Wikipedia:Templates for discussion/Log/2021 August 26

2016 Republican Primary Templates

edit
The following discussion is an archived debate of the proposed deletion of the template(s) or module(s) below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the template's talk page or in a deletion review).

The result of the discussion was keep , noting that these templates are each used on multiple pages. Primefac (talk) 11:20, 4 September 2021 (UTC)[reply]

All 56 of these templates should be substituted on the articles it's used on. Don't see any reason for keeping these templates as is. One-off election results should not be on their own templates. --WikiCleanerMan (talk) 21:33, 26 August 2021 (UTC)[reply]

The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the template's talk page or in a deletion review).
The following discussion is an archived debate of the proposed deletion of the template(s) or module(s) below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the template's talk page or in a deletion review).

The result of the discussion was relisted on 2021 September 4. plicit 04:44, 4 September 2021 (UTC)[reply]

The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the template's talk page or in a deletion review).
The following discussion is an archived debate of the proposed deletion of the template(s) or module(s) below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the template's talk page or in a deletion review).

The result of the discussion was delete. plicit 04:12, 3 September 2021 (UTC)[reply]

merged with the "French Catholic School Board election" sections of the 2010 and 2014 articles. Frietjes (talk) 18:03, 26 August 2021 (UTC)[reply]

The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the template's talk page or in a deletion review).
The following discussion is an archived debate of the proposed deletion of the template(s) or module(s) below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the template's talk page or in a deletion review).

The result of the discussion was delete. plicit 04:12, 3 September 2021 (UTC)[reply]

merged with the "French Public School Board election" sections of the 2010 and 2014 articles Frietjes (talk) 17:59, 26 August 2021 (UTC)[reply]

The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the template's talk page or in a deletion review).
The following discussion is an archived debate of the proposed deletion of the template(s) or module(s) below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the template's talk page or in a deletion review).

The result of the discussion was delete. plicit 04:12, 3 September 2021 (UTC)[reply]

Not enough items to warrant a template, even more so when only the parent articles belong. Amaury15:56, 26 August 2021 (UTC)[reply]

The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the template's talk page or in a deletion review).
The following discussion is an archived debate of the proposed deletion of the template(s) or module(s) below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the template's talk page or in a deletion review).

The result of the discussion was delete. plicit 04:09, 3 September 2021 (UTC)[reply]

Unused after it was replaced on the articles it was used on. --WikiCleanerMan (talk) 15:39, 26 August 2021 (UTC)[reply]

Delete per nom. Gonnym (talk) 10:16, 1 September 2021 (UTC)[reply]
The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the template's talk page or in a deletion review).
The following discussion is an archived debate of the proposed deletion of the template(s) or module(s) below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the template's talk page or in a deletion review).

The result of the discussion was relisted on 2021 September 5. Primefac (talk) 00:14, 5 September 2021 (UTC)[reply]

The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the template's talk page or in a deletion review).

Political shortname and color templates

edit
The following discussion is an archived debate of the proposed deletion of the template(s) or module(s) below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the template's talk page or in a deletion review).

The result of the discussion was merge to Module:Political party. Planning in TFDH or the module talk page as necessary. Izno (talk) 12:40, 6 September 2021 (UTC)[reply]

Propose merging Political shortname and color templates with Module:Political party.
This is a nomination for the ~6000 Template:.../meta/shortname and ~10k Template:.../meta/color templates that currently exist on Wikipedia. Over the years there have been dozens of discussions (e.g. 1, 2, 3) in which unused templates are deleted and used templates are kept. There is an almost constant turnover (it seems, anecdotally) of these templates being created, used, and then deleted.

My proposal here is to delete all of these templates after merging their information into one centralised table (and no, I'm not going to tag them all, because that's a waste of time, but I will be cross-posting to the Village Pump and a few other locations). This has worked for sports team colours, country team names, and other huge-group-of-one-word-info type content. Since all of these meta templates are pulled from other templates like Template:Election box candidate with party link it should be fairly straightforward to update these templates following creation of the storage module. Following that, party names can be added or subtracted as necessary in the Module without wasting time at TFD.

And to head off a couple perennial complaints about this sort of thing: yes, it won't happen overnight (that's why we have WP:TFDH). No, there should not be any disruption or loss of functionality after implementation (we're just replacing a template call with a module call). It will be a huge investment of time, but the benefits over the long-term will be appreciable. Primefac (talk) 11:32, 26 August 2021 (UTC)[reply]

  • VPR, E&R, and PLT notified. Primefac (talk) 11:41, 26 August 2021 (UTC)[reply]
  • I would imagine this module will be massive (I'd guess a minimum of 500kb). Will it be a drag on servers to constantly look up this module whenever someone accesses an article using it? Some (such as those with large election results tables or lists of constituency results) may have dozens of meta templates transcluded, which might mean multiple lookups to it (I don't really know how this works, so happy to be assured it's not an issue)!
Also, this will need many templates and modules to be recoded to use the module instead. Someone will need to compile a list of these to ensure they are sorted out.
One possible alternative – previously the idea of transferring the colour templates to Wikidata has been suggested, which would help ensure consistency across different language Wikipedias, as different colours being used on different languages causes issues for election maps and parliamentary diagrams on commons. Might this be worth exploring for the colours (this would not work for shortname templates due to the language issue)? Cheers, Number 57 12:04, 26 August 2021 (UTC)[reply]
I imagine that even the shortname can be moved to wikidata and have a localized version for any language, however the issue is that en.wiki editors don't want to use wikidata. Gonnym (talk) 12:11, 26 August 2021 (UTC)[reply]
I don't think the main module itself will be all that large, as it can be split into subpages, much like Module:Sports color has a dozen subpages that hold each individual sport. It could be done alphabetically (as seen in places like Module:High-use) or some other split method; this way each individual call will be relatively small since it will be pulling from a (relatively) small pool of values. Primefac (talk) 13:08, 26 August 2021 (UTC)[reply]
I'd like to get a definitive answer on whether this will cause server drag. Even if split into smaller modules of say 10kb, would an article with 100 uses be an issue for the server?
Also, Gonnym, I'm not sure about the issue with some editors not wanting to use wikidata. If it's agreed in this discussion that the colours are moved to wikidata and then they are moved across and then drawn automatically from there, what would it matter if some editors here didn't want to use it? Is there an actual guideline on Wikipedia that we shouldn't use wikidata? If not, then I can't see a problem with making a decision to do so. I think this is already done for a few things, like official website or twitter templates? Number 57 13:54, 26 August 2021 (UTC)[reply]
Johnuniq would be able to give a more definitive answer, but based on my experience with {{country alias}} and the replacement of the Country IOC alias XYZ pages in the past, I would say the answer to your question is "no". Primefac (talk) 14:18, 26 August 2021 (UTC)[reply]
Another thing to note is that (assuming proper implementation), the data module will only be parsed once per page, regardless of how many different colors are used, due to use of mw.loadData. Also, why are we worrying about things like server drag? * Pppery * it has begun... 17:32, 26 August 2021 (UTC)[reply]
Wikidata would be interesting, but party colors are usually original and are only decided by wikipedians for illustrating purposes; so I'm not sure it would be a good idea to put these colors on wd. I would however suggest working on a module that would later be ported to Abstract Wikipedia (once it's done), which would also allow to centralize the colors from all wikis onto a single module. Julio974 (Talk-Contribs) 13:59, 26 August 2021 (UTC)[reply]
I'm not sure that's a major issue, as the colours tend to be used on many different language versions – creating consistency is a good thing. Also, while I agree some of them definitely are OR, many are properly based on party logos or flags. Number 57 14:08, 26 August 2021 (UTC)[reply]
Support the new module, replacement and the deletion of everything else. I actually noted this myself as a comment in a recent TfD. As these templates just hold one piece of data, ideally, a system similar to wikidata would have been used, but since en.wiki does not trust that data source then creating a module to hold that data is the second best option we have. How exactly the coding details of such a template are done (a submodule for each country or one single module) I leave for the implementer. As this has already been done before (replacing hundred/thousand of templates with one single module) and has proven to be a successful system, I see it as great improvement over the current system. Gonnym (talk) 12:09, 26 August 2021 (UTC)[reply]
If we were to have sub-modules, I think doing them by alphabetical order would be the best way – I don't think by country would work as I'm not sure how the module would understand that a party name (like Likud) was associated with a certain country. Also, there would be a huge number of parties starting with some letters (e.g. M for Movement), so it might end up being subsets of letters. Number 57 13:54, 26 August 2021 (UTC)[reply]
To make my support comment a bit more clearer, I support both a Wikidata implementation option and a module option, whichever route the community here wishes to go. Both have their own advantages and disadvantages but both are a far more superior system than the current one. Gonnym (talk) 11:13, 31 August 2021 (UTC)[reply]
Perhaps it would help to have a draft module containing at least a few of the to be replaced templates? Regarding Wikidata I am quite neutral, to be honest I don't think OR on Wikidata is worse than OR onwiki. Jo-Jo Eumerus (talk) 15:27, 26 August 2021 (UTC)[reply]
If I get a chance in the next few days, I might throw up a proof-of-concept. My only concern is putting too much work into it and having that effort wasted. Primefac (talk) 18:55, 26 August 2021 (UTC)[reply]
  Done. {{#invoke:Political party|fetch|Liberal Democrats (UK)|color}} and {{#invoke:Political party|fetch|Liberal Democrats (UK)|shortname}} give #FAA61A and Liberal Democrats, same as {{Liberal Democrats (UK)/meta/color}} and {{Liberal Democrats (UK)/meta/shortname}}. The invocations would be wrapped up either in a single template that allowed for output type switches, or multiple templates (one for each type of output, e.g. {{Political party/shortname}}). Primefac (talk) 01:05, 27 August 2021 (UTC)[reply]
Conditional support provided that it doesn't cause server drag. It would be far easier than the current system of thousands of disparate templates, which are formatted in many different ways and have various naming conventions. I'd like the wikidata option to be explored further if possible though, as that seems like a sensible solution considering each political party will have a wikidata item anyway. PinkPanda272 (talk/contribs) 16:08, 26 August 2021 (UTC)[reply]
  • My first preference would be for the information to be centralized on Wikidata, but if we're not ready to go multilingual yet, I guess a module is the next best option. Module:College color might be precedent. The merger sounds fine so long as it doesn't cause issues. {{u|Sdkb}}talk 17:31, 26 August 2021 (UTC)[reply]
  • Assuming there is no good reason to oppose this, the merge would be desirable as having all related information in one place is much better as it allows easy monitoring of bold edits and other vandalism. Using Wikidata is the worst option as it makes detection of bad changes virtually impossible, and the performance and error handling can be bad. The reason to worry about performance is that editing an article that pushes the servers too hard is a PITA—a single "show preview" can take over 30 seconds. I don't know how large the central module with all the information would be, but it would probably be manageable and not a performance problem. If this merge goes ahead, there should be a discussion somewhere for planning because getting a good design would be important. One drawback of a central list is that if the data needs to be frequently updated, each tiny adjustment purges the cache for thousands of articles (all those that use the central module). I don't know if that would be a problem. Johnuniq (talk) 02:23, 27 August 2021 (UTC)[reply]
  • This proposal will require significant and complex changes. Ideally I would have wanted a bit more testing and work with regards to the implementation, but it is completely understandable not wanting to spend significant time doing something that may never be used. I trust Primefac to do this well in which case I don't see any reason this can't be a significant improvement with easier maintenance without performance issue. One thing I feel we definitely should bring up however is that the module will be template protected while the current subpages are mostly unprotected. I think this is fine and will remove vandalism or unsourced changes on these templates but it will also require most changes to go through edit requests. --Trialpears (talk) 13:59, 27 August 2021 (UTC)[reply]
  • I would like this to happen, but I am somewhat concerned with implementation. For one, a lot of places use these templates in their current form and many existing templates will need to be reworked to accommodate the change. Secondly, many of these templates are created every week - having all of that go through TPERs would be a somewhat painful process, as well as potentially risking breaking tens of thousands of pages whenever updating the data module (every edit introduces a chance to make a mistake). I'd be willing to help keep on top of this, just something to consider. If someone is willing to actually implement the changes necessary, and write a script to do the migration, then I don't have a strong objection to this. Elli (talk | contribs) 19:50, 27 August 2021 (UTC)[reply]
    On your first concern - this is something we do almost every time a template is listed at WP:TFDH: we plan out what needs doing, and make sure everything is working before and during implementation. Step 1 might be creating the module, but Step 2 is identifying where the extant templates are being used. Step 3 is figuring out how to modify those templates to not give any major impact, and THEN we get to Step 4 of actually implementing/using the new module. Throughout the process there are usually checks and double-checks. Very, very rarely does anything go so sideways we can't quickly fix it.
    On your second concern: I am not particularly bothered with being "flooded" with TPERS. {{rint}} is hit with about one TPER per week, and there might be the occasional backlog, they all get processed eventually. In this case it's even easier, because it's literally just adding a name to a list, and we might even end up just doing "once-a-week updates" if they really do get that bad. The folks who do TPERs are trusted to know how to properly amend a template, and I'm envisioning a check-and-balance system much like {{rint}} where proposed changes must be sandboxed and tested before they'll be implemented. Maintenance is the least of my concerns! Primefac (talk) 01:33, 28 August 2021 (UTC)[reply]
    @Primefac: fair. It's not that I don't trust people implementing TPERs - everyone makes typos on occasion, so having to routinely edit a Lua module transcluded in tens of thousands of pages really just doesn't seem optimal. (also, I know how the holding cell works - I just wanted to make sure everyone was aware of the magnitude of the changes here) Elli (talk | contribs) 02:13, 28 August 2021 (UTC)[reply]
  • found Template:Political party color (and the less broad-in-scope Template:Canadian party colour), which seems to be a non-templated version of what is proposed? although it just does the coloring. Frietjes (talk) 21:45, 27 August 2021 (UTC)[reply]
    Yeah, that's essentially how {{rint}} works, and I did think about just making one mega-switch, but with the sheer number of these party names I feel like it would be a massive template. If it's not going to be that large I'd be okay with a Template instead of a Module. Primefac (talk) 01:33, 28 August 2021 (UTC)[reply]
    At 2700 lines long, {{rint}} (in my opinion) is already too long for template code. Gonnym (talk) 08:53, 28 August 2021 (UTC)[reply]
    Might be time to kick it to a module as well... Primefac (talk) 10:34, 28 August 2021 (UTC)[reply]
  • I agree with sdkb that it would be nice if a lot of this could be moved to and imported from wikidata. Regardless, seems like a good idea and thanks Primefac for taking this on. Wug·a·po·des 17:57, 28 August 2021 (UTC)[reply]
The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the template's talk page or in a deletion review).