Wikipedia:Bots/Requests for approval/Monkbot 17

The following discussion is an archived debate. Please do not modify it. To request review of this BRFA, please start a new section at Wikipedia:Bots/Noticeboard. The result of the discussion was

Approved.

Monkbot 17[edit]

Operator: Trappist the monk (talk · contribs · SUL · edit count · logs · page moves · block log · rights log · ANI search)

Time filed: 15:49, Sunday, October 4, 2020 (UTC)

Automatic, Supervised, or Manual: auto

Programming language(s): awb/c#

Source code available: yes

Links to relevant discussions (where appropriate): Help talk:Citation Style 1 § last-author-amp= (now archived)

Edit period(s): primarily one-time with possible occasional cleanup runs after

Estimated number of pages affected: ~61k

Namespace(s): all namespaces categorized by cs1|2

Exclusion compliant (Yes/No): yes

Function details: see bot task 17 page

Discussion[edit]

According to the bot task's description this is exactly what is needed to perform the desired transition of CS1 citation parameters as discussed in the linked thread, and it therefore has my support.

I have a small issue with one of the listed ancillary tasks, though, and would be delighted if this could be addressed before running the bot (the change would be minor): |url-status= should not be deleted if it is set to values other than "live", even if |archive-url= and |archive-date= are empty or missing. While it is true that the template currently changes its behaviour only with archives defined, semantically, the |url-status= parameter belongs to |url= and only secondarily to |archive-url=, and if |url-status=dead indicates a dead URL, this is important information that should not be deleted just because there would, at present, be no archive to switch to.

Otherwise, everything appears to be fine.

--Matthiaspaul (talk) 10:48, 9 October 2020 (UTC)[reply]

Despite its name, and despite that I have argued in the past that it is more closely linked to |url= than it is to |archive-url=, |url-status= is, in reality, a display-control parameter that dictates how cs1|2 renders when given |archive-url=. There were suggestions for something like |archive-display= but, as you see, not adopted. It is not the purpose of cs1|2 to duplicate the functionality of other templates and their en.wiki infrastructures. For dead external links, we have {{dead link}}.

This simple search finds about 75,300 articles that have cs1|2 templates with |url-status=live|archive-url=|.... For all of these, |url-status=live is wholly meaningless. Compare that to this simple search that finds about 165 articles with |url-status=dead|archive-url=|.... These might actually be dead but the 'dead' urls in the two articles from that search that I checked were not dead. Yeah, I know, hardly a sufficient sample, but sufficient to show that the 'dead' designation cannot be trusted. I'm sure that among the cs1|2 templates with 'live' designations, there are in fact, dead urls.

I see no reason not to treat |url-status=<anything> as empty when |archive-url= is empty or missing.

—Trappist the monk (talk) 15:22, 9 October 2020 (UTC)[reply]

Of course, I am aware that the parameter is, at present, only used to switch the rendering of URLs/archive-URLs (if both are present). Still, semantically, the parameter is more than this, and the possible keywords accepted (dead, usurped, etc.) also indicate a potentially more general use to record the status of the URL at the time of inspection. While there is some overlap with the external {{dead link}} template, I consider the purposes to be different - and independent of each other.

There have been proposals and I expect that the functionality of the parameter will be expanded (not necessarily only display-related) somewhen in the future. F.e., if |url-status=dead indicates a permanently dead URL, we could disable the external link so that it cannot accidently point to something already known to be no longer related to the citation. The citation template's new auto-linking feature could then take over and link to some identifier instead while the former |url= would be preserved on source-code level as kind of a pseudo-identifier (and in meta-data, or in a tool-tip) to maintain the integrity of the citation. Of course, I don't know if this particular usage example will be implemented in the future, or another, my point is more that it could, and if the editor who provided the citation deliberately set |url-status=dead to indicate the status (because he used the URL to retrieve the document before it went dead) that info would get lost due to your bot's ancillary task.

Things would be somewhat different if your bot task would actively check and adjust the status, or if each citation would be manually inspected and evaluated by a human. However, mechanically removing this parameter (if it is set to non-default values) blindly, without individual inspection and evaluation, will cause at least some collateral damage. In many cases, this may be don't care (as you suggest above), but I am aware of at least a few citations where I would care and therefore assume that there will be more of them around. I maintain the opinion that bots should be used only for tasks they can reliably & predictably solve to 100% without causing collateral damage (minus an the occasional accidental coding error).

The issue could be solved easily enough simply by removing this particular ancillary task, after all, the bot's main task, for which it should be run and where it cannot cause any damage, is completely unrelated to this.

--Matthiaspaul (talk) 16:44, 13 October 2020 (UTC)[reply]

This venue is not the place to discuss any of the possible cs1|2 futures.

Clearly you and I are not going to agree on the value (or lack of value) obtaining from |url-status=dead when |archive-url= is left empty or is omitted. The template documentation for |url-status= (and its predecessor, |dead-url=) does not and, as far as I know, never has suggested that this parameter may be used as an alternative to {{dead link}} or as some sort of personal aide-mémoire or advertisement to other editors that |archive-url= needs filling; nor should it. Editors who, for whatever reason, use |url-status=dead outside of its documented purpose should not be surprised when the undocumented use is deleted or converted to proper use.

But, in the interest of getting this BRFA unstuck, I will have the bot skip any cs1|2 template that it might have edited if that template has some sort of  on the offhand chance that editors doing this left some sort of note.

—Trappist the monk (talk) 21:11, 13 October 2020 (UTC)[reply]

One hundred manual edits using the bot's code in my history ending at 2020-10-11T1802Z.

—Trappist the monk (talk) 18:32, 11 October 2020 (UTC)[reply]

—Trappist the monk (talk) 14:21, 13 October 2020 (UTC)[reply]

Why are empty parameters being deleted? They serves as prompts that certain parameters look useful, but are not to hand at the time of composition, or are not adequately understood at the time of composition. --RichardW57 (talk) 10:54, 16 October 2020 (UTC)[reply]

You can believe those myths if want. These days, most cs1|2 templates aren't written by hand but by tools like ve and reftoolbar. There are, no doubt, some still written by editors who copy the skeleton from a template doc page and fill the parameters they need. Those kinds of editors likely cleanup after themselves. Gone are the days of many many editors manually writing cs1|2 templates. Leaving empty parameters without also leaving an  to explain why they are empty is meaningless to other editors so only serve to clutter the wikitext. Such clutter should be removed.

—Trappist the monk (talk) 12:43, 16 October 2020 (UTC)[reply]

Approved for trial (50 edits). Please provide a link to the relevant contributions and/or diffs when the trial is complete. Primefac (talk) 16:43, 16 October 2020 (UTC)[reply]

Trial complete.

Nothing out of the ordinary though I did discover that the skip-when-has-html-comment skipped-out too early; that has been fixed. The test run was supposed to be 50, but awb did 51 edits. I've seen that before. Results here.

—Trappist the monk (talk) 18:09, 16 October 2020 (UTC)[reply]

@Trappist the monk: could you exclude pages containing {{Taxonbar? I'm planning on piggybacking some general & WP:Tree of Life-specific trivial edits on the back of this (discussion here). ~ Tom.Reding (talk ⋅dgaf) 12:22, 18 October 2020 (UTC)[reply]

That can be accomplished without modification to the bot's code.

—Trappist the monk (talk) 13:53, 18 October 2020 (UTC)[reply]

Approved. Primefac (talk) 13:26, 19 October 2020 (UTC)[reply]

The above discussion is preserved as an archive of the debate. Please do not modify it. To request review of this BRFA, please start a new section at Wikipedia:Bots/Noticeboard.