biases so in citations I fill in |publisher={{ill|Ta Kung Pao|zh|大公報|preserve=1}} which produces "Ta Kung Pao".

Finally, for parameters like |authorlink= you can fill in a link to a language on a different wiki, you just have to begin it with a colon: for example |authorlink=:id:Daniel Rudi Haryanto for Daniel Rudi Haryanto, an Indonesian documentary film maker who despite winning international awards currently only has an article on Wikipedia bahasa Indonesia and not English Wikipedia.

p.s. a benefit of using an appropriate language template or parameter that may not be visible in your own browser is that the HTML code is marked up to indicate the language being rendered, which may help other peoples' browsers or screen readers show it properly. --▸₷truthious Ⓑandersnatch◂ 17:16, 7 October 2020 (UTC)

Umm, no. Do you not see the at the top of the {{ill}} documentation? What your example really produces (as far as the cs1|2 templates are concerned) is this:

{{ill|Ta Kung Pao|zh|大公報|preserve=1}} → [[Ta Kung Pao]]<span class="noprint" style="font-size:85%; font-style: normal; ">&nbsp;&#91;[[:zh:大公報|zh]]&#93;</span>

All of that ends up in the citation's metadata which should only hold (for your example) the publisher's name; no html markup, no wiki markup, just the name.

{{ill}} may be the template that the OP was thinking about. That, in my opinion is a badly thought out design. We as editors may know what ru, or es, or kl, or sq mean, but it is doubtful that readers know. We are here for the readers and there is no limit on space; use {{in lang}} so that you can use the IETF language tags and so that readers know what you mean without having to decode the tag.

{{in lang|ru|es|kl|sq}} → (in Russian, Spanish, Greenlandic, and Albanian)

—Trappist the monk (talk) 17:41, 7 October 2020 (UTC)

"We as editors may know what ru, or es, or kl, or sq mean, but it is doubtful that readers know" – agreed, that's why there is {{Interlanguage link info}} to explain it. Martin of Sheffield (talk) 19:17, 7 October 2020 (UTC)

Citing a section of a web page?

Transmission of COVID-19 cites https://www.cdc.gov/coronavirus/2019-ncov/faq.html in multiple places. The problem is, the page is hard to navigate, with the collapsed sections. At least in Chrome, if you search on a page for a word in a collapsed section, it won't be found. So, is there some way, analogous to {{rp}}, to cite a specific section on a web page? -- RoySmith (talk) 02:02, 8 October 2020 (UTC)

@RoySmith:, if you use shortened footnotes ({{sfn}}), you can use |loc= instead of |p=. If you are using {{cite web}}, you can use |at= instead of |p=. And if you're hand-rolling the citation, you can use plain text. In any of those cases, I'd just put the closest header title above the section of interest. Naturally, if it's an anchor, then include the fragment in the link. Mathglot (talk) 11:35, 8 October 2020 (UTC)

Here's an example:

Can wild animals spread the virus?

Note that in this example, the term '§wild animals' in the short footnote is a direct link to the FAQ question. Hope this helps. Mathglot (talk) 11:48, 8 October 2020 (UTC)

Works cited

• "Coronavirus (COVID-19) frequently asked questions | CDC". CDC. 18 September 2020.

References

Citing documents in proprietary formats

Is it permissible to cite a document in a proprietary format, and what is the proper way to do so? The case that I am thinking of is in IBM BookManager format. Shmuel (Seymour J.) Metz Username:Chatul (talk) 09:14, 29 September 2020 (UTC)

Whether the format is proprietary is somewhat irrelevant, though it does decrease the likelihood that someone can read it (consider that MS Word file formats were proprietary for some time). What does matter is whether the document in question was published in a place where it can be accessed by company-outsiders. In a citation template, file formats can be specified in the badly named |format=, but only if you have a URL to go with it (it should be something like |url-file-format=). --Izno (talk) 12:10, 29 September 2020 (UTC)

Do you mean something like this? |url-file-format= isn't valid. Shmuel (Seymour J.) Metz Username:Chatul (talk) 20:17, 29 September 2020 (UTC)

User:Chatul, your citation seems fine if you just omit the url-file-format parameter. I take Izno's comment to mean that |format= won't work if there is no URL, and that Izno doesn't think "format" is a wise name for the parameter; Izno thinks it would have been wiser to name it "url-file-format". Jc3s5h (talk) 20:58, 29 September 2020 (UTC)

Indeed. --Izno (talk) 21:00, 29 September 2020 (UTC)

Totally acceptable, as long as someone independent of the creator(s) of the file format can read the document in question in one way or another. Glades12 (talk) 09:44, 14 October 2020 (UTC)

References

RfC: Whether YouTube is a reliable source

YouTube allows uploading anything, and some possibility on one supporting oneself is here. -- PythonSwarm ^{Talk | Contribs | Global} 23:33, 10 September 2020 (UTC)

See Wikipedia:Reliable_sources/Perennial sources#YouTube. —David Eppstein (talk) 00:07, 11 September 2020 (UTC)

Youtube in a sense isn't really publisher itself but a distribution system or media "format". Hence the question whether it is a reliable source or not doesn't make much sense and is a bit like asking, whether a book, video cassette pr website is a reliable source. Instead you need to look athe publisher/author of an individual video (being available on Youtube). Now if the video is a legal copy and the publisher/author meets the requirement for reliable sources, then youtuve video is usually a reliable source otherwise it is not. However it is also worth to note that even in cases, where the reliability criteria is met, text sources are normally preferred to video sources (assuming both are available).--Kmhkmh (talk) 21:07, 12 September 2020 (UTC)

I could create an account and upload virtually anything on YouTube right now, and it would honestly have a decent chance of surviving even if was intentionally shocking/misleading or a copyright violation. In other words, YT is only a reliable source if the (verified) uploader in question is. Glades12 (talk) 11:46, 18 October 2020 (UTC)

Citing an encyclical

What would be the correct and appropriate cite template for referencing a Catholic Church encyclical? For example, Fratelli tutti. The general structure is numbered paragraphs and numbered footnotes. They are eventually published as books, but most people don't access them that way. The primary thing I'd want is accurate citation of a paragraph number or footnote, because these things get long. Elizium23 (talk) 05:15, 18 October 2020 (UTC)

Because WP:SAYWHEREYOUGOTIT, you might use {{cite web}}:

{{cite web |url=http://www.vatican.va/content/francesco/en/encyclicals/documents/papa-francesco_20201003_enciclica-fratelli-tutti.html#DARK_CLOUDS_OVER_A_CLOSED_WORLD |title=Encyclical ''Fratelli tutti''|author=Francis |website=The Holy See |at=¶11}}

Francis. "Encyclical Fratelli tutti". The Holy See. ¶11.

Alas, you can't link directly to the paragraph but you can link to the nearest heading.

—Trappist the monk (talk) 12:12, 18 October 2020 (UTC)

You can always generate a citation with a template, and append whatever you want:
<ref>{{cite web |url=http://www.vatican.va/content/francesco/en/encyclicals/documents/papa-francesco_20201003_enciclica-fratelli-tutti.html#DARK_CLOUDS_OVER_A_CLOSED_WORLD |title=Encyclical ''Fratelli tutti''|author=Francis |website=The Holy See |at=¶11}} Search for "transsubstantiation".</ref>
Also, remember that a citation isn't necessarily a template; you could, for example, use the template to generate the citation, copy it from the list of citations, edit the copy and enter it without template (Entered as
<ref>Francis. "[http://www.vatican.va/content/francesco/en/encyclicals/documents/papa-francesco_20201003_enciclica-fratelli-tutti.html#DARK_CLOUDS_OVER_A_CLOSED_WORLD Encyclical Fratelli tutti]". ''The Holy See''. Typed in pointer to exact location here. ¶11.</ref>
Less convenient than a quick template. I expect other reasons not to do this will be appended.

Best wishes, Pol098 (talk) 14:04, 18 October 2020 (UTC)

Okay, thank you both, this is very helpful. Is "¶" the accepted emoji? The docs say to use abbreviated words such as "para." I assume that I can also cite a footnote with at=? Elizium23 (talk) 16:29, 18 October 2020 (UTC)

Not an emoji; pilcrow, very often used to mark or reference a paragraph. Yes, |at=fn. 8 or:

|at=[http://www.vatican.va/content/francesco/en/encyclicals/documents/papa-francesco_20201003_enciclica-fratelli-tutti.html#_ftn8 fn. 8]

The list of things that can be used in |at= is not a prescriptive list; it is just a list of common in-source locators.

—Trappist the monk (talk) 16:45, 18 October 2020 (UTC)

Comments appreciated regarding proposed citation-related maintained template

See Wikipedia_talk:WikiProject_Citation_cleanup#New_maintenance_template_idea:_Source_title_missing. --_{Piotr Konieczny aka Prokonsul Piotrus| reply here} 04:22, 21 October 2020 (UTC)

consistent citation style

reference info for Airbus A300

unnamed refs	46
named refs	31
self closed	138
cs1 refs	73
cs1 templates	60
cs2 refs	1
cs2 templates	1
rp templates	87
cleanup templates	2
webarchive templates	5
use xxx dates	dmy
cs1|2 dmy dates	34
cs1|2 dmy access dates	28
cs1|2 ymd access dates	3
cs1|2 dmy archive dates	13
cs1|2 last/first	12
cs1|2 author	8
List of cs1 templates cite book (2) Cite book (5) cite journal (2) cite magazine (4) Cite news (5) cite news (7) cite web (26) Cite web (9)
List of cs2 templates Citation (1)
explanations

Hello, we had a discussion on Talk:Airbus_A300#Citation_style_conversion? about citation style consistency, as I relied on WP:CITEVAR stating Generally considered helpful: imposing one style on an article with inconsistent citation styles (e.g., some of the citations in footnotes and others as parenthetical references): an improvement because it makes the citations easier to understand and edit). Other editors thought I went too far. What would be the correct interpretation of this quote? Thanks!--Marc Lacoste (talk) 17:22, 24 October 2020 (UTC)

When making citation styles consistent, it should be to the format that is used most predominately or from the standard style first established with the article, not to a style that only has less frequent use in the article. If you want to propose that minor style to become the predominate, that should require a consensus -based discussion. EG As with spelling differences, it is normal practice to defer to the style used by the first major contributor or adopted by the consensus of editors already working on the page, unless a change in consensus has been achieved. --Masem (t) 17:28, 24 October 2020 (UTC)

Thanks for your reply. There were 66 refs in the article, mostly inline, and 8 books refs with multiple calls. Those 8 books were cited in shortened footnotes, and I replaced then with inline citations. I was waiting for m:WMDE Technical Wishes/Book referencing, but this was delayed to 2021.--Marc Lacoste (talk) 06:31, 25 October 2020 (UTC)

Replacing references to specific pages in books by generic references to large page ranges in books, as you did, is a disimprovement, because it makes it more difficult for anyone who actually wants to use the reference to verify the information that was referenced. The purpose of references is not to make the article look pretty and referency and all-ducks-in-a-row, it is to make the information in the article verifiable, and your edits go against that purpose. The article reference format was not problematic before, and should not have been changed to a worse format merely to promote your ideas of consistency of format. I think you should revert those changes. Consider an analogy: would it be ok to require that all citations be {{cite web}}, and to remove all print sources, merely for the reason that using a single type of citation is more consistent than using a mix of sources of different types? If not that, then why should what you did be considered any better? If you are going to enforce a consistent format, it needs to be a format flexible enough to allow for multiple separate references to separate pages in book sources; the mix of long+short references in place before had that flexibility, and your new straitjacket does not. Meanwhile, you overlooked a genuine inconsistency of format that could have been fixed without such problems: the "AirDisasters" reference is in citation style 2 and the rest are in citation style 1. —David Eppstein (talk) 06:50, 25 October 2020 (UTC)

I did restore page numbers for the ref with the widest page range, but I'm going to do the same for the others, with 2-3 page ranges. Meanwhile, a bit of civility would be pleasant.--Marc Lacoste (talk) 21:03, 25 October 2020 (UTC)

Civility is about not making remarks about you as a person, which I have not. It is not and should not be about silencing criticism of your edits. —David Eppstein (talk) 21:54, 25 October 2020 (UTC)

"straitjacket". And "you overlooked a genuine inconsistency..." is a personal remark on my capability, not an answer to my question. Constructive criticism would have been saying shortly page numbers should be kept.--Marc Lacoste (talk) 06:35, 26 October 2020 (UTC)

Both are remarks about content. "Straitjacket" = the heavy restraints that article referencing would be subject to if we were unable to make citations to individual pages of books, as your push for consistency in this instance has done. "And you overlooked" is again a (completely accurate) description of your edits in this specific case. Constructive criticism requires pointing out problems with edits, not staying silent about them and meekly suggesting that maybe something else would also have been possible. —David Eppstein (talk) 06:42, 26 October 2020 (UTC)

Also note that according to :m:WMDE Technical Wishes/Book referencing "All previous methods for referencing are still possible without restriction" so that still wouldn't give carte blanche to change the reference style of an article.Nigel Ish (talk) 10:25, 25 October 2020 (UTC)

Possible but inconsistent, then.--Marc Lacoste (talk) 21:03, 25 October 2020 (UTC)

A technical point about references with page numbers: there were comments and criticisms here about reference editing which lost page numbers. I would remind that the template {{rp|pagenum}} can be used; it is placed after a reference to indicate a page number or range, so that a single detailed reference can be used to support points on different pages, It also has what I consider the advantage that it is clear to the user that different text is supported by the same reference, as the number shown in read mode is the same.

Best wishes; Pol098 (talk) 10:58, 26 October 2020 (UTC) P.S. references cited in this comment are spurious examples, no need to look them up!

My personal opinion is that {{rp}} is an abomination. I won't use it in articles I create. But it is of course a consistent citation style and should not be removed without discussion. —David Eppstein (talk) 16:31, 26 October 2020 (UTC)

"Work" versus "agency" versus "publisher"

One mistake that I made consistently during my early wiki career was to misuse the "publisher" and "agency" fields for e.g. The New York Times, when what I was looking for was actually the "work" field. "Work" just isn't a very good label. I notice that there's a question mark in the form that displays a clarifying message—perhaps that was added more recently, helping newer editors more than I was, or perhaps it just took me a while to notice because it's so subtle (in which case newer editors are probably going through the same thing). Is there anything we could do to help people learn about these fields more easily? {{u|Sdkb}} ^talk 19:15, 13 October 2020 (UTC)

Courtesy ping Maile66, who I notice asked about this at VPT three years ago. {{u|Sdkb}} ^talk 19:21, 13 October 2020 (UTC)

There are multiple alternative parameter names for various specific types of works, eg |website= or |newspaper=. Based on your VPT link it seems like the best way to help people learn about these in that specific case would be a change in RefToolbar? Nikkimaria (talk) 20:27, 13 October 2020 (UTC)

Nikkimaria, yeah, I'm thinking primarily about RefToolbar, since really that's the only way any newer editor (using source code) ought to add references. {{u|Sdkb}} ^talk 21:06, 13 October 2020 (UTC)

It is probably true that "work" could be confusing new editors who use structured citation helpers. A better label might be "source" since that is what "work" represents, and that is what is cited.

In a citation context, "source" is any published, discoverable, free-standing item that unambiguously verifies, or includes subitems that unambiguously verify, in real time, claims (in this case in wikitext), in whole or in part. Sources unrelated to the editor of such claims are preferred, though this is not necessarily a disqualification. Proven past reliability applies only to the related past citations, and no source should be considered a priori reliable (or unreliable) on any subject.

Good luck! 65.88.88.69 (talk) 21:12, 13 October 2020 (UTC)

"Source" is ambiguous, since a typical news or journal article has an article title and a publication title. That is probably why |source= is not a supported parameter in any of the Citation Style 1 "Cite xxx" templates. "Publication" or "Periodical" might be more helpful than work, although the latter might be too jargony for the average editor. For those of us who do not use RefToolbar, a screen shot or two, showing the placement of this word "work", would be helpful. – Jonesey95 (talk) 21:46, 13 October 2020 (UTC)

You are confusing the source (the journal) with an in-source location (the article). Until a few years ago journal articles were practically non-discoverable by the general public. Many if not all reference databases would index journals, not articles. That was the singular, free-standing, searchable item. Granted that with digital indexing it is easier to zero in on a location rather than to the contents of the including source. But there is no such structured helper in Wikipedia. All of them, whether they are templates or addons are focused on sources (the enclosing works), not in-source locations. So there is no "cite article". The reason a parameter "source" does not exist in one part (or anywhere) of Wikipedia's citation environment has more to do with entrenchment of faulty design and of the attendant (erroneous) terminology than anything else. 135.84.167.61 (talk) 23:04, 13 October 2020 (UTC)

1. I am not confused (this time). Journal articles have been discoverable at public libraries for as long as I can remember. I learned how to use an index of periodical articles in the early 1980s as a young student, before there were general-purpose computers in school libraries; here's one from 1922.

2. {{Cite journal}} is specifically for citing individual journal articles, and will give an error message if the title of the article is not provided. {{Cite web}} and {{Cite news}} are the same.

3. The phrase "in-source location" is used by the Citation Style 1 documentation to refer to a page number or other specific location within the cited item. It would help this discussion if you referred to the documentation before making claims about how templates work and what they are for. – Jonesey95 (talk) 00:22, 14 October 2020 (UTC)

I am not discussing any specific part of Wikipedia's citation environment. Neither am I discussing documentation that in certain cases perpetuates erroneous approaches to citing materials. The documentation of the templated application of citation style 1 falls in that category, but that is often because it documents flawed design, or in this case outright novel and wrong information. The point is that citations are about published sources. Journal articles are not published (when they are, they are no longer "articles"). Journals are published, and they are classified accordingly, in libraries or elsewhere. Until very recently, you could not just search for a journal article. The first question always was, what journal is it published in? Because as stated earlier journals and their TOCs was what was indexed. There are/were databases of journals, not articles. To find the relevant part (article) one would first find the (work) journal, again because that was the published entity, and therefore that was what could be easily found. Citations are there to provide an easy, quick path to verification, they are not research material. Sure, you could find an article not knowing the journal. If you had a LOT of time in your hands, for something so pedestrian. 64.18.9.208 (talk) 04:04, 14 October 2020 (UTC)

Jonesey95, you can look at GorillaWarfare's video at Wikipedia:RefToolbar/2.0 (or just enable RefToolbar and test it for yourself). When you hover over the question mark near "work", you get the tooltip Name of journal, magazine, newspaper, periodical or website, encoded here. {{u|Sdkb}} ^talk 23:35, 13 October 2020 (UTC)

Thanks. I looked at the video, and at 0:53, the "Cite journal" pop-up box shows "journal" as an option, but I don't see "work". Can you be more specific about how to reproduce the problem that you are experiencing? – Jonesey95 (talk) 00:22, 14 October 2020 (UTC)

Jonesey95, make sure you have RefToolbar enabled (as is default). Go edit a page in source code, go to the cite tab, and under the templates menu click "cite news". The box that pops up will have a number of fields, including "Work", "Agency", and "Publisher". The discussion here is about the potential confusion among those fields. {{u|Sdkb}} ^talk 00:49, 14 October 2020 (UTC)

The definition of "work" present in the toolbar is correct, even if it is not complete. The confusion between the fields mentioned above is more a social issue. Understandable, correct documentation should be provided, that clarifies each field's use; hopefully editors will apply it. 68.173.79.202 (talk) 12:35, 14 October 2020 (UTC)

An aside on "source" as parameter label for the work: if I remember correctly (this goes back to the early days of Wikipedia), "source" was rejected because editors thought it would be confused as referring solely to "source code". The majority of early editors were technically inclined. If they were dealing daily with accounting instead, "journal" would probably never be used as a citation template name or parameter. 68.173.79.202 (talk) 12:48, 14 October 2020 (UTC)

An idea: add tooltips for "agency" and "publisher"

How about this: let's add an explanation next to the "agency" and "publisher" fields the same as the one next to the "work" field. Going off of the documentation at {{cite news}}, for publisher I'd suggest parent company that published the work being cited and for agency I'd suggest news agency (wire service) that provided the content published elsewhere. Thoughts? {{u|Sdkb}} ^talk 17:54, 14 October 2020 (UTC)

As in a tooltip in RefToolbar itself? I'd suggest that be proposed at the tool's talk page rather than here. Or did you mean something else? Nikkimaria (talk) 20:28, 14 October 2020 (UTC)

Sounds like a good idea. Do we have any stats on how often the toolbar is used.....do new editors not see it? Why is it not used by the majority? Never seen "agency" paramater used.--Moxy 🍁 23:59, 14 October 2020 (UTC)

Not the most commonly used cs1|2 parameter, but it is used in about 104k articles

I expect that most 'new' editors don't see WP:RefToolbar because that abomination that is ve is the default so must be disabled or editors must intentionally switch – both are things that new editors likely know nothing about.

I hold the opinion that WP:TemplateData's attempt to combine control and documentation into one crippled tool is a failure. But, I think that if the 'documentation' that TemplateData holds is considered in the same light as the suggested RefToolbar tool tip, then ve gets it more-or-less right for |agency=: 'The news agency (wire service) that provided the content; examples: Associated Press, Reuters, Agence France-Presse'.

—Trappist the monk (talk) 00:29, 15 October 2020 (UTC)

Nikkimaria, I put an invitation there. {{u|Sdkb}} ^talk 02:18, 15 October 2020 (UTC)

• Update: I requested implementation. {{u|Sdkb}} ^talk 19:45, 27 October 2020 (UTC)

Citing an image

I'm a bit rusty, and can't easily find the answer to my problem. I want to cite an image as a reference. Note that the information i am referencing is only contained in the image, not in the text of the article that the image is displayed in. Do i still use 'cite web', and use the image's title text as the title attribute? Thanks in advance.Julianhall (talk) 19:51, 21 October 2020 (UTC)

Can you provide a link to the article containing the image? That would make it easier for others to understand the issue. Glades12 (talk) 18:13, 26 October 2020 (UTC), updated 19:12, 27 October 2020 (UTC)

The article is here:[1]. The Wikipedia page in question is List of Leeds Rhinos players, and the specific information contained only in the image is the heritage number which can be seen on the player's shirt. Thanks. Julianhall (talk) 19:51, 27 October 2020 (UTC)

I would make a normal citation to the article that the image is displayed in, with an annotation (either in the |at= parameter or immediately following the citation if |at= would interfere with the page numbers of the citation) describing which image from that article is being cited. —David Eppstein (talk) 19:19, 27 October 2020 (UTC)

Thanks for the reply. I'll investigate that. Julianhall (talk) 19:51, 27 October 2020 (UTC)

citing personal communication confirming a cited fact?

I worked on a page for a person who plays the organ for sports teams. After the recent Dodgers World Series win, someone pointed out on Twitter that he's played for three championship teams in three separate sports. I thought this was notable and while I have an authoritative citation that confirms he was the staff organist during the time those games were played, I reached out to him to get confirmation that he was actually playing AT those games, a fact he confirmed. I think this is OK under WP:SELFSOURCE. Is that correct? If so, how do I cite that personal communication? Thank you. 18:18, 1 November 2020 (UTC)

You cannot use a personal communication to source some fact because it is not published in the sense required by Wikipedia. See WP:PUBLISH. --Izno (talk) 20:43, 1 November 2020 (UTC)

• @Jessamyn: Izno is correct regarding our policies/guidelines, so if you were trying to get Ruehle's page recognized as a good article or featured article, a non-published confirmation would not be sufficient. However, given where the page is right now, no one is likely to complain if you just include the fact as original research. You can add the {{Better source needed}} tag to indicate that your existing authoritative source doesn't confirm that he was actually playing at those games, and perhaps a source will be published/found providing confirmation sometime in the future. Cheers, {{u|Sdkb}} ^talk 22:28, 1 November 2020 (UTC)

Thank you both. Jessamyn (talk) 22:52, 1 November 2020 (UTC)

Expand WP:DEADREF

I have been doing RC patrol recently and I've noticed that tons of people seem to, for some reason, think that the official protocol is to simply remove material from articles (sometimes entire sections) because "the link is dead". While WP:DEADREF does mention this as bullet point 6 (the final point in the list), I think it might be better for that section to mention up-front that removing large referenced sections from articles because "the URL didn't load" is not a proper application of policy. I might just go ahead and do it if nobody objects. jp×g 09:45, 5 October 2020 (UTC)

What specifically would you propose adding? Nikkimaria (talk) 12:15, 5 October 2020 (UTC)

I ended up doing it with this diff, which wasn't as major a change as I had expected it to be. jp×g 10:00, 27 October 2020 (UTC)

I was just reading point 6 and wondered - how doesn't it contradict Assume Good Faith? I could be trusted to add a newspaper or magazine print citation but for some reason online is expected to work forever. Why was point 6 added in the first place? I improved an article to GA in 2007 and it was a Featured Article Candidate in 2008. The links were checked by several people in the FAC process and they worked then. Now many links are dead and I bet many are hopelessly lost. I am considering another Featured Article attempt. Royalbroil 04:08, 18 October 2020 (UTC)

Royalbroil, if the scenario you describe, the process probably stops at item 4, "Remove convenience links: If the material was published on paper (e.g., academic journal, newspaper article, magazine, book), then the dead URL is not necessary. Simply remove the dead URL, leaving the remainder of the reference intact." In other words, if you cited that newspaper article, and the URL dies, but some library probably has the original print copy on microfiche, then remove the URL (only) and keep the content. The citation about "Big News"] in The Daily Paper becomes a citation to "Big News" in The Daily Paper, sans URL, and everything's fine.

If the content isn't available anywhere (e.g., an online-only source), then we have a problem with Wikipedia:Published#Accessible. One of the main points of citing content is so that a dedicated person could check to make sure that you weren't making it up. If your source completely disappears from the face off the Earth, and no other source contains the same claims, then the content stops being verifiable; our dedicated person is literally no longer "able to verify" that any source contained that claim. In that case, the content no longer complies with the core policies and needs to be removed. WhatamIdoing (talk) 05:44, 19 November 2020 (UTC)

Videography/Music videos section

When creating a list of official music videos of a musical artist, can we link direct YouTube Music videos link as the references, or must it be from a third party source? Thank you.--TerryAlex (talk) 21:03, 11 November 2020 (UTC)

@TerryAlex, if you are still looking for an answer, then I encourage you to copy your question to Wikipedia:Reliable sources/Noticeboard, which exists for the purpose of deciding whether a source is strong enough for specific claims. WhatamIdoing (talk) 05:54, 19 November 2020 (UTC)

RfC: Citation tools

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

---

Should citation tools be allowed?

Background:

• Wikipedia:Citing sources#Citation templates and tools
• Help:Citation tools
• Category:Wikipedia citation/link tools

Currently, tools such as AutoEd, AutoWikiBrowser and Reference Organizer may be used to make changes to citation styles on individual pages.

(This is about using such tools on single pages and carefully checking the results before publishing. Using such tools on many pages in an automated fashion is far more controversial and would require a separate discussion.)

Lithopsian (talk · contribs · deleted contribs · logs · filter log · block user · block log) has decided that edits made with such tools should not be allowed[2] and has started reverting them.[3][4]

Proposed: citation tools may be used on individual Wikipedia articles. The person using the tool should preview the page and carefully check the results before publishing. Such edits should not be reverted unless they introduce errors, and in particular should not be reverted simply because someone does not like the resulting citation style.

--Guy Macon (talk) 22:01, 11 November 2020 (UTC)

Survey

• Support as proponent. --Guy Macon (talk) 22:01, 11 November 2020 (UTC)
• Support the general use of software to format references, obviously. However, the actual dispute that triggered this appears to be mass conversion of reference names from an unquoted style <ref name=refname> to a quoted style with modified names <ref name="Ref Name">. This violates WP:CITEVAR (which applies to source formatting of references, not just their appearance in the rendered articles) and should not have been done without discussion and a good reason. The unquoted style works perfectly well (because ref tags are not html) and is a perfectly consistent style for formatting references. If your citation tools cannot handle unquoted refnames, you should fix that limitation before applying them to articles that use that style. —David Eppstein (talk) 22:11, 11 November 2020 (UTC)

  @David Eppstein: WT:Citing sources/Archive 39#RFC: Is a change in citation markup method a change in citation style? should be reviewed. I might draw a different conclusion based on the RFC summary. (Consider also the followon RFC about vertical versus horizontal.) --Izno (talk) 23:34, 11 November 2020 (UTC)

  You might want to change that link to WT:Citing sources/Archive 40#RFC: Is a change in citation markup method a change in citation style?. --Guy Macon (talk) 00:24, 12 November 2020 (UTC)

  The vertical and horizontal RfC was quite clear: "There is a clear consensus that the usage of vertical and horizontal templates does not fall within the purview of WP:CITEVAR." and "...the established decision made last year which stated that non-systemic changes of coding that occur while updating the content of a citation and/or adding citations do not require consensus." --Guy Macon (talk) 01:49, 12 November 2020 (UTC)

  As for the specific claim about quoted versus unquoted, I assert that quotations are safer for every end user, whether a person or a bot. Adding them falls into the "this is a good edit regardless of any question of CITEVAR". --Izno (talk) 23:39, 11 November 2020 (UTC)

  Quotations are awkward unnecessary cruft that makes the source harder to read without adding any value. Adding them falls into the "you are making things worse for no good reason" category of edits and should be reverted on sight. Also this sort of disagreement is exactly why we have CITEVAR. —David Eppstein (talk) 00:06, 12 November 2020 (UTC)

  I have seen examples where an new editor saw a <ref name=NYT>, changed it to <ref name=NYT 2017> and added a <ref name=NYT 2018> cite somewhere else. If the original had been <ref name="NYT"> they would have avoided the error. --Guy Macon (talk) 00:14, 12 November 2020 (UTC)

  I have seen too many cases of curly quotation marks causing reference breaks to agree with "making things worse for no good reason". No, quotation marks are good categorically and are always an improvement. Either way, I'm sure we can have an RFC on that question if you feel that's valuable. That said, you have not responded to the more substantive comment, which indicates that either you did not see or that you accept that you are wrong in the case of the changes made in context. --Izno (talk) 00:15, 12 November 2020 (UTC)

• Of course they should be used and allowed. literally thousands and thousands of edits are done on a daily basis. So status quo"..as I am simply puzzled about this RFC to begin with.--Moxy 🍁 22:37, 11 November 2020 (UTC)

  @Moxy: You didn't close you italics tag. I'm not sure what you meant to be italicized, so don't want to touch it and change your meaning. Feel free to delete this comment once you've fixed it. — Wug·a·po·des​ 00:09, 12 November 2020 (UTC)

Thanks for the notice.--Moxy 🍁 00:25, 12 November 2020 (UTC)

• The alternative would be edit warring to restore a perfectly acceptable edit. Better to post an RfC so that it is clear that the community does not support reverting an edit that changes <ref> to <ref name="example"> or even one that changes <ref name=example> to <ref name="example">. --Guy Macon (talk) 00:14, 12 November 2020 (UTC)

Why would people editwar over something the end reader won't even see let alone notice. Should be chalked up to disruptive editing and block.--Moxy 🍁 00:25, 12 November 2020 (UTC)

• Comment. The quoted form is much less easy to type for those of us working with carpal tunnel syndrome. I never use quotation marks in refnames for this reason. Espresso Addict (talk) 05:42, 12 November 2020 (UTC)
• Close - Do we really need an RfC, posted to CENT no less (by a third party FWIW), asking a question nobody disagrees with, in order to indirectly win a personal dispute? If someone is repeatedly reverting productive edits, take them to a noticeboard. If a tool makes edits that aren't in line with guidelines, bring that up in particular. If there's disagreement about whether certain tool-assisted edits are ok to edit war over, like quotes around ref names, ask that question. — Rhododendrites ^talk \\ 14:57, 12 November 2020 (UTC)

Threaded discussion

This RFC appears to be absurd. There is a long-standing precedent that these tools are useful, and they can do other things besides fix citations. If someone is violating CITEVAR or another guideline, or a policy, that person should be asked to stop doing so. When someone uses a tool to violate a guideline or policy, it does not necessarily mean that the tool should be banned. That's like banning hammers and screwdrivers because someone built a house that did not comply with local building codes. – Jonesey95 (talk) 22:53, 11 November 2020 (UTC)

• I agree that this RFC is absurd. I don't agree with the premise that edit violates CITEVAR, however, in any meaningful sense whatsoever, especially given the RFC I linked above. --Izno (talk) 23:38, 11 November 2020 (UTC)

WP:CITEVAR concerns attempts to change an article's established citation style:

"When an article is already consistent, avoid: switching between major citation styles, e.g., parenthetical and <ref> tags, or replacing the preferred style of one academic discipline with another's".

It does not say that if a citation starts with <ref> you are not allowed to change it to <ref name="example">. --Guy Macon (talk) 23:57, 11 November 2020 (UTC)

• I almost early closed this discussion, as I do not think the question as asked is necessary and several folks have pointed that out. Of course we let folks use citation tools. That's not going to change. Wikipedia would be unworkable without citation tools. That is so commonsense that we don't need to enshrine that in policy, that is WP:CREEP. However, there does seem to be some actually valid underlying questions here, they just need to be asked right. I would encourage @Guy Macon: to rephrase this RfC or introduce some new sections that cover what appear to be the core issues, such as <ref> to <ref name=> conversion, horizontal v. vertical refs, and CITEVAR if need be. But in terms of allowing the use of tools, broadly? That is an absurd question and should be withdrawn. CaptainEek ^{Edits Ho Cap'n!}⚓ 04:31, 12 November 2020 (UTC)

  Lithopsian (who was pinged but so far has not contributed to this discussion) appears to oppose all uses of the Reference Organizer tool and has reverted me twice. If this is closed without an answer and he keeps reverting, what do you suggest I should do? IMO if using the tool is allowed, Lithopsian should be warned to stop reverting uses of the tool, and if the tool is not allowed, the page that has the tool should be replaced with a message saying to stop using it. --Guy Macon (talk) 05:40, 12 November 2020 (UTC)

  Guy Macon, Well it appears that Lithopsian was being disruptive in the first place, and if they keep it up they should be warned, and if they keep it up, then blocked. Now, if there is a definitive policy issue at stake, it ought be discussed. But the use of citation tools is not an issue. Lithopsian doesn't like citation tools? Well tough bananas. CaptainEek ^{Edits Ho Cap'n!}⚓ 05:46, 12 November 2020 (UTC)

  Thanks! In that case, I think this can be closed. --Guy Macon (talk) 05:53, 12 November 2020 (UTC)

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

Call for re-opening of RFC

@Sdkb, CaptainEek, Moxy, David Eppstein, Izno, Espresso Addict, Rhododendrites, and Jonesey95:

It is clear from the discussion below that at least one editor believes that this RfC did not settle the question of whether the Reference Organizer tool is allowed, that what the tool does is "bad edits", and that editors who use the tool may be reverted for making "bad edits". I want to give the RfC process one last chance to settle the question before going to ANI over these reverts.

This of course does not shortcut the ongoing discussion below regarding quotations marks and reference names. Obviously if a consensus emerges that the Reference Organizer tool should not add quotation marks the author of that tool will be asked to modify the tool to stop doing that. This is specifically about the sequence of events "Editor A uses the Reference Organizer tool, Editor B decides that this is a "bad edit", Editor B reverts editor A." --Guy Macon (talk) 03:34, 21 November 2020 (UTC)

Guy, are you arguing that editors who make automated or semi-automated edits ought not to be reverted? That they can only be reverted if they edit manually? SarahSV ^(talk) 04:02, 21 November 2020 (UTC)

I am having a hard time believing that you aren't purposely misunderstanding me, but per WP:AGF I am going to assume that I have been unclear and try to explain it again. Let me turn the question around. Let's assume for the sake of argument that I made a new rule: all edits by anyone with "SV" in their signature are bad edits and should be reverted. Assume that when you objected I asked "are you saying that edits with SV in the sig can't be reverted? That they can only be reverted if the SV is removed?" Clearly the issue in my hypothetical case is not some imaginary rule against ever reverting you, but rather the issue would be "have I established that, in fact, all posts with SV in the sig are bad posts?" Or how about if I decided to revert every use of template:unsigned as a "bad edit"?" Again the issue wouldn't be some imaginary claim that edits using template unsigned cannot be reverted, but rather whether I would be allowed to revert them simply because I decided, without seeking consensus, that they are "bad edits". Likewise, the issue here is that neither you or anyone else has actually established that the output of Reference Organizer is bad. One aspect of it (adding quotation marks) is being discussed, but we have not reached a consensus on that issue. --Guy Macon (talk) 06:27, 21 November 2020 (UTC)

Your story seems to be: Someone hates this script, someone reverts it on sight, that someone is wrong and bad and nobody should be allowed to do that.

My story sounds like: I like scripts, but this script screwed up once (exactly one time, AFAIK), so I should be able to revert that one-time problem, exactly the same as I would revert someone who manually screwed up an edit one time. And, as proof of the fact that this actually was an objective screw-up on the part of the script, I remind you that you've filed a bug report about it. It's not just a matter of personal preferences.

So I see a one-time problem, and I think that one-time problem should be handled exactly like any other one-off accident, but you're filling this page with proposals that nobody should ever revert anything done to citations by any script, even if it's objectively bad and the result of a bug in the script, because someone might be biased by the method. This doesn't sound like a good approach to this problem.

I think that a good approach is to be tool-agnostic, and to keep good (and neutral) edits no matter how those edits were ostensibly made, and to revert bad edits, no matter how those edits were ostensibly made. WhatamIdoing (talk) 01:26, 22 November 2020 (UTC)

The above RfC shouldn't be re-opened, as its a badly formed question. But I have no opposition to opening a new RfC that is well formulated and specific to the issue at hand. Of course reference tools are allowed. But you should tackle: how should this specific tool be used/should it be used. Don't make it a referendum on all reference tools because that's silly. CaptainEek ^{Edits Ho Cap'n!}⚓ 04:22, 21 November 2020 (UTC)

Fair enough. How about "should single edits using the Reference Organizer tool be reverted in the sole basis that the reverting editor has decided that every edit using the Reference organizer tool is by definition a 'bad edit'?" Or do you have some other suggested wording that will tell me whether these reverts are allowed? --Guy Macon (talk) 05:57, 21 November 2020 (UTC)

Ok, as a general rule, I don't see that edits to references using a tool are any different from hand edits, and should be retained or reverted depending on the quality of the edit. I don't know what exactly Reference Organizer does, but if the default is to change well-formed reference tags of the form <ref name=Guardian2020> or <ref name=Guardian_2020> to <ref name="Guardian 2020"> in an article that largely uses the former style, then I feel the page creator or any regular editor to the page is entitled to revert, if they desire. As I have said, I have a strong personal preference for the former, partly owing to ease of typing for someone with long-term RSI problems, partly because I find them easier to spot in the editing pane. However, I'd be equally chary of a tool that changed them the other way; it seems to fall within the range of referencing preferences, where the creator/regular editor is entitled to the choice of what makes editing easier and drive-by editors should respect that choice. Espresso Addict (talk) 04:24, 21 November 2020 (UTC)

I believe that you have have inadvertently made a minor error. It is not possible using any option for Reference Organizer to change <ref name=Guardian2020> or <ref name=Guardian_2020> to <ref name="Guardian 2020">. Reference organizer changes <ref name=Guardian2020> to <ref name="Guardian2020"> and <ref name=Guardian_2020> to <nowiki><ref name="Guardian_2020">. It is a small difference, but we should discuss what the tool actually does. Also, I am having trouble understanding in what way someone else changing <ref name=Guardian2020> to <ref name="Guardian2020"> causes a person with RSI to type even a single extra character. All you have to do is leave it alone. Any new citations you add can be formatted any way you prefer. Even in the rare case where you add another link to the citation somewhere else, you wouldn't manually type out each character. You would cut and past the ref and add a space and a forward slash. --Guy Macon (talk) 06:51, 21 November 2020 (UTC)

As I wrote, I have not investigated precisely what the tool does, as I have no intention of using it. It's often easier with RSI to just type everything long hand; copy and paste can cause injury. I prefer to use the same style throughout the article, so I would feel encouraged to put in quotation marks in future, and " is one of the most difficult characters to type. Thus said, as I'm having an RSI flare up now, I won't respond further. Cheers, Espresso Addict (talk) 07:04, 21 November 2020 (UTC)

Espresso Addict, one of the things that this particular script does is to turn the manually/deliberately chosen <ref name="Guardian 2020"> into plain, unnamed <ref>, if the current version of the article doesn't cite that ref at least twice. This is likely a violation of CITEVAR, and even if it's not, it's still a bad idea. WhatamIdoing (talk) 01:30, 22 November 2020 (UTC)

WhatamIdoing Ah, I see. I'd agree that's a very poor idea; it's often useful to set up refnames, even if not immediately needed, to ease reuse of the reference and reduce the widespread problem of inexperienced editors repeating references often with variations. Espresso Addict (talk) 01:36, 22 November 2020 (UTC)

Or to stop the visual editor from turning them into <ref name=":0"> if someone re-uses it, because it's still not possible to manually name refs from inside the visual editor. (Dealing with that is probably a job for the m:Community Wishlist 2021.) WhatamIdoing (talk) 02:01, 22 November 2020 (UTC)

This 2017 ArbCom decision may be relevant: "For the purposes of dispute resolution, whether an editor is engaging in 'high-speed editing' (that is, the number of edits per minute) is irrelevant. Where editors have made a number of similar edits in a short time space and other editors have raised concerns about those edits, the editor is to stop making the edits and engage in discussion." SarahSV ^(talk) 04:56, 21 November 2020 (UTC)

All of the reverted edits we are discussing were done to single pages. The RfC above specified that the question was not about automated or repetitive edits. One use of the tool on Monday and another use a week later is hardly "a number of similar edits in a short time space". --Guy Macon (talk) 06:05, 21 November 2020 (UTC)

Continued discussion

(Please don't re-open this discussionn for me.) I'm about to fall asleep over my keyboard, and if I weren't so sleepy, then work-me would feel obliged to tell you that <ref name=refname> is technically supposed to have quotation marks around it. It seems that if you leave them out, then the servers secretly stick them in every time it parses the page, with the result that it would be a tiny better for performance if quotation marks were always included. Please Wikipedia:Don't worry about performance, and note that I'm not asking you to add quotation marks, but it would probably be best not to deliberately remove quotation marks from ref names. (Also, please remember to be nice to editors who haven't read this, because most editors don't know that there actually is a preferred/correct way from the servers' POV.) WhatamIdoing (talk) 06:04, 19 November 2020 (UTC)

WhatamIdoing, please explain what you mean by the servers secretly adding quotes around ref names. A reference like <ref name=UCSDSC/> gets served up to the reader in HTML as <sup id="cite_ref-UCSDSC_5-1" class="reference"><a href="#cite_note-UCSDSC-5">[5]</a></sup> whether there are quotes around the ref name or not. StarryGrandma (talk) 06:43, 19 November 2020 (UTC)

The servers secretly convert lots of wiki-markup that is not actually html into html. Ref tags are one of those pieces of not-html that they convert in this way. They more closely resemble html than some other kinds of converted wiki-markup, but they are not html. So trying to claim that they should follow rules for how html should be formatted, rather than the rules that the server uses to parse wiki-markup, is nonsensical. In particular, the server parses them just fine without quotes, so quotes are patently not required for them. —David Eppstein (talk) 06:57, 19 November 2020 (UTC)

@StarryGrandma, that is exactly what I mean: If you don't add the quotation marks, then the parser adds them for you. I've been told that this automatic addition, over the course of zillions of edits, has a small but non-zero negative effect on server performance. This effect is very small; you should not worry about it in everyday editing. But if you were asking that the nominally correct approach is, then inclusion is correct and removal is (very very very very very slightly) harmful. One would, therefore, encourage people to include quotation marks (if and when it's easy to do so) and deplore efforts to remove them (e.g., if someone wrote a bot to remove them on the grounds that it's 'patently not required', even though it is patently not required, since the parser copes with what it thinks is your mistake). Because the effect is tiny, it would not be proportional to yell at people for not typing quotation marks, create user-warning templates about this, etc. A proportional response is merely to keep in mind that it's slightly better to have them than to not have them. WhatamIdoing (talk) 20:01, 19 November 2020 (UTC)

WhatamIdoing, that is a completely different set of quote marks, added by the parser when it generates the HTML from the wikicode. If the reference had been <ref name="UCS DSC"/>, the parser would have removed the quotes and converted the space into an underline giving <sup id="cite_ref-UCS_DSC_5-1" class="reference"><a href="#cite_note-UCS_DSC-5">[5]</a></sup> as it generated the HTML with the HTML-reqired quotes. In fact, putting quotes when not necessary makes more work for the parser, not less. The quotes are needed only if the name has spaces or certain other characters. They are not required by the parser. StarryGrandma (talk) 20:48, 19 November 2020 (UTC)

@StarryGrandma, the devs who work on this part of MediaWiki tell me that it's better and more efficient for the software to include this. Do you have any experience with this code, or are you speculating hypothetically? WhatamIdoing (talk) 21:29, 19 November 2020 (UTC)

I want to know where you're getting these bizarre ideas about code performance. If the parser can handle unquoted strings or quoted strings, and converts both of them both into some different syntax when it's done (as it does), it's not going to take any measurably greater effort for it to handle unquoted ones. Or if it does, some programmer has done something very stupid. Do you worry that, when you're commenting in discussion threads like this one, you're indenting with too many colon characters and that each colon causes more work for the parser so you'd better outdent before it gets overworked? That's the level of thing you're worrying about with quotes in refs. —David Eppstein (talk) 06:39, 20 November 2020 (UTC)

David, I "get these bizarre ideas" by talking to the devs who work on that code. I have been working with the Editing team for the last seven years. The Editing team is the WMF team with primary responsibility for mw:Extension:Cite (which is the extension that we're talking about), and they work very closely with the Parsing team. I have every reason to believe that they know what they're talking about. It happens that "performance" was their answer to my complaints to them about VisualEditor always adding quotation marks to ref names. So I'm telling you: The devs say that having the quotation marks present in the wikitext is very slightly better for performance compared to having them absent.

As for whether someone "has done something very stupid", I understand that what works in theory and on small-scale demo systems is not always quite what's best when you are running one of the largest websites in the world, which leads to things like Google's main page being "unnecessarily" hand-coded and some "normal" MediaWiki features and extensions not being permitted at the English Wikipedia. If you think you can make this particular bit of code be performance-neutral (or even to prefer unquoted ref names), then I recommend that you start with mw:How to become a MediaWiki hacker. It's an open source project, and patches are welcome.

(In these discussions, I don't worry about the number of colons, because I mostly use the Reply tool, which handles that automatically. But for manual edits, (a) creating WP:LISTGAP errors is far more important, and (b) I don't personally happen to know whether it has an effect on performance.) Whatamidoing (WMF) (talk) 20:33, 20 November 2020 (UTC)

WhatamIdoing, my experience has just been with HTML and programming with strings. Upon delving into PHP I think I can see what is going on. Wikitext is not HTML, but the parser makes sure there are quotes around groupname and refname when it parses a reference so it can use its regular HTML parsing. That means a new section of code didn't have to be developed to handle it. The tradeoff is a small amount of processor time rather than development and maintenance time, which in my experience with programming projects is a good thing. This maintains compatibility with usual Wikitext syntax (as in no quotes needed around fields in templates). So <ref group=note> has the same requirements as the corresponding {{reflist|group=note}}. The quotes all vanish of course in the parsing process as the new string is created. StarryGrandma (talk) 19:21, 20 November 2020 (UTC)

I've never considered whether it's the best tradeoff (although my guess is that you're right); I can only tell you that it's the situation that we're living with, and that there are no current plans to change it. Whatamidoing (WMF) (talk) 20:37, 20 November 2020 (UTC)

Just to be clear, I strongly disagree that it's okay to go around changing ref name=x to ref name="x". It's very fiddly to have to keep adding quotation marks when you're writing an article. They're not needed if there are no spaces. SarahSV ^(talk) 05:41, 20 November 2020 (UTC)

Do you also strongly disagree that it's okay to go around changing ref name="x" to ref name=x? You say that they are not needed, which is true, but they do reduce errors. Newbies often do things like changing ref name=McGargle to ref name=McGargle 1925 when adding another citation with ref name=McGargle 1936. I have also seen newbies do things like changing ref name=McGargle to ref name=Eustace McGargle when adding a citation to Cuthbert McGargle. Having the quotation marks already in place avoids that particular error -- it allows an inexperienced editor to edit the reference name without having to think about there being two different formats, one for McGargle and another for McGargle 1936. --Guy Macon (talk) 20:31, 20 November 2020 (UTC)

Sarah has a long history of taking a principled stand on this issue, so I think your pointed question needs to find a new target. WhatamIdoing (talk) 21:46, 20 November 2020 (UTC)

Evasion noted. The above argument is relevant and should be addressed on its merits. --Guy Macon (talk) 18:22, 21 November 2020 (UTC)

CITEVAR

The whole point of CITEVAR is to stop article writers from being bothered by drive-by editors who might make a fiddly change to citation style before moving on to the next page, leaving you to carry on working with a style not of your choosing that might increase your workload. The RfC says: "Such edits ... should not be reverted simply because someone does not like the resulting citation style." But that's a valid reason to revert, if the person writing or maintaining the article doesn't want to work with a style someone else using a tool is trying to impose.

CITEVAR is good at stopping disputes, but it has to be remembered and nurtured. Don't let it die by neglect. SarahSV ^(talk) 06:14, 20 November 2020 (UTC)

If you disagree with "Such edits ... should not be reverted simply because someone does not like the resulting citation style." then propose changing the wording and see if there is consensus for your change. It appears that the above argument could also be used as an argument against WP:ENGVAR: "But that's a valid reason to revert, if the person writing or maintaining the article doesn't want to work with a language variant someone else is trying to impose." I say no. Neither of those are valid reasons to revert.

As for tools, if you don't like what a tool does, you should start by discussing the issue with the author to see if they agree, and if that doesn't work post an RfC to find out what the community consensus is. Neither reverting uses of the tool or haranguing users of the tool is helpful. If a tool does something that is not allowed, change the tool. --Guy Macon (talk) 20:31, 20 November 2020 (UTC)

@Guy Macon, why should I have to do that? Why should I have to even care whether an editor is using a script in the first place, much less to track down the author of that script, figure out whether the author is currently active, help him understand the problems, get him to agree to fix them, and see whether the author is capable of fixing them?

I reverted this ref-mangling edit yesterday. It took actions that we'd yell at an editor for doing manually. It might have been generous of me to figure out who wrote that script and tell him that his script is screwing up, but I think it was entirely appropriate for me to revert script-based edits that remove manual ref names from this high-traffic, high-activity article right away.

I have just looked up the script. The author hasn't edited for a couple of months, which usually means that the odds of getting it fixed aree low. Maybe someone should send it to MFD. WhatamIdoing (talk) 21:45, 20 November 2020 (UTC)

Bug reported:

User talk:Kaniivel/Reference Organizer#Bug report

et:Kasutaja arutelu:Kaniivel#Bug report

The author last edited on the 16th. His home Wiki is the Estonian Wikipedia but he should get notifications of edits made to his talk pages on the English Wikipedia. Just to be sure I posted a notice on his home wiki.

I would strongly advise against reverting users who use citation tools after Wikipedia talk:Citing sources#RfC: Citation tools was closed with a result of "Sentiment among participants is that a large discussion to affirm the acceptability of using citation tools is not needed" and "questions of editor behavior should be addressed as needed at noticeboards". --Guy Macon (talk) 00:04, 21 November 2020 (UTC)

Wait, are you really saying that if someone accidentally makes a bad edit with AWB, that you can revert it, but if someone accidentally makes a bad edit with AWB that touches citations, then you should leave the bad edit in place and try to get the editor punished instead? Maybe we should re-open that RFC again, and have people vote on what they actually want to see, rather than some bland language about whether scripts are generally acceptable. WhatamIdoing (talk) 00:11, 21 November 2020 (UTC)

Re: "are you really saying...", of course not. Please avoid silly debating tactics like interpreting what people write as if they are idiots or robots. What I am saying is that you can't all by yourself decide that the edits that Reference Organizer makes are "bad edits" and revert them. I asked whether that was allowed in the above RfC and the result was that what Reference Organizer does is so obviously allowed that "questions of editor behavior [such as d4eciding that what Reference organizer does is "bad edits" and reverting them] should be addressed as needed at noticeboards". --Guy Macon (talk) 02:50, 21 November 2020 (UTC)

No, that's silly. Primefac (talk) 00:12, 21 November 2020 (UTC)

...as well as being a purposeful distortion of my actual position. --Guy Macon (talk) 02:50, 21 November 2020 (UTC)

And yet Guy "would strongly advise against reverting users who use citation tools", with no sensible exceptions, such as "unless it's a bad edit that you would revert even if they'd made the same edit manually". WhatamIdoing (talk) 00:54, 21 November 2020 (UTC)

Again, you can't just decide that all edits using the Reference organizer tool are "bad edits" and revert them. That is not a "sensible exception". I asked that exact question in the above RfC and the result was that you aren't allowed to do that, I am allowed to use the tool, and that you should seek seek consensus if you want to forbid particular Reference Organizer actions such as adding quotation marks. --Guy Macon (talk) 02:58, 21 November 2020 (UTC)

Wait, isn't adding quotation marks in refnames considered objectively an improvement? Is it equally valid or invalid to add quotation marks where there aren't any and to remove them where there are? El Millo (talk) 03:17, 21 November 2020 (UTC)

Most people consider them to be an improvement. A few people hate them, but they know that the consensus is against them, which is why they don't want to settle the issue through an RfC. --Guy Macon (talk) 22:56, 21 November 2020 (UTC)

Please take the point that everything depends on what you do with the tool. If you violate CITEVAR, whether manually or with a tool, don't be surprised if someone reverts. SarahSV ^(talk) 03:05, 21 November 2020 (UTC)

CITEVAR doesn't say what you think it says. --Guy Macon (talk) 22:56, 21 November 2020 (UTC)

Guy, I never said anything about "all edits using the References organizer tool". I said that one (1), single, individual, isolated edit was a bad edit. In general, I'm pro-script, just not to the extreme of being pro-screwing-up-with-scripts. WhatamIdoing (talk) 01:15, 22 November 2020 (UTC)

Guy Macon, your RfC wasn't closed, it was halted. There is no consensus that changing ref name= is an acceptable thing to do per CITEVAR, with or without a tool, unless someone has actually made an error. I wonder whether the developers could make quotation marks with refname redundant entirely, so that if someone did write <ref name=NYT 2020>, it would be fine. SarahSV ^(talk) 00:30, 21 November 2020 (UTC)

@Guy Macon: There was certainly no consensus in the RfC for "bad edits by tools must be left unreverted". That's not what the RfC even asked. Bad edits are bad edits and should be reverted regardless of whether they are manual or automated. And @Facu-el Millo: No, adding quotation marks is not objectively an improvement. That's just, like, your opinion, man. I happen to believe that their addition, to an article that has been consistently formatted without them, is detrimental: fussy, and more of a pain than a benefit for future editing. —David Eppstein (talk) 05:17, 21 November 2020 (UTC)

@David Eppstein: According to WP:REFNAME: Quotation marks are preferred but optional if the only characters used are letters A–Z, a–z, digits 0–9, and ... all printable ASCII characters except some specific ones. They quotation marks required if spaces or any other character not listed there is used. Doesn't all that make it objectively an improvement to add them? Perhaps the optional in that sentence makes it a part of CITEVAR, in which case is still shouldn't be added, even if it is an objective improvement. Is that the case? El Millo (talk) 06:17, 21 November 2020 (UTC)

That "preferred" wording was added in 2019 by WhatamIdoing in this edit. I can find no discussion of this change at Template talk:Refname rules nor from up to that date in Help talk:Footnotes/Archive 2. I don't see any evidence that it ever actually had consensus. And it is only that one word that provides any evidence of one method being better than another; it's certainly not "objective", it's just a baseless assertion. —David Eppstein (talk) 06:43, 21 November 2020 (UTC)

Well, there's still the fact that the quotation marks work in every case and no quotation marks doesn't, but I'm interested in seeing what WhatamIdoing has to say in this matter. Perhaps there is a good reason for her edit and we just don't know it yet. El Millo (talk) 06:56, 21 November 2020 (UTC)

Facu-el Millo, search the section above for "WMF". WhatamIdoing (talk) 01:37, 22 November 2020 (UTC)

Ok, so it seems it isn't just a baseless assertion, David Eppstein. It actually comes from the developers who work on the code on Meta. El Millo (talk) 01:50, 22 November 2020 (UTC)

I am still not convinced. It comes second- or third-hand from developers, who could have easily said something like "there is a tiny slowdown in making the code flexible enough to handle both quoted and unquoted forms, but both forms are parsed equally quickly" that has then been amplified through a game of telephone into the current "we must add quotes because they slow things down so much that the inconvenience to editors of repeatedly typing them is worth the speedup obtained by omitting them". —David Eppstein (talk) 02:03, 22 November 2020 (UTC)

David Eppstein, it seems that WhatamIdoing has communicated directly with them. According to her, she got this information by talking to the devs who work on that code. Now you seem to be not assuming good faith, but I'll just AGF myself and assume you just forgot that she specified that. Either way, the claim that the developers could have easily said... anything other than what they actually said is merely hypothetical and not a valid argument. Lastly, any justified preference the developers express for quotation marks is enough to warrant putting preferred in the guideline. El Millo (talk) 02:20, 22 November 2020 (UTC)

Prioritizing developers over editors is as wrongheaded as prioritizing editors over readers. —David Eppstein (talk) 02:24, 22 November 2020 (UTC)

Sure many dont believe adding 2 quote marks is really chaging a "major citation styles" WP:CITESTYLE.--Moxy 🍁 06:34, 21 November 2020 (UTC)

And others do. As said above, it is a pain for editors who are actually working on the content of an article and have evolved a style where they don't like all those xtra keystrokes to deal with drive-by gnomes making their editing practice more difficult, for no benefit to anyone. —David Eppstein (talk) 06:43, 21 November 2020 (UTC)

Your going to go crazy if your fighting aggest the growing number of bots and new and old tools used by more and more that do code corrections of this nature. G L with that view of CITEVAR.--Moxy 🍁 08:55, 21 November 2020 (UTC)

As for "bad edits", I would prefer some evidence that adding two quotation marks is a "bad edit". Where was that decided? Finally, I typically only use the tool on pages that are on my watchlist. I don't much like drive by edits of any kind other than noncontroversial typo fixes and reverting spam added to multiple pages. --Guy Macon (talk) 07:02, 21 November 2020 (UTC)

I would prefer some evidence that the community doesn't want us to revert edits made by this script that do not involve adding two quotation marks, but do involve causing problems serious and objective enough that you had to file a bug report about them. WhatamIdoing (talk) 01:39, 22 November 2020 (UTC)

What does CITEVAR actually say?

I am seeing the oft-repeated argument that WP:CITEVAR forbids changing...
<ref name=Zapatopi>
...to...
<ref name="Zapatopi">.
What does CITEVAR actually say?

"Editors should not attempt to change an article's established citation style merely on the grounds of personal preference, to make it match other articles, or without first seeking consensus for the change."

The page then goes on to give multiple examples:

To be avoided

When an article is already consistent, avoid:

• switching between major citation styles, e.g., parenthetical and <ref> tags, or replacing the preferred style of one academic discipline with another's;
• adding citation templates to an article that already uses a consistent system without templates, or removing citation templates from an article that uses them consistently;
• changing where the references are defined, e.g., moving reference definitions in the reflist to the prose, or moving reference definitions from the prose into the reflist.

I contend that adding two quotation marks in the Wikimarkup that the reader will never see is not "switching between major citation styles" or any of the other to be avoided items.

I can, of course, post an RfC asking this exact question, but I am pretty sure what the result will be. --Guy Macon (talk) 08:41, 21 November 2020 (UTC)

The thing is, in order to avoid a potential edit war, or maybe a pointless discussion in a specific article, we must come to a consensus on which one is better. If everyone can change it from Zapatopi to "Zapatopi" and everyone can do it the other way as well, then that's what can (and probably will happen). We must set that changing from one to the other is an improvement, so doing it the other way around would necessarily be a change for the worse and it wouldn't be allowed. If we can't arrive to a consensus on that, then it must be included as part of CITEVAR. El Millo (talk) 17:42, 21 November 2020 (UTC)

As far as I am concerned I already have an answer to that. From Wikipedia talk:Citing sources#RfC: Citation tools:

START OF QUOTE

• Lithopsian (who was pinged but so far has not contributed to this discussion) appears to oppose all uses of the Reference Organizer tool and has reverted me twice. If this is closed without an answer and he keeps reverting, what do you suggest I should do? IMO if using the tool is allowed, Lithopsian should be warned to stop reverting uses of the tool, and if the tool is not allowed, the page that has the tool should be replaced with a message saying to stop using it. --Guy Macon (talk) 05:40, 12 November 2020 (UTC)

• Guy Macon, Well it appears that Lithopsian was being disruptive in the first place, and if they keep it up they should be warned, and if they keep it up, then blocked. Now, if there is a definitive policy issue at stake, it ought be discussed. But the use of citation tools is not an issue. Lithopsian doesn't like citation tools? Well tough bananas. CaptainEek ^{Edits Ho Cap'n!}⚓ 05:46, 12 November 2020 (UTC)

• Thanks! In that case, I think this can be closed. --Guy Macon (talk) 05:53, 12 November 2020 (UTC)

END OF QUOTE

There is our answer. Reverting users of Reference Organizer because you don't like the way it edits pages is disruptive. Anyone doing that should be warned, and if they keep it up, blocked. I can't see any way the answer could be clearer.

This does not change the fact that it is possible that some time in the future there may be a consensus that Reference Organizer is doing the wrong thing, but so far no such consensus has been established. Nor does CITEVAR forbid the kinds of edits Reference Organizer makes. --Guy Macon (talk) 18:22, 21 November 2020 (UTC)

I don't think you really understood my question. I'm not talking specifically about Reference Organizer, I'm talking about the change itself: from Zapatopi to "Zapatopi" and vice versa. If making changes both ways is acceptable, then it should be included in CITEVAR, so as to not have an article be constantly changed from one to another, allowing neither A to B nor B to A . If making changes one way is acceptable and the other way around isn't, then we should state which one's which, and from then on A to B is allowed and B to A isn't. El Millo (talk) 18:43, 21 November 2020 (UTC)

I understand you just fine. I simply disagree with your premises. You are assuming that we have two choices when we actually have at least four:

1. Fix it if you notice it, don't revert the fix. Examples: "Joe Jorgensen and Jo Biden were on the ballot", spam, vandalsm.
2. Fix it if you notice it, report at WP:SUPPRESS so nobody can revert it even if they wanted to. Example: His credit card number is..."
3. Either way is fine, but don't revert or edit war over it. Example: minor formatting changes done while making a more substantial edit.
4. Either way is fine, but don't change it, and it's OK to revert somome who does (Example: ENGVAR and CITEVAR (the real CITVAR that talks about major citation styles, not the made up one that supposedly talks about adding and removing quotation marks from ref names) ).

--Guy Macon (talk) 22:52, 21 November 2020 (UTC)

Nobody is talking about Reference Organizer except you. Stop using it as a stick to beat me with. I objected to your edits, not your use of a helper script. You just decided to open an RfC about something else and misrepresent my position in an attempt to get your own way. You were told the RfC as worded was foolish but instead choose to selectively quote from the responses to avoid the main issues. Everybody except you is trying to find a sensible set of guidelines for which reference edits might be considered improvements or which might be considered CITEVAR violations. There, I've replied. Now you can also stop implying that I'm somehow skulking in shame. Lithopsian (talk) 17:14, 22 November 2020 (UTC)

You reverted (not just objected to) the edits that the Reference Organizer script made when I ran it -- the exact same kind of edits that it has made thousands of times for hundreds of users. And an administrator commented on your revert by saying "Well it appears that Lithopsian was being disruptive in the first place, and if they keep it up they should be warned, and if they keep it up, then blocked" and "Lithopsian doesn't like citation tools? Well tough bananas." (In case you didn't notice, Reference Organizer is a citation tool -- the specific citation tool that you reverted.) If you really think that I am the one being disruptive here, go ahead and report me at ANI and see how that works out for you. I will post a warning on your talk page the next time you revert the edits that the Reference Organizer script makes, and if you do it again after the warning we will be discussing your behavior at WP:ANI. --Guy Macon (talk) 18:42, 22 November 2020 (UTC)

As I have pointed to above, we have already had the general RFC that says that minor formatting is not protected by CITEVAR. I linked it above inside the purple archive box. This is minor formatting. Ergo, it is not covered by CITEVAR. Now, which is preferred is a reasonable question that I'm willing to RFC about (I'm fairly confident the pros of quotation marks in ref statements will outweigh any cons) if certain editors above are willing to insist the opposite, or that CITEVAR does cover this. I'm even happy to launch the RFC myself, though I'm sure I'll be lambasted out of the house for doing such a thing over such a small thing. :^) --Izno (talk) 18:52, 21 November 2020 (UTC)

Izno's link does not lead where it purports to lead. Jc3s5h (talk) 19:28, 21 November 2020 (UTC)

The correct link to the RFC about whether markup falls under CITEVAR is WT:Citing sources/Archive 40#RFC: Is a change in citation markup method a change in citation style?. The relevant phrase from the closing statement is "There's not consensus for a clear yes-or-no answer to the proposed question, but...." Nothing in the closing statement endorses (or disparages) the practice of hopping from article to article, changing the markup to one's preferred style of markup, if the change does not change the visual appearance of the article as displayed to the reader. (But if you did that to computer code in a corporate environment I'll bet you'd be looking for a job.)Jc3s5h (talk) 19:38, 21 November 2020 (UTC)

The full closes:

There's not consensus for a clear yes-or-no answer to the proposed question, but two useful tidbits have come out of this discussion:

1. There is consensus that any change in coding that alters the visual output whatsoever requires consensus. This could be as minor as a visual change in whitespace or a comma changing into a period.
2. Non-systemic changes of coding that occur while updating the content of a citation and/or adding citations do not require consensus. Editors on the "no" side were concerned that the use of VisualEditor would fall afoul of CITEVAR and/or editors adding new citations would be required to use a specific citation style. The "yes" side is more concerned with systemic changes without consensus.

The first numbered section above refers to systematic changes across an entire article changing from one consistent citation style to another consistent citation style. When making such a systematic change, any change in visual output requires consensus. This is not to be interpreted as needing consensus for any of the following changes:

1. Making the visual output of citations consistent within an article where there has been no history of consistency.
2. Changing the coding of a single citation while writing content, altering it, etc. This specifically does not require consensus as per the second numbered section above.

In short, the first point above was intended to be read very narrowly. An editor hopping from article to article converting everything to a template would be a "no" without consensus, but that's the only situation that I can think of where the first point would apply.

The followon RFC was:

There is a clear consensus that the usage of vertical and horizontal templates does not fall within the purview of WP:CITEVAR.

The main argument made by the "no" side is that the inclusion of wikitext formatting within a style guideline is a form of WP:CREEP as the coded structure of the citation does not visually alter the article and provides no difference to the reader. They also point to the established decision made last year which stated that non-systemic changes of coding that occur while updating the content of a citation and/or adding citations do not require consensus. Additionally, the use of the VisualEditor can lead to changes in citation formatting that the user may be unaware of, which could directly violate CITEVAR if it were to be expanded to include such situations

The "yes" side's main concern was with codifying instruction in order to prevent edit wars, and that users should not change the formatting without prior consensus. The existence of established policies such as WP:BRD, WP:EW, WP:OWN, and WP:BUREAU eliminates the need to codify something as specific as this.

While the code structure does not require consensus to change, all participants agreed that no part of this discussion gives license to WP:EDITWAR or participate in widespread WP:TENDENTIOUS editing.

Now, are you really going to argue that addition of apostrophes in ref names is more a major change to article formatting than whether I remove the entirety of all return characters in citations? Are you going to argue it is so in combination with the tool? If only the latter, the previous RFC, clearly going for a SNOW close of "of course the tools are permitted", will take winds out of those sails. If the former... I can't lead a horse even to water, clearly. --Izno (talk) 20:11, 21 November 2020 (UTC)

The link to the RFC which, in my opinion, answers the issue is the one quoted, but not linked to, by Izno at 20:11, 21 November 2020 (UTC). The link is WT:Citing sources/Archive 43#RfC: Should usage of vertical and horizontal templates fall within WP:CITEVAR. Indeed, if CITEVAR does not apply to removing all the new-lines from all the citations in an article (or putting them in after every parameter), then it shouldn't apply to adding or removing quote marks around reference names in instances where the Cite extension does not require quotes. Jc3s5h (talk) 20:31, 21 November 2020 (UTC)

Hmm? --Izno (talk) 21:14, 21 November 2020 (UTC)

"What does CITEVAR actually say?" isn't really the best way to address this question. CITEVAR does not provide a comprehensive list of every permissible (or prohibited) edit to citation formatting, and was never intended to.

There are multiple questions here. "Can I mass-add quotation marks to an article's ref names?" is not necessarily going to get you the same answer as "Can I mass-delete manual ref names?" or "Am I allowed to revert an edit made by a buggy script?" WhatamIdoing (talk) 01:44, 22 November 2020 (UTC)

"Mass adding" and running the script once a week or so while in the process of editing a page are two different things. And yes, you certainly are allowed to revert an edit made by a buggy script. You will have to pardon me for asking for at least a shred of evidence that it actually is buggy. --Guy Macon (talk) 18:42, 22 November 2020 (UTC)

Not understanding the technical aspects of the issue, so asking for clarification

For the benefit of those who just copy and paste, but don’t understand, the magic glyphs that make things work around here: could someone explain the DIFFERENCE between <ref name=Mcguffin> and <ref name=“Mcguffin”>? What does adding or removing the quotation marks DO (in layman’s terms). As best I can make out, there is a problem with having a space in the refname (ie: <ref name=Mcguffin 123>) that adding the quote marks avoids... but even here I am unclear. Please explain. Blueboar (talk) 21:01, 21 November 2020 (UTC)

Yes, a ref with name as <ref name=Mcguffin 123> will actually be <ref name="Mcguffin"> (plus some 123 text the system does not care about because it is an unnamed parameter). As it happens, since you used them, curly quotes are also wrong, so <ref name=“Mcguffin 123”> will actually be <ref name="“Mcguffin"> (plus some of the same). Adding the quotation marks as in <ref name="Mcguffin 123"> makes it clear to the system that you want a reference with the name "Mcguffin 123" and interpreted accordingly. --Izno (talk) 21:07, 21 November 2020 (UTC)

Ok... so are you saying that, as long as you don’t have a space in your refname, there ISN’T any difference? Does it MATTER whether you use quotes or not?

(BTW... I am editing via my phone ... which does not HAVE an option for straight quotes... only curly. So THAT is another problem we need to address... but perhaps separately). Blueboar (talk) 21:22, 21 November 2020 (UTC)

Correct, in a page which does not change, there is no difference to the system between a ref name with the standard alphabet plus numbers and no quotes and the same with quotes (I am only limiting it to that alphabet because I do not know how well the name attribute deals with non-Latin alphabets or those Latin alphabets with diacritics of whatever sort; I have no reason to believe those would behave differently). This is a pure technical statement.

In a less technical statement, if you've ever gnomed or cared about the various categories related to broken references (see Help:Cite errors), you will inevitably find issues where some editor has, in good faith, changed the name such that it includes a space now instead (or changes it such that it includes curly quotation marks, for a relevant example to you). These changes inevitably break those references without quotation marks. (Sometimes it works by chance that this does not break the page, but it is by chance in almost all such cases.) This problem is the primary reason I assert that quotation marks are always beneficial. --Izno (talk) 21:36, 21 November 2020 (UTC)

Izno, the ref tags accept just about anything, including non-Latin names, such as <ref name="مولد تلقائيا2">. I think, but I haven't verified, that you can use pretty much any Unicode character except for the handful of reserved characters such as < and = (all of which will be obvious to you), including curly quotation marks (which will be treated like any other character, not like quotation marks). WhatamIdoing (talk) 01:54, 22 November 2020 (UTC)

Izno, well technically a space, a single quote, a double quote, an equals sign, a less than, or a greater than character. Standard HTML convention however is attributename="attribute value".. Since this wikicode is inspired by HTML and because most systems/software that would want to parse wikitext will expect that style, that is the preferred style and that is the style that any software is likely to use. That humans want something else is not something you should be expecting the software to be able to understand.

Can't believe people still fight about pointless differences in wikitext like this. Just ignore the changes and be happy. —TheDJ (talk • contribs) 21:59, 21 November 2020 (UTC)

Just to expand a bit on Izno's excellect explanation ("Yes, a ref with name as <ref name=Mcguffin 123> will actually be <ref name="Mcguffin"> (plus some 123 text the system does not care about because it is an unnamed parameter).) this causes real problems when an editor sees <ref name=Mcguffin 2004> and adds another cite to <ref name=Mcguffin 2019>. The editor expects them to be two different names (and they would be if enclosed in quotation marks) but instead Wikipedia reads them as the same name. --Guy Macon (talk) 22:15, 21 November 2020 (UTC)

Blueboar: I think Izno does a good job at explaining the technical aspects of this concisely. To try to pare it all down to even more practical terms: it is a good habit to develop to use the quotes when writing from scratch, because even though there are scant few cases on Wikipedia where not doing so would cause problems, the situations where you do encounter problems have the potential to be confusing and frustrating to resolve. In the situation you describe where you're editing from your phone, I'd probably be inclined, personally, to stick to cutting and pasting as-is.

But apart from that practical consideration we should say for the record that—if you are only ever going to be writing code in Wikipedia pages—the developers of the MediaWiki software have ensured that, like, 99.99% of the problems not using quotes would cause in other environments are taken care of automatically.

(As someone who started off coding non-wiki stuff where it matters much more, I have to admit that seeing the no-quotes version is like nails on a chalkboard. But, sigh, that doesn't really matter here.) --‿Ꞅtruthious 𝔹andersnatch ͡ |℡| 22:33, 21 November 2020 (UTC)

Thanks folks. Final question: since I do a lot of editing on my phone... which does not give me the OPTION to include straight quotes... I would probably have to convert a broken <ref name=Mcguffin 1920> to an unspaced <ref name=Mcguffin1920>. I assume this would be OK, given the technical limitations of my devise. Blueboar (talk) 23:41, 21 November 2020 (UTC)

I don't think anyone would object to that even if you just did it just because you felt like it, and certainly not if your phone has that limitation. And if they did object they could simply add the quotation marks themselves. --Guy Macon (talk) 23:52, 21 November 2020 (UTC)

Rather than developing the habit of typing unnecessary quote marks, I've developed the habit of only using alphanumeric characters in my ref names. In fact until this discussion I didn't even realize that hyphens were allowed. —David Eppstein (talk) 00:32, 22 November 2020 (UTC)

Thanks all... I think I understand things better, and can now opine on the issue at hand. Obviously, when someone breaks a refname citation by adding a space (or some other “non-compatible” character) it needs fixing. And it looks like there are several acceptable ways to do that - removing the space being one, adding quote marks being another (and I am sure there are yet others we have not discussed). Which to use will depend on how other (non-broken) citations are formatted. Don’t change all, just to fix one. Make the one adhere (as best you can) to the others.

As for refname citations that ARN’T broken (ie have no spaces or “non-compatible” characters) ... don’t add OR remove quote marks. Don’t fix things that don’t need to be fixed. Leaving acceptable citation formatting “as is” adheres best to the spirit of CITEVAR (and its underlying goal of preventing potential edit wars). Blueboar (talk) 01:20, 22 November 2020 (UTC)

Yes, I agree. It's also a complete waste of resources, since it takes up processor time and hence uses energy, just to make changes like <ref name=Smith20> to <ref name="Smith20">, which I see happening all the time. (Editors trying to up their edit count??) Peter coxhead (talk) 10:37, 22 November 2020 (UTC)

The motivation for such changes has been clearly stated several times: to replace a reference name that is a land mine that causes problems when a new editor makes a reasonable-looking change and replace it with a robust reference name that works perfectly no matter what name the user chooses. Please AGF and stop accusing people of doing things to up their edit count after they have clearly expressed their actual reasons for doing those things. --Guy Macon (talk) 12:05, 22 November 2020 (UTC)

Also, as an XML attribute, it should be quoted in the first place to be proper XML, though Wikimedia is configured to ignore these types of errors, for now. --Masem (t) 18:51, 22 November 2020 (UTC)

The sfn template, and editors rather than authors

Hi. It appears that {{sfn}} doesn't work if there are editors listed but no authors. An example is Gilbert's honeyeater; the Higgins footnotes don't "jump" properly to the ref. Should they? Or is there a reason we've chosen for it not to work with editors? MeegsC (talk) 11:16, 28 November 2020 (UTC)

You need to list all the editors in the {{sfn}} (i.e. Higgins, Peter and Steele in Higgins, Peter J.; Peter, Jeffrey M. & Steele, W. K., eds. (2001). Handbook of Australian, New Zealand and Antarctic Birds. Volume 5: Tyrant-flycatchers to Chats. Melbourne: Oxford University Press. ISBN 0-19-553258-9., not just the first - now fixed.Nigel Ish (talk) 11:32, 28 November 2020 (UTC)

Ah! Thanks for that Nigel Ish. You're a star. MeegsC (talk) 12:06, 28 November 2020 (UTC)

A tool for converting citation styles

Please see the proposal at meta:Community_Wishlist_Survey_2021/Citations/Convert_tools_for_different_citations_or_citation_styles. --_{Piotr Konieczny aka Prokonsul Piotrus| reply here} 03:57, 30 November 2020 (UTC)

Citation

Material scientist could you guide me how to provide citations Eaageshwari (talk) 10:02, 6 December 2020 (UTC)

Eaageshwari, I would highly suggest reading through WP:REFB. If you want more help, stop by the Teahouse, or Wikipedia's live help channel, or the help desk to ask someone for assistance. Primefac (talk) 15:40, 6 December 2020 (UTC)

Italics in citations (again)

FYI

 – Pointer to relevant discussion elsewhere.

Please see Help talk:Citation Style 1/Archive 72#Italics 2 – another round of recurrent dispute over what to italicize in citations, and whether to cite just the publisher if the work title and the publisher are essentially the same. This was addressed in a previous RfC fairly recently, and many prior discussions, but there still seems to be some doubt among some editors what the answers are.  — SMcCandlish ☏ ¢ 😼  22:23, 6 December 2020 (UTC)

Shortened footnote syntax vs inline references

Shortened footnotes are in my opinion godsent. They make it much easier to add precise footnotes with page numbers and other metadata. They also significantly reduce text clutter. Though the guideline says

Editors should not attempt to change an article's established citation style merely on the grounds of personal preference, to make it match other articles, or without first seeking consensus for the change.

What would "first seeking consensus" mean in this case? A lot of Wikipedia articles are very rarely edited so finding a consensus one way or the other is often not easy. And what does the guideline mean for articles in which shortened footnotes coexist with inline references? ImTheIP (talk) 10:41, 10 December 2020 (UTC)

To answer your second question, inline/Harvard-style references were recently deprecated, so you can remove those without discussing first. As far as "seeking consensus" goes, it generally means starting a talk page discussion. Primefac (talk) 11:01, 10 December 2020 (UTC)

Just a point of clarity, short-form references inside <ref>...</ref> tags are not deprecated so all of these are acceptable:

<ref>Smith 1999, p. 24</ref>

<ref>{{harvnb|Smith|1999|p=24}}</ref>

{{sfn|Smith|1999|p=24}} – this template does not require <ref>...</ref> tags

—Trappist the monk (talk) 12:02, 10 December 2020 (UTC)

I should clarify. What I mean by "inline references" is wiki code like this:

<ref name="Aharoni2006">{{cite book|author=Yohanan Aharoni|title=The Jewish People: An Illustrated History|url=https://archive.org/details/jewishpeopleillu00ahar_0/page/99|date=15 September 2006|publisher=A&C Black|isbn=978-0-8264-1886-9|pages=[https://archive.org/details/jewishpeopleillu00ahar_0/page/99 99–]}}</ref><ref name="FahlbuschBromiley2005">{{cite book|author1=Erwin Fahlbusch|author2=Geoffrey William Bromiley|title=The Encyclopedia of Christianity|url={{Google books|C5V7oyy69zgC|page=PA15|keywords=|text=|plainurl=yes}}|year=2005|publisher=Wm. B. Eerdmans Publishing|isbn=978-0-8028-2416-5|page=15}}</ref><ref>{{cite journal|volume=94|issue=885|date=Spring 2012|journal=International Review of the Red Cross|title=The law of belligerent occupation in the Supreme Court of Israel|first=David|last=Kretzmer|author-link= David Kretzmer|doi=10.1017/S1816383112000446|url=https://www.icrc.org/eng/assets/files/review/2012/irrc-885-kretzmer.pdf|quote=This is probably the longest occupation in modern international relations, and it holds a central place in all literature on the law of belligerent occupation since the early 1970s|pages=207–236}}</ref>

It is, in my opinion, preferable to use a bibliography section and then reference the works like this: {{sfn|Aharoni|2006|p=99}}{{sfn|Fahlbusch|Bromiley|2005|p=15}}{{sfn|Kretzmer|2012|pp=207-236}} ImTheIP (talk) 14:21, 10 December 2020 (UTC)

Sometimes that's a good thing, yes, though I would argue that it's not strictly necessary to use sfn (I personally prefer named references). I think if an article has huge blocks of refs like that one, it would be relatively uncontroversial to shift all of them to the references section and replace them with <ref name=Aharoni2006 /><ref name=FahlbuschBromiley2005 /><ref name=Kretzmer2012 />. Primefac (talk) 14:34, 10 December 2020 (UTC)

I would just post in the talk page about what I wanted to do, and give it a week. If nobody objected, I'd go ahead.

As far as how controversial it would be to shift from what Primefac described to using the sfn template, the safest approach would be to post in the talk page as I mentioned. There would be a medium risk of complaints if someone who edits the article on a regular basis, or who at least edits articles on the same topic, were to make the change without seeking consensus. There would be a high risk of complaints if someone with no interest in the topic made the change as part of a campaign to introduce the sfn template on as many articles as possible as fast as possible, without regard to the topic of the article. In other words, an editor who cares much more about the sfn template than the articles. Jc3s5h (talk) 16:07, 10 December 2020 (UTC)

I would be one of those people, for what it's worth. It's one thing to say "there are too many references in the body of the text, let's move the bulk of them to the {{reference}} table, it's another thing to say "let's blindly convert every <ref>...</ref> to use {{sfn}} because we can." Primefac (talk) 16:18, 10 December 2020 (UTC)

Alright, and just to clarify for myself and others. What I'm calling "inline references" is the "syntax gunk" in my previous comment. What Primefac is describing with "relatively uncontroversial to shift all of them to the references section" is, I believe, called list-defined references which is not the same thing as shortened footnotes. I take it to mean that introducing list-defined references in articles where there is none is, generally, less controversial than introducing shortened footnotes in articles where there is none? I should add that I'm thinking about articles which I've already edited a lot. I can understand how it would be rude for someone who has not otherwise edited the article in question to change the referencing style. ImTheIP (talk) 12:25, 11 December 2020 (UTC)

introducing list-defined references in articles where there is none is, generally, less controversial than introducing shortened footnotes in articles where there is none is not necessarily an accurate statement - in either case you should propose on talk first and proceed only if there are no objections. Nikkimaria (talk) 13:28, 11 December 2020 (UTC)

There are other ways of doing shortened footnotes besides list-defined or using the sfn template. You can do them easily with <ref>Author p. X</ref> or <ref name=A>Author p. X</ref> or variations. -- Ealdgyth (talk) 13:36, 11 December 2020 (UTC)

feature request bultin reference management system

for real after all these years with all these strict sourcing rules and wikipedia/mediawiki hasn't yet a proper reference management system. it's a mess with duplicated stuff everywhere. there should be a repository (probably using wikidata) for creating entries for reference works. then being able to cite them on wikimedia projects like wikipedia... — Preceding unsigned comment added by TRANSviada (talk • contribs) 15:36, 8 December 2020 (UTC)

I feel like that's on one of the perennial requests lists, like the Wishlist. Primefac (talk) 15:44, 8 December 2020 (UTC)

do you know of any issue regarding this on the bug tracker? i couldn't find anything in specific. talk@TRANSviada 16:13, 8 December 2020 (UTC)

I personally do not; while I'm heavily involved in template editing I'm not so much involved in the specifics of referencing and the citation templates. Primefac (talk) 16:19, 8 December 2020 (UTC)

just saw this on the wishlist; they also didn't find a ticket. So yeah, seems like it needs to be taken initiative to open an issue on the bug tracker. talk@TRANSviada 18:40, 8 December 2020 (UTC)

Suggest taking a look through meta:WikiCite. Nikkimaria (talk) 23:07, 8 December 2020 (UTC)

I think it is actually possible currently to keep references on Wikidata and then cite them indirectly, using {{cite Q}}. It doesn't seem to be very widely used relative to direct citations, though, and I don't know an easy way to push the reference information onto Wikidata. —David Eppstein (talk) 00:41, 9 December 2020 (UTC)

It doesn't work in practice because cite Q formats differently to all the other cite templates. So, to be consistent, articles have to be all one or all the other, and cite Q is less easy to police for changes. DrKay (talk) 17:34, 9 December 2020 (UTC)

Cite Q claims to format the same as {{citation}}, i.e. Citation Style 2. Is that not the case? —David Eppstein (talk) 19:53, 9 December 2020 (UTC)

{{cite q}} calls {{citation}} for rendering. To get it to render as cs1, |mode=cs1

—Trappist the monk (talk) 20:06, 9 December 2020 (UTC)

It wasn't in my sample size of 1[5], and judging from the cite Q documentation, there is only an 'author' parameter, not 'firstname', 'lastname' parameters. So, the output of authors names will not be uniform unless standardized across all cite templates. DrKay (talk) 20:08, 9 December 2020 (UTC)

|first=Joe|last=Schmoe works with {{cite Q}}, but currently breaks if author1 is in the data source. Sandbox version of the template fixes this, and hopefully will be deployed in a couple days. —Michael Z. 16:53, 11 December 2020 (UTC)

Propose to be able to officially cite application source code

Sometimes there are articles which are about an application or piece of software where it can be difficult to find reliable sources describing application features, known issues, etc, as they may not exist in reliable sources. I propose the idea that if actual source code is publicly available, that it should be allowed as a reliable source. The source code doesn't have a point of view, and cannot lie or make exaggerated claims (the comments might, but the actual code cannot). --Thoric (talk) 19:36, 11 December 2020 (UTC)

This isn't the right page to discuss this. Appropriate pages would be the orignial research talk page or the reliable sources talk page. This page is about how to cite, not what to cite. Jc3s5h (talk) 20:44, 11 December 2020 (UTC)

Okay, thanks :) I'll move it over there... --Thoric (talk) 20:50, 11 December 2020 (UTC)

Citing Wikidata

I could have sworn that, about a year ago, we had a huge RFC that closed saying we should NOT cite Wikidata (although a side bar link was OK). Has something changed? Did I miss a subsequent RFC saying it was now OK? Would someone point me to the most recent decision? Blueboar (talk) 18:34, 22 December 2020 (UTC)

Not a year ago, but perhaps you're thinking of one of the discussions linked from Wikidata's entries at WP:RSP? Nikkimaria (talk) 22:27, 22 December 2020 (UTC)

Not what I was thinking of, but it points me in the right direction. I have amended this guideline to note what is said at RSP ... no direct citations to WD itself, but transclusion of “items” (citations) via WD can be done in certain circumstances (with caution). It’s complicated. Blueboar (talk) 12:09, 23 December 2020 (UTC)

I saw a not-unreasonable proposal recently to sorta-kinda-not-exactly cite Wikidata in a list (classical artwork, maybe? It was something along those lines). The specific Wikidata items in question were more comprehensive than the list would (or should) be, and they were sourced. We decided that putting general information in the list, with an "external link" to Wikidata, would not be unreasonable in comparison to the alternative, which would have required changing "c. 1780" to "Source A says 1779-1780, Source B says 1782, Source C says early 1780s" for a long list. WhatamIdoing (talk) 02:08, 28 December 2020 (UTC)

Does CITEVAR prohibit using last/first instead of author for new citations?

I have been reverted (example) for adding new citations with the author's last name split between last and first when all other citations in the article used author or did not cite an author at all.

I understand that "Editors should not attempt to change an article's established citation style merely on the grounds of personal preference". But for example, if I am only adding new citations (not modifying any existing author fields) with a citation tool, and these new automated citations use last and first (for individuals, not organizations), is this grounds to be reverted?

I already asked one user personally about this (since I am new here), and he suggests that my behavior was acceptable, but I would really like to hear everyone's thoughts on this.

"The only reason this user seem to be reverting your edits is because of WP:CITEVAR. They think that |author= should not be changed to |last= and |first=, which seems like a non-issue to me since 'author' is at best used when citing books and journals, while the first-last name combo is used in most situation especially news and other things. Even then if you use this tool, it uses the 'first-last name' by default. Check for example Ed Stelmach FA article I chose randomly. It uses first-last combo as you can see. As for the date formats here. I don't know what kind of commonality they are following since the ref dates are not all of same format. I believe the warning they issued you is uncalled for, 'bullying' and displays some sort of WP:OWN behavior. - Fylindfotberserk (talk) 07:25, 26 December 2020 (UTC)"

TimSmit (talk) 05:10, 29 December 2020 (UTC)

If the article has been deliberately maintained with a consistent style of using first-name-first surname-last author name spelling, then you should not be changing that style without discussion and consensus per WP:CITEVAR. In the more common case where author name ordering is haphazard and inconsistent, then I think it would be ok to standardize it, preferably in the more common surname-first order. The harder cases to judge are where the author names happen to be first-name-first but not for any reason or with any intent, just the way they turned out. In those cases you can get away with reversing the order, but if someone objects (as appears to have happened) then you know you're in the CITEVAR case instead. —David Eppstein (talk) 06:00, 29 December 2020 (UTC)

@David Eppstein: thank you. More generally, my question is: if all existing citations use |author=, is it unacceptable (per CITEVAR) to add any new citations that contain |first= and |last=? TimSmit (talk) 17:50, 29 December 2020 (UTC)

People add inconsistently formatted citations all the time. It's sloppy but usually better than adding no citations at all. It would be more polite for you to take the effort to make your added citations conform, and for someone objecting to the formatting to format them correctly, rather than being sloppy or reverting the addition. But that you know the citation format is important in this case, why are you still even continuing to argue or look for excuses to justify your sloppiness? —David Eppstein (talk) 17:55, 29 December 2020 (UTC)

@David Eppstein: I'm sorry if I came across as sloppy. I had been under the impression that last/first should always be used for individuals (with author being reserved only for groups). But now, thanks to you, I understand that this distinction does not matter, and the citation format should always follow other citations in the article. TimSmit (talk) 18:45, 29 December 2020 (UTC)

The original post does not correctly describe the edit that was reverted. Before this edit the article contained a mixture of hand-formatted citations and citations that used Citation Style 1 (CS1) templates. The edit changed one of the hand-formatted citations to a CS 1 template, and added a CS 1 template. I have not reviewed the history of the article, so I don't know how long this inconsistency in citation methods existed, nor do I know what the article looked like when it was consistent (if it ever was). Jc3s5h (talk) 18:10, 29 December 2020 (UTC)

Raw unformatted URLs, like the Twitter ones that were reverted aren't really OK - however. In that case even if templated references don't fit with the article style, then the relevant detail should have been retained in the manually formatted version.Nigel Ish (talk) 19:44, 29 December 2020 (UTC)

(For the record, the reverting editor restored the {{cite tweet}}s a few minutes later.) WhatamIdoing (talk) 17:04, 12 January 2021 (UTC)

If the issue is as asked, last first is always preferred to author (save for exceptions such as organizational authors). Period and end of story. No, this issue is not covered CITEVAR, not least due to the prior RFC that I continue needing to cite. David is simply incorrect on that point. --Izno (talk) 22:16, 29 December 2020 (UTC)

Izno, your post at 22:16 is your first contribution to this thread, so I don't know which RFC you may be referring to. The original post states in part "when all other citations in the article used author or did not cite an author at all." That implies that Citation Style 1 or 2 are accepted as the citation style for the article. If that's true, then the issue isn't addressed by this "Citing sources" page, but rather by help:Citation Style 1. That page makes it pretty clear that the style is last, first, if the author is a natural person with a first and last name.

Citation styles exist that use first, last. One such style is Chicago Manual of Style 17th ed., when note numbers are placed in the text and endnotes are at the end of the article. See, for example, ¶ 14.19. This is an allowable citation style. Jc3s5h (talk) 23:02, 29 December 2020 (UTC)

Yes, the context of the question was CS1/2 templates. Is that not obvious to you? --Izno (talk) 23:34, 29 December 2020 (UTC)

It is not obvious to me the context was CS1/2 templates because Help talk:Citation Style 1 would be a more appropriate page to discuss an issue that is limited to CS1/2, and by placing the discussion on this page, the statements can be understood to apply to all citations. Jc3s5h (talk) 00:40, 30 December 2020 (UTC)

In what cases are general sources preferred?

When I read WP:GENREF, I understood that specific references are preferred over general references, but I see that, in the edit Special:Diff/997480277 to Wolf, a number of references and the statements they supported were deleted with the edit summary sticking to gerneral sources. As I'm fairly new to Wikipedia and its behind-the-scenes world of internal documentation, I'd like to ask: where should I look for a page that would explain why one would prefer "sticking to [general] sources"? —2d37 (talk) 14:51, 2 January 2021 (UTC)

2d37, you could ask the user to clarify, but based on that diff my assumption is that general sources here refers not to GENREF but to the more generic meaning of "broad" sources rather than more specialized ones. Nikkimaria (talk) 15:52, 2 January 2021 (UTC)

@2d37, a GENREF is the opposite of an Wikipedia:Inline citation. The main use of general references seems to be in articles about mathematics. It's not unreasonable to expect one chapter in a textbook to be capable of verifying the entire contents of a Wikipedia article about some math concept. In that case, why add little blue clicky numbers at the end of every sentence, when you could just list it once at the end?

I agree with Nikkimaria that this doesn't seem to be what the editor is talking about, though. I think you should consider asking the editor about this on the article's talk page. WhatamIdoing (talk) 17:11, 12 January 2021 (UTC)

WP:INTEXT for opinions

From WP:INTEXT:

In-text attribution should be used with direct speech (a source's words between quotation marks or as a block quotation); indirect speech (a source's words modified without quotation marks); and close paraphrasing. It can also be used when loosely summarizing a source's position in your own words.

I was under the impression that in-text attribution should be used with anything controversial, anything that's an opinion, and ideas from partisan and somewhat unreliable sources. Basically, use Wikivoice for facts, and in-text attribution for opinions.

In my opinion, the current wording doesn't convey this. Any interest in adding a sentence about this, or modifying the sentences above to reflect this? –Novem Linguae (talk) 17:02, 10 January 2021 (UTC)

Are you referring to WP:ATTRIBUTEPOV? There is already a reference there, I think but adding some language could help, yes.--JBchrch (talk) 18:02, 11 January 2021 (UTC)

JBchrch, yes, WP:ATTRIBUTEPOV sums up what I'm talking about. Thanks for pointing it out. I went ahead and added a sentence to WP:INTEXT borrowing some wording from the WP:ATTRIBUTEPOV section. –Novem Linguae (talk) 10:13, 12 January 2021 (UTC)

Novem_Linguae, thanks. I added a hyperlink on "biased statements of opinion" to WP:ATTRIBUTEPOV.--JBchrch (talk) 10:38, 12 January 2021 (UTC)

I have some concerns about including the word "biased" in here. Who gets to decide which opinions are the biased ones? The unpopular opinions? The opinions that an editor claims are subjective? Maybe the opinions about art and culture are biased, but opinions about whether a notable scientific study was well-designed are not? WhatamIdoing (talk) 17:19, 12 January 2021 (UTC)

WhatamIdoing, thanks for the feedback. I copy/pasted the phrase "biased statements of opinion" from the WP:ATTRIBUTEPOV section. I was hoping by copy pasting existing text that it would be uncontroversial. If we start debating this wording, we'd probably have to get a big consensus, then change the text in both sections. –Novem Linguae (talk) 20:29, 12 January 2021 (UTC)

I think you're right. It would be much better to have this addressed in the policy first. WhatamIdoing (talk) 21:11, 12 January 2021 (UTC)