Help talk:Citation Style 1

Template:NewMusicBox

Would anyone care to make {{NewMusicBox}} a wrapper for a suitable CS1 template? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 20:56, 26 July 2016 (UTC)

Doesn't this question first belong on that template's talk page? Or, in lieu of that, a notice to the template's author about the conversation here?

—Trappist the monk (talk) 09:59, 28 July 2016 (UTC)

No. It needs someone with an understanding (better than mine, or I would do it) of CS1. This is the place where such people congregate. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 17:41, 31 July 2016 (UTC)

@Pigsonthewing: I've made {{NewMusicBox/sandbox}}, which is a wrapper for {{Cite interview}}. Note that if this sandbox version is used, someone should go through the articles using this template and change the parameter |composer=[[Composer article|Composer]] (or |composer=[[Composer]]) to |composer=Composer|composer-link=Composer article - Evad37 [talk] 06:07, 1 August 2016 (UTC)

@Evad37: Thank you. I've asked for an AWB operator to assist. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 17:43, 30 August 2016 (UTC)

New access-date check leads to bad bot proposal

I understand that the date validation code now produces an error message if the access date is not precise to the day:

• {{cite journal | title = Three-body problem, the measure of oscillating types. A short review | url = http://journals.cambridge.org/action/displayIssue?jid=IAU&volumeId=9&issueId=S310 | access-date = August 4, 2016 | last1 = Marchal | first1 = Christian | journal = Proceedings of the International Astronomical Union | date = July 2014 | pages = 43–44 | volume = 9 | issue = S310}}

  Marchal, Christian (July 2014). "Three-body problem, the measure of oscillating types. A short review". Proceedings of the International Astronomical Union. 9 (S310): 43–44. Retrieved August 4, 2016.

• {{cite journal | title = Three-body problem, the measure of oscillating types. A short review | url = http://journals.cambridge.org/action/displayIssue?jid=IAU&volumeId=9&issueId=S310 | access-date = August 2016 | last1 = Marchal | first1 = Christian | journal = Proceedings of the International Astronomical Union | date = July 2014 | pages = 43–44 | volume = 9 | issue = S310}}

  Marchal, Christian (July 2014). "Three-body problem, the measure of oscillating types. A short review". Proceedings of the International Astronomical Union. 9 (S310): 43–44. Retrieved August 2016. {{cite journal}}: Check date values in: |access-date= (help)

The unfortunate response to these error messages is a proposal to falsify all the old imprecise access dates by using a bot to falsely assert the access occurred on the first of the month: Wikipedia:Bot requests#Fix thousands of citation errors in accessdate

Of course, the only ways to "correct" an old access date is to have the editor who accessed the source to provide the day of the month the access occurred, or to revisit the source and provide the date of the revisit (provided, of course, the source in its current form still supports the claim(s) in the article. Jc3s5h (talk) 16:31, 4 August 2016 (UTC)

Or remove the access date for articles with DOI or other identifiers, per the {{cite journal}} documentation. – Jonesey95 (talk) 18:29, 4 August 2016 (UTC)

Or have the bot access the url in the citation and if error=0 input the date of the bot access. If any webpage-related http errors appear (404 etc.) remove |access-date= altogether. Then, in phase 2, the bot could check to see whether in pages that it successfully accessed, the content actually verifies the respective citations. That 1. could keep bot writers busy for a few years/decades, and therefore, 2. keep CS1 relatively unsullied. 65.88.88.127 (talk) 21:15, 4 August 2016 (UTC)

Use the date the URL was added to the article which the BOT could extract from the history. Keith D (talk) 00:21, 5 August 2016 (UTC)

That may or may not be the date the source was accessed. Revision date does not automatically imply access date. Also, inputting a prior (or forward) date violates WP:SAYWHEREYOUGOTIT. Only the editor who accesses the source can make such corrections after the fact. 72.43.99.146 (talk) 00:37, 5 August 2016 (UTC)

One solution would be to only output the error warning if the MM YYYY date is Sept 2016 or later (or a year of 2017 or later) - thereby catching (mostly) new instances, but not historic ones. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 16:30, 6 August 2016 (UTC)

Why should past instances be ignored? They are still wrong. Not flagging such obvious incogruities diminishes confidence in the encyclopedia. 64.134.101.16 (talk) 19:25, 6 August 2016 (UTC)

Firstly, they are not wrong, although they may be imprecise. And they should not generate an error for that reason, and for the reasons stated above, in this section. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:51, 11 August 2016 (UTC)

Since we are discussing the date when an anonymous editor claims to have verified the particulars of a citation by accessing it, not entering the exact, full date this happened is a bit more than imprecise, or careless, or just not following the correct procedure. Entering such exact information is as easy as not entering it. Therefore, other questions now arise. Afaic, something is wrong with such a citation and it should be flagged. 65.88.88.127 (talk) 19:43, 11 August 2016 (UTC)

Partial access dates

CS1 is generating an error if |accessdate= is partial. For example these generate an error:

|accessdate=2015

|accessdate=May 2015

This is good:

|accessdate=May 1, 2015

The number of articles impacted is 14665 in 32255 citations. The error is a new feature added recently from Help_talk:Citation_Style_1/Archive_19#incomplete_access_dates. This led to a proposal for a bot fix: Wikipedia:Bot_requests#Fix_thousands_of_citation_errors_in_accessdate .. However the Oppose camp do not believe an access-date with two parameters (Month Year) should generated a CS1 error. There is no way to fix 32255 citations without a bot, and no way to do a bot without consensus for the error message.

1. Should CS1 generate an error in accessdate's with only a year |accessdate=2015 ?
2. Should CS1 generate an error in accessdate's with only a month year |accessdate=May 2015 ?

Pinging previously involved parties: @Trappist the monk, 72.43.99.138, Matthiaspaul, 65.88.88.62, Rfassbind, Jonesey95, Jc3s5h, Cyberpower678, Redrose64, Magioladitis, PBS, Tom.Reding, BabbaQ, and Finnusertop:

Note: this is not an RfC please consider discussion rather than vote-style. If we need an RfC it will become evident and we can announce to the community a separate RfC here or Village pump. -- GreenC 16:56, 30 August 2016 (UTC)

It should. Having accessed an article in August 2015 doesn't help you if the page got updated on August 17th of that year. You the day needs to be specified for that reason, as recommended by every style guide out there. Bots can retrieve the date a link got added (with some fancy logic to rule out corner cases) to take care of most of those. Headbomb {talk / contribs / physics / books} 17:10, 30 August 2016 (UTC)

The templates should not generate error messages if the access date is earlier than the date the "feature" was added to the templates. Jc3s5h (talk) 17:30, 30 August 2016 (UTC)

@Headbomb In practice there is little point to the precise date to the day because in the example you gave let us suppose that the accessdate was 2 August 2015 and the page was updated on 17 August. Then August 2015 is sufficient to check for the nearest update in the archives of a page, because the day is normally irrelevant as one is usually lucky if there is more than a couple of archives a year for most web sites. As a practical example take the article Prussian Trust and the archive links Polish Foreign Affairs German Foreign Office, as by chance there is an additional access date on a third citation set to "December 2006" which is not strictly necessary as the date of the article is given as 18 December 2006, but to red flag the accessdate is I think unnecessary. This an interesting page to use as an exaple because the comments by the foreign office/affairs change relatively frequently so the only hope of capturing the text is via web archives. -- PBS (talk) 18:19, 30 August 2016 (UTC)

• Comment' Help talk:Citation Style 1/Archive 19#incomplete_access_dates: Two user accounts and an IP address should not be making changes to templates that end in a bot making mass changes to thousands of articles. As there are relativity few people changing the Module:Citation/CS1 because few know how to modify such code as few editors know Lua, those that do ought not to make changes like this without wide consultation first. There is far too cavalier attitude to making changes to the LULA functions that affect thousands of articles with next to no wider consultation. I think that changes like this to the CS1/CS2 templates ought not to be made unless there is a well attended RfC to agree the change, otherwise the status quo should prevail. -- PBS (talk) 18:36, 30 August 2016 (UTC)

The discussion was initially about year-only of which there were around 100 cases. That is reasonable. Somehow that morphed into requiring full dates without thinking through the scale of errors generated by Month Year cases. -- GreenC 21:40, 30 August 2016 (UTC)

• I'm impartial to whether or not they should be complete or not. I know I have a way of reasonably extrapolating the true access-date if we want a bot to do this, and no, it's not by putting a 1 as the day.—^cyberpower_{Chat:Limited Access} 19:08, 30 August 2016 (UTC)
• Comment—the dates should be complete to the day, and if they aren't, they should flag an error. Major style guides follow this best practice, and so should we. Imzadi 1979 → 22:47, 30 August 2016 (UTC)

Here is a bit of history. The current cs1|2 documentation defines |access-date=:

• access-date: Full date when the content pointed to by url was last verified to support the text in the article; do not wikilink; requires url; use the same format as other access and archive dates in the citations. Not required for linked documents that do not change. For example, access-date is not required for links to copies of published research papers accessed via DOI or a published book, but should be used for links to news articles on commercial websites (these can change from time to time, even if they are also published in a physical medium). Note that access-date is the date that the URL was checked to not just be working, but to support the assertion being cited (which the current version of the page may not do). Can be hidden or styled by registered editors. Alias: accessdate.

For this discussion, the important bit of that definition is the first two words. I wondered when the 'full date' requirement entered the documentation so I trolled through the template doc histories of the five most commonly used cs1|2 templates:

1. {{citation}}

   3 Jun 2007 – first documentation of access date does not require full date

   14 Jan 2012 – addition of {{Citation Style documentation/url}} requires full date

   12 Jan 2012 – first version of {{Citation Style documentation/url}} requires full date

2. {{cite web}}

   31 Aug 2006 – first template documentation requires full date

3. {{cite journal}}

   12 Sep 2006 – first template documentation requires full date

4. {{cite news}}

   16 Aug 2006 – first template documentation does not require full date

   7 Oct 2008 – documentation change requires full date

5. {{cite book}}

   9 Sep 2006 – first template documentation requires full date

—Trappist the monk (talk) 11:24, 31 August 2016 (UTC)

Free-to-read icon

In Module:Citation/CS1/Configuration/sandbox:135, I vertically aligned the green lock icon that appears after PMC identifiers () and added an entry to {{cite journal/testcases}}. The change is minor and straightforward, but given the propagation of the change, I’m checking in here first. Thank you. —LLarson (said & done) 18:42, 5 August 2016 (UTC)

I'm going to undo that part of your change. We did this comparison earlier at Help talk:Citation Style 1#adding free-to-read icons (I've modified that test here to use the same image as is used in the Module:

qK

and here's another:

[]

and with your change:

qK

[]

—Trappist the monk (talk) 19:08, 5 August 2016 (UTC)

Thanks as always, Trappist! What about this though?

Current:

qK

[]

Proposed:

qK

[]

—LLarson (said & done) 19:22, 5 August 2016 (UTC)

Looks to me like the {{open access}} lock is too high. It should be lowered so that it more-or-less spans the distance between the lowest descender and the highest ascender.

—Trappist the monk (talk) 20:36, 5 August 2016 (UTC)

I also find it more natural to have the main circle at the same height as a "o" letter (so, LLarson's proposal) − Pintoch (talk) 20:50, 5 August 2016 (UTC)

I'm on Team LLarson for height. Headbomb {talk / contribs / physics / books} 20:22, 9 August 2016 (UTC)

Quick question, but what is the difference in the meaning of the green "free to read" icon () and the orange "open access" icon ()? --bender235 (talk) 16:46, 24 August 2016 (UTC)

As I understand it, the orange icon is meant to identify sources that are free to reuse. The green only indicates that the source may be read without cost to the reader. Reuse rights are determined by agreement between the publisher and the author(s).

—Trappist the monk (talk) 16:51, 24 August 2016 (UTC)

Hmm, well if so, then the "open access" icon suggested for Newspapers.com citations is wrong, isn't it? --bender235 (talk) 17:50, 24 August 2016 (UTC)

I think you're right, that use of the orange icon there is inappropriate. I don't think that newspaper.com urls in a cs1|2 template need either of these icons because the norm for urls that link title-holding parameters in cs1|2 templates is that the linked source can be read without the reader jumping through registration or paywall hoops.

—Trappist the monk (talk) 19:00, 24 August 2016 (UTC)

The suggestion for Newspapers.com links is to use clippings which do not require registration. Others can read the clipping, but they'd need an account to see the rest of the page from which it originated. Imzadi 1979 → 21:14, 31 August 2016 (UTC)

Wadewitz memorial proposal

Recently a discussion at Talk:Jane Austen has revolved around the linchpin issue of template formatting. The original version of the article from 10 years had initially been started in MLA Handbook format (without templates, see below) by the late Wadewitz, and it was clearly her strong desire (as evidenced by her comment in the references section) that that MLA format be retained. However, a regrettable situation arose: since Wadewitz passed away, 5-6 editors have edited the article in different cite formats (or even "freestyle format", perhaps). Manual editing resulted in a massively inconsistent references section that I described as a "steaming mess of wrongness".

I strongly support the use of reference templates to create consistency in formatting within any given article, and to provide the many benefits of COinS.

However, neither of these advantages is available to any editors who wish to edit in MLA (or APA, or Chicago/Turabian, or Bluebook) as expressly supported under WP:CITEVAR.

WP:CITEVAR does not by extension mean that such template options must be created, but I suggest that the proper way to show courtesy and respect to all editors working within those extremely common (outside of Wikipedia) formats would be to create them. The reason these formats are rare in Wikipedia is precisely because template options for them are completely unavailable. If these template options had existed before, the "steaming mess of wrongness" at Jane Austen would never have come to be.

If possible, I propose (perhaps using |mode=) the creation of |style-guide=mla, |style-guide=apa, |style-guide=chicago and |style-guide=bluebook as alternate parameters that reformat displayed template text.   Lingzhi ♦ (talk) 09:33, 23 August 2016 (UTC)

The reason why MLA style is not suited to use in Wikipedia has been explained at length, by user:RexxS, in the discussion at Talk:Jane Austen; not least - in great detail - in Talk:Jane_Austen#MLA. Attempting to bypass that discussion here, and framing it in such a way as to imply that anyone opposing your proposal is both disrespecting other editors and disrespecting the memory of a much missed former colleague, is facile, and in the latter case is an abuse of her memory. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 11:44, 23 August 2016 (UTC)

@Lingzhi: Rather than wade through that lengthy talk page, I'd rather refer anyone to the essay presently at User:RexxS/NoCiteBar, where I've tried to collect my thoughts and explore the issues related to MLA-type and similar referencing schemes. That does rather damn MLA-type short cites as inadequate for our needs. If it's any consolation, I have been in the process of working through Module:Citation/CS1, with a view to creating a version that allowed a parameter to act as a switch for the displayed output. That, of course, would only affect the long citations, and would do nothing to fix the fundamental problems with MLA-type short citations. When I can get an uninterrupted stretch of time to do some programming (hint, hint), I'd hope to have something sandboxed that might go some way to meeting your request. Any pointers to definitive on-line references for the different styles, MLA, APA, Chicago, etc. would be appreciated. --RexxS (talk) 12:05, 23 August 2016 (UTC)

I think that having an MLA style option for the CS1 {{cite xxx}} templates is a good idea. It would allow editors to make use of the error checking that is available in CS1 templates but still format full citations in their discipline's preferred format. (As Andy says above, MLA style, e.g. "Smith, 101", is unsuitable for Wikipedia, where anyone can edit an article and easily introduce ambiguity. We would have to adopt a modified MLA style for short footnotes, e.g. "Smith 2005a, 101", linked to the full citation.) – Jonesey95 (talk) 13:23, 23 August 2016 (UTC)

I think a |mod=MLA wouldn't be a bad idea for full citations, but the existing {{harvnb}} templates should be be used for any short even though MLA itself doesn't call for the inclusion of the year there. Other mode options could be added in the future for other styles. Imzadi 1979 → 20:02, 23 August 2016 (UTC)

• We need short form MLA templates (and APA, and Chicago, and...) similar to sfn etc. Why oh why oh why do I have to come here and beg? Why do I have to kiss the ring of a small group of editors? Give me permission to edit templates, show me how you've implemented things in the years since I last edited a template, and I'll do it. This is nothing but flat WP:OWNership, and that's a fact.  Lingzhi ♦ (talk) 04:19, 24 August 2016 (UTC)

  A good way to get me to abandon this mla project is for you to continue to make unsupported accusations of WP:OWNership and the like:

  • Cite Book is a wikipedia-only standard that is in practice shoved down everyone's throats because the template maintainers flatly refuse to make MLA and APA templates. (you wrote that here)
  • I DO think that the template maintainers of cite book are massively remiss for flatly refusing to produce MLA, APA and Chicago flavors of cite book; they have an untouchable cast-iron WP:OWN on the issue, in flagrant violation of policy at WP:OWN. (you wrote that here)

  Show me where template maintainers or the maintainers of {{cite book}} have refused to make MLA and APA templates.

  —Trappist the monk (talk) 12:18, 24 August 2016 (UTC)

  @Trappist the monk: I admit to having become consistently embittered against those who maintain templates... I've been around Wikipedia off and on (with a nearly three year "off" period) for ten years. I have come to various template-related forums asking for various things, and have always been shown the door... even when I asked nicely (I ceased being so nice after a while). I don't recall when I asked last (either as User:Ling.Nut or User:Ling.Nut3) for these particular APA/MLA/Chicago template alterations, or what specific template-related forum I asked on, but I do recall asking for them, and I do recall being treated like an intruding idiot, to put it nicely. So my attitude is embittered, but it is a response to past arrogance. I do not recall the usernames of all of those who were rude in the past, but I am quite certain that you were not one of them. I do recall one user (still quite active) who treated me rudely in the dim past, but I will not repeat that username in this particular moment. So. Is this a non-apology apology? I'm not sure. But it is perhaps true that nicer people are around now than I have dealt with in the past, and if I have unfairly tossed you and others here into the bucket with the relatively unreasonable ones, then I do apologize for that. But having said that, I still feel bitter. So it's a bit of a dilemma. :/  Lingzhi ♦ (talk) 14:51, 24 August 2016 (UTC)

  Perhaps you are mis-remembering. I can find no discussions on template-related talk pages (the Template:, Module:, and Help: namespaces) where any of User:Lingzhi, User:Ling.Nut, or User:Ling.Nut3 asked for templates to write references in a style other than cs1|2. I did not look at user talk pages because those are not generally public venues and did not look into article or other namespace talk pages because those are clearly not template-related fora so perhaps you asked in one of those places. I am interested in knowing the reasons that were given when requests for alternate styles were denied.

  —Trappist the monk (talk) 16:14, 24 August 2016 (UTC)

(←) I also spent 10 or 15 minutes looking for that incident and have not found it yet either. I will continue looking. Meanwhile, I am in the present case obviously in the wrong, though (in my opinion) somewhat humanly so. But still in the wrong.  Lingzhi ♦ (talk) 02:35, 25 August 2016 (UTC)

• You don't need any special permission to create a new template or to edit a template's sandbox code. Go for it! And all template code is available via the Edit or View Source links at the top of the page. You might start at Module:Footnotes. – Jonesey95 (talk) 05:17, 24 August 2016 (UTC)

  @Lingzhi: Let's assume that you have a ref like <ref>Smith, 101</ref> and you want that linking to a long-form cite template. There are two things to do here, and neither of them is restricted by the edit protection on the templates.

  The first thing to do is to make a link inside that ref, this could be as simple as <ref>[[#Smith|Smith]], 101</ref> but as there may be a section in the article titled "Smith" it's a good idea to add something distinguishing, such as the letters "ref" - so you might use <ref>[[#refSmith|Smith]], 101</ref>.

  The second thing to do is to create an anchor on the long-form citation template, this might be {{citation}}, {{cite book}} or one of the others. Use the |ref= parameter for that, i.e. |ref=refSmith.

  You can see it in action at Wikipedia:Sandbox. You could make a template for the [[#refSmith|Smith]] part, but it's not essential. --Redrose64 (talk) 10:12, 24 August 2016 (UTC)

  @Lingzhi: Further to the above: I've found that you can in fact do it using existing templates, see Wikipedia:Sandbox. Here, we have <ref>{{harvnb|Smith|loc=101}}</ref> in the text, and |ref={{harvid|Smith}} in the cite template. The two templates used here are {{harvnb}} and {{harvid}}; notice that I didn't use the year parameter on either one, and with {{harvnb}} I used |loc= instead of |p= in order to suppress the "p." that would otherwise have been included. --Redrose64 (talk) 11:29, 24 August 2016 (UTC)

  @Redrose64: I think the short MLA can be finessed using existing |loc= in sfn, plus text for the year (since sfn treats everything as a string), plus and ref={{harvid}} except for the hard-coded comma in Module:Footnotes (in the args.location). See the very first cite in my sandbox for a fake example.  Lingzhi ♦ (talk) 12:15, 24 August 2016 (UTC)

For a long time I have wanted to restructure the code that assembles the miscellaneous constituent parts into a recognizable whole. Perhaps this is the burr under the saddle.

To help me understand what kind of changes are needed, I have hacked the sandbox so that |mode=mla controls a couple of the most obvious differences between cs1 and mla, date and editor placement and style, for {{cite book}}:

Author; Author2; Author3 (2016). "Chapter". In Editor; Editor2 (eds.). Title. Location: Publisher. pp. 12–34. {{cite book}}: |author= has generic name (help)CS1 maint: numeric names: authors list (link)

Author; Author2; Author3 (2016). "Chapter". In Editor; Editor2 (eds.). Title. Location: Publisher. pp. 12–34. {{cite book}}: |author= has generic name (help); Invalid |mode=mla (help)CS1 maint: numeric names: authors list (link)

—Trappist the monk (talk) 11:15, 24 August 2016 (UTC)

And last author / editor separator. —Trappist the monk (talk) 13:27, 24 August 2016 (UTC)

I don't think this is currently feasible in practice, but if you're looking at switches to control citation style, please consider the possibility of the displayed citation format being controlled by a user preference. At least half the clamour for, e.g., MLA is due to dissonance when reading a cite in an unfamiliar format; as opposed to that caused when having to enter a cite in an unfamiliar format (for which, those with MLA bred in their bones will dislike any template-based solution). Letting them choose to see all cites in MLA format would help a lot (and the right gizmo in VisualEditor might eventually help with the other issue). --Xover (talk) 14:46, 24 August 2016 (UTC)

Aye, that would be the thing. But, I think that for such a thing, we would need to compel all editors to write citations with templates. I think that it is a bit much to expect some bit of code at MediaWiki to read free-form contents of <ref>...</ref> tags and then correctly render that reference according to a user's Special:Preferences. It would also require the server to render the citations at the time they are served or to cache multiple versions of the rendered page.

—Trappist the monk (talk) 15:35, 24 August 2016 (UTC)

We might not be able to "compel all editors to write citations with templates". It would be progress to stop editors from blocking the conversion of (for want of a better description) "plain-text" references to use citation templates. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 16:05, 24 August 2016 (UTC)

@Lingzhi: The changes that Trappist the monk has made to Module:Citation/CS1/sandbox are producing something very close to what you want for long citations:

• {{cite book/new |last=Southam |first=Brian C |chapter=Grandison |editor-last=Grey |editor-first=J David |title=The Jane Austen companion |year=1986 |publisher=Macmillan |location=New York |isbn=0025455400 |mode=mla}}

produces:

• Southam, Brian C (1986). "Grandison". In Grey, J David (ed.). The Jane Austen companion. New York: Macmillan. ISBN 0025455400. {{cite book}}: Invalid |mode=mla (help)

and

• {{cite book/new |last=Tomalin |first=Claire |author-link=Claire Tomalin |title= Jane Austen: A Life |location=New York |publisher= Alfred A. Knopf |date=1997 |isbn=0-679-44628-1 |mode=mla}}

produces:

• Tomalin, Claire (1997). Jane Austen: A Life. New York: Alfred A. Knopf. ISBN 0-679-44628-1. {{cite book}}: Invalid |mode=mla (help)

Surely that is appreciable progress. Thank you very much, Trappist.

Before anybody starts thinking about "faking" MLA-style short citations, I'd like to hear your refutation of my arguments in User:RexxS/NoCiteBar against using <<Author Page>> or sometimes <<Author "made-up Short title">> short citations in Wikipedia articles. It is clear that authors regularly produce multiple works on the same topic, which require disambiguating via a user-constructed "short title" in MLA-style (e.g. about 70 times out of 146 short references in Jane Austen 18 February 2016), whereas it is equally clear that authors rarely produce multiple works on the same topic in the same year (none in Jane Austen today). For Wikipedia use, MLA-style short citations are a non-starter: different editors make different short titles, and adding another work by the same author requires all the previous <<Author page>> cites to be found and disambiguated. No, no, no, no (sounds familiar?). Harvard-style short cites (as already implemented in sfn) have the same format throughout, so any editor new to the page can copy the same style without effort by using the template {{sfn}}, and Ucucha's script can catch any errors that creep in. There's really no case for using anything else. --RexxS (talk) 21:21, 24 August 2016 (UTC)

First/last name order:

Last1, First1; Last2, First2; Last3, First3 (2016). "Chapter". In EditorLast1, EditorFirst1; EditorLast2, EditorFirst2 (eds.). Title. Location: Publisher. pp. 12–34.{{cite book}}: CS1 maint: numeric names: authors list (link)

Last1, First1; Last2, First2; Last3, First3 (2016). "Chapter". In EditorLast1, EditorFirst1; EditorLast2, EditorFirst2 (eds.). Title. Location: Publisher. pp. 12–34. {{cite book}}: Invalid |mode=mla (help)CS1 maint: numeric names: authors list (link)

—Trappist the monk (talk) 00:36, 25 August 2016 (UTC)

I have a few quick thoughts about MLA, and maybe these are items yet to come as this effort proceeds. If so, I apologize for jumping the gun.

• At least the more recent versions, the style specifically omits URLs and assumes that anyone can search for the source online. Of course we can link to a source within our citations because our formatting explicitly includes hyperlinks while MLA is designed for print. I assume we'd want to continue the practice of actually linking items, and probably continue to insert the file type as we do now.
• Also, MLA would append a medium designation, something like "Print" or "Web" to the end of the citation. For online sources, the "Web" would be followed by the access date. If it were a source reprinted online, that would be "Web. <republisher>. <access date>." Perhaps a little coding to assume that this would be "Print" unless a link is supplied when it would be "Web", but we'd need to either use |type= or another new parameter to allow editors to override this for DVDs and other media.

Imzadi 1979 → 01:42, 25 August 2016 (UTC)

I would have thought that the addition of a hyperlink when a source is available online is such a convenience for the reader that there's no case for failing to do that in our citations. MLA is indeed designed for printed work (including instructions on margins, double line spacing and the name of the Works Cited section), so we have to adapt to some extent. The whole point of offering an MLA-style long citation is so that readers accustomed to that style don't find it jarring when they see the citations in CS1 default style, not to precisely mimic a print format within an online encyclopedia. I can therefore see little point in appending "Web" or "Print" - that tells readers of the printed article that it's worth searching for online or not. In a Wikipedia article, it is surely obvious that if there's a hyperlink, it's available online, and if there's not, it isn't. --RexxS (talk) 02:11, 25 August 2016 (UTC)

It seems to me that it would be a bit disingenuous to add the capability to code a citation with |mode=mla and not actually output what MLA says is their citation format. That said, I've seen something that makes it seem as though they've dropped the medium indicator from the 8th edition published this year, so that might be something unique to the 7th edition that's now superseded. Imzadi 1979 → 15:22, 26 August 2016 (UTC)

Original year and edition:

Author; Author2; Author3 (2016) [1920]. "Chapter". In Editor; Editor2 (eds.). Title (2nd ed.). Location: Publisher. pp. 12–34. {{cite book}}: |author= has generic name (help)CS1 maint: numeric names: authors list (link)

Author; Author2; Author3 (2016) [1920]. "Chapter". In Editor; Editor2 (eds.). Title (2nd ed.). Location: Publisher. pp. 12–34. {{cite book}}: |author= has generic name (help); Invalid |mode=mla (help)CS1 maint: numeric names: authors list (link)

—Trappist the monk (talk) 09:48, 25 August 2016 (UTC)

Translators:

Author; Author2; Author3 (2016) [1920]. "Chapter". In Editor; Editor2 (eds.). Title. Translated by Translator; Translator2 (2nd ed.). Location: Publisher. pp. 12–34. {{cite book}}: |author= has generic name (help)CS1 maint: numeric names: authors list (link)

Author; Author2; Author3 (2016) [1920]. "Chapter". In Editor; Editor2 (eds.). Title. Translated by Translator; Translator2 (2nd ed.). Location: Publisher. pp. 12–34. {{cite book}}: |author= has generic name (help); Invalid |mode=mla (help)CS1 maint: numeric names: authors list (link)

—Trappist the monk (talk) 13:49, 25 August 2016 (UTC)

Contributors:

Contributor (2016) [1920]. Introduction. Title. By Author; Author2; Author3. Editor; Editor2 (eds.). Translated by Translator; Translator2 (2nd ed.). Location: Publisher. pp. 12–34. {{cite book}}: |author= has generic name (help)CS1 maint: numeric names: authors list (link)

Contributor (2016) [1920]. Introduction. Title. By Author; Author2; Author3. Editor; Editor2 (eds.). Translated by Translator; Translator2 (2nd ed.). Location: Publisher. pp. 12–34. {{cite book}}: |author= has generic name (help); Invalid |mode=mla (help)CS1 maint: numeric names: authors list (link)

Contributor (2016) [1920]. Introduction. Title. By Author; Author2; Author3. Editor; Editor2 (eds.). (2nd ed.). Location: Publisher. pp. 12–34. {{cite book}}: |author= has generic name (help); Invalid |mode=mla (help)CS1 maint: numeric names: authors list (link)

Contributor (2016) [1920]. Introduction. Title. By Author; Author2; Author3. Translated by Translator; Translator2 (2nd ed.). Location: Publisher. pp. 12–34. {{cite book}}: |author= has generic name (help); Invalid |mode=mla (help)CS1 maint: numeric names: authors list (link)

Contributor (2016) [1920]. Introduction. Title. By Author; Author2; Author3 (2nd ed.). Location: Publisher. pp. 12–34. {{cite book}}: |author= has generic name (help); Invalid |mode=mla (help)CS1 maint: numeric names: authors list (link)

Contributor (2016). Introduction. Title. By Author; Author2; Author3 (2nd ed.). Location: Publisher. pp. 12–34. {{cite book}}: |author= has generic name (help); Invalid |mode=mla (help)CS1 maint: numeric names: authors list (link)

Contributor (2016). Introduction. Title. By Author; Author2; Author3. Location: Publisher. pp. 12–34. {{cite book}}: |author= has generic name (help); Invalid |mode=mla (help)CS1 maint: numeric names: authors list (link)

—Trappist the monk (talk) 11:07, 1 September 2016 (UTC)

Break

@Trappist the monk, RexxS, and Lingzhi: I know nothing about templates and haven't read the above, so perhaps this isn't feasible, but I would love to see those templates converted so that we could simply place the punctuation where we needed it. At the moment, I write manually:

• John Rawls, A Theory of Justice, Cambridge: Harvard University Press, 1971, 1.
• and thereafter: Rawls 1971, 2.
• Chantal Zabus, "The Excised Body in African Texts and Contexts," in Merete Falck Borch (ed.), Bodies and Voices: The Force-field of Representation and Discourse in Colonial and Postcolonial Studies, New York: Rodopi, 2008, 1.
• and thereafter: Zabus 2008, 2.
• Nicky Woolf, "Stingray documents offer rare insight into police and FBI surveillance", The Guardian, 26 August 2016.

When I'm editing with people who use templates, I can copy their style manually, even though it leads to internal inconsistency with newspapers (date after name if there's a byline, and date at the end if not; it's frustrating to add an inconsistency deliberately just to copy the template style). But if people edit an article where I've added manual cites in the style above, they can't copy my style (or any other) using templates; they have to do it manually because the templates are so limiting.

Is it not possible to create a set of templates where the punctuation is more flexible? SarahSV ^(talk) 00:07, 27 August 2016 (UTC)

Is it really punctuation that you're talking about or is it element placement or a combination of both? Your apparently preferred style looks to be a combination of cs2 (comma separated elements) and mla (publication date moves to the end):

• John Rawls, A Theory of Justice, Cambridge: Harvard University Press, 1971, 1. – your original
• John Rawls (1971), A Theory of Justice, Cambridge: Harvard University Press, 1. {{cite book}}: Unknown parameter |nopp= ignored (|no-pp= suggested) (help) – {{cite book}} with |mode=cs2
• John Rawls (1971). A Theory of Justice. Cambridge: Harvard University Press. 1. {{cite book}}: Invalid |mode=mla (help); Unknown parameter |nopp= ignored (|no-pp= suggested) (help)CS1 maint: postscript (link) – {{cite book}} (sandbox) with |mode=mla

A primary purpose of the templates is to take on the burden of punctuation so that editors don't have to worry about it – punctuation just happens and it is the same, citation-to-citation, editor-to-editor. One could, I suppose, create punctuation-specific parameters:

|title= – normal or standard title parameter follows the cs1|2 rules according to the rules of the enclosing template

|title,= – override the normal cs1 element separator, full stop, with a comma

|qtitle= – force a normally italic title to be quoted

|ititle= – force a normally quoted title to be italic

|ptitle= – force a normally italic or quote title to be plain text

One could extend this idea and enumerate the parameters so that they would render in order specified in the template:

{{cite book |1author,=John Rawls |2title,=A Theory of Justice |3location=Cambridge |4publisher,=Harvard University Press |5date,=1971 |6page.=1 |nopp=yes}}

I don't think that either of these ideas should be pursued. Better, I think is to support a handful of commonly used and well documented styles – mla, apa, cms.

The above discussion that you didn't read is mostly about implementing mla in the cs1|2 templates. I have adapted {{cite book}} so that it can render basic mla style. I intend to similarly adapt {{cite journal}}, {{cite news}}, and {{cite web}} and then when real life gets out of the way, take what I've learned from this experiment and rewrite a large chunk of Module:Citation/CS1 so that other similarly documented styles can be supported.

—Trappist the monk (talk) 10:38, 27 August 2016 (UTC)

Trappist the monk, thank you for the detailed response. The style I use is close to Chicago, but I've pared it down to all commas, no brackets, to keep it simple. I usually use the long cite on first reference in the text and the short thereafter, and usually no separate bibliography section.

Can cite news at least be changed so that the dates don't move? We had an RfC that supported the change, but it was never implemented. SarahSV ^(talk) 16:16, 27 August 2016 (UTC)

Above I mentioned a rewrite of a large chunk of Module:Citation/CS1. The difficulty that Editor Dragons flight mentioned in the RfC still exists. I would expect that at the end of the rewrite, it will be easier to position the publication date as the second element in the rendered citation.

—Trappist the monk (talk) 09:50, 28 August 2016 (UTC)

@Trappist the monk: I strongly disagree with both creating punctuation-specific and enumerating parameters as this would make total mess over months/years and references would be not unified/uniform but – mess. I'm even surprised that you proposed this as goal of CS1 was to standardize – not to allow "citeshakes"?

Moreover, I thought you would simply ignore SarahSV's comment as this user is not using CS1 and says ... they can't copy my style (or any other) using templates; they have to do it manually because the templates are so limiting. – I can't even comment on this one.

@SlimVirgin: Templates and CS1 are not "so limiting" but a perfect solution not to disable people from "copying each other styles" but to introduce one style for all references that can be altered in future if there is need to, because citation style and Wikipedia style generally should be uniform and consistent with guidelines; if every editor would cite in a different way – "chaos mode" would be turned on in references section.--Obsuser (talk) 23:07, 27 August 2016 (UTC)

You must have missed this line in that post: I don't think that either of these ideas should be pursued. It is the prerogative and the responsibility of the cs1|2 templates to handle formatting details. I do not think that we should be changing that. What I wrote was merely a thought experiment.

When an editor specifically asks a question of me, as was done here, common courtesy requires me to respond. No one benefits if questions that can be answered are left unanswered.

—Trappist the monk (talk) 09:50, 28 August 2016 (UTC)

Just for the record, I also "strongly disagree with both creating punctuation-specific and enumerating parameters" for similar reasons. We already have a frequent problem with people abusing the |publisher= parameter to "get their way or else" in their WP:GREATWRONGS campaign against ever italicizing the name of a cited website, for example. This nonsense battlegrounding has to stop, not be given a whole new navy of battleships with which to engage in style-micromanagement editwarring. I also agree with the points above about the benefits of a consistent citation style. I don't think we should be doing any work at all to integrate MLA, ALA, AMA, etc. citation styles into CS1 or any template system; it's a non-productive waste of editorial time to do it, and an much worse waste of many editors' combined editorial time to deal with the mess it creates, and, worst of all, all the "don't you dare touch my precious citation formatting" drama festivals it generates. If we just ignore these off-WP citation styles, at some point the sheer pressure of more and more articles being done in default CS1 style, and more and more articles converted to it without objection, will result (finally) in consensus that WP should have a consistent citation style, not permit every imaginable, inconsistent one. That may take 5 years, or 10. Even if it never happened, what has to go – no doubt – is the nonsense WP:LOCALCONSENSUS over at WP:CITE that any made-up on the spot out of your own imagination citation "style", not found in any source anywhere, can be ruthlessly enforced at an article as long as you made it the consistent "style" at the article at some point, no matter how irrational and confusing the "style" is. WP:CITEVAR was intended to permit people professionally steeped in a particular, real-world citation style, like AMA or MHRA or Vancouver, to use it here without hissy-fits erupting about it (and this has proven to be a mistake). Whether or not that was a good idea, it's since then been totally hijacked to permit idiosyncratic WP:OWN chaos that has to go. This will probably require a WP:VPPOL RfC, since prior attempts to deal with it at WT:CITE itself have met with tagteam stonewalling.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  10:10, 28 August 2016 (UTC)

mla in cite journal

Adapting {{cite journal}} to use mla:

Author; Author2; Author3 (2016). "Title". Journal. Volume (Issue): 12–34. {{cite journal}}: |author= has generic name (help); |issue= has extra text (help); |volume= has extra text (help)CS1 maint: numeric names: authors list (link)

Author; Author2; Author3 (2016). "Title". Journal. Volume (Issue): 12–34. {{cite journal}}: |author= has generic name (help); |issue= has extra text (help); |volume= has extra text (help); Invalid |mode=mla (help)CS1 maint: numeric names: authors list (link)

—Trappist the monk (talk) 12:12, 27 August 2016 (UTC)

If I make a tangential comment, but seeing how this is formatted above MLA style, reminds me of a minor request related to {{cite magazine}}. That template is capitalizing "Vol." when I think it really should be lowercasing it as your MLA example does. Maybe a minor change for a future module update? Imzadi 1979 → 14:01, 29 August 2016 (UTC)

Compare these; one is {{cite magazine}} (cs1) and the other is {{citation}} (cs2):

"Title". Magazine. Vol. 23, no. 4. Archived from the original on 2016-08-29. Retrieved 2016-08-29.

"Title", Magazine, vol. 23, no. 4, archived from the original on 2016-08-29, retrieved 2016-08-29

The 'style' defined for cs1 is to use full stops between citation elements compared to commas for cs2. Because each citation element in cs1 is essentially a new sentence, the element's static text is capitalized. Because cs2 is a single sentence, the static text is not capitalized.

—Trappist the monk (talk) 15:04, 29 August 2016 (UTC)

Except that "p." and "pp." aren't capitalized when used, so I don't buy the "separate" sentence argument. Why isn't the volume ended by a period when issue and page number are?

"Title". Magazine. Vol. 23, no. 4. p. 6. Archived from the original on 2016-08-29. Retrieved 2016-08-29.

I think it would be better to be a bit more consistent here, and oddly MLA shows a bit of a better way forward. Imzadi 1979 → 15:36, 29 August 2016 (UTC)

Are p. and pp. ever capitalized outside of Wikipedia? I don't think that those two have ever been capitalized in cs1|2. 'Vol.' has only recently been made available. The thing that I notice about your example is the separator character following the issue number. If there isn't a separator character between volume and issue, should there be a separator character between issue and page?

—Trappist the monk (talk) 10:03, 30 August 2016 (UTC)

mla in cite news

Adapting {{cite news}} to use mla:

Author; Author2; Author3 (3 September 2016). "Title". Newspaper. Vol. Volume, no. Issue. pp. D12+. {{cite news}}: |author= has generic name (help); |issue= has extra text (help); |volume= has extra text (help)CS1 maint: numeric names: authors list (link) – cs1 live

Author; Author2; Author3 (3 September 2016). "Title". Newspaper. Vol. Volume, no. Issue. pp. D12+. {{cite news}}: |author= has generic name (help); |issue= has extra text (help); |volume= has extra text (help)CS1 maint: numeric names: authors list (link) – cs1 sandbox

Author; Author2; Author3 (3 September 2016). "Title". Newspaper. Vol. Volume, no. Issue. pp. D12+. {{cite news}}: |author= has generic name (help); |issue= has extra text (help); |volume= has extra text (help); Invalid |mode=mla (help)CS1 maint: numeric names: authors list (link)

—Trappist the monk (talk) 10:33, 3 September 2016 (UTC)

I didn't get it quite right. When the mla version of {{cite news}} (and {{cite journal}} and {{cite web}}) did not have |author=, the title would not display. That has been remedied.

without author; with editor

Editor; Editor2; Editor3, eds. (3 September 2016). "Title". Newspaper. Vol. Volume, no. Issue. pp. D12+. {{cite news}}: |editor= has generic name (help); |issue= has extra text (help); |volume= has extra text (help)CS1 maint: numeric names: editors list (link) – cs1 live

Editor; Editor2; Editor3, eds. (3 September 2016). "Title". Newspaper. Vol. Volume, no. Issue. pp. D12+. {{cite news}}: |editor= has generic name (help); |issue= has extra text (help); |volume= has extra text (help)CS1 maint: numeric names: editors list (link) – cs1 sandbox

Editor; Editor2; Editor3, eds. (3 September 2016). "Title". Newspaper. Vol. Volume, no. Issue. pp. D12+. {{cite news}}: |editor= has generic name (help); |issue= has extra text (help); |volume= has extra text (help); Invalid |mode=mla (help)CS1 maint: numeric names: editors list (link)

without author or editor

"Title". Newspaper. Vol. Volume, no. Issue. 3 September 2016. pp. D12+. {{cite news}}: |issue= has extra text (help); |volume= has extra text (help) – cs1 live

"Title". Newspaper. Vol. Volume, no. Issue. 3 September 2016. pp. D12+. {{cite news}}: |issue= has extra text (help); |volume= has extra text (help) – cs1 sandbox

"Title". Newspaper. Vol. Volume, no. Issue. 3 September 2016. pp. D12+. {{cite news}}: |issue= has extra text (help); |volume= has extra text (help); Invalid |mode=mla (help)

—Trappist the monk (talk) 11:02, 8 September 2016 (UTC)

Edited collections and chapters

Is there a difference between edited collections and books with chapters with different authors? Template:Cite book says: "For edited collections, use {{cite encyclopedia}}". But 'cite book' also gives the option for 'Citing a chapter in a book with different authors for different chapters and an editor'. This seems identical to 'cite encyclopedia'. Is there a difference? Some collections are arranged in chapters, some are not. Is that the main difference? That non-chapter collections should use 'cite encyclopedia' and collections arranged in chapters can use either? (In passing, people do sometimes use 'contribution' for chapters in a collection, when 'contribution' should really be for "afterword, foreword, introduction, or preface"). Carcharoth (talk) 12:06, 23 August 2016 (UTC)

Everything makes sense when you realize that the doc is actually a story by Lewis Carrol. 108.55.199.227 (talk) 12:43, 23 August 2016 (UTC)

Am I supposed to find a queen somewhere on the documentation page? Clearly I'm looking in the wrong places. --Izno (talk) 13:08, 23 August 2016 (UTC)

I'm not sure where the use-{{cite encyclopedia}}-for-edited-collections notion comes from. Perhaps it is a documentation artifact from the early days of the cs1 templates when those templates were much less capable than they are today. Scholarly works on some topic are often edited collections of writings by many authors compiled into a book of chapters. I see no reason to prefer {{cite encyclopedia}} over {{cite book}} for these works.

There is no requirement that states that |contribution= is specifically limited to afterword, foreword, introduction, and preface. The requirement that I think you allude to is for |contributor= which requires |contribution=. The contribution can be anything but is typically one of the afterword, foreword, etc.

—Trappist the monk (talk) 13:18, 23 August 2016 (UTC)

Puzzling. I use 'cite encyclopedia' all the time. Quoting from {{cite encyclopedia}}:

"This Citation Style 1 template is used to create citations for articles or chapters in edited collections such as encyclopedias and dictionaries, but more generally any book or book series containing individual sections or chapters written by various authors, and put together by one or more editors."

Example of my use of 'cite encyclopedia': [1], [2]. Another example is the Dictionary of Scottish Architects, which could be cited as a collection (more strictly, it is an online database). Many of these online collections would be encyclopedias if printed. Carcharoth (talk) 13:39, 23 August 2016 (UTC)

Certainly your first example could use {{cite book}}. Your second could use {{cite web}} especially if the online version is the source that you consulted:

{{cite web|last=Brody|first=David|author-link=David Brody|title=Meany, George|url=http://www.anb.org/articles/15/15-01098.html|website=American National Biography Online|publisher=Oxford University Press and American Council of Learned Societies|subscription=yes|date=February 2000|access-date=12 August 2016}}

Brody, David (February 2000). "Meany, George". American National Biography Online. Oxford University Press and American Council of Learned Societies. Retrieved 12 August 2016. {{cite web}}: Unknown parameter |subscription= ignored (|url-access= suggested) (help)

—Trappist the monk (talk) 14:16, 23 August 2016 (UTC)

I don't think that any editor actually makes use of the information contained in the name of the cite template. {{Cite encyclopedia}} treats the container |encyclopedia= rather like MLA's style of using quotes, while {{cite book}} treats it as an alias for |work= so it is italicised. Chacun à son goût.

• Cavell, Richard (2015). "Remembering Canada: The Politics of Cultural Memory". In Sugars, Cynthia (ed.). The Oxford Handbook of Canadian Literature. Oxford University Press. pp. 64–79. ISBN 9780199941865. - cite encyclopedia
• Cavell, Richard (2015). Sugars, Cynthia (ed.). Remembering Canada: The Politics of Cultural Memory. Oxford University Press. pp. 64–79. ISBN 9780199941865. {{cite book}}: Unknown parameter |encyclopedia= ignored (help) - cite book with |encyclopedia=
• Cavell, Richard (2015). Sugars, Cynthia (ed.). Remembering Canada: The Politics of Cultural Memory. Oxford University Press. pp. 64–79. ISBN 9780199941865. {{cite book}}: |work= ignored (help) - cite book with |work=

--RexxS (talk) 14:27, 23 August 2016 (UTC)

I don't think a chapter in a book is supposed to be rendered in italics. The citation immediately above should be rendered as:

• Cavell, Richard (2015). "Remembering Canada: The Politics of Cultural Memory". In Sugars, Cynthia (ed.). The Oxford Handbook of Canadian Literature. Oxford University Press. pp. 64–79. ISBN 9780199941865. - cite book with |chapter=

– Jonesey95 (talk) 14:37, 23 August 2016 (UTC)

Jonesy is right. 'Title' in cite encyclopedia is equivalent to 'chapter' in cite book. This can confuse people. Carcharoth (talk) 15:10, 23 August 2016 (UTC)

Quite. I'm sorry I wasn't clearer: I was trying to show what would happen if you were to use |encyclopedia= in {{cite book}}, not advocate its use! There is a case you could make for using {{cite book}} for all works written on paper with covers on the front and back. You just learn one set of parameters and the output they produce. Personally, I don't care if chapter titles are in quotes or offset in superscripted Comic Sans with a pink shadow. As long as we all agree on something recognisable and use it consistently. --RexxS ^{Call me Dino} (talk) 16:39, 23 August 2016 (UTC)

I realise they produce the same output. It is more a case of when I look at a source, I think to myself "what type of source is this?". If it is a printed book, that is obvious. If it is is a collection, I tend to reach for 'cite encylopedia'. If it is a news publication, then 'cite news'. When it is a webpage, I know I should really use 'cite web', but there is a distinction in my mind between webpages with no named author and no obvious date on a website with no clear structure, and more organised online websites clearly intended to replicate the organisation of a collection of entries under a named work (sometimes with a named author, sometimes not). The DSA and ANB are examples of the latter, while an example of the former would be this. In those cases, it is sometimes only really possible to identity the title of the webpage and the publishing organisation (usually obvious from the website home page). The authority (i.e. 'reliability') of a citation with no named author rests on the reputation of the publisher. I'm not consistent, though, as citations to various online databases I do often format using cite web. What I think people struggle with is the idea that the webpage is the thing with the 'title' and the 'website' is the 'work', when in the case of a book, the 'work' is the 'title' and bits within it (the webpages in the case of the website) are pages or chapters and so on. What I am saying is that in some cases, it is obvious that the website as a whole is the 'work', but for other websites, they are more a collection of pages that sometimes bear little relation to each other except for being on the same domain name. This is a well-defined website, but this is more a loose grouping of pages published by the same organisation. Does that make sense of the difference I'm trying to describe? Carcharoth (talk) 15:04, 23 August 2016 (UTC)

The documentation is inconsistent and confusing, but sometimes this results from design flaws, such as the dual use of |title= as remarked. Not that this hasn't come up before, it has, several times in the past 5 years (or more). Fixing these flaws (there are several) should be the first order of business. In my mind, this is much more pressing than adding nice little icons, or making sure that machines can exchange metadata, or making the native citation system comfortable for users of other systems. A moratorium on any new features until a clean easy-to-use design emerges, would not be a bad idea imo. Otherwise the rabbit hole may keep getting larger. 72.43.99.146 (talk) 23:31, 23 August 2016 (UTC)

Agree the docs are contradictory and this should be resolved. I would advocate re-documenting {{Cite encyclopedia}} as for being for tertiary sources composed of numerous encyclopedia- or dictionary-style entries (whether it's a multi-author work or not is irrelevant, as a large number, maybe even a majority, of topical specialist encyclopedias and dictionaries are single-author). For my part, any time I'm citing a multi-author academic book composed of papers by authors, presented as chapters, with a volume editor, I use {{Cite book}} and treat the paper/chapter as |chapter=. I one case I ran into trouble because the paper itself had chapters, and I dealt with this with |at= to site both the subchapter and the page number, though technically it was overkill since a page number was probably sufficient. I have not monkeyed with |contribution= much, so I'm not sure how nice it plays with |chapter=; I've always reserved it for forewords, afterwords, and introductions.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  09:40, 28 August 2016 (UTC)

Url validation

@Trappist the monk: and others: Why is http://нижнийновгород.рф/references/inter/tampere.html not recognized as good url? I found it on one page and error is raised. However, if I copy it in Chrome and paste it – I get form converted into standard latin characters: http://xn--b1acdfjbh2acclca1a.xn--p1ai/references/inter/tampere.html. --Obsuser (talk) 10:17, 25 August 2016 (UTC)

Because the domain name is not written using the Latin character set. When you paste that url into Chome, it translates it into an internationalized domain name so that the internet infrastructure can understand it. Module:Citation/CS1 uses the standardized rules for domain name validation which requires domain names to be in the Latin character set.

—Trappist the monk (talk) 11:23, 25 August 2016 (UTC)

Using Firefox, I don't find any difference between the two types of url: both work well. --Mauro Lanari (talk) 11:44, 25 August 2016 (UTC)

The translation is still done; firefox apparently doesn't show that it is done.

—Trappist the monk (talk) 11:48, 25 August 2016 (UTC)

We have a magic word, {{urlencode:}}, that could be used to wrap URLs like that provided by the OP. Can this be built into CS1's Lua module? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 17:30, 30 August 2016 (UTC)

Different kind of encoding. Percent-encoding is not ASCII Compatible Encoding. Here is what the magic word produces:

http%3A%2F%2F%D0%BD%D0%B8%D0%B6%D0%BD%D0%B8%D0%B9%D0%BD%D0%BE%D0%B2%D0%B3%D0%BE%D1%80%D0%BE%D0%B4.%D1%80%D1%84%2Freferences%2Finter%2Ftampere.html

—Trappist the monk (talk) 17:59, 30 August 2016 (UTC)

Is there some other template that will do the job, then? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 18:30, 30 August 2016 (UTC)

Sections of websites

Is there a correct way to link to a section of a website? Which of these three is best?

• [3] put both section and title information in the 'title' parameter
• [4] put the section information outside the citation template
• [5] use the 'at' parameter to describe the section

I should provide the full examples here, but am not sure how to do that. Carcharoth (talk) 18:20, 25 August 2016 (UTC)

If the destination website provided an anchor to a section, then I'd use:

• {{cite web |url=http://venn.lib.cam.ac.uk/Documents/acad/lists/P.html#Aeronautical_Engineering,_Francis_Mond_Professorship_of |title=Francis Mond Professor of Aeronautical Engineering |website=A Cambridge Alumni Database |publisher=University of Cambridge |accessdate=15 September 2013}}
• "Francis Mond Professor of Aeronautical Engineering". A Cambridge Alumni Database. University of Cambridge. Retrieved 15 September 2013.

On the grounds that nobody actually cares what the title of the container is (i.e. page title) when they have the title of the section linked via the fragment in the url. But that's just me. As for the actual venn.lib.cam.ac.uk site, they haven't realised yet that the HTML5 spec at https://www.w3.org/TR/html-markup/a.html#a-constraints has now made obsolete the "name" attribute that the web-designers are using to create an anchor for each section - which is why the above doesn't actually work. Even dinosaurs like me have caught on to that. --RexxS (talk) 20:21, 25 August 2016 (UTC)

The spec that I normally refer to shows that name= is obsolete, but words it as "should not", which is weaker than "must not". It then describes how the name= attribute can be used in such a way that it won't conflict with other name= or id= attributes. Later on we find "The following attributes are obsolete (though the elements are still part of the language), and must not be used by authors ... name on a elements (except as noted in the previous section)". Here we have the strong "must not", but conditionally weakened. --Redrose64 (talk) 22:47, 25 August 2016 (UTC)

I would use the |at= solution here. The above suggestion of |title=Francis Mond Professor of Aeronautical Engineering is a made-up pseudo-title not appearing in the work, thus making it harder to find and verify the exact source material. I encounter this problem frequently, and fix it, of people trying to use the |title= parameter to describe or retitle a work because they think the actual title isn't clear enough or something. E.g., often I'll seen dictionary entries done as |title='anthropogenesis' entry, and this is simply incorrect; the correct value is |title=anthropogenesis. Don't second-guess the title even if you don't like it.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  09:31, 28 August 2016 (UTC)

Title2 or AltTitle

Is it possible to incorporate something into template:cite web to reflect when a piece has more than one title?

For example as you can see at http://web.archive.org/web/20160729014127/http://www.macleans.ca/news/justice-for-black-canadians-not/ the original title of the July 28 piece by Domise is "Justice for Black Canadians (not)" but the present version of the article at http://www.macleans.ca/news/justice-for-black-canadians-not/ has changed the title to "Why we can’t trust SIU to probe death of Abdirahman Abdi" even though the URL still reflects the old title.

I would like a way to convey in the citation both the original title of the piece and then the replacement title for the piece.

To give another example of what I mean, take this piece by Stephanie McCrummen in the Washington Post:

• 22 March 2005 http://www.washingtonpost.com/wp-srv/metro/specials/homefront/khan_article.html "Looking for Logic Amid the Pain: Grieving Father Struggles to Understand"
• 28 July 2016 https://www.washingtonpost.com/national/khizr-khans-loss-a-grieving-father-of-a-soldier-struggles-to-understand/2016/07/28/18e8139a-552d-11e6-bbf5-957ad17b4385_story.html "Khizr Khan's loss: A grieving father of a soldier struggles to understand"

It's the same piece, but was given a new title. Ranze (talk) 06:32, 26 August 2016 (UTC)

I would use |orig-year=originally published 22 March 2005 as "Looking for Logic amid the Pain". 64.134.66.58 (talk) 11:43, 26 August 2016 (UTC)

I would not advise what the IP above suggests. In this case, I would only use the title for the version of the article that was consulted and cited. Yes, that could mean a mismatch between the title and the version of it repeated within the URL, but so be it. If necessary, it may be possible to additionally cite an archived copy of the old version of the article under the old title, but I would not try to combine what are really separate sources together. Imzadi 1979 → 15:19, 26 August 2016 (UTC)

I disagree. This is for all practical purposes a reprint, except for the article's title. For purposes of verification, nothing changed materially, and discovery of the source has not been affected. It is convenient to provide prior history for the title, but this is not necessary in this case. Why make it any more complicated? It would be a different story if this was a different title in a different edition. 72.43.99.138 (talk) 15:32, 26 August 2016 (UTC)

When you add the content to the article, with a ref, in that ref you should use the title as it is shown at that time, the date as it is shown, and the accessdate is today's date. If the title changes tomorrow, leave it alone together with the date, accessdate and anything else. --Redrose64 (talk) 19:20, 26 August 2016 (UTC)

If the title changes, the citation should be edited because this affects discovery of the source, and/or may confuse readers. This may be more pertinent when sources are online (eventually, all of them may be). Something should promptly alert the reader that they are looking at the right source, albeit with a different title. The use of |orig-year= was suggested in cases where the new title is cited, in order to provide title history as a convenience. 72.43.99.138 (talk) 14:47, 27 August 2016 (UTC)

If you went and looked (how else could you know the title changed?), then you have accessed it again, so should update the title to the current one and change the access-date. The purpose of the access-date parameter is to mark the date that something was most recently verified, not to freeze in time when a source was first added. In complex situations involving offline sources, you can just add a note after the citation template and before the closing </ref>. Another thing I've done when a work as two titles in different markets and they're they same down the page numbers, just with different covers and nominal publishers, is to do <ref>{{Cite book|...}} Also published as: {{Cite book|...}}</ref> There is no reason that every possible detail must be fitted into a single citation template.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  09:21, 28 August 2016 (UTC)

Adding a "disable italics on work" parameter?

I know there's a history about the use of italics for websites or not (which comes from MOS:TITLE) and whether to use the work= or publisher= parameter to achieve the right effect. We're having a discussion on WT:VG for websites like Rotten Tomatoes and Metacritic which seem to have, in common practice around WP, to not be italicized as websites (as they aren't creative work websites but more as services). Past discussion on CS1 suggests they should be entered as publisher=, but this is not true as they have a separate publishing entity.

The only trick with CS1 that can make these appear non-italic in citations is to italic the name in the template entry , but this has this data passed into wikidata so it is undesirable.

Is it possible if we could get a field like "workni=y" or "wni=y" added to CS1 that, if set, disables the italics on the work= entry, so that we respect the general styling these types of sites have in prose, while also respecting how they are entered as wikidata? If the field is not set (eg applying to all existing templates), then there should be no change in behavior, the work entry is still italicized.

This type of change probably would affect less than 1% of the citations out there, while also making these templates consistent with the openness of MOS:TITLE which does not fully prescribe how to handle websites. --MASEM (t) 16:26, 27 August 2016 (UTC)

Oppose. We need to put to bed any further coddling of people who want to force-format things to suit their personal peccadilloes. People are just going to have to live with the fact that WP is not their personal blog, and has it's own style guide. Every citation style in the world is different in minor details, zero of them satisfy 100% of the people, and probably no one is 100% satisfied with every nit-pick any any of them. People just get over it. If someone is abusing the |publisher= parameter for the |website= title just to force it to not be italic, this is an error and should be reverted.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  06:09, 28 August 2016 (UTC)

I looked over that thread, and the entire thing is just because some people do not think clearly about the difference between a publication, a [re]distribution, and a publishing company, nor bother to figure out what the template parameters are. There is no use case at all in which we could need some |workni=. I've laid this out as a mini-tutorial at Wikipedia talk:WikiProject Video games#This really isn't difficult once you sort it out.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  09:09, 28 August 2016 (UTC)

The problem is that CS1 is forcing a choice that MOS:TITLE does not fully define. MOS:TITLE only speaks to using italics to a specific type of website but does not describes for all websites, and practice (which defines policy not the reverse) clearly shows that we don't italicize certain types of websites in running prose. So CS1 should be flexible to the openness that MOS:TITLE has, and not force a style choice nor encourage poor use of parameters to make a style choice work. --MASEM (t) 14:40, 28 August 2016 (UTC)

??? CS1 is a style, and it is optional. Styles must be internally consistent. Templated styles like CS1 must be standardized. In this optional, templated style, |work= is always italicized. Unless you want to change the font-style rendering of the parameter (a different discussion, imo) I suggest using a different citation style, maybe a non-templated one. 72.43.99.138 (talk) 15:35, 28 August 2016 (UTC)

New categories

I have just created:

• Category:Templates that wrap Cite book
• Category:Templates that wrap Cite encyclopedia
• Category:Templates that wrap Cite journal
• Category:Templates that wrap Cite map
• Category:Templates that wrap Cite news
• Category:Templates that wrap Cite web

These are applied automatically, by {{cs1 wrapper}}. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 17:45, 27 August 2016 (UTC)

Cite interview title and italics

I searched in the CS1 archives regarding {{cite interview}} and couldn't find a reason for why cite interview italicizes its title.

I couldn't find anything in our Manual regarding the title of an interview. MLA and Chicago recommend (not official links!) to place the title of a published interview in quotation marks, while APA doesn't seem to have a recommendation (presumably because they consider published interviews to take the format of the publication type in question--official blog from 2009--APA then seemed to have the practice of italicizing their titles regardless, so maybe it's not worth considering).

Would there be support for changing how interview titles are cited to use quotation marks instead of italics? (I did note Help talk:Citation Style 1/Archive 5#The "title" parameter in the Cite interview, which seems to work around a local issue and which happens to use the italics as a feature, not a bug, but as cite interview is only used about 3k times, I doubt that we couldn't fix them locally as necessary.) --Izno (talk) 22:46, 27 August 2016 (UTC)

Yeah, should be quotation marks. The major work in which the interview is published gets the italics.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  09:14, 28 August 2016 (UTC)

{{cite interview}} is one of those templates that needs a rewrite. To make it work with {{citation/core}}, |program=, |callsign=, and |city= are all concatenated into the meta-parameter ID. For some reason, |type=, which is media descriptor, is used to modify |interviewer(s)= (an alias of |others=). When I converted the template to use Module:Citation/CS1, I retained these peculiarities because the task at the time was to faithfully reproduce {{citation/core}}.

—Trappist the monk (talk) 11:09, 2 September 2016 (UTC)

I'm happy to have a discussion actually about remapping/rewriting for cite interview, but I suspect the change needed for |title= would be simple.

Regarding the other stuff in your comment, the template talk page's history may be an interesting view into template's history and may explain some of the peculiarities.

• Type looks to be explained: it looks like the name just overlaps and so cleanup is needed for it. We might want to do a survey (add a category) of the pages using type in cite interview.
• Program should probably alias |work=; usage seems to indicate that the field might need some cleanup.

City should probably map to |location= or |publication-place=. |callsign= mapping to ID doesn't seem like a bad choice, but it shouldn't override ID in case another is identified. In our current example at Template:Cite interview, the call sign seems to be an alias for the |publisher= or possibly the republisher, as in the second example (which would go in |via=). NPR for some reason is the program, which doesn't make sense to me. Maybe call-sign should wait for another day. :D

Another personal peculiarity: I find the mapping of interviewee to author peculiar, but not interviewer to author--ostensibly, they should share credit, because the interviewer thought of and published the questions asked. Of course, now we have |contributor=, if we don't think the interviewer should share credit... just some thoughts.

To get around the issue in the H:CS1 archive, I might suggest the following order regarding the URL: title, work/program, and if neither are available (unlikely--the interview may as well be unpublished at that point!) the text "interview with interviewer" should probably get the url. Of course, |work= can usually take a wikilink, so I'm not sure about that suggestion... So many cut out linking work, and simply: if title is missing, then link the text "interview with" or possibly "interview". --Izno (talk) 11:41, 2 September 2016 (UTC)

This search suggests that there aren't too many {{cite interview}} templates that use |type=.

In looking through that list, I notice stuff like this ('video' and the 'Interview' static text in separate parentheses – because there is no |interviewer=):

{{cite interview |subject= Sia|subjectlink= Sia Furler|type= video|title= Sia—Interview|url= http://www.bbc.co.uk/later/artists/sia/|program= ''[[Later... with Jools Holland]]''|callsign= BBC Two|city= London|date= October 2008 |accessdate=14 November 2008 }}

Sia (October 2008). "Sia—Interview" (video). Retrieved 14 November 2008. {{cite interview}}: Unknown parameter |callsign= ignored (help); Unknown parameter |city= ignored (|location= suggested) (help); Unknown parameter |program= ignored (help); Unknown parameter |subjectlink= ignored (|subject-link= suggested) (help)

So that needs fixing.

Remapping |program= → |work= seems obvious as does |city= → |location=. |callsign= → |publisher=?

I wonder if we shouldn't treat interviewers the same way we treat translators: create enumerated |interviewern=, etc; have dedicated static text so instead of Translated by $1 we use Interviewed by $1; |type= if set overrides the predefined meta-parameter TitleType value (just like all of the other cs1 templates that use predefined TitleType).

Titles are required for all cs1|2 templates. I'm not sure we should break that rule here.

—Trappist the monk (talk) 13:36, 2 September 2016 (UTC)

The template currently supports |interviewer= and |interviewers=. I don't see a problem with adding the enumerated case, and perhaps interviewer-first/last; would the intent be to add that to the metadata as I suggested? Otherwise, doesn't seem worth it.

I'm wondering if we can't just remove the non-standard parameters; namely, |call-?sign= (179 articles), |city= (465 articles), and |program= (563 articles).

Should we choose not to remove them:

• For callsign, it looks like it's subject to the standard work/publisher confusion, but indicating perhaps a higher-level work. I don't think capturing the higher-level work is necessary or desirable, so setting it to publisher would be okay in the majority of cases.
• I think we're agreed re city and program.

Regarding titles, don't we have a keyword lying around when the title is unknown or is not present for the lesser work in question? It's not documented in the generic documentation, if so. --Izno (talk) 16:08, 2 September 2016 (UTC)

For the sake of consistency, interviewer names would have the same parameter variants ans other namelists.

The only names in the metadata are author names. Translators, editors, interviewers, others, coauthors; none of these get into the metadata. When the metadata standard changes or we adopt another standard, then those names can be included.

We should deprecate before we remove, but I am in favor of removing those three parameters.

Only {{cite journal}} allows |title=none because there is an abbreviated 'style' used in some scientific disciplines that omits article titles in journal citations.

—Trappist the monk (talk) 11:04, 3 September 2016 (UTC)

The only point in response, because I'm fine with the rest: the handful of links above for the other styles each noted the case of the untitled interview. I'm personally fine with an interview being untitled and having an error, but it seems to me like the other information present would be sufficient (namely, the combination of interviewer, interviewee, and work). Maybe it should be questionable from our point of view whether the interview is legitimately published if it contains no title. *shrug* I think the other changes clearly have my agreement. :D --Izno (talk) 23:12, 3 September 2016 (UTC)

Google Books and page URLs

The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section. A summary of the conclusions reached follows.

Redundant thread; my bad.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  09:13, 28 August 2016 (UTC)

---

Moved to #About WP:PAGELINK

User Mauro Lanari (talk), at 02:06, 25 August 2016 (UTC), opened a WP:TFD about {{Cite book}}, and it was speedily closed as being in the wrong venue. As a procedural matter, I'm posting it to the correct one for him, in slightly clarified wording, though I haven't formulated a firm opinion about it myself:

About WP:PAGELINK: The problem seems to have arisen nearly 6 years ago, when it was decided to link Google Books pages in a way that distorts the purpose of |titleurl=, rather than adding a |pageurl= parameter to go along with |titleurl= and |chapterurl=. Since then, |titleurl= can also no longer be addressed to the front cover of a Google Book. Personally I [Maura Lanari] solve this by placing a link in the page or quote parameter, but some editors disagree and revert me. I mean, what do you think of updating the template {{cite book}}?

For Mauro Lanari (talk)  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  06:05, 28 August 2016 (UTC)

• For my part, I'm uncertain we have an encyclopedic need to link to the front of a Google Book when citing a page in it. Most of us do not use |titleurl= but |url= (I didn't even know |titleurl= existed). A possible solution would be to enable a |pageurl=, and treat |url= as an alias of it if a separate |titleurl= is given, but otherwise treat |url= as synonymous with |pageurl=. A problem with treating any Google Books |url= as a |pageurl= is that people expect to see the title linked, not a tiny page number, and another is that we'd need the Lua to parse for books.google.com URLs in particular.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  06:05, 28 August 2016 (UTC)
  • @SMcCandlish: look at #About WP:PAGELINK above. He started another discussion here at about the same time. Imzadi 1979 → 06:23, 28 August 2016 (UTC)

  There is no |titleurl= to know about:

  Title. {{cite book}}: Unknown parameter |titleurl= ignored (help)

  —Trappist the monk (talk) 08:35, 28 August 2016 (UTC)

Okay!  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  09:13, 28 August 2016 (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.

Porting into another wiki with localisation

I've tried to port the module code into the local wiki and found out that the $1 value in element 'mult_names' in this module may not be localisable. Is anyone know how to localise the value for that? (typically the $1 maybe "authors list", "editors list" or similar.) Shinjiman ⇔ ♨ 05:58, 30 August 2016 (UTC)

I think that you need to better explain what the problem is. What and where is the local wiki? Does the module properly render author lists, editor lists?

—Trappist the monk (talk) 09:41, 30 August 2016 (UTC)

Forgot to mention the module would be modified was listed in zh-yue:Module:Citation/CS1/Configuration. And seems by the current code, the category name which cannot be localised (like that one listed below). :D Shinjiman ⇔ ♨ 14:00, 30 August 2016 (UTC)

['mult_names'] = 'CS1 maint: Multiple names: $1', -- $1 is authors or editors

As an experiment, I've created a special_case_translation table in Module:Citation/CS1/Configuration/sandbox, provided the English words that replace $1 in the category names for mult_names and disp_auth_ed, and made the appropriate changes to extract_names(), name_has_mult_names(), and get_display_authors_editors():

Cite book comparison

Wikitext	{{cite book|author=blue, black, brown|title=Title}}
Live	blue, black, brown. Title.{{cite book}}: CS1 maint: multiple names: authors list (link)
Sandbox	blue, black, brown. Title.{{cite book}}: CS1 maint: multiple names: authors list (link)

Cite book comparison

Wikitext	{{cite book|author2=black|author3=brown|author=blue|display-authors=6|title=Title}}
Live	blue; black; brown. Title. {{cite book}}: Invalid |display-authors=6 (help)
Sandbox	blue; black; brown. Title. {{cite book}}: Invalid |display-authors=6 (help)

Does this work in your application?

—Trappist the monk (talk) 15:42, 30 August 2016 (UTC)

Seem this one works, verified in zh-yue:鄧芝 as an example. :) Shinjiman ⇔ ♨ 01:43, 1 September 2016 (UTC)

Good. I'll leave the translation table in the en.wiki module.

—Trappist the monk (talk) 11:12, 2 September 2016 (UTC)

Unwanted semicolon

When a book or article has more than one article or editor, with Template:Cite book a semicolon is placed between the first and second author, or between the first and second editor. For example "Dominic Baker-Smith; Cedric Charles Barfoot". I think this should be a comma. In the ISBD, a semicolon is used when contributors to a publication have different roles, which are mentioned in the description, for examples an editor, author, illustrator or translator. A comma or the word 'and' are usually used however between contributors with the same role (like two authors). In most referencing styles I see elsewhere, like APA style, the same applies. Bever (talk) 21:20, 1 September 2016 (UTC)

CS1 follows the "Last, First" convention for most names. Comma separators between contributors would not work well with the current scheme. There is also some facility for indicating roles in CS1 already. Is there any specific reason for CS1 to comply with ISBD? 65.88.88.127 (talk) 21:33, 1 September 2016 (UTC)

@Bever: you can include |last-author-amp=yes which inserts an ampersand between the last two authors (or editors) in a list. In your example, you'd get "Dominic Baker-Smith & Cedric Charles Barfoot". If you used |first=Dominic |last=Baker-Smith |first2=Cedric Charles |last2=Barfoot, you'd get "Baker-Smith, Dominic; Barefoot, Cedric Charles", which makes clear why we use the semicolon without the optional parameter to use an ampersand between the last two entries in an author list. Imzadi 1979 → 01:55, 2 September 2016 (UTC)

Inconsistency between older documentation and template data re: location vs. place?

Hi, in the older documentation it says that place is the preferred parameter, and that location is an alias of place. However, in most of the examples, location is used, and also the template data uses location as the parameter and does not use place at all (which is funny because the copied over description in TD claims it's an alias). Which is it? Should the templatedata be fixed to reflect the older documentation, or should the description of location/place be fixed? Mvolz (WMF) (talk) 13:56, 2 September 2016 (UTC)

It looks like the documentation used to list |location= as the preferred identifier. The documentation was changed on 6 May 2013 by Gadget850, who added documentation for |publication-place= in the same edit. I was unable to find a corresponding discussion on the talk page.

Almost every template I have seen that populates |location= or |place=, perhaps 99%, uses |location=. I recommend changing the documentation to show |location= as the preferred parameter and |place= as an alias, although functionally, it makes no difference. – Jonesey95 (talk) 14:53, 2 September 2016 (UTC)

The cs1|2 templates rarely, if ever, have 'preferred' parameters. In your example, |place= was the term that the contributing editor chose to 'define' (you will also notice that in the same definition, the editor uses the alias |location=).

Should template data be fixed? Hell, yes! Keeping and maintaining two separate and wholly incompatible sets of documentation is madness.

—Trappist the monk (talk) 14:54, 2 September 2016 (UTC)

cs3 for a new separator.

A recent technical change to Template:nlab to use cite-web broke the display style there. It dropped the word "in", and I can't figure out how to reinstate it. Viz. it now reads, for example, Pointed object at the nLab whereas it used to read, last week, Pointed object in nLab. (without the quotes, with the word "in"). The new format is awkward and visually ugly, and (most importantly) confusing to look at (to non-afficionados who may not know what ncatlab is). So... I'd like to get the old display style back, while keeping whatever the technical reasons are for converting the the cite-web template. It would seem that proposing a mode=cs3 that drops the quotes, and adds the word "in" is the only way to do this. Is there another way? How can we get the technical advantages(?) of using cite-web, and still get the prettier formatting of the older template? 67.198.37.16 (talk) 19:03, 2 September 2016 (UTC)

That would be a heck of a change to accommodate one external link template that is used just a few times. The more straightforward way to get what you want would be to discuss the change on the template's talk page or at the relevant WikiProject and decide whether the template should use the cite web template at all. Since it is used to link to a wiki, using the cite web template may be misleading to editors and may lead them to use it in references instead of just in external links. – Jonesey95 (talk) 20:52, 2 September 2016 (UTC)

that would result in us loosing all the other advantages of wrapping {{Cite web}}, such as validation, and COinS. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 21:02, 2 September 2016 (UTC)

OK, never mind, I'm just being stupid, there is an obvious alternative. 67.198.37.16 (talk) 17:09, 3 September 2016 (UTC)

Yes; you've canvassed another editor to revert my changes. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:11, 7 September 2016 (UTC)

Protected edit request on 4 September 2016

Prospect Reservoir - just edited up the access dates because the day had been left off two. They now show correctly, were showing a red error for check dates.

The 4th reference, a cite journal, the previous two are cite web, it will show the error if it is present, but when the syntax is correct it doesn't show in the displayed text.

Dave Rave (talk) 08:16, 4 September 2016 (UTC)

 Not done

No specific request made. The template appears to be rendering correctly.

—Trappist the monk (talk) 09:41, 4 September 2016 (UTC)

The cite journal reference that I think you are concerned with, has a doi id. Access date is superfluous. Perhaps the doc should make it clearer that some ids exclude access dates. 72.43.99.138 (talk) 14:28, 4 September 2016 (UTC)

Cite Polish law

I've made {{Cite Polish law}} a wrapper for {{Cite web}}, but it has |volume= and |number= parameters, which I've temporarily shoe-horned into the |website= parameter. Is there a better solution? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 11:58, 7 September 2016 (UTC)

I've solved that by making it wrap {{Citation}}, instead. Please let me know if my edits can be improved further. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:09, 7 September 2016 (UTC)

A technicality, but AFAIK, {{citation}} is Citation Style 2, not CS1, so your wrapper template is not quite accurate. Keep up the bold work. – Jonesey95 (talk) 13:59, 7 September 2016 (UTC)

Why not use cite journal instead of cite web? Headbomb {talk / contribs / physics / books} 15:01, 7 September 2016 (UTC)

Someone has reverted your changes but left the documentation inconsistent. Before I realised that I created a sandbox and testcases to experiment with using the mode parameter, but so far that has not worked as I expected. I've not touched the live template. --Mirokado (talk) 22:00, 7 September 2016 (UTC)

Thank you, all. I've made it wrap {{Cite journal}}, instead. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 11:31, 8 September 2016 (UTC)

I've been reverted, again, apparently because the template did not match "the legal Polish format". I wasn't aware that we were subject to Polish law. It would be interesting to review the internal consistency of citations in articles using the template. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 19:03, 13 September 2016 (UTC)

Language template

A side issue: I understand now why this revert of part of my edit was made, but how else can we show the language of a non-English title? (As opposed to the |language= of the target page.) Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:11, 8 September 2016 (UTC)

We've discussed a |title-lang=es/|title-es= in the past, derived from discussion at the feature request for it. Aside: I doubt there would be support for displaying the language of the title, as that's really just fluff for a citation. --Izno (talk) 12:32, 8 September 2016 (UTC)

It's not about displaying the language of the title (my use of "show" was imprecise, sorry), but marking it up correctly. For instance, {{lang|pl|Twoje zdrowie!}} displays as Twoje zdrowie! and emits the HTML: <span xml:lang="pl" lang="pl">Twoje zdrowie!</span> . Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 19:10, 8 September 2016 (UTC)

Then you may want to review the archive discussion; contributors do desire this function, but were a) split on the parameter set and b) per the Html 5 specification, specifying only the lang and not the xml:lang e.g. <span lang="pl">Twoje zdrowie!</span>. --Izno (talk) 19:50, 8 September 2016 (UTC)

BioRxiv support

bioRxiv, similarly to arxiv, is a preprint repository for biology. They don't quite have a dedicated identifier, but they do make use of a DOI system, which is not the same as the published version's DOI. I propose we add support for Biorxiv as an identifier

Option 1

• |biorxiv=047720 → bioRxiv 047720
• |biorxiv=10.1001/047720 → bioRxiv 047720

e.g. Luallen, Robert J.; et al. (30 June 2016). "Discovery of a Natural Microsporidian Pathogen with a Broad Tissue Tropism in Caenorhabditis elegans". PLOS Pathogens. 12 (6): e1005724. bioRxiv 047720 . doi:10.1371/journal.ppat.1005724.

Option 2

• |biorxiv=047720 → bioRxiv 10.1101/047720
• |biorxiv=10.1001/047720 → bioRxiv 10.1101/047720

e.g. Luallen, Robert J.; et al. (30 June 2016). "Discovery of a Natural Microsporidian Pathogen with a Broad Tissue Tropism in Caenorhabditis elegans". PLOS Pathogens. 12 (6): e1005724. bioRxiv 10.1101/047720 . doi:10.1371/journal.ppat.1005724.

Option 1 has the merit of being in line with how bioRxiv presents their pseudoidentifier (click citation tools on the right of that website), but option 2 makes the doi nature of the link clearer, and would be more useful in print. Personally I lean towards option 1, but I'd rather let people more experienced with bioRxiv than me decide. I'll advertise the discussion at relevant wikiprojects. Headbomb {talk / contribs / physics / books} 21:49, September 7, 2016 (UTC)

• I'm fine with Option 1. I don't think it's important to draw attention to the DOI resemblance. --Tryptofish (talk) 21:59, 7 September 2016 (UTC)

These citations are confusing (to me). You are citing a pre-print and the published, peer-reviewed version in the same citation. Those are two different articles. Just as DOI is not allowed in {{cite arxiv}}, we should not allow a DOI and biorxiv value in the same citation. Help us understand what you are trying to achieve here. – Jonesey95 (talk) 23:24, 7 September 2016 (UTC)

The journal is cited, the bioRxiv copy is linked as convenience, just like we do with arXiv (e.g. W.-M. Yao; et al. (Particle Data Group) (2006). "Review of Particle Physics: Pentaquark Update" (PDF). Journal of Physics G. 33 (1): 1–1232. arXiv:astro-ph/0601168. Bibcode:2006JPhG...33....1Y. doi:10.1088/0954-3899/33/1/001. {{cite journal}}: Unknown parameter |displayauthors= ignored (|display-authors= suggested) (help)). Headbomb {talk / contribs / physics / books} 23:41, 7 September 2016 (UTC)

I agree these could be two different articles, depending on the peer-review process. What might be useful is the linked biorxiv copy where the final version is non-OA. In my experience of mostly compbio papers, this isn't usually the case, but it could be different for other biology subfields? Amkilpatrick (talk) 00:16, 8 September 2016 (UTC)

(edit conflict)The identifier 10.1101/047720 is a doi simply because of its structure. Put that number in |doi= and it works:

{{cite journal |title=Discovery ... |doi=10.1101/047720}}

"Discovery ...". doi:10.1101/047720. {{cite journal}}: Cite journal requires |journal= (help)

This is just a proposal for a biorxiv identifier, right? We aren't contemplating a {{cite biorxiv}} here are we? If just an identifier, is the identifier anything but digits? Because it is and can (does) use the doi mechanism, anything after the first virgule can be any printable character.

—Trappist the monk (talk) 00:07, 8 September 2016 (UTC)

Here it's for adding the identifier support. We might develop a {{cite biorxiv}} template in parallel to this, but that's not what this proposal is about specifically. Headbomb {talk / contribs / physics / books} 12:06, 8 September 2016 (UTC)

• option 1 seems preferable--Ozzie10aaaa (talk) 00:03, 8 September 2016 (UTC)
• Thanks for the heads up at WP:COMPBIO. It would definitely be good to have bioRxiv supported similarly to arXiv. I think Option 1 is fine here, in line with their own style. Amkilpatrick (talk) 00:13, 8 September 2016 (UTC)
• Option 1 seems good to me. It is more useful than using the |doi= field because this field might be used for the published version (and adding a custom |biorxiv= parameter enables us to add the free-to-read lock silently). I am happy to implement that in the sandbox if it helps. − Pintoch (talk) 07:46, 8 September 2016 (UTC)

Here is the exemplar citation rendered by the sandbox. This rendering uses |biorxiv=047720 which is the simplest form and should work for all as long as the prefix is and remains 10.1101/. If ever that changes, then this scheme will stop working.

Luallen, Robert J.; et al. (30 June 2016). "Discovery of a Natural Microsporidian Pathogen with a Broad Tissue Tropism in Caenorhabditis elegans". PLOS Pathogens. 12 (6): e1005724. bioRxiv 047720. doi:10.1371/journal.ppat.1005724. {{cite journal}}: Check |biorxiv= value (help)CS1 maint: unflagged free DOI (link)

—Trappist the monk (talk) 12:20, 8 September 2016 (UTC)

Should/does it support both 047720 and 10.1101/047720 as inputs? Headbomb {talk / contribs / physics / books} 12:27, 8 September 2016 (UTC)

As I wrote in my post, the sandbox does not support the doi form. Should it? I think not. If editors want to use the doi form there is |doi=.

—Trappist the monk (talk) 14:57, 8 September 2016 (UTC)

Just throwing it out there as an idea. The output would be the same (truncated) in all cases though. Headbomb {talk / contribs / physics / books} 15:50, 8 September 2016 (UTC)

Non-standard citation templates

This search finds templates whose names begin "Template:Cite...", that do not wrap one of the standard CS1 citation templates (currently finding 144 results). I've already upgraded all those that were straightforward, to do so. This give all the usual advantages, such as error trapping and embedded COinS.

I invite anyone interested and capable to look at upgrading the rest (or to consider whether they should be deleted, or merged - I nominated several such templates at TfD in recent days). Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 11:42, 8 September 2016 (UTC)

You may be overreaching here. CS1 is a citation style. It is not a "standard". Just because a template attempts to cite something, does not mean it has to follow CS1, an unfinished spec to be sure. A more pertinent approach would be to fix meta- and specific-source templates that purport to be based on CS1 but either are not, or do so out of spec. 71.247.146.98 (talk) 11:56, 8 September 2016 (UTC)

Same thoughts here. If there are non-standard citation templates, it's probably for a reason. For example, some templates may reflect country-specific legal requirements for citing legal documents. At the very least, such changes should be first discussed on talk pages of the templates in question. — Kpalion^(talk) 12:29, 8 September 2016 (UTC)

[reply to both] Well, I didn't ask anyone to immediately "upgrade", I said "look at upgrading". And CS1 templates are more configurable in their presentation, thanks to |mode=, than anything hard-coded to a single presentation. After that, WP:BOLD applies; prior discussion is not mandated. if you have a list of "templates that purport to be based on CS1 but either are not, or do so out of spec", please feel free to link to it. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 14:42, 8 September 2016 (UTC)

Editing is not the issue. Editing just to comply with a specific style is. Especially when the style is wrongly claimed a "standard". Write similar templates that apply your preferred style. Then let editors choose which iteration to use. There has been previous discussion here about CS1-based templates that need attention or de-categorizing. The relevant categories include several examples. 64.134.65.76 (talk) 17:36, 8 September 2016 (UTC)

Thanks for the advice; I don't find it compelling; not least because you wrongly refer to "editing just to comply with a specific style"; and misquote me. And if you/ the other editor (or are you the same?) want me to work on something else, I expect you to provide the link, not to send me to search for it. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 18:22, 8 September 2016 (UTC)

Well your heading is "Non-standard citation templates". And this is a talk page for some citation templates (CS1), that you obviously exclude from the ones referred to in the heading, unless I'm mistaken. I don't think I misquoted. You mention "mode" but this is a CS1 parameter, and therefore not style-neutral. Also, I was not asking/telling you to work on anything, but laying out how things should be properly handled, imo. There are several subcategories in Category:Citation Style 1 templates. There are templates in them that don't belong. To me that is a better approach, but I wasn't trying to convince anyone to do anything about it. 65.88.88.127 (talk) 21:24, 8 September 2016 (UTC)

The Times

It looks as though {{Cite newspaper The Times}} could be made a wrapper for {{Cite news}}. The former's |column= appears to be unused, while the rather unnecessary |day_of_week= is used only once. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 16:20, 8 September 2016 (UTC)

There seems to have been some lag in applying the tracking categories'; but those parameters still appear to be used on fewer than ten articles each. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 18:25, 8 September 2016 (UTC)

Looking at the documentation for {{Cite news}} it appears that |department= could be used for the name of a column; it could also be used for other areas within a newspaper, such as editorials. Jc3s5h (talk) 20:37, 8 September 2016 (UTC)

Now updated to max. 59 pages using |column=; max. 85 pages using |day_of_week=. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 15:15, 9 September 2016 (UTC)

Currently 118 page and 154 pages respectively. Andy, I think a lot more pages are using this template with those parameters than you think. Category population can be very slow when the job queue has high lag (you know that). Can you please wait until you are sure the categories are fully populated before doing anything. Previous discussion here (at the template talk page) is relevant. Also, you really should be pointing people to this discussion when they ask you what is going on. When Mjroots asked you here what was happening, your reply failed to point to this discussion, that is verging on misdirection. Your reply should have been along the lines of "I created the categories to help generate some statistics for the discussion I started where I am proposing to make this template a wrapper for cite news" (with a link to the discussion). @Iridescent: who I know has strong views on how this template should be used. Carcharoth (talk) 21:33, 9 September 2016 (UTC)

Now 136 pages and 194 pages respectively. It is clear that populating these tracking categories is going to take a long time. The transclusion count is 3017. Those that know how this template works and why it has 'column' and 'day of the week' parameters (The Times was published in a chapbook format for most of its history, hence this different citation style) know that most of the transclusions of this template will include use of these parameters, hence what Andy said above about the parameters being unused (or almost unused) makes no sense at all. I am going to ask at the technical village pump for advice here. Reversing the edits will just re-add to the job queue. Currently the issue is readers being served up with nonsensical category names as red-links (e.g. at Beckton Gas Works). I checked logged out, and readers are seeing those category names. I think we will need to create the categories and mark them as hidden, unless there is a way to 'hide' red-linked categories. Carcharoth (talk) 03:01, 10 September 2016 (UTC)

Short answer: no there isn't. Redlinked cats can never be hidden. --Redrose64 (talk) 07:47, 10 September 2016 (UTC)

Update: I have asked for advice here at the technical village pump. I wouldn't normally do this, but I think the issue needs some wider attention and I'll not be around for the next day or so. I'll add a similar note over at the template talk page. Carcharoth (talk) 03:24, 10 September 2016 (UTC)

@Carcharoth: No. When asked:

what is the purpose of these uncreated categories? The are appearing on all articles and lists that use this template.

My reply, was:

As the name indicates, they are TEMPORARY. One tracks TEMPORARY Cite newspaper The Times using the 'column' parameter; the other tracks TEMPORARY Cite newspaper The Times using the 'day of week' parameter.

There was no "misdirection" - borderline or otherwise - whatsoever. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:15, 10 September 2016 (UTC)

I fully understood what they did, but I asked what the purpose of the categories was. If there is to be a proposal to do away with {{cite newspaper The Times}}, I for one will be opposing it. Mjroots (talk) 14:25, 10 September 2016 (UTC)

Andy, your reply to Mjroots lacked information. You gave a literal reply (explaining what 'temporary' means) rather than pointing to this discussion. I think that omission of a link to this discussion is far from ideal. I apologise for saying it verged on misdirection. I think we should now focus on trying to work out when the 'temporary' categories will be fully populated. Currently the numbers are 260 and 356 respectively. I think we need the numbers to be stable for at least a full day before being confident that the categories have fully populated. On the wider question of the use of the parameters, I think they are only really useful before a certain date. After that date, The Times became a 'normal' newspaper. Not sure when that was though. Hopefully someone else will be able to provide that information. Carcharoth (talk) 09:16, 11 September 2016 (UTC)

File:Times May 10 1830.jpeg

Page 2 of The Times as it was formatted in 1830

Regarding the comments above from Carcharoth, it might make it easier to explain why citing early issues of The Times differs from other newspapers to see what it looked like in chapbook days. Each issue consisted of a single large sheet of paper, which was folded in half to create four nominal "pages". The front was filled with adverts, and the rear was filled with official notices (law reports, auction notices, births, marriages & deaths, stock prices etc), leaving only the two inner "pages" for actual news. These two pages (one shown to the right for illustrative purposes) were filled with very, very small type so as to cram everything in, and most headings in the same small type as a space-saving measure. Thus someone checking a reference who wants to find it in the original will effectively have to read 50% of the newspaper to find it since it will always be on page 2 or 3 so the page number isn't particularly useful, and there aren't 'headlines' in the modern sense for the eye to skim over. As a consequence, it's conventional when citing early issues of the Times to provide a column reference as well as the page, which as things stand the existing {{cite news}} template can't handle. Regarding when they stopped using this format, it was a gradual process; in the early 1830s they went from one sheet of paper to two (e.g. 8 pages per issue), circa 1850 they went to three sheets (12 pages), and circa 1860 they went to four sheets (16 pages) and started to abandon their formatting eccentricities (although not entirely; they stubbornly stuck to some conventions like "no news on the front page" until the late 1960s). ‑ Iridescent 13:49, 12 September 2016 (UTC)

My quick thoughts: keep the column indication as is, but drop the day of the week. I can't find it in the MOS now, but I recall reading that we don't include the day of the week in giving dates unless it's significant to the mention of the date in prose. As for including it in citations, I don't know of any style guide that includes it either. Imzadi 1979 → 17:37, 12 September 2016 (UTC)

Imzadi1979's proposal is acceptable to me. It'd be a bit less work for editors using the template, and a bot run could do the removal of the parameter from existing uses of the template. Mjroots (talk) 19:11, 12 September 2016 (UTC)

I'd agree with that. I suspect that "day of the week" field stems from the fact that the Sunday Times and Times have always had different editors and different editorial lines (and in some periods, different proprietors) yet are usually archived as a single entity. The other idiosyncratic field I'd be reluctant to lose from this template is "section", because the Times is both a general newspaper and an official gazette, it's sometimes important to make it clear if we're citing general editorial content or the formal paper-of-record stuff like the Law Reports or the Court Circular. ‑ Iridescent 20:12, 12 September 2016 (UTC)

The |section= could probably be mapped to the |department= of {{cite news}}. As for the day of the week, I have a random thought based on Iridescent said above: should the uses with dates that are Sundays give |work=The Sunday Times instead of |work=The Times? If so, that would retain the distinction while dropping the day of the week from date. Imzadi 1979 → 22:19, 12 September 2016 (UTC)

In late 2014, I made {{Cite newspaper The Times/sandbox}} that pretty much adheres to Editor Imzadi1979's suggestion:

"The Queen Mary Back In Port". The Times. No. 51269. London. 3 January 1949. col E, p. 4. template uses deprecated parameter(s) (help) – live

"The Queen Mary Back In Port". The Times. No. 51269. London. 3 January 1949. col E, p. 4. template uses deprecated parameter(s) (help) – sandbox

There are differences: period after the article title; |day_of_week= is ignored; |location= (not a parameter in {{Cite newspaper The Times}}) is in a different location as is |issue=.

—Trappist the monk (talk) 00:08, 13 September 2016 (UTC)

This all sounds good. Can anyone identify the uses of this template that are not using the day of the week parameter? Those will have to be checked to see if they are Sundays or not. Might also be worth looking around for other 'Sunday Times' citations elsewhere in Wikipedia and seeing if they are made distinct from 'The Times'. I am sure 'cite newspaper' has had this problem in the past with other newspapers with weekend editions. PS. I think the numbers are stable now - 1,274 for the column parameter and 2,935 for the day of the week parameter. Another check in a day or so should confirm that. Carcharoth (talk) 06:11, 13 September 2016 (UTC)

Re the Sunday Times, wouldn't it be simpler to create a separate template for that newspaper? Mjroots (talk) 07:55, 13 September 2016 (UTC)

Not necessarily. We can do something like this addition to the sandbox:

|newspaper={{#ifeq:{{{day_of_week|}}}|Sunday|Sunday Times|Times}}

which would do this:

"Steamer lost off The Lizard". The Times. No. 40718. London. 6 December 1914. col E, p. 4. template uses deprecated parameter(s) (help) – live

"Steamer lost off The Lizard". The Times. No. 40718. London. 6 December 1914. col E, p. 4. template uses deprecated parameter(s) (help) – sandbox

If this is not a good way to distinguish the Sunday edition, then yeah, a separate template will answer.

—Trappist the monk (talk) 11:24, 13 September 2016 (UTC)

The discussion above is leaning towards the abolition of the day parameter, so a separate template would seem to be the way to go. Mjroots (talk) 09:43, 14 September 2016 (UTC)

No need for a separate template. It's what |newspaper= is for. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 13:11, 14 September 2016 (UTC)

Improving HTML citation element semantics

Cite

Given the following wiki markup as example:

{{Cite web |url=https://support.apple.com/en-us/HT202944 |title=TCP and UDP ports used by Apple software products |date=2014-11-08 |website=Apple Support |publication-date=2016-02-05 |archive-url=https://web.archive.org/web/20160913023842/https://support.apple.com/en-us/HT202944 |archive-date=2016-09-13 |dead-url=no |access-date=2016-09-13}}

produces output similar to this:

<cite class="citation web"><a rel="nofollow" class="external text" href="https://support.apple.com/en-us/HT202944">"TCP and UDP ports used by Apple software products"</a>. <i>Apple Support</i> (published 2016-02-05). 2014-11-08. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20160913023842/https://support.apple.com/en-us/HT202944">Archived</a> from the original on 2016-09-13<span class="reference-accessdate">. Retrieved <span class="nowrap">2016-09-13</span></span>.</cite>

My concern is that the HTML citation element (<cite>) is wrapped around the whole template. The WHATWG HTML Standard says in section 4.5.6 that A person's name is not the title of a work and and the element must therefore not be used to mark up people's names. Further examples are given in that section for correct usage according to that specification. HTML5 specification published by W3C (which was developed by WHATWG, but later diverged by W3C) allows using a name as a citation, but I believe the examples given mostly point at the direction of giving a very specific citation based on work name (in our case, |title=).

The correct output should look something like this (untested, I mind you):

<p class="citation web"><a rel="nofollow" class="external text" href="https://support.apple.com/en-us/HT202944">"<cite>TCP and UDP ports used by Apple software products</cite>"</a>. <i>Apple Support</i> (published 2016-02-05). 2014-11-08. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20160913023842/https://support.apple.com/en-us/HT202944">Archived</a> from the original on 2016-09-13<span class="reference-accessdate">. Retrieved <span class="nowrap">2016-09-13</span></p>.

Time

It may be a good idea to change the <span> for dates to <time> element as well, for further improving the semantics. I also question if the quotation marks " need to be around the <cite> element, as if they are only for visual appearance then they should be added in CSS.

My view is biased towards WHATWG as the current status of HTML as the right thing to do, because updating to the WHATWG specification would satisfy requirements of both W3C's HTML5 and WHATWG's HTML standard. The current template does not allow satisfying HTML standard requirements.

References

Not quite so relevant citations

Hickson, Ian. "The cite element". In Pieters, Simon; van Kesteren, Anne; Jägenstedt, Philip; Denicola, Denicola; et al. (eds.). HTML Standard. Acknowledgements. (Living Standard ed.). WHATWG. 4.5.6. Archived from the original on 13 September 2016. Retrieved 13 September 2016. {{cite book}}: External link in |others= (help); Unknown parameter |dead-url= ignored (|url-status= suggested) (help) Hickson, Ian. "Is this HTML5?". In Pieters, Simon; van Kesteren, Anne; Jägenstedt, Philip; Denicola, Denicola; et al. (eds.). HTML Standard. Acknowledgements. (Living Standard ed.). WHATWG. 1.2. Archived from the original on 13 September 2016. Retrieved 13 September 2016. {{cite book}}: External link in |others= (help); Unknown parameter |dead-url= ignored (|url-status= suggested) (help) Sheppy; Sebastianz; JohnnyWalker; Jeremie; openjck; et al., eds. (7 August 2006). "<cite>". HTML element reference. Mozilla Developer Network. Mozilla (published 2 May 2016). Archived from the original on 13 September 2016. Retrieved 13 September 2016. {{cite encyclopedia}}: Unknown parameter |dead-url= ignored (|url-status= suggested) (help)CS1 maint: numeric names: editors list (link) Hickson, Ian; Berjon, Robin; Faulkner, Steve; Leithead, Travis; Navara, Erika Doyle; O'Connor, Edward; Pfeiffer, Silvia, eds. (28 October 2014). "The cite element". HTML5 (W3C Recommendation ed.). W3C. 4.5.6. Archived from the original on 13 September 2016. Retrieved 13 September 2016. {{cite book}}: Unknown parameter |dead-url= ignored (|url-status= suggested) (help)

80.221.159.67 (talk) 11:50, 13 September 2016 (UTC)

Discussion

We changed cite deliberately to use cite per Help_talk:Citation_Style_1/Archive_8#HTML5 bait-and-switch: The cite element again. Please review that discussion. --Izno (talk) 12:40, 13 September 2016 (UTC)

And the discussion linked therein at MediaWiki_talk:Common.css/Archive_17#The_cite_element_needs_to_not_auto-italicize_any_longer. --Izno (talk) 12:43, 13 September 2016 (UTC)

And finally, the actual implementation discussion at Help_talk:Citation_Style_1/Archive_9#<cite> has been fixed, so we can now use it for entire citation. --Izno (talk) 12:46, 13 September 2016 (UTC)

Regarding the styling (issue with quotation marks), we have already defaulted the use of cite to not-italicize based on the issue that sometimes the citation provides only work vice work and sub-section title. Long works use italics while sub-work-level usually use quotation marks. So I believe there would not be consensus (even if it made sense to change our use of cite, which it doesn't) to change those points of styling in CSS. No comment regarding the use of the time element. --Izno (talk) 12:57, 13 September 2016 (UTC)

Any updates on adding open access links to references?

For simplicity's sake, I suggest we implement the coloured version of the locks, with |access= (paired for whatever link is in |url=) and |id-access= to append locks after the identifiers (replace 'id' with 'bibcode'/'doi'/etc.). We can then refine the behaviour later after we see what issues come up. Headbomb {talk / contribs / physics / books} 14:27, 14 September 2016 (UTC)

I totally agree! I think that's what is currently implemented in the sandbox (but only for id=doi, not for the others). We are ready to submit OAbot to WP:BAG and we can discuss there which icons we want the bot to add (if the current implementation is pushed to live, otherwise there is no discussion to have about that as the bot would not have any option to display the status of the links it adds). − Pintoch (talk) 21:27, 14 September 2016 (UTC)

I also agree that the colored versions are more intuitive and also more consistent with usage outside of Wikimedia. OAbot can do some pretty nifty work, so I think the current implementation is definitely "good enough" to get started with, and as noted above doesn't prevent us from tweaking icons in the future as we progress, get feedback, and learn. Cheers, Jake Ocaasi (WMF) (talk) 22:27, 14 September 2016 (UTC)

Colored versions of the lock are only more intuitive if you as a reader have color sight. For those who don't, the color is more-or-less meaningless; perhaps slightly differing shades of gray. As before, I remain opposed to implementing colored versions of the various locks.

As before, I remain opposed to highlighting the norm. Because named identifiers typically require registration or payment, we should not be adding closed-lock icons to those links. Throughout all of Wikipedia, links in |url= typically link to a readable source; we should not be highlighting those by adding open-lock icons.

I am not opposed to including lock icons in rendered citations where their use is appropriate, distinguishable in color and gray-scale against white or black background, and constrained in their application.

—Trappist the monk (talk) 11:27, 15 September 2016 (UTC)

They are more intuitive, and those with color blindness can still tell them apart by different shapes. Let's implement them, and if we can find a better way of showing this, then we'll refine. We're not saying ignore the 1% here, but to deny the 99% easier access because the information is conveyed in one extra way that the 1% doesn't have access to is silly.Headbomb {talk / contribs / physics / books} 17:36, 15 September 2016 (UTC)

OK. I will not write again here my arguments against your decision to forbid certain locks at certain places as they have not changed either. Are there any chances the change could still be implemented if you acknowledge that there is a consensus for it, or is your own position a definitive veto against it? All I want is to know when to submit the bot for approval: in the first case, we'll try to reach that consensus, in the latter, I'll submit the bot without waiting for CS1 to change. (I am only concerned about which parameters are allowed - the pictures do not have any influence on how the bot should work.) − Pintoch (talk) 17:18, 15 September 2016 (UTC)

Trappist's concerns are legitimate and I will comment that he is not the only one who holds them. "A consensus" does not seem to exist with such firmness as you think, and I suspect that an RFC would likely resolve in Trappist's favor on those points. --Izno (talk) 17:28, 15 September 2016 (UTC)

Fine! I thought most editors agreed on the mockup that I followed for my implementation in the sandbox (hence the "consensus"). Trappist was the only one to voice his concerns about this issue after my implementation. Good to see others finally chiming in! I just need the discussion to actually take place, because this detail is holding us back on another front, and I'm happy to see any decision taken about this. I hope this is clearer. − Pintoch (talk) 19:55, 15 September 2016 (UTC)

Let's implement the technicalities, and we'll work out the best practices. I'm not super thrilled about plastering |access=free myself, mostly because the url SHOULD be free by default. If it's not free, then the link adds nothing over |doi=, and should be removed. So the only time you really need to flag anything is if it's not free, and no id/doi are present. Likewise, DOIs should be closed by default, and we flag the free ones (with automated links when that is the case).

So that could mean only allowing |access=subscription/registration/ignoring |access=free and allowing |doi-access=free/ignoring |doi-access=subscription/registration. 17:45, 15 September 2016 (UTC)

They appear to have been refuted; and accessibility guidelines (our own, and the W3C's) suggest not relying on colour alone to distinguish items; they do not deprecate its use at all. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 19:20, 15 September 2016 (UTC)

I join others above in expressing doubts that there is sufficient consensus for the proposals as they stand. Both the icons to be used and the circumstances of their use have been debated and changed back and forth several times. There should be a clear proposal, or a clear small set of alternative proposals, and then a much wider RfC. Nothing less will be satisfactory for a change with such a significant visual impact all over Wikipedia. Peter coxhead (talk) 19:43, 15 September 2016 (UTC)

Let me join in the chorus of people here who are skeptical about this proposal. It cluttters up the citations, is not always easy to define (is a site that limits you to a certain number of downloads per month open access? is a link that works when you follow it from a specific other link, but not when you copy and paste the same url into Wikipedia, open access?), is about a specific link rather than the whole citation but is proposed to be parameterized and displayed in a way that does not indicate which links are open and which are not (the same paper can be open access through one access method and not open access through another), and is of dubious value to most readers. We can continue to discuss it and mock things up, of course, but I remain unpersuaded. Just because we can do something doesn't mean we should. —David Eppstein (talk) 20:03, 15 September 2016 (UTC)

The locks would apply to the links provided. doi:10.1234/whatever may sometimes resolve to different places, but never one that's open access and another that isn't. The lock on that doi would reflect the state of that link. We might need some more discussion, but right now what I'm proposing, and that I think is likely to be agreed on, is that |url= link shows when they ARENT free, while identifier links shows when they ARE free. But let's at least have a working sandbox version so we can debate and tweak behaviour as needed. And then we can RfC for deployment or something.Headbomb {talk / contribs / physics / books} 20:22, 15 September 2016 (UTC)

Re the different potential expansions of a doi never having different open-access statuses: [citation needed]. And what do you propose to do with links like the ones generated by {{MR}} that already give you useful information when accessed openly but give enhanced information to subscribers? —David Eppstein (talk) 22:15, 15 September 2016 (UTC)

Re doi: You can't prove a negative, and you know that. But I do citation cleanup like few people do, and I have never once ran into a doi that resolves to both an open and a closed access provider. That's because articles are either openly licensed, in which case they will be openly licensed to both providers, or they aren't, in which case neither providers will provide it.

Re MR: MR is closed/subscription based, and non-subscribers do not have access to the review. The options available are, mark it locked (which I think is unecessary/overkill, but others might differ), or do like we do now (which I think is the best behaviour, but again others might differ) and leave the MR link plain. Headbomb {talk / contribs / physics / books} 22:48, 15 September 2016 (UTC)

I have to go along with Headbomb, that we need a fully functional sandbox version so that we can compare the output with the existing templates to see what works best. Keith D (talk) 23:04, 15 September 2016 (UTC)

I have changed the sandbox to forbid |access=free and |doi-access=subscription. Intermediate access levels are currently allowed on both |access= and |doi-access= (but it's straightforward to change). |id-access= is not implemented for other ids. I can add them if you think that it would ease the discussion (but as far as I can tell they would work just like |doi-access=, right?). So I believe the sandbox is functional now. − Pintoch (talk) 06:15, 16 September 2016 (UTC)

Current state of the sandbox

Cite journal comparison

Wikitext	{{cite journal|access=subscription|date=2007-09-29|first=Charles S.|issue=1|journal=Origins of Life and Evolution of Biospheres|last=Cockell|pages=87–104|title=The Interplanetary Exchange of Photosynthesis|url=http://link.springer.com/article/10.1007/s11084-007-9112-3|volume=38}}
Live	Cockell, Charles S. (2007-09-29). "The Interplanetary Exchange of Photosynthesis". Origins of Life and Evolution of Biospheres. 38 (1): 87–104. {{cite journal}}: Unknown parameter |access= ignored (|access-date= suggested) (help)
Sandbox	Cockell, Charles S. (2007-09-29). "The Interplanetary Exchange of Photosynthesis". Origins of Life and Evolution of Biospheres. 38 (1): 87–104. {{cite journal}}: Unknown parameter |access= ignored (|access-date= suggested) (help)

Cite journal comparison

Wikitext	{{cite journal|arxiv=1412.8548|date=2014-12-28|doi-access=free|doi=10.4204/EPTCS.172.23|first1=Krzysztof|first2=Jamie|journal=Electronic Proceedings in Theoretical Computer Science|last1=Bar|last2=Vicary|pages=316–332|title=A 2-Categorical Analysis of Complementary Families, Quantum Key Distribution and the Mean King Problem|volume=172}}
Live	Bar, Krzysztof; Vicary, Jamie (2014-12-28). "A 2-Categorical Analysis of Complementary Families, Quantum Key Distribution and the Mean King Problem". Electronic Proceedings in Theoretical Computer Science. 172: 316–332. arXiv:1412.8548. doi:10.4204/EPTCS.172.23.
Sandbox	Bar, Krzysztof; Vicary, Jamie (2014-12-28). "A 2-Categorical Analysis of Complementary Families, Quantum Key Distribution and the Mean King Problem". Electronic Proceedings in Theoretical Computer Science. 172: 316–332. arXiv:1412.8548. doi:10.4204/EPTCS.172.23.

Cite journal comparison

Wikitext	{{cite journal|access=free|date=1992-10-01|doi-access=subscription|doi=10.1139/f92-220|first1=K. K.|first2=R. C.|first3=J. R.|issue=10|journal=Canadian Journal of Fisheries and Aquatic Sciences|last1=English|last2=Bocking|last3=Irvine|pages=1982–1989|title=A Robust Procedure for Estimating Salmon Escapement based on the Area-Under-the-Curve Method|url=https://www.researchgate.net/profile/James_Irvine5/publication/233865122_A_Robust_Procedure_for_Estimating_Salmon_Escapement_based_on_the_Area-Under-the-Curve_Method/links/09e4150c65bae92991000000.pdf|volume=49}}
Live	English, K. K.; Bocking, R. C.; Irvine, J. R. (1992-10-01). "A Robust Procedure for Estimating Salmon Escapement based on the Area-Under-the-Curve Method" (PDF). Canadian Journal of Fisheries and Aquatic Sciences. 49 (10): 1982–1989. doi:10.1139/f92-220. {{cite journal}}: Invalid |doi-access=subscription (help); Unknown parameter |access= ignored (|access-date= suggested) (help)
Sandbox	English, K. K.; Bocking, R. C.; Irvine, J. R. (1992-10-01). "A Robust Procedure for Estimating Salmon Escapement based on the Area-Under-the-Curve Method" (PDF). Canadian Journal of Fisheries and Aquatic Sciences. 49 (10): 1982–1989. doi:10.1139/f92-220. {{cite journal}}: Invalid |doi-access=subscription (help); Unknown parameter |access= ignored (|access-date= suggested) (help)

− Pintoch (talk) 06:15, 16 September 2016 (UTC)

Personally, if |access/doi-access= are set to any 'valid' access options (free/registration/subscription), I wouldn't output error messages for the options we disallow. Those would act as a sort of edit-window comment of "yeah, I've actually checked this link, and it is freely accessible" / "yeah I've checked this doi, and it isn't freely accessible". Error messages should, IMO, only be displayed if |access/doi-access= is set to something like "24$" or "Smith 2005 et al.". Headbomb {talk / contribs / physics / books} 07:56, 16 September 2016 (UTC)

PMC error checking adjustment

QuackGuru has edited the Help:CS1 errors page to note that |pmc= values higher than 5000000 are being issued by PubMed. See, for example:

S. Klotz; et al. (26 Aug 2016). "Ice VII from aqueous salt solutions: From a glass to a crystal with broken H-bonds". Sci Rep. 6: 32040. doi:10.1038/srep32040. PMC 5000010.

The above message generates an error message at this writing, because our error check for |pmc= looks for PMC IDs between 1 and 5000000 (five million). I have adjusted the sandbox code to allow for values up to 6000000 (six million):

Questions or comments are welcome. – Jonesey95 (talk) 00:01, 16 September 2016 (UTC)