Template talk:Author

From Wikisource
Jump to navigation Jump to search


  • September 2006: Problem with section edit link | New header style? | Transition to new author template (2 subtopics) | intersource cs:.


I've been looking at the source for the template, and it looks like we are not automatically categorizing author pages anymore? Did I read the source code correctly?—Zhaladshar (Talk) 05:09, 14 September 2006 (UTC)Reply[reply]

That should be fixed now. // [admin] Pathoschild (talk/map) 20:53, 15 September 2006 (UTC)Reply[reply]


Please add interwiki for Slovenian Wikisource: sl:Predloga:Avtor Thanks, -- 06:36, 29 November 2006 (UTC).Reply[reply]

Done. —{admin} Pathoschild 18:18, 29 November 2006 (UTC)

Defaultsort key not working?[edit]

Looking at Category:Authors-B, it looks as if the "defaultsort" parameter on Template:Author is not working, as many (if not all) of the author pages listed in this category have the defaultsort parameter but are categorized all out of order.—Zhaladshar (Talk) 21:50, 14 February 2007 (UTC)Reply[reply]

The default sort magic word defines the sort key only for categorization when no other sort key is specified. The problem is the line {{{category|[[Category:Authors-{{{last_initial}}}|{{{last_initial}}}]]}}}. That line tells the software to override the default sort key and sort on the last initial. So all author pages in Category:Authors-B is sorted as "B". The order of all those "B" sort keys is more or less arbitrary. /EnDumEn 22:25, 14 February 2007 (UTC)Reply[reply]
That is fixed in the new version proposed at Wikisource:Scriptorium#Overhaul_author_template. —{admin} Pathoschild 18:52:08, 28 February 2007 (UTC)

"Quotations" not "quotes"[edit]

"Quote" is a verb, and it is very poor English usage to use it in place of the proper noun. Can somebody fix this? Brisvegas 03:55, 25 April 2007 (UTC)Reply[reply]

Quote is a valid informal noun form of quotation. The interwiki links box needs to be as small as possible, since it pushes aside the author description. —{admin} Pathoschild 19:55:24, 28 April 2007 (UTC)

Chinese names[edit]

Is there a way to manage Chinese names, which place the family name first? Take a look at the page Author:Li Bai for an example of the difficulty; Li is the family name.Easchiff 23:21, 16 June 2007 (UTC)Reply[reply]

Yep; use the following parameters. Please leave in the comment in the name field.
 |firstname      = Bai
 |lastname       = Li
 |name           = Li Bai <!-- override parameter, do not use for most author pages -->
{admin} Pathoschild 05:36:19, 14 August 2007 (UTC)
Isn't that an improvement over the advice the template explanation currently gives? LlywelynII (talk) 06:47, 15 October 2011 (UTC)Reply[reply]
Except the parameter name isn't there, and it wasn't working. — billinghurst sDrewth 13:10, 15 October 2011 (UTC)Reply[reply]

Categorizing with "Category:Authors"?[edit]

I've noticed that the template every now and then gives an author page Category:Authors. I'm not sure why it does that, but it's got to be in the template somewhere, as the actual author pages do not categorize themselves with that category.—Zhaladshar (Talk) 15:27, 30 July 2007 (UTC)Reply[reply]

I don't see anything in the template that would cause that; if you see it again, link to it from here so I can take a look. —{admin} Pathoschild 05:38:27, 14 August 2007 (UTC)


The author template on French Wikisource is much better looking (see, for example, Charles Baudelaire), although it has less details. It would be nice if the English template could have its look improved. I would also point out that the image option is sort of pointless in the author template, since invariably the image is too big. Wouldn't it make more sense if the images were just added separately to the pages? — The Man in Question 11:03, 4 January 2008 (UTC)Reply[reply]


When a birth and/or death year isnt known, it is customary to provides years when the person flourished (fl.), or was was known to exist. I think a "flourished" override would be useful, especially for the more obscure authors that we tend to find as editors and translators. John Vandenberg (chat) 11:13, 8 June 2008 (UTC)Reply[reply]

  • Support, it would allow us to document what date ranges the author is active in to avoid confusion especially when firm birth and death dates are not available. Jeepday (talk) 11:42, 4 July 2008 (UTC)Reply[reply]

Single names[edit]

The "name" option mentioned in the Chinese name discussion above does not seem to actually work. I do not see it in the template parameters, and it is ignored in favor of firstname and lastname. This should be included for use with authors only having a single name such as Plato or Socrates as well. --Parsa 17:16, 4 July 2008 (UTC)Reply[reply]

Last initial guidance[edit]

One or two letters?[edit]

I have started to see the field for Last_initial having two letters 'Xx'. If this is now the preferred methodology, if so would it be possible to amend the guidance in the template. Further, is there a plan to convert those existing with single initial? -- billinghurst (talk) 00:48, 22 September 2008 (UTC)Reply[reply]

I have raised this at Wikisource:Scriptorium#Two character Author initial. John Vandenberg (chat) 02:18, 28 September 2008 (UTC)Reply[reply]

How about zero?[edit]

How about scrapping it entirely? It seems utterly superfluous for any use and the current template guidance — viz., "use the initial letters" — invites simply typing the last name over again? Certainly clarify that, preferably with some justification for using the field in the first place. LlywelynII (talk) 06:50, 15 October 2011 (UTC)Reply[reply]

It creates link to [[Wikisource:Authors-X#Xx]] and filters to [[Category:Authors-Xx]] — billinghurst sDrewth 13:14, 15 October 2011 (UTC)Reply[reply]


Can the optional Signature field be added? Such as Image:Bill Clinton signature2.png.—Markles 11:57, 5 December 2008 (UTC)Reply[reply]

Have you thought about just adding it to the Notes field? For the vast bulk of the authors, we are not even go to have an image, so a signature file would be even less used. -- billinghurst (talk) 15:02, 8 August 2009 (UTC)Reply[reply]
I actually think it’s a good idea to have as an optional field Lemuritus (talk) 22:15, 1 January 2020 (UTC)Reply[reply]

Image description[edit]

Is there a way to change the default image description? T0lk (talk) 14:31, 8 August 2009 (UTC)Reply[reply]

Having a look at the script, you should be able to use | image_caption =One mean dude -- billinghurst (talk) 15:01, 8 August 2009 (UTC)Reply[reply]
  • I don't see the point in having a thumb style with a caption, or i don't like it - whatever. Can we have image_caption as an option instead. Cygnis insignis (talk) 09:10, 18 December 2009 (UTC)Reply[reply]
??? I was informing of what is the existing name field for images, by default it picks up the author name, and this is the means to override the default. Nothing about thumb style captions. If you want to blank it I am sure that there is a means to utilise that field. billinghurst (talk) 09:30, 18 December 2009 (UTC)Reply[reply]
I think that adding a caption should be the option, the page title is already given: big, bold and in the url. Cygnis insignis (talk) 10:00, 18 December 2009 (UTC)Reply[reply]
I am ambivalent, though given a choice if it is there, I would centre it, however, the care factor is lacks significance to pursue it.

Link to Hebrew[edit]

Hebrew Wikisource has a parallel template. You may want to link to it:he:תבנית:מחבר

Done billinghurst (talk) 12:47, 1 September 2009 (UTC)Reply[reply]

Hidden maintenance categorization[edit]

Would it be possible to have the the template place the pages into hidden categories if the template is missing some parameters? Because of the large number of author pages we have, it is becoming difficult to browse through them to find those that are missing links to Wikipedia, Commons, or Wikiquote, or ones that don't have pictures. I think to help make this aspect of maintenance more scalable, we should have the template place respective hidden categories for these missing parameters.—Zhaladshar (Talk) 13:58, 13 September 2009 (UTC)Reply[reply]

Good thoughts, and I would add to this missing birth and death years, especially where there is not a death date, and the birth date is >110 years ago.-- billinghurst (talk) 14:43, 13 September 2009 (UTC)Reply[reply]

Minor tweak[edit]

"Indexes" should be "indexes". Moondyne (talk) 04:45, 23 December 2009 (UTC)Reply[reply]


How do people feel about this recently added feature? My view is that it is nearly always non-functional, plus it masquerades as an interwiki link. I'd prefer it wasn't there. Hesperian 11:28, 13 January 2010 (UTC)Reply[reply]

There does need to be more tweaking of the Special:IndexPages capability before it becomes useful through such a link. I think a discussion on how we provide better local customised finding aids would be beneficial, it may be indexes, or it may be a combination of items. billinghurst sDrewth 21:25, 13 January 2010 (UTC)Reply[reply]

use of empty (defaultsort = ) is problematic[edit]

I have seen the behaviour where defaultsort is added to the header, yet left empty. This causes the name to be sorted under A presumably for Author. Is it possible that where the parameter is added, but left empty that it is ignored? billinghurst sDrewth 16:36, 21 January 2010 (UTC)Reply[reply]

Centering image text?[edit]

Wondered on people's opinion on to have the text that sits with an author's image to be centered rather than having the text left aligned. billinghurst sDrewth 03:11, 31 January 2010 (UTC)Reply[reply]

No objection from me. Hesperian 04:48, 1 February 2010 (UTC)Reply[reply]
Donebillinghurst sDrewth 12:50, 20 February 2010 (UTC)Reply[reply]

1911 link[edit]

Is it possible to add a parameter here to link to 1911 Encyclopædia Britannica? --Amir E. Aharoni (talk) 14:57, 28 February 2010 (UTC)Reply[reply]

Which way for the link? About from the page, or inbound to the page? There is {{EB1911 link}} from Author page to EB1911 articles. We would generally do that in a +++Works about author+++ section. — billinghurst sDrewth 15:13, 28 February 2010 (UTC)Reply[reply]
I referred to what you did at Author:Heinrich Friedrich Wilhelm Gesenius (thanks). If linking from the author page to 1911 Britannica is common, maybe it can be made more structured by putting it straight into the Author template. --Amir E. Aharoni (talk) 15:18, 28 February 2010 (UTC)Reply[reply]
I think that this is the preferred means, as giving priority to EB1911 isn't always superior to DNB, SBDEL, JE, CE, ..., see Category:Internal link templatesbillinghurst sDrewth 16:44, 28 February 2010 (UTC)Reply[reply]

Worldcat identities[edit]

Has adding a field to the template for Worldcat Identities been considered? We use a template for that when appropriate in Wikipedia (in External links), but I would think it would be even more useful in Wikisource for Authors. Flatterworld (talk) 13:27, 18 August 2010 (UTC)Reply[reply]

Sounds like a possible thing to do, how accurate and how many variations would be required? I am thinking that we have variations in speelingwink, dates and pen names, so we may need several variables to do the job accurately. What output would you want displayed? Also, should we also consider openlibrary.org in that mix? — billinghurst sDrewth 08:01, 19 August 2010 (UTC)Reply[reply]
Addendum, you can always add it as a standalone template that can be entered into the Notes field, just like {{OCLC}}. — billinghurst sDrewth 08:04, 19 August 2010 (UTC)Reply[reply]
For an example of what it would look like (wherever it lives) I added '[/lccn-n94-112934 Barack Obama] at WorldCat' to 'See also' at the end of Author:Barack Obama. I also added the relevant WorldCat Identity link to William Shakespeare and Author:Samuel Langhorne Clemens, the latter as an example of how WorldCat Identities apparently combines real names and pen names into a single 'identity'. Adding openlibrary.org (or anything similar) is fine with me, as long as it isn't just replicating the WorldCat search under its own name. I did find http://errol.oclc.org/laf/n94-112934.html is the OCLC 'Linked Authority File' for Barack Obama, and noted the similarity with the WorldCat link. For whatever that's worth. ;-) Flatterworld (talk) 23:59, 19 August 2010 (UTC)Reply[reply]
Forgot: I couldn't find an example of a page with a 'Notes' field for an author, just for specific works, so I'm not sure what you meant by that. Flatterworld (talk) 00:14, 20 August 2010 (UTC)Reply[reply]
My oops. The _description_ field is what I was meaning, as it pretty much covers additional notes, and additions of sister links and additional components like {{edition}}. — billinghurst sDrewth 21:47, 20 August 2010 (UTC)Reply[reply]
I'm sorry, but I don't know what you're talking about. 'Edition' has nothing to do with an author, it's only relevant to a specific work. I'm talking about a parameter for Template:Author because WorldCat Identities are for people (including every author), not individual works. Flatterworld (talk) 22:38, 20 August 2010 (UTC)Reply[reply]
Yep, it was early morning and the cells were freewheeling and not gripping. We put a number of descriptive components in _description_. I believe that the requested addition is a significant move within the author header. As the link can be added without addition as a field, that for the discussion for the addition that there should be an indication of the benefits that would be expected by the change, and if the discussion is to take place that it should be announced to the community at Wikisource:Scriptorium. My major concern is that at WorldCat they have duplicates of some authors, and that component of Identities is still in beta. — billinghurst sDrewth 13:22, 21 August 2010 (UTC)Reply[reply]
I emailed OCLC to ask how long they expect to remain in beta. Do you have an example of duplicate identities for some authors? I was looking for such, but I found the opposite: a list of 'other names' in the righthand column of the one Identity page. Flatterworld (talk) 16:33, 21 August 2010 (UTC)Reply[reply]
1 for 1 … http://www.worldcat.org/identities/find?fullName=william+trollope
I see the confusion. :-) He only has one - http://www.worldcat.org/identities/lccn-nr93-40387 - although there are many search results. That's the only one I'm talking about (lccn in the url, orange 'Controlled Personal Identity'). The other icons are just ordinary 'Personal Identity' icons. ('Corporate Identity' and 'Controlled Corporate Identity' are other icons I've noticed.) btw - I found this about NACO, which may be what the Authority File is about. There's a link about pseudonyms listed as 'New'. Flatterworld (talk) 18:47, 22 August 2010 (UTC)Reply[reply]
At this point in time, my suggestion would be to build a template that can sit within the description field of the author template, that locates itself nicely and unobtrusively, and does not impact upon the sister links. If it works; is useful; and meets community approval, then in can be incorporated into the author header. At that time we would get a bot to do a merge of the separate template and header to make it seamless. If it doesn't work in the header, then we can amend the header and get a bot to run through and put within a See also section. — billinghurst sDrewth 23:48, 22 August 2010 (UTC)Reply[reply]
Sounds great! We now have Template:WorldCat id, example at William Shakespeare, based on Template:OCLC as used in Fables (La Fontaine/Wright). Thanks! :-) (edited: I posted on George Orwell III's Talk page because after he reverted my Barack Obama example, I thought he would leave it alone once he knew we were having a discussionabout it. Since he insists on deleting it, I changed the example. Hard to have a discussion about an example without the actual, um, example. ;-) ) Flatterworld (talk) 15:27, 23 August 2010 (UTC)Reply[reply]

 Oppose -- This sort of external pointer belongs at the end of the list of hostable works if at all. George Orwell III (talk) 15:34, 23 August 2010 (UTC)Reply[reply]

Oppose to add any new link inside the author template. It must go, iff we accept that sort of link, in the body of page. Phe (talk) 15:38, 23 August 2010 (UTC)Reply[reply]

C-SPAN et al[edit]

I would also suggest adding fields to the template for any major sources of video or audio speeches. For example, many U.S. national and state politicians are on C-SPAN. For example, Barack Obama. I would think this would be helpful for anyone wanting to watch a speech as well as read the wikified transcript. Flatterworld (talk) 14:56, 18 August 2010 (UTC)Reply[reply]

In the Author template, we have generally created links to works to sister sites for collective reference, where there is the chance for each field to be applied to each author. Alternatively in the body (outside of the template) we have had specific reference to specific works. Where a specific work is available at Commons in audio or video, there are templates {{listen}}, {{speaker}}, {{watch}} that can be used to identify such works. To add to the template it needs to almost be a universal option for all authors, and I don't think that your suggestion can be universally applied. There are other options &heelip;
We have allowed generous SEE ALSO links to specific collections, and I would think that would be the means top achieve what you are after, rather than addition to the template, especially where the vast bulk of our (dead) authors are never going to have their vision or audio, and never going to be available at C-SPAN. — billinghurst sDrewth 04:35, 19 August 2010 (UTC)Reply[reply]
I like the SEE ALSO idea - didn't realize that existed. :-) As these are collections of 'works', they do belong on the Author page and that sounds like a good grouping label. Since I started this discussion using Barack Obama as an example, I went to Author:Barack Obama and added 'Barack Obama at C-SPAN' at the end of the page (I changed the 'Sources' to 'See also', assuming it was left over from an older format - I also wikified some fields in the previous links.) Is that what you meant? I can create a Template:C-SPAN (I assume the format is about the same as those in Wikipedia), or one multi-purpose template depending on how many 'collections' we might want to include. The advantage of the latter - something like Template:Collections - is that it would act as a sort of 'checklist' so nothing important is left out. otoh, at this point I don't know of any similar or even vaguely related links, so unless you know of any I should perhaps just create Template:C-SPAN for now and we can keep the alternative in mind if things change? Flatterworld (talk) 00:10, 20 August 2010 (UTC)Reply[reply]
created {{cspan link}} and added it to Category:External link templates. Just the basic information that you had, which can be amended as necessary. — billinghurst sDrewth 22:35, 20 August 2010 (UTC)Reply[reply]
Thanks. Just as well, as Wikisource input and output formatting for Templates is apparently quite different from Wikipedia's. Flatterworld (talk) 16:34, 21 August 2010 (UTC)Reply[reply]
Not sure what you mean by that as numbers of our templates are direct lifts at a point of their evolution at WP, though sometimes they don't evolve once lifted; so if you think it is important, you may wish to elucidate. — billinghurst sDrewth 12:28, 30 August 2010 (UTC)Reply[reply]

Updating to embed Template:plain sister?[edit]

The template {{plain sister}} has recently been embedded into {{header}}, and it has been suggested that we could look to embed into {{author}} to allow for a few extra sister links and to synchronise on the shorter linking format (sans _link). I don't have a concern, though maybe the architects of the template have opinions on adv/disadv. — billinghurst sDrewth 12:28, 30 August 2010 (UTC)Reply[reply]

The more that I see plain sister in action, I do prefer its output compared to the earlier presentation form we are currently using in Author. — billinghurst sDrewth 11:54, 1 November 2010 (UTC)Reply[reply]
I would also like to see this changed. The way I would execute the change-over is as follows:
  • Modify the template so that both "wikipedia_link" and "wikipedia", etc are valid parameters, and add all fields piped into {{plain sister}} Done
  • Also cause using "_link" parameters to sort into temporary tracking categories, so we can pick up any stragglers.
  • Maybe also consider having tracking categories for the new parameters, so we can use category intersection to find pages with and without links to other wikis. Done
  • Modify the preloader to use the new parameters in all new author pages. Done
  • Change the author templates (of which there are about 8000) to use the new parameters. Done
  • This would probably be done by bot, since the changes are straightforward. I have a nice script for extracting and parsing header templates, it wouldn't be hard to convert to formatting author templates.
  • In the process, "commons_link" entries would be made into "commonscat" or "commons" depending on whether or not it is preceded by "Category:"
  • Once all author templates are done (check the cats), change again so that "_link" templates are no longer valid. Perhaps a changeover period when using "_link" templates throws an error to "train" users. Done
Inductiveloadtalk/contribs 05:30, 9 December 2010 (UTC)Reply[reply]
I must confess, I'm not partial to either implementation, but I'm not opposed to a change if others like the new one. I like the smaller real estate used in the original implementation ("See biography" vs. "sister projects: Wikipedia"). But if we do decide to implement it, I think Inductiveload's is a pretty good manner of doing so.—Zhaladshar (Talk) 14:42, 9 December 2010 (UTC)Reply[reply]
The change is complete. Pages using the old categories will sort into Category:Author pages using deprecated parameters Inductiveloadtalk/contribs 02:08, 11 January 2011 (UTC)Reply[reply]
All pages are also categorised according to which interwiki links they do or don't have. This will presumably be helpful for people wishing to compile lists of authors without links. See Category:Author maintenance. The old parameters no longer work, but they do throw an error message. This can be removed at some future time when Category:Author pages using deprecated parameters has been empty for long enough. Inductiveloadtalk/contribs 02:31, 11 January 2011 (UTC)Reply[reply]

Parameter "invert_names"[edit]

I have added the new parameter for the cases where we want to display the author's name in the order of "Lastname Firstname" like Author:Li Xishen. The parameter takes a "yes" answer to work, all other responses should have no effect. Examples of the tests are in Template:Author/testcasesbillinghurst sDrewth 05:30, 5 June 2011 (UTC)Reply[reply]

Note. I have not identified other author pages are going to need this correction made to them. Others may wish to undertake that for pages that they know exist. I do not believe that there is any easy means to identify such pages beyond the eyeball test. — billinghurst sDrewth 05:33, 5 June 2011 (UTC)Reply[reply]

Isn't it kind of superfluous to add a new field, since the current template already manages without it? (See the advice above.) Although such a suggestion is a good example of why the template advice should be corrected. LlywelynII (talk) 06:52, 15 October 2011 (UTC)Reply[reply]
It didn't, which is why I did the adaptation. — billinghurst sDrewth 13:08, 15 October 2011 (UTC)Reply[reply]

Recoding birth and death year catgorisation[edit]

I have changed the way this template categorises by birth/death year. It now runs through a subroutine/subpage of the template (Template:Author/year) which itself calls several other templates in order to cope with the various formats used with this template. I made this change because, while randomly browsing authors, I noticed that any year before 1000 AD wasn't being processed properly (it needed to be a four digit number and "BCE" was causing an error). I've taken the opportunity to enhance the functionality of the template while fixing this bug. The subroutine currently handles all years including BCE, decades, centuries and some approximate dates (all should be categorising by era as well, rather than dumping all non-numeric dates into Ancient authors).

There are still a few more things I have to do. However, I have made this change live now because the areas that need more work were broken in the previous version anyway. This version at least fixes some things that were broken without making the situation worse. The subroutine can technically handle "circa YYYY" dates but this needs to be implemented properly (it may require entering those years in a "circa/YYYY" format to be readable). The "YYYY?" format is not supported yet either.

If there is any other problem caused by this, please let me know. - AdamBMorgan (talk) 00:01, 1 August 2012 (UTC)Reply[reply]

Microformat issue[edit]

The use of class="label" in

<span id="ws-description" class="label">{{{description}}}</span>

is causing the wrong data to be included in the template's emitted microformat. I'd like to change it to class="note". Would that conflict with anything else? Is it used for styling? Pigsonthewing (talk) 14:56, 18 January 2013 (UTC)Reply[reply]

This is still an issue. Pigsonthewing (talk) 22:47, 1 June 2014 (UTC)Reply[reply]
@Pigsonthewing: I don't see that class=Label" is being used in common.css, or subsidiary files. I suppose the lack of reaction has been due to the lack of information to the nature error/problem, and the nature of the solution. I don't see an issue with the proposal, though css is not my area of expertise. — billinghurst sDrewth 16:32, 2 June 2014 (UTC)Reply[reply]
The change was made here, and it referenced oldwikisource:Wikisource:Microformat, though that has no classes given as examples. fr:Modèle:Titre is cited for comparison at that page, though it uses different class names. I am not certain if any of them are used, or were added in anticipation of some use. I would think that if we were going to have classes that we would look to have familiarity of terms through the interlanguage wikis. If there is no dispute to that, then I would think that we could update. My question is though, is the problem the use of "label" or the requirement for "note"? I think that frWS use of classes that align more closely with the field has some value. — billinghurst sDrewth 16:49, 2 June 2014 (UTC)Reply[reply]
[reply to both] This markup forms a Microformat; it conveys semantic meaning, and is not used by CSS for styling. Pigsonthewing (talk) 17:27, 2 June 2014 (UTC)Reply[reply]
Done see new reply below - although I'm pretty sure a class definition named "notes" will turn up to be an issue one day, so I replaced "label" with "desc" AND "notes" for now. Personally, I'd rather mirror the {{{ }}} param in question and use that as the basis for "naming" - I'd go with just "desc" ( for (((description}}} ) in spite of the fact this case dealt with the blue notes field of this (& most other) header templates. -- George Orwell III (talk) 23:07, 2 June 2014 (UTC)Reply[reply]
Thank you, but that should be class="note" (singular) not class="notes" (plural). The class names are defined in the hCard specification. Pigsonthewing (talk) 00:05, 3 June 2014 (UTC)Reply[reply]
Ah-ha! After reviewing the spec., I totally get it now and made the correction to note accordingly. Scratch my previous comment in light of that review as well. -- George Orwell III (talk) 01:42, 3 June 2014 (UTC)Reply[reply]

{{Str left}} - returning substrings since 2009[edit]

So, it seems that {{str left}} has been around since 2009, though the functionality may have been around in 2006. This template returns an arbitrary number of characters from the beginning of a text string. For example, the first two letters of the lastname parameter. D'oh!

I'm not sure of any real reason not to implement this code. It will increase server load a little bit on page saves, but the template/magicword is there to be used in cases like this, after all. It will save the silly busywork of re-entering the first two letters of the surname, and will also eliminate associated typos and mistakes. However, I'd like to ask if anyone sees something I'm missing. --Eliyak T·C 12:14, 9 January 2014 (UTC)Reply[reply]

The first two letters are auto-generated in the Author creation gadget, so I am not certain what is the issue that you raise. If you are talking about automating the presentation of those letter, then no, they are overwritten on occasions, so the lost of that functionality would not be desirable. — billinghurst sDrewth 12:48, 9 January 2014 (UTC)Reply[reply]
You are of course correct. Obviously, I don't create that many author pages. So, never mind! -Eliyak T·C 21:25, 9 January 2014 (UTC)Reply[reply]

Organizations? VIAF?[edit]

Hi, I'm curious how to handle authors that are not natural persons, but organizations (e.g., [[Author:Wikimedia Foundation]], [[Author:Olmsted Brothers]]).

Also -- should there be a paramater for VIAF ID in the Author template? -Pete (talk) 23:59, 21 January 2014 (UTC)Reply[reply]

Use override_author = corporate body for the parameter, and if you think it worthwhile create a portal = Corporate body and list on such a page; whereas joint authors would have separate pages as each person's date of death individually impacts duration of copyright. We have done such an approach for publishers of works.

Re VIAF, we use {{authority control}} like most sites, and it appears at the bottom of a page, before the copyright tag, and there is a gadget available to make it easier. I thought that it was mentioned on the style guide, though it may not be, I will check it when I am able. — billinghurst sDrewth 04:04, 22 January 2014 (UTC)Reply[reply]

Thanks, that helps. -Pete (talk) 18:14, 22 January 2014 (UTC)Reply[reply]

Persondata fetching from Wikipedia[edit]

I am sure that the template used to fill in the birthyear, deathyear and wikipedia parameters automatically if there is an Wikipedia page with the same name (except the author prefix), the first two probably by the w:Wikipedia:Persondata. But it seems to be no longer working. Anybody knows why? Solomon7968 (talk) 16:58, 11 February 2014 (UTC)Reply[reply]

That worked with a gadget, MediaWiki:Gadget-TemplatePreloader.js, using Wikipedia's API. I can't see why it would have stopped, the gadget hasn't been changed and the API appears to be working (as far as I can tell). (FYI, it actually seems to search the page for the Category:xxxx births, for example, to get the birth/death data). - AdamBMorgan (talk) 17:40, 11 February 2014 (UTC)Reply[reply]
As a note, Phe has recently updated his script, so if people have turned on components of this in their common.js file, and utilise the gadget, then they will be better served using the gadget than specially calling the file. — billinghurst sDrewth 16:54, 2 June 2014 (UTC)Reply[reply]

Image adding from WD if not prescribed, plus ...[edit]

I am thinking that we should be auto linking to an image at WD if one exists (WD field exists), which can be overridden by locally assigning the parameter, which maintains the status quo.

With that we can identify where we have the parameter used locally, and where the field does exist at WD, and look to remove locally used image calls; and to also identify where we have an image parameter, and where WD does not have the image loaded, and such backfill the data to WD. — billinghurst sDrewth 00:50, 4 August 2015 (UTC)Reply[reply]

I have done some fiddling, and the presentation here for Author:Harriet Martineau shows with the parameter image existing through pulling from WD. If the image parameter is blank and used in the template no image will show, and if the image parameter is used then that is used. It will need some rewriting of the main template to fix that logic, but conceptually functioning. — billinghurst sDrewth 11:10, 5 August 2015 (UTC)Reply[reply]
Note that it fails if there is more than image listed. — billinghurst sDrewth 13:03, 5 August 2015 (UTC)Reply[reply]
Got it, parameter wins, empty parameter or no parameter then use Wikidata image, rolling to production. — billinghurst sDrewth 14:08, 5 August 2015 (UTC)Reply[reply]

N-ve sister link categorisation[edit]

I have removed the negative categorisation of author pages not having links to WP/WQ/Commons by use of active parameters. For the moment the positive categorisation can remain so we can find where it is being used. We can later look to query processes to identify where authors are listed on WP, though have no author pages. — billinghurst sDrewth 03:07, 27 August 2015 (UTC)Reply[reply]

Proposing on adding DESCRIPTION call to Wikidata where local field is empty[edit]

I am finding numbers of empty description fields when editing authors. As per other empty fields in the template, I am proposing that if the field is empty that we will call the English language description that exists in Wikidata. — billinghurst sDrewth 00:24, 28 October 2015 (UTC)Reply[reply]

Red link if two occurrences on Wikidata[edit]

  • d:Q5791238 [1] have two values for d:Property:P18, resulting in a redlink on Author:William Torrey Harris. Lugusto 19:47, 5 April 2016 (UTC)Reply[reply]
    Mainly as a personal confirmation I altered the ranking between the two available choices and the higher preference one is now being displayed without link problems. Please revert if you need to re-establish the demonstration. AuFCL (talk) 22:22, 5 April 2016 (UTC)Reply[reply]
    @555: As AuFCL states we have a conundrum, we can either take the first image, or the preferred image, hoping that WD manages their data well. I chose to use the preferred image route, and we have category tracking in place where the image load is problematic which I occasionally review and resolve. In short, if it happens again, it can be resolved by editing the WD item. Opinions welcomed. — billinghurst sDrewth 00:04, 6 April 2016 (UTC)Reply[reply]
    To be fair to 555/Lugusto the link processing (essentially assuming [[File:{{#invoke:Wikidata|getValue|P18|FETCH_WIKIDATA}} is always a valid link based only upon whether getValue returns a value or no) is a bit simplistic. If this is readily do-able it might be best to isolate and present only the first link (and set a maintenance category, maybe Category:Author pages with ambiguous Wikidata images?) pending manual setting of a WD preference.

    By the by I simply chose the first portrait on the basis a fuller face might present better on our Author page. There is no guarantee this will not conflict with the current or future needs of a sister project, so having a selection criteria local to the utilising wiki would be far superior to the "Commons"-style preference mechanism currently implemented at Wikidata. But this is probably a fight for another day… AuFCL (talk) 09:19, 6 April 2016 (UTC)Reply[reply]

    We still have the capability to override the WD preference by specifically stating an image within the header template. — billinghurst sDrewth 12:31, 6 April 2016 (UTC)Reply[reply]
    Useful, but still missing the point. I shall try again. The LUA code states:
    -- This is used to get a value, or a comma separated list of them if multiple values exist
    p.getValue = function(frame)
    In other words if this function is to be persisted with; and if a given usage results in a value which contains a comma; then it is programmatically pretty irresponsible to enclose said return within [[]] and hope that the result will still work as a valid link (Big hint: as 555 above has pointed out it does not.) And as far as my (minimal) inspection of {{Author}} extends there is no maintenance category other than Category:Author pages with Wikidata image raised under these circumstances. In other words the error case is buried amongst many pages of valid ones. Isn't this just making unnecessary work for the reviewers? AuFCL (talk) 22:22, 6 April 2016 (UTC)Reply[reply]

Prefix/Suffix for titles/honorifics.[edit]

{{edit protected}}

Add a prefix/suffix parameter, so authors that are primarily known with an honorific or title can be included wihtout needing to put the honorific in the name of page, examples being people with with names of the form like Rev A. Theist , Rt Hon. B. Ritish MP etc..

If this would not be a good idea, I'd appreciate the objections as well :) ShakespeareFan00 (talk) 21:16, 13 August 2016 (UTC)Reply[reply]

I think suffix makes sense for someone like, for example, Martin Luther King Jr or someone with a III (or other numeral) in their name. Instead of prefix, I'd go for "title", but I don't think that really has a place in the actual template. (Arthur Conan Doyle or Sir Arthur Conan Doyle?) --Mukkakukaku (talk) 21:40, 13 August 2016 (UTC)Reply[reply]
What do most works use would be the normal consideration I think. ShakespeareFan00 (talk) 00:23, 14 August 2016 (UTC)Reply[reply]
Not done—the omission of titles and honorifics is deliberate. These items are not a part of someone's name. As a result we disambiguate authors by their dates (see Help:Disambiguation) and not by the various prefixes and suffixes that a person can go through in their publishing career (e.g. Rev.; Rt Rev.; Dr.; Rev. Dr.; Rt. Rev Dr. for an archbishop who has taken a doctorate at some time in their career. And three of those could be extended by Hon. or Rt. Hon. if the same gentleman sat in the Upper House and/or was a member of the Privy Council.) Post-nominals are equally (if not more) problematic. Both of these items can be covered in the notes field, if they are significant. Beeswaxcandle (talk) 00:45, 14 August 2016 (UTC)Reply[reply]
As per Beeswaxcandle. People's honorifics easily change in time, pre and post knighthood, before and after noble titles, etc. Then with Sr/Jr, only usually became pertinent when the son published, if they did, and they are not unique descriptors anyway. Accordingly we decided long ago to keep things simple, and not play guess the pick the right name game … too many instances of duplication were occurring, and the determining of which name was precedent. Add them to the notes if you so wish. — billinghurst sDrewth 02:31, 14 August 2016 (UTC)Reply[reply]
In an ideal world (i.e. one in which we do not actually live) the primary Author: page ought to be the most-fully qualified name of the concerned person known—and in the case of duplicates with any luck appending (birth year-death year) will break the conflict—but without any time-or-context-dependant honorifics. However in the case of commonly-know pseudonyms or titles liberal use of appropriate ancillary redirect records is a nice touch and ought to address the stated objective? (Which "Queen Mary"? (Country? Roman Numerals?) Female authors who published in later life as "Mrs. H. Jones" after having used their maiden names in earlier publications, or who remarried? Per above career progressions: Lieut./Capt. etc.?) AuFCL (talk) 03:17, 14 August 2016 (UTC)Reply[reply]
Thanks for the explanations, request withdrawn. ShakespeareFan00 (talk) 10:59, 14 August 2016 (UTC)Reply[reply]

Refactor date-handling?[edit]

Would love any feedback on: Wikisource:Scriptorium#Refactoring_dates_in_the_author_template. Thanks! Sam Wilson 06:55, 17 January 2017 (UTC)Reply[reply]

Broken microformat[edit]

The microformat emitted by this template does not work. The microformat is marked up (using Author:George Grove as an example) in this code:

<div id="ws-data" class="vcard ws-noexport" style="display:none; speak:none;"><span id="ws-article-id">478885</span><span id="ws-wikibase-id">Q708101</span><span id="ws-name" class="fn">Sir George Grove</span><span class="n"><span class="given-name">Sir George</span><span class="family-name">Grove</span></span><span id="ws-key">Grove,_Sir George</span><span id="ws-image">George-grove.jpg</span><span id="ws-birthdate" class="bday">1820</span><span id="ws-deathdate" class="dday">1900</span><span id="ws-description" class="note">Engineer and later English author specialized in music, authority on Beethoven, Schubert and Mendelssohn. Secretary of the Sociey of Arts; first Secretary of the Crystal Palace Company; editor of <i><a href="/wiki/Macmillan%27s_Magazine" title="Macmillan's Magazine">Macmillan's Magazine</a></i> from 1868 to 1883; first Director of the Royal College of Music (1882-1894)</span>

Specifically, classes: vcard, fn, n, given-name, family-name, bday, dday, note. for others, see the template's documentation.

Note that the code includes style="display:none; speak:none;"

Microformat parsers ignore content which is styled to not display.

The markup (i.e. the classes listed above) should be applied to:

<td class="gen_header_backlink">←<a href="/wiki/Wikisource:Authors-Gr" class="mw-redirect" title="Wikisource:Authors-Gr">Author Index: Gr</a></td>
<td class="gen_header_title"><b>Sir George Grove</b><br />
<td class="gen_header_forelink"></td>
<table class="author_notes">
<ul id="plainSister" style="display:inline-block; font-size:93%; line-height:normal; list-style-type:none; list-style-image:none; list-style-position:outside; border:1px solid #AAA; float:right; clear:right; margin:0.5ex 0.5ex 0.5ex 0.5ex; padding:0.0ex 0.0ex 0.0ex 0.0ex; background:#FFFFFF; background-color:#FFFFFF;">
<li class="sisitem"><span class="sisicon" style="padding-right:1ex;"><a href="/wiki/Special:SiteMatrix" title="Special:SiteMatrix"><img alt="Sister Projects." src="//upload.wikimedia.org/wikipedia/commons/thumb/8/81/Wikimedia-logo.svg/18px-Wikimedia-logo.svg.png" width="18" height="18" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/8/81/Wikimedia-logo.svg/27px-Wikimedia-logo.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/8/81/Wikimedia-logo.svg/36px-Wikimedia-logo.svg.png 2x" data-file-width="1024" data-file-height="1024" /></a></span><a href="/wiki/Special:SiteMatrix" title="Special:SiteMatrix">sister projects</a>: <a href="https://en.wikipedia.org/wiki/George_Grove" class="extiw" title="w:George Grove">Wikipedia article</a>, <a href="https://commons.wikimedia.org/wiki/Category:George_Grove" class="extiw" title="commons:Category:George Grove">Commons category</a>, <a href="https://en.wikiquote.org/wiki/George_Grove" class="extiw" title="q:George Grove">quotes</a>, <a href="https://www.wikidata.org/wiki/Q708101" class="extiw" title="d:Q708101">data item</a>.</li>
Engineer and later English author specialized in music, authority on Beethoven, Schubert and Mendelssohn. Secretary of the Sociey of Arts; first Secretary of the Crystal Palace Company; editor of <i><a href="/wiki/Macmillan%27s_Magazine" title="Macmillan's Magazine">Macmillan's Magazine</a></i> from 1868 to 1883; first Director of the Royal College of Music (1882-1894)
<hr width="5%" />
This <a href="/wiki/A_Dictionary_of_Music_and_Musicians/List_of_Contributors" title="A Dictionary of Music and Musicians/List of Contributors">author</a> wrote articles for <a href="/wiki/A_Dictionary_of_Music_and_Musicians" title="A Dictionary of Music and Musicians">A Dictionary of Music and Musicians</a>.<br />
Articles written by this author are designated in the DMM by the initials "G."</td>

-- Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:33, 26 March 2017 (UTC)Reply[reply]

@Pigsonthewing: which formatting are you suggesting to be used that won't be ignored, though isn't visible from our point of view? — billinghurst sDrewth 13:06, 3 March 2018 (UTC)Reply[reply]

Deactivate addition of "Category:VIAF not on Wikisource‎"[edit]

I think that the VIAF data is now firmly entrenched at Wikidata, and that we can stop monitoring for its absence there. I can see maybe we may still wish to monitor for the presence of that data locally so that it can be removed.

Problem is that I cannot find where we add that categorisation here. Checked the plain sister template and module, and author template and module, without success, so would someone less blind than e please look and see if we an easily remove it. Thanks. — billinghurst sDrewth 13:04, 3 March 2018 (UTC)Reply[reply]

@Billinghurst: Didn't we store all that information in {{Authority control}} when we had it? --16:25, 20 July 2018 (UTC)Reply[reply]

Wikidata image link data islimited to one picture stated[edit]

The automatic link data provided by Wikidata using image(s) stated in the respective item does not work, if there are multiple wikidata:Property:P18 statements. Example: On Author:Nelson Mandela, no image is displayed, as this template tries to display one image using the link of two sperate images on commons stated as P18 on the Nelson Mandela Wikidata item. Rather then manually overriding pages with multiple P:18 statements on the respective Wikidata item, the template could be fixed. Regards --Πτολυσϙυε .-- .. -.- .. 07:30, 20 July 2018 (UTC)Reply[reply]

I'm not seeing this problem, but I don't know whether that's because somebody fixed it, or a browser/operating system-related issue, or what. Is it fixed for you as well? -Pete (talk) 16:17, 20 July 2018 (UTC)Reply[reply]
Hrishikes solved the problem on Wikidata by setting one image to "preferred" rank. --EncycloPetey (talk) 16:23, 20 July 2018 (UTC)Reply[reply]

Living authors category[edit]

Authors with the date of the death missing in Wikidata are categorized as "living authors" in Wikisource. As a result the authors who were active centuries or millenia ago and whose date of death is unknown, such as Author:Zeuxippus, are also categorized as "living". For this reason I suggest people over 90 should not be categorized based just on the fact that Wikidata do not have any record of their death. If needed, they might be categorized manually.

I tried to discuss this at Wikisource:Scriptorium#Living authors category, and was told to raise the topic here. Can User:Samwilson help with it? --Jan Kameníček (talk) 09:56, 13 September 2018 (UTC)Reply[reply]

@Jan.Kamenicek: Yes, I can totally have a look at it. Let me see what's what. I think assuming death over a certain threshold is a good idea. In the case of someone with an approximate birth date, I guess we'd want to assume the latest possible point in time wouldn't we? Anyway, I'll have a play in the sandbox and add some test cases. Sam Wilson 10:13, 13 September 2018 (UTC)Reply[reply]
@Samwilson: Thanks for looking at it. Some people do not have their birth date at Wikidata either, but they can have floruit date (such as Author:Chval Dubánek) or work period start (such as Author:Mansur II), or date of baptism in early childhood. --Jan Kameníček (talk) 10:32, 13 September 2018 (UTC)Reply[reply]
@Jan.Kamenicek: Ah, baptism dates too? I'll have a look. In the meantime, the living-authors bug is fixed. It turns out we were already accounting for people born more than 110 years ago, however there was a bug in how that was calculated (it was using a two-digit date instead of four-digit! how silly; my fault). Fixed now, see what you think. I'll have a look at baptism things; how should that be expressed, with a "bap." suffix? Sam Wilson 11:05, 13 September 2018 (UTC)Reply[reply]
I may have fixed one bug, but I think there is another with relation to 'Living authors'. I'll keep digging. Sam Wilson 11:21, 13 September 2018 (UTC)Reply[reply]
Ah, it was that the death date calculation was adding 'living authors' too, and was being done after birth dates. Fixed now. Sam Wilson 11:28, 13 September 2018 (UTC)Reply[reply]
@Samwilson: The reason why I mentioned the floruit and baptism dates was that they can help to find dead people whose date of birth is unknown too. E.g. Author:Zeuxippus is still categorized as living at this moment. I think that we can assume that people whose floruit or baptism date + 90 < current date are not living, too. --Jan Kameníček (talk) 11:38, 13 September 2018 (UTC)Reply[reply]
The same could be also applied to "work period (start)" or "work period (end)" date. --Jan Kameníček (talk) 11:40, 13 September 2018 (UTC)Reply[reply]
@Samwilson: Now I have noticed that the problem with Zeuxippus was that the floruit date was mentioned only locally at Wikisource and not added to Wikidata. I added it and now it works fine, so thank you very much for fixing the issue! Could you please add also the above mentioned work periods to fix cases such as Author:Mansur II? Maybe the wikidata properties work period (start)=997 and work period (end)=999 could also be manifested at Wikisource as "fl. 997–999"; what do you think? --Jan Kameníček (talk) 12:13, 13 September 2018 (UTC)Reply[reply]

Bibliowiki has changed its name back to Wikilivres[edit]

Could somebody please change the Bibliowiki link back to Wikilivres? Preferably with a capital W. Thank you. Simon Peter Hughes (talk) 13:55, 18 January 2019 (UTC)Reply[reply]

DoneBeleg Tâl (talk) 14:04, 2 March 2019 (UTC)Reply[reply]

Pull description from Wikidata[edit]

Can we amend this template, so that it will pull |description= from Wikidata? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:08, 9 April 2019 (UTC)Reply[reply]

I'd say the template should do this if and only if "description" is empty. Until and unless we can do a good analysis on how existing Wikidata and WS:Author descriptions differ, it would be hasty to assume that one is better than the other (or that both sites have the same needs). But going forward, it would certainly be nice to obviate the step of creating a new description when a perfectly good one exists. Good suggestion. -Pete (talk) 18:00, 2 January 2020 (UTC)Reply[reply]

Bug in template[edit]

In this case fail: Author:Anicius Manlius Severinus Boethius, because from wikidata can be have several files for images, and the template should be fetch only one. Thanks Shooke (talk) 14:35, 30 December 2020 (UTC)Reply[reply]

@Shooke: Head over to Wikidata and select the best image of the multiple to be preferred. — billinghurst sDrewth 04:54, 31 December 2020 (UTC)Reply[reply]

Taking the first image from WD is the wrong option[edit]

Please amend the change where the first linked image at WD is the preferred image. *If* we have to have preference then it should be based on the preference set, not the first listed, especially as changing the order is just a PITA. At least when I introduced the image filter, I discussed the process and sought a consensus, I didn't just change it without consultation. Personally I was happy to fix and set preferences once a week where the images were of equal rank, but now we are often left with the worse image. — billinghurst sDrewth 04:35, 7 March 2021 (UTC)Reply[reply]

@Billinghurst: If you're referring to this change, it was to fix then-broken code. The old version simply dumped out all the images defined on Wikidata, leading to broken image markup (red link instead of an image displayed). The new code makes sure we always get a valid image if there is one. We can certainly investigate ways to also pick the one with the highest precedence on Wikidata rather than just pick the first one, but the problem here is that none of the existing modules for Wikidata access gives us this "for free" so doing that requires a non-trivial amount of new code. Of course, having a function like "giveMeTheHighestPrecedenceImageForWikidataItem(QID)" would seem to be a pretty obvious and common need so getting it added to one of the WD access modules should be feasible.
PS. But keep in mind, the precedence on Wikidata doesn't mean "This is the nicest image". That toggle is related to the truth value of the associated data (image): images marked as lower priority are definitionally incorrect, inaccurate, or for other reasons "less true". The inverse is also true: if all but one portrait of Shakespeare are of plausible but unverified provenance, but one painting has been verified as painted from life (the Kesselstadt death mask, for example), the others would be "normal" and the verified one would be "preferred". Regardless of whether the "preferred" one is a really crap painting in æstethic, artistic, or technical (resolution, contrast, colour balance, etc.) sense. It is dangerous to rely on these priorities as if they are switches to let us (Wikisource) control which image to use when Wikidata has a different definition of what it means, and when other projects (enWP) also run around flipping these switches (one in particular is running around setting a black&white engraving after the painting instead of the portrait itself as "clearly superior". Compare the lede image in Samuel Johnson compared with the current version.) --Xover (talk) 10:20, 7 March 2021 (UTC)Reply[reply]
I think that our templates should take the image with the highest preference rank from WD and if we are not happy with the image chosen in this way, we should be able to overwrite it locally. If there are two or more pictures in WD without their preference rank given, it is better to pick the first one than to have an error message instead. However the page could e. g. be placed into some maintenance category so that some volunteers could check it and set the preference in WD. --Jan Kameníček (talk) 10:38, 7 March 2021 (UTC)Reply[reply]
I know these things, and mentioned bits of this in opening the discussion. Way back when we introduced the image import and its automated selection I said that it would break when there are multiple images, and I said that I would fix these, and I have regularly done these nearly every week, or thereabouts, since. They are categorised into category:Pages with missing files. I have also mentioned this above and at WS:S on multiple occasions. The category itself mentions specifically the means to fix, so can we please return to that functionality, and I will continue with my fixing so that we get the better images. Thanks. PS. We already can override the parameters, and have been able to do so since we converted to data import; it was a pre-condition of the migration process. — billinghurst sDrewth 11:48, 7 March 2021 (UTC)Reply[reply]
Yes, I wanted to point out those bits that are good ideas + suggest categorizing instead of the error message. It is great and very useful that you are willing to check it regularly every week. Checking and fixing the pages in the category would eliminate the several days of displaying the error message. --Jan Kameníček (talk) 13:15, 7 March 2021 (UTC)Reply[reply]
@Billinghurst: Change reverted since there's prior consensus. But you can take some obligatory grumbling and a "Bah. Humbug!" as read. --Xover (talk) 21:36, 7 March 2021 (UTC)Reply[reply]
@Billinghurst: No, I take that back! A lightbulb went on just as I was reverting the prior changes. I think I've now changed it to work the way we want: if there's a preferred image on WD it uses that, and if there isn't it'll use a normal rank image. If there are multiple images of the given rank it'll use the first one of those. Have a look and see if it now does what we want (and if it's still broken just revert it: there's no magic or intervening edits so a plain undo should be safe). --Xover (talk) 21:53, 7 March 2021 (UTC)Reply[reply]

subpages used to have author in the top line[edit]

A recent change of the template has moved the author from the top line next to the title on subpages, to being on the second line of the header, on its own. I don't believe that I seen this change discussed. It used to be on the top line to keep the header depth shorter. — billinghurst sDrewth 12:26, 16 March 2021 (UTC)Reply[reply]

@Billinghurst: I assume this is supposed to be at Template talk:Header? Relax, it was a typo in the module and it's fixed. Inductiveloadtalk/contribs 12:43, 16 March 2021 (UTC)Reply[reply]
D'oh, yes. Thanks. Distracted juggling other things. — billinghurst sDrewth 12:49, 16 March 2021 (UTC)Reply[reply]

Move last_initial to Module:Author?[edit]

I'd like to suggest moving the |last_initial= logic to Module:Author. This is the change that I've got in mind. I've made a new function in that module that handles the linking and categorization, and also looks up family name (P734) from Wikidata if no |last_initial= is given. Sam Wilson 08:53, 27 September 2022 (UTC)Reply[reply]

Actually, while doing this I think we can get rid of the !NO_INITIALS convention, because it doesn't seem to be used anywhere. Or am I missing something? Sam Wilson 09:26, 27 September 2022 (UTC)Reply[reply]
I was missing something, of course. It's used in {{disambiguation}}, and so must be kept. :-) Sam Wilson 02:36, 28 September 2022 (UTC)Reply[reply]
I disagree. The |last_initial= is for indexing in a list of names and does not always use the first two letters of the |lastname=. A good example is Author:Leonardo da Vinci, who is indexed using Le, and not da, or Author:Alcaeus of Mytilene, who is indexed using Al and not of. Likewise, pulling the last name will not work for Author:Plato, who has no |lastname=, nor is there a family name on Wikidata because Plato was not known by a family name. --EncycloPetey (talk) 02:24, 29 September 2022 (UTC)Reply[reply]
@EncycloPetey: That's a good point. But for those situations, it'd still be possible to specify |last_initial= as is currently done. It'd just make it possible to leave it out for cases where it is exactly the same as the first two letters of the last name. Sam Wilson 02:51, 29 September 2022 (UTC)Reply[reply]
Also, we have to consider pen names (versus birth names) where the family name is not part of the indexing name, hyphenated names of married women (potentially the same issue), names of Spanish authors who have two family names, etc. There are enough exceptions that I think it is better to require the parameter for correct indexing. --EncycloPetey (talk) 14:14, 29 September 2022 (UTC)Reply[reply]
@EncycloPetey: There are a bunch of exceptions to the rule, it's true. Those would still need to use the |last_initial= parameter. I'm only proposing that if that that parameter is not set then a value will be looked up from family name (P734). I've had a look, and of the total 35,711 authors there are 1,632 (4.6%) where our |last_initial= does not match the first two characters of family name (P734) (e.g. Author:Adriaen Cornelissen van der Donck we have as 'Do' but Wikidata would have incorrectly as 'Va'). Some of these are errors, I think, such as Author:Albert-Auguste Cochon de Lapparent who we have as 'La' but as the family name is actually 'Cochon de Lapparent' should be 'Co' — but most look like correct exceptions. Still, the number of exceptions is considerably smaller than the number of correctly calculable initials, and it still seems like it'd be good to add this Wikidata look-up to make things easier for people creating author pages. — Sam Wilson 22:16, 28 November 2022 (UTC)Reply[reply]
Would it be possible to prompt the editor with "this name will be indexed as . . . " whenever the header is edited or the template is added / created with a page? --EncycloPetey (talk) 01:53, 29 November 2022 (UTC)Reply[reply]
@EncycloPetey: I'm not sure where that would be displayed? But certainly, previewing (or saving) the page would show them what index letters were to be used; would that be sufficient? We could also add a tracking category. — Sam Wilson 02:58, 29 November 2022 (UTC)Reply[reply]
@Samwilson, @EncycloPetey: Did this discussion peter out?
I know some modules on enWP does something clever to show a big red warning above the edit box on preview, but I don't know how hacky that is. Regular preview also shows what the categories will be so you can see it there too.
It also occurs to me that we (well, mostly Billinghurst, but...) do a whole lot of patrolling of newly created Author: pages, so if the proportion that needs a manual override is reasonable then that might not be an excessively large additional burden on that task. Xover (talk) 13:55, 5 January 2023 (UTC)Reply[reply]

additional logic required => last_initial[edit]

I have recently seen at least one case of use of last_initial causing a redlink. Would it please be possible to add some logic to also have a maintenance category where the cat does not exist. Thanks. — billinghurst sDrewth 05:12, 5 January 2023 (UTC)Reply[reply]

@Billinghurst: I took a stab. Please test whether it behaves as expected: Category:Author pages with missing initials category (1). Note that checking for the existence of anything is computationally expensive in MediaWiki, so for high-transclusion code like this it may have undesirable side-effects (slower updates of categories and transclusions, higher server load, hitting hard limits, etc.). Xover (talk) 09:25, 5 January 2023 (UTC)Reply[reply]
Not certain that it is working. I can understand that the old pages would take time to iterate, I would think that new pages would show quickly, and Author:Sand box doesn't show. — billinghurst sDrewth 13:28, 5 January 2023 (UTC)Reply[reply]
@Xover: Maybe I am just better running a petscan query. petscan:23617963 seems to work fine on this issue. — billinghurst sDrewth 13:37, 5 January 2023 (UTC)Reply[reply]
@Billinghurst: Meh. Yes, you'll have to use Petscan for now because I'm an idiot. See the thread above: last_initial is implemented in the module, but is never called from the template. Implementing this logic in the template is possible, but so painful to do and hard to maintain that it's not something I'd recommend trying. I'll leave the tracking cat code in there so that if we decide to use the Lua version we'll get that functionality too. When I migrate the rest of the template to Lua I'll need to revisit the last_initial code anyway so I'll try to remember this tracking cat issue then. Xover (talk) 13:48, 5 January 2023 (UTC)Reply[reply]
<shrug> It is a dozen pages and who knows how long that took to build-up. I've added it to my occasional task list, so it just becomes a push check, rather than having a polled category. — billinghurst sDrewth 20:57, 5 January 2023 (UTC)Reply[reply]