Jump to content

Wikipedia:Templates for discussion/Log/2014 March 15

From Wikipedia, the free encyclopedia

March 15

[edit]
The following discussion is an archived debate of the proposed deletion of the template 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). No further edits should be made to this section.

The result of the discussion was delete Plastikspork ―Œ(talk) 03:33, 23 March 2014 (UTC)[reply]

Template:Location map start (talk · history · transclusions · logs · subpages)
Template:Location map marker (talk · history · transclusions · logs · subpages)
Template:Location map end (talk · history · transclusions · logs · subpages)

Not used in any articles. Redundant to Template:Location map+ Jackmcbarn (talk) 23:12, 15 March 2014 (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). No further edits should be made to this section.
The following discussion is an archived debate of the proposed deletion of the template 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). No further edits should be made to this section.

The result of the discussion was delete Plastikspork ―Œ(talk) 03:34, 23 March 2014 (UTC)[reply]

Template:Location map skew (talk · history · transclusions · logs · subpages)

Unused and unneeded, as the {{location map}} family of templates can now handle the skew parameter. Jackmcbarn (talk) 23:06, 15 March 2014 (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). No further edits should be made to this section.
The following discussion is an archived debate of the proposed deletion of the template 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). No further edits should be made to this section.

The result of the discussion was moving to a sandbox, since discussion has continued at Template talk:Location map, and since there is consensus here to improve the current template, rather than use this one. Plastikspork ―Œ(talk) 22:12, 8 April 2014 (UTC)[reply]

Template:Location map quick (talk · history · transclusions · logs · subpages)

Orphaned. Created as a less-featured fork of Template:Location map for better performance, but since a rewrite in Lua, Template:Location map is now better performance-wise than this template. Jackmcbarn (talk) 21:55, 15 March 2014 (UTC)[reply]

The {location_map_quick} is used in prior revisions of pages where it was removed without consensus. -Wikid77 10:04, 6 April 2014 (UTC)[reply]
  • Keep to support all browsers as faster/simpler than Lua: The broken, slower Template:Location_map, although re-written to use Lua, still shifts the map-markers lower on some browsers, with the shift worse near the bottom edge of maps. In performance timing tests, of course the simpler Template:Location_map_quick has often run faster than Lua because the markup is extremely simple, as needed years ago to be quick, and cranking up the Lua/Scribunto interface is a slow, time-consuming process, beyond just the extraordinary complexity to bypass a few lines of markup which rapidly scales and overlays a marker div section on a map. To examine another faster-than-Lua template, see: Template:Str_left which has never used a slow Lua module; it has been a pervasive myth that "Lua is always faster" (not true), because Lua must handle the slow Scribunto interface, which is limited to the 10-second Lua timeout cut-off. As for "templates not used" then beware people who systematically remove templates and then claim hey, "No page is using that template" as among key "wp:Reasons to avoid in a deletion discussion". I've found people remove a template from 80 or 220 pages to slant the claim, "No one uses the template anyway". So finagling the trick of "template not used" is one of the most deceptive methods by which users are deceived into thinking a template has no real purpose, because it is very difficult to show the distortion from where a template was formerly used, before being frantically removed from dozens/hundreds of pages to slant the viewpoint. Anyway, rather than deleting this quick map-marker template which works for all browsers, the other map templates should be fixed first to work correctly, even if they are slower. -Wikid77 (talk) 12:19, 16 March 2014 (UTC)[reply]
Just as in Wikipedia:Templates for discussion/Log/2014 March 14#Template:Location map all, I've still yet to see a concrete example of Template:Location map being broken in any of the ways you describe. If you demonstrate with a side-by-side example that it does something wrong that one of your forks does right, I'll fix it immediately. Jackmcbarn (talk) 14:59, 16 March 2014 (UTC)[reply]
See side-by-side example below. Wikid77 10:04, 6 April 2014 (UTC)[reply]
  • delete or move to a sandbox or to userspace (although, it probably already exists in userspace). Frietjes (talk) 19:48, 24 March 2014 (UTC)[reply]
  • Comment with example maps: Below is a comparison of mapping Syracuse, Sicily (Italy), to show how the locator dot is too low in the Lua-based map (left-side) on some browsers:
{{Location map quick | Italy
| width      = 200
| float      = right
| caption    = Syracuse in Italy (Location map quick)
| alt        = Map of Italy with dot at Syracuse
| label      = Syracuse
| label_size = 100 <!--100% of normal size-->
| position   = left
| background = orange
| mark       = Orange_pog.svg
| marksize   = 6 <!--size in pixels-->
| lat_deg    = (37+5/60)
| lon_deg    = (15+17/60)
}}

{{Location map/sandbox quick| Italy | width = 200 | float = right | caption = Syracuse in Italy (Location map quick) | alt = Map of Italy with dot at Syracuse | label = Syracuse | label_size = 100 <!--100% of normal size--> | position = left | background = orange | mark = Orange_pog.svg | marksize = 6 <!--size in pixels--> | lat_deg = (37+5/60) | lon_deg = (15+17/60) }}

Map of Italy with dot at Syracuse
Map of Italy with dot at Syracuse
Syracuse
Syracuse in Italy (Location map)
Also, the Lua-based map might trigger "Script error" for the latitude "lat_deg = (37+5/60)" whereas the extremely fast {location_map_quick} allows the expression for latitude 37+5/60, while running faster, and displaying the map dot correctly on any browser. -Wikid77 10:04, 6 April 2014 (UTC)[reply]
Actually, it appears that it's too low in Location map quick (in the water in fact), and correct in Location map, at least in Firefox 28 and Chrome 33. I'll get a screenshot momentarily. See File:Location map quick.png. Jackmcbarn (talk) 13:31, 6 April 2014 (UTC)[reply]
I just tried a similar test, noticing some of the examples on the documentation page are seriously out, comparing one of them to Location map. And the same's happening, Location map quick is getting positions wrong: User:JohnBlackburne/maptest. This is with Safari 7.0.3.--JohnBlackburnewordsdeeds 15:25, 6 April 2014 (UTC)[reply]
  • Algorithm is in {Location_map_any} for any browser: The placement of the dot by {location_map_quick} is slightly lower on Firefox browsers, so I will adjust the markup to use the algorithm from {location_map_any}.
    Syracuse (Italy)
    Syracuse
    Syracuse
    Syracuse in Italy (Location map many/sandbox2)
    All these massive, catastrophic deletions have convoluted the discussion, and the support for any browser is in Template:Location_map_any, and I will update {Location_map_quick} to use that method. The solution is really quite trivial, but I guess I will have to write it. -Wikid77 (talk) 15:51, 6 April 2014 (UTC)[reply]
Per previous TfD consensus, Template:Location map all was moved to Template:Location map many/sandbox2. Your move of it back to regular template space as Template:Location map any went against that consensus. In any case, if you're going to fix another template, please fix the main one, Module:Location map (although it doesn't seem broken), rather than fixing more of your forks. Jackmcbarn (talk) 15:57, 6 April 2014 (UTC)[reply]
Also, Template:Location map many/sandbox2 is slightly too high (though admittedly a lot better than Template:Location map quick. Jackmcbarn (talk) 16:04, 6 April 2014 (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). No further edits should be made to this section.
The following discussion is an archived debate of the proposed deletion of the template 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). No further edits should be made to this section.

The result of the discussion was no consensus Plastikspork ―Œ(talk) 06:18, 8 April 2014 (UTC)[reply]

Template:Infobox Rome episode (talk · history · transclusions · logs · subpages)

Only 22 transclusions. Redundant to {{Infobox television episode}} (Transclusion count: 8,119). Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 20:15, 19 February 2014 (UTC)[reply]

  • Delete and replace with standard.--eh bien mon prince (talk) 12:44, 20 February 2014 (UTC)[reply]
  • Subst and replace with {{Infobox television episode}}. "Delete" is a simplistic solution that removes two encyclopaedic parameters, |setting= and |time_frame=. These should still be available when using Infobox television episode. However, these are low use so including them "as is" is probably undesireable. At {{Infobox television season}} I solved this by adding some generic parameters. --AussieLegend () 04:26, 15 March 2014 (UTC)[reply]
  • Delete I've boldly added some generic parameters to {{Infobox television episode}}[1] and converted all Rome episodes to use that template so this infobox is now unused. Ironically, this is something the nominator could have done. --AussieLegend () 06:07, 15 March 2014 (UTC)[reply]
    • I reverted your addition of three blank fields to the television episode infobox. Making similar changes to an infobox with 8252 transclusion without first starting a discussion was ill-advised, especially since using blank parameters is itself controversial.--eh bien mon prince (talk) 02:39, 16 March 2014 (UTC)[reply]
      • The addition of those fields was an extremely minor edit that doesn't break any articles. Unfortunately, by reverting you've broken all of the infoboxes in articles that used {{Infobox Rome episode}}. If you want to revert the changes, please make sure you revert the episode article changes first. As for blank parameters, these are used in many infoboxes, such as {{Infobox character}}. --AussieLegend () 03:01, 16 March 2014 (UTC)[reply]
        • It would be a good thing if you fixed your own mess, but if you don't of course someone else will have to take care of it. Substituting all instances of a template while the TfD is still ongoing was, again, a very poor decision, especially when the outcome of the discussion is by no means obvious.--eh bien mon prince (talk) 03:24, 16 March 2014 (UTC)[reply]
          • Why would you revert that? It would appear to me that if Aussie took the time to do all the work on this and it basically solves all the problems, then how was it controversial to you? It rids the need of using the infobox Rome episode template and uses the regular template but also keeps the parameters needed for those particular articles. JOJ Hutton 03:42, 16 March 2014 (UTC)[reply]
            • This is something that needs to be discussed before rolling out the change in template space. Aussie should have used the sandbox to illustrate his proposal, as is commonly done before making non-trivial changes to widely used templates.--eh bien mon prince (talk) 03:49, 16 March 2014 (UTC)[reply]
              • That's misleading at best. Infobox television episode has only 39 watchers and a proposition wouldn't go anywhere. The last discussion there was over a year ago and Pigsonthewing's request from 11 months ago has gone unanswered because TV project members (the end users of this template) aren't really interested. We've seen this before at {{Episode list}} and {{Infobox television season}}. The changes were extremely minor and didn't break any of the articles that I checked so it shouldn't have been a problem. --AussieLegend () 03:59, 16 March 2014 (UTC)[reply]
                • All kinds of high-visibility templates have low-traffic talk pages, that doesn't give you the permission to 'boldly' change them however you please, only to threaten an edit war and falsely plead that they were "extremely minor" changes when someone reverts you.--eh bien mon prince (talk) 04:35, 16 March 2014 (UTC)[reply]
                  • Please do not make baseless allegations. I did NOT threaten an edit-war. Anyone can make bold edits. In fact we encourage it on a daily basis. The change to the template was simply the addition of three generic parameters that do not interact with any other parameters and did not break any articles (unlike your reversion) so the change was extremely minor. Please, try sticking to the facts. --AussieLegend () 08:00, 16 March 2014 (UTC)[reply]
  • keep as a wrapper, the substituted version is a mess, especially the repeated large episode list. Frietjes (talk) 17:18, 15 March 2014 (UTC)[reply]
Calling it a mess is an exaggeration. It's only a couple of extra parameters and as the series is ended, it's not something that will need to be changed again or added to new articles. Editors seem to have no trouble with something similar in multiple season articles for television series that used to have individual infoboxes with low transclusion counts prior to conversion to use {{Infobox television season}}, and most of these series are still running. As for the episode list, this was a temporary measure that has now been fixed. I should probably also suggest that if you want to see a mess, have a look at the state of the infoboxes prior to the substitution. Now that was a mess. --AussieLegend () 19:49, 15 March 2014 (UTC)[reply]

Relisted to generate a more thorough discussion so a clearer consensus may be reached.
Please add new comments below this notice. Thanks, Plastikspork ―Œ(talk) 21:06, 15 March 2014 (UTC)[reply]

If this is turned into a wrapper for IB television episode, I have no objections to keeping it.--eh bien mon prince (talk) 02:41, 16 March 2014 (UTC)[reply]
There have been numerous discussions about this sort of thing in the past and you've generally supported the "delete" option. In this case so do I, as we really don't need a wrapper that only caters for 22 articles. --AussieLegend () 03:01, 16 March 2014 (UTC)[reply]
It's a little creepy that someone would keep track of my voting record, but no, I don't usually support the deletion of wrapper templates.--eh bien mon prince (talk) 03:55, 16 March 2014 (UTC)[reply]
Please don't be paranoid. We've been involved in many of the same discussions. --AussieLegend () 04:03, 16 March 2014 (UTC)[reply]
  • Keep As the changes that I made have been reverted, I have no choice but to change my vote at this time. At the TV project we've had previous discussions and wrappers and other forks have received no support. Numerous forks of {{Episode list}} were deleted and the same was done at {{Infobox television season}} so we should be consistent across the TV project. If we have to have this template, we have to have it. --AussieLegend () 08:33, 16 March 2014 (UTC)[reply]
  • we should really be asking whether or not the information presented using the blank parameters is needed at all. for example, in the Rome episode infobox, we have setting, time frame, and link. time frame could be merged with setting, since the setting includes both the time and the place. the link is not needed, since it is in the external link section, and most of them are actually broken in this particular case. so, I could see adding a single additional 'setting' parameter, which would then make the Rome box redundant. I would also question the need for the episode list, since there is already the previous/next links and the navbox at the bottom of the page. as a side note, most of the Rome articles are unsourced and pure plot summary and would probably not survive AfD. so, I now think we could probably delete this one. my main objection to deleting it was the addition of the blank parameters to the television episode infobox. Frietjes (talk) 17:55, 16 March 2014 (UTC)[reply]
    • If we merge time frame and setting as you suggest (which I oppose because the contents of each field is different - they should remain separate), we still have the problem that Infobox television episode doesn't have an equivalent parameter. Inclusion of the episode list is supported by Infobox television which includes a parameter, |Season list= specifically for linking to "a template containing a list of episodes in that season". That the episode articles would not survive AfD is not a given, you'd be surprised what survives and I've seen far worse articles survive. In any case, it's not something that is relevant to this discussion, which is about the template and not the articles. What is your opposition to the blank parameters? There are any templates that use blank parameters and editors don't seem to have problems with them. In recent years the TV project has generally rejected wrappers and other forks of the templates we use so blank parameters seem the only way around this. --AussieLegend () 01:53, 17 March 2014 (UTC)[reply]
  • Delete. I see no reason this series is different than any other. 117Avenue (talk) 05:20, 19 March 2014 (UTC)[reply]
It's not, but I tried converting this to use Infobox television episode and it was reverted. --AussieLegend () 06:32, 19 March 2014 (UTC)[reply]
You were reverted because you made undiscussed changes to {{Infobox television episode}} and tried to use the excuse that all the Rome templates would be broken if your change was not accepted, not because this infobox cannot be deleted.--eh bien mon prince (talk) 14:12, 21 March 2014 (UTC)[reply]
Apparently undiscussed changes are acceptable when somebody else does it.[2][3] I note that you haven't reverted those. However, you've missed the point that until we can actually replace this template with Infobox television episode we can't really delete it. --AussieLegend () 15:14, 21 March 2014 (UTC)[reply]
Yes we can. Like I said, this series is no different, so it needs no special fields. 117Avenue (talk) 02:23, 23 March 2014 (UTC)[reply]
Discussion at WT:TV supports the inclusion of special fields. --AussieLegend () 03:18, 23 March 2014 (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). No further edits should be made to this section.
The following discussion is an archived debate of the proposed deletion of the template 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). No further edits should be made to this section.

The result of the discussion was delete Plastikspork ―Œ(talk) 03:34, 23 March 2014 (UTC)[reply]

Template:Cleveland sports curse (talk · history · transclusions · logs · subpages)

This template is pure WP:NOR Delete Secret account 18:41, 15 March 2014 (UTC)[reply]

Delete per nominator....William 22:47, 15 March 2014 (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). No further edits should be made to this section.
The following discussion is an archived debate of the proposed deletion of the template 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). No further edits should be made to this section.

The result of the discussion was delete Plastikspork ―Œ(talk) 03:56, 23 March 2014 (UTC)[reply]

Template:CBE Classic MVP navbox (talk · history · transclusions · logs · subpages)

MVP of Run of the mill preseason college basketball tournament - there are probably 30-40 that occur each year. WP:CBBALL has never supported templates of this type and, frankly, get the complaint that too many college basketball templates exist as it is. Also, there would never be a parent article for the CBE Classic MVP and there is no guarantee the subjects would be notable as not all college athletes are. Rikster2 (talk) 14:58, 15 March 2014 (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). No further edits should be made to this section.
The following discussion is an archived debate of the proposed deletion of the template 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). No further edits should be made to this section.

The result of the discussion was delete Plastikspork ―Œ(talk) 03:56, 23 March 2014 (UTC)[reply]

Template:Conference USA Men's Basketball First Team (talk · history · transclusions · logs · subpages)

Subject of templates (an all-conference team for a single year) is not significant enough for an infobox. We have previously deleted templates for a year's All-NBA team in the past, which is much more significant. There is no article for a single year's All-Conference team, nor would such an article pass AfD. Rikster2 (talk) 14:51, 15 March 2014 (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). No further edits should be made to this section.
The following discussion is an archived debate of the proposed deletion of the template 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). No further edits should be made to this section.

The result of the discussion was delete Plastikspork ―Œ(talk) 03:56, 23 March 2014 (UTC)[reply]

Template:2013 ITF Men's Circuit (talk · history · transclusions · logs · subpages)

I merged the function in with Template:ITF Men's Circuit seasons so that we don't have 16 additional navigational templates when one would suffice. Aureez (Talk) 12:41, 15 March 2014 (UTC)[reply]

  • Delete. Agree. Good solution. 12:47, 15 March 2014 (UTC)
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). No further edits should be made to this section.
The following discussion is an archived debate of the proposed deletion of the template 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). No further edits should be made to this section.

The result of the discussion was delete Plastikspork ―Œ(talk) 03:57, 23 March 2014 (UTC)[reply]

Template:Aqui se encuentra Iglesia Bíblica Casa de Dios) (talk · history · transclusions · logs · subpages)

An unsalvageable text in Spanish. GregorB (talk) 11:45, 15 March 2014 (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). No further edits should be made to this section.
The following discussion is an archived debate of the proposed deletion of the template 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). No further edits should be made to this section.

The result of the discussion was userfy Plastikspork ―Œ(talk) 22:22, 8 April 2014 (UTC)[reply]

Template:Cite web/auto (talk · history · transclusions · logs · subpages)

Created as a fork of Cite web, its author created an essay wp:AUTOFIX advocating for it and been similarly pushing it on User talk:Jimbo Wales but has attracted little interest and no support. It's now being added to articles, totally unnecessarily, as errors should be fixed. It's incorrectly documented (it uses Cite web's documentation) so anyone finding in an article it will have a very hard time finding out why it exists and when it should be used. No way is this ready for deployment. JohnBlackburnewordsdeeds 11:42, 15 March 2014 (UTC)[reply]

  • Keep with new documentation: I have created a specific documentation subpage (at 16:00, 15 March 2014) to explain the features of Template:Cite_web/auto, which is a co-template of wp:CS1 Lua-based template {{cite web}}, supporting all the same parameters in the same order. There has been some confusion because {cite_web/auto} was intended to be merged as the next release of {cite_web}, but due to prolonged discussions about wp:autofixing cites, the current version has remained a co-template with parallel additions of the same new changes from the /sandbox version of {cite_web}. Hence, {cite_web/auto} is ready now for deployment, but just as a separate co-template for additional testing among users who wish to suggest changes, without any impact to the live version of {{cite web}}. The discussions have indicated how a direct merge into {cite_web} would raise concerns at this point. -Wikid77 (talk) 16:38, 15 March 2014 (UTC)[reply]
  • delete, forking the template just adds to confusion. Frietjes (talk) 17:41, 15 March 2014 (UTC)[reply]
    • There has been some confusion, and Template:Cite_web/auto is not a "fork" but rather a co-template of wp:CS1 Lua-based template {{cite web}}, supporting all the same parameters in the same order; there is no "forking" involved. The disconnect came because {cite_web/auto} was intended to be the next release of {cite_web}, but discussions have raised concerns about the amount of autofixing to perform with cite parameters. -Wikid77 23:55, 15 March 2014 (UTC)[reply]
  • Comment. Why isn't this template explicitly marked as experimental (with {{Experimental template}} or otherwise)? GregorB (talk) 17:47, 15 March 2014 (UTC)[reply]
      • Well, {cite_web/auto} is not really "experimental" but rather a complete template and the intended next release of {cite_web}, being tested as a co-template to avoid reformatting the 1.63 million pages which use {cite_web} until options have been finalized. -Wikid77 (talk) 23:55, 15 March 2014 (UTC)[reply]
    • As Wikid77 created it, and as usual has decided it's ready for deployment, without any testing, without gaining consensus, without even trying to determine if there's any need for it or whether there's a better solution, such as improving existing templates not forking them.--JohnBlackburnewordsdeeds 22:01, 15 March 2014 (UTC)[reply]
      • My point being: if this in fact is an experimental template, why not mark it as such, rather than delete it? GregorB (talk) 22:23, 15 March 2014 (UTC)[reply]
        • Because that doesn't work: it was done before with one of these templates but it still kept being used, requiring a second lengthy TfD to deal with it: Wikipedia:Templates for discussion/Log/2013 June_27#Template:Cite quick. Better to deal with this once and for all, delete or move so it can't be used in articles.--JohnBlackburnewordsdeeds 22:32, 15 March 2014 (UTC)[reply]
          • Now I'm a bit confused because in the nom you said the template is not "ready for deployment"; of course, it will never be ready if it gets deleted. I believe what you meant to argue here was that the template (more precisely, its implementation) stands no chance of being accepted by community. Would this be your position? GregorB (talk) 23:14, 15 March 2014 (UTC)[reply]
            • If it's deleted but then someone wants to work on it it can be undeleted to user space, and I've no problem with it being userfied instead of deleted. I think it stands no chance of being accepted as a template, but someone might find it useful in some way - it has been suggested a bot might be a better solution, so the work done researching this problem won't be wasted, while other fixes might be rolled into existing templates.--JohnBlackburnewordsdeeds 23:53, 15 March 2014 (UTC)[reply]
  • Userfy to allow development and experimentation in sandboxes; and use to upgrade/ replace the existing version, only once there is consensus. Otherwise, delete. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 20:34, 15 March 2014 (UTC)[reply]
    • Because I developed the original module for the Lua-based cite templates, this template {cite_web/auto} is a careful extension of the current {cite_web}, rather than a sandbox for "experimentation" or such. The intention is to allow more users to try using {cite_web/auto}, and further expand its autofix methods, then upgrade the current {cite_web} to reformat the related 1.63 million pages over a period of 6 weeks, but allow {cite_web/auto} to use newer, better algorithms for autofixing cite parameters, in advance of upgrading {cite_web} at each stage. The logistics of upgrading a mega-template, used in almost 2 million pages, require using co-templates to deploy improved features while not triggering continual reformat of millions of pages all year long. -Wikid77 (talk) 23:55, 15 March, 17:03, 17 March 2014 (UTC)[reply]
  • Keep for now or userfy. I don't necessarily support the idea of autofixing. Browsers utilize HTML autofixing, often comprising a large portion of code. XHTML DTDs were supposed to help remove autofixing. I still think such errors should be fixed in the transclusion, but if they can be autofixed in the meantime, preferably adding a tracking category, (without really reducing performance for non-autofixed cites,) I could see keeping this as an option for field testing purposes. Eventually, these autofixes could be merged temporarily or permanently, or simply dropped as "not worth it." I think it should be done on a per autofix basis, (possibly grouping similar ones, such as those correcting misspellings,) because it will be easier on those trying to make a decision. —PC-XT+ 01:14, 17 March 2014 (UTC)[reply]
    • Tests have revealed the autofixing will correct multiple errors: The topic of autofixing started last year, as a general concept, but the live template {cite_web/auto} has shown how autofixing a first cite error can help to autofix other secondary errors, and completely salvage a cite with multiple issues. A common case is a URL parameter missing the "url=" keyword, and then triggering error "accessdate requires url" as a secondary error. Another common case is a failure to find parameter "last=" perhaps invalid as upcased "Last=" and then all authors fail to be listed because of the first author's name parameter. The overall goal is to handle a user's written cite and turn it into a workable, logical cite to verify the associated text, but then a sub-goal might be to edit the page and change parameter names to match formal spelling as autofixed to appear to readers. -Wikid77 (talk) 17:03, 17 March 2014 (UTC)[reply]
      • Right. I think there is a place for it, possibly even as a separate template only used to replace citations with errors. There will always be creative errors that a human must figure out, but some of these autofix errors can be hard for some editors to even identify. Of course, this will only help cite web, but it is one of the more common templates to have these errors, so it is a good candidate for autofix field testing. —PC-XT+ 00:00, 18 March 2014 (UTC)[reply]
        • But why would an editor, rather than fix an error, replace {{cite web}} a template that leaves the error in place, "fixing" it so it looks OK which means other editors won't notice it? A web citation is the easiest error to fix as you can check the source yourself. An invalid URL is much easier to identify than an invalid ISBN, doi or other code. A dead link can often be replaced with a live one after a search, or an archive link. Sometimes a better source can be found in the page history, or on another page. If an error can't be fixed by an editor considering all these possibilities it certainly can't be fixed by a template.--JohnBlackburnewordsdeeds 00:19, 18 March 2014 (UTC)[reply]
          • There is no excuse for relying on this template if you know how to fix the error. I understand testing it out in preview, or saving for demonstration purposes, but it shouldn't be left that way. That would just be lazy, and is why I am not necessarily in favor of autofixing. The errors should be fixed. Also, automated fixes should be checked. This template seems to categorize autofixes for humans to check, so I don't think it is asking to be misused. —PC-XT+ 15:16, 18 March 2014 (UTC)[reply]
Even veteran editors make typos in cite parameters: Along with checking for common misspelled parameter names, I also checked how often newcomers were causing errors, but I found many cite errors from long-term users, perhaps because they add more cites to pages (than newcomers do). I, personally, have saved some pages with a typo in the cites, when lacking time to completely proofread the whole page. All the typos add up, and soon 20-40 more pages, each day, have small cite typos. Plus, for the past few years, some Bot programs have been renaming duplicate parameter names as "unused_data=" in over 350 pages. Hence, the autofixing can also reduce cite error messages in pages edited by veteran editors or Bot programs. -Wikid77 01:11, 22 March 2014 (UTC)[reply]
I'm starting to lean towards merging this into the main template, as it was originally intended. I think it adequately warns when it applies the autofixes. The only question I have is: how much server load is there for unfixed citations? I gathered a collection of references that use cite web, then converted them to use test templates to compare:
Extended content

The list of references I used in the test is at User:PC-XT/sandbox/cite web test. I tried each template 5 times:

This is with cite web, mostly for calibration purposes, as I found a closer control:

CPU time usage	2.068 seconds
Real time usage	2.133 seconds
Preprocessor visited node count	4857/1000000
Preprocessor generated node count	19662/1500000
Post-expand include size	305647/2048000 bytes
Template argument size	2472/2048000 bytes
Highest expansion depth	11/40
Expensive parser function count	3/500
Lua time usage	0.519/10.000 seconds
Lua memory usage	2.66 MB/50 MB

CPU time usage	2.424 seconds
Real time usage	2.497 seconds
Preprocessor visited node count	4857/1000000
Preprocessor generated node count	19662/1500000
Post-expand include size	305647/2048000 bytes
Template argument size	2472/2048000 bytes
Highest expansion depth	11/40
Expensive parser function count	3/500
Lua time usage	0.775/10.000 seconds
Lua memory usage	2.66 MB/50 MB

CPU time usage	2.148 seconds
Real time usage	2.223 seconds
Preprocessor visited node count	4857/1000000
Preprocessor generated node count	19662/1500000
Post-expand include size	305647/2048000 bytes
Template argument size	2472/2048000 bytes
Highest expansion depth	11/40
Expensive parser function count	3/500
Lua time usage	0.780/10.000 seconds
Lua memory usage	2.66 MB/50 MB

CPU time usage	2.412 seconds
Real time usage	2.485 seconds
Preprocessor visited node count	4857/1000000
Preprocessor generated node count	19662/1500000
Post-expand include size	305647/2048000 bytes
Template argument size	2472/2048000 bytes
Highest expansion depth	11/40
Expensive parser function count	3/500
Lua time usage	0.780/10.000 seconds
Lua memory usage	2.66 MB/50 MB

CPU time usage	1.540 seconds
Real time usage	1.598 seconds
Preprocessor visited node count	4857/1000000
Preprocessor generated node count	19662/1500000
Post-expand include size	305647/2048000 bytes
Template argument size	2472/2048000 bytes
Highest expansion depth	11/40
Expensive parser function count	3/500
Lua time usage	0.484/10.000 seconds
Lua memory usage	2.66 MB/50 MB

The following measurements use my sandboxed test version of cite web (not the autofix one) User:PC-XT/sandbox/cite web as a control for the following test:

CPU time usage	2.364 seconds
Real time usage	2.437 seconds
Preprocessor visited node count	4857/1000000
Preprocessor generated node count	19685/1500000
Post-expand include size	305647/2048000 bytes
Template argument size	2472/2048000 bytes
Highest expansion depth	11/40
Expensive parser function count	3/500
Lua time usage	0.734/10.000 seconds
Lua memory usage	2.66 MB/50 MB

CPU time usage	2.428 seconds
Real time usage	2.530 seconds
Preprocessor visited node count	4857/1000000
Preprocessor generated node count	19685/1500000
Post-expand include size	305647/2048000 bytes
Template argument size	2472/2048000 bytes
Highest expansion depth	11/40
Expensive parser function count	3/500
Lua time usage	0.706/10.000 seconds
Lua memory usage	2.66 MB/50 MB

CPU time usage	2.388 seconds
Real time usage	2.464 seconds
Preprocessor visited node count	4857/1000000
Preprocessor generated node count	19685/1500000
Post-expand include size	305647/2048000 bytes
Template argument size	2472/2048000 bytes
Highest expansion depth	11/40
Expensive parser function count	3/500
Lua time usage	0.784/10.000 seconds
Lua memory usage	2.66 MB/50 MB

CPU time usage	2.188 seconds
Real time usage	2.244 seconds
Preprocessor visited node count	4857/1000000
Preprocessor generated node count	19685/1500000
Post-expand include size	305647/2048000 bytes
Template argument size	2472/2048000 bytes
Highest expansion depth	11/40
Expensive parser function count	3/500
Lua time usage	0.744/10.000 seconds
Lua memory usage	2.66 MB/50 MB

CPU time usage	1.928 seconds
Real time usage	1.985 seconds
Preprocessor visited node count	4857/1000000
Preprocessor generated node count	19685/1500000
Post-expand include size	305647/2048000 bytes
Template argument size	2472/2048000 bytes
Highest expansion depth	11/40
Expensive parser function count	3/500
Lua time usage	0.663/10.000 seconds
Lua memory usage	2.66 MB/50 MB

And these measurements are using User:PC-XT/sandbox/cite web auto to avoid including the TfD template in the measurements:

CPU time usage	2.492 seconds
Real time usage	2.561 seconds
Preprocessor visited node count	4858/1000000
Preprocessor generated node count	19687/1500000
Post-expand include size	305965/2048000 bytes
Template argument size	2472/2048000 bytes
Highest expansion depth	11/40
Expensive parser function count	3/500
Lua time usage	0.925/10.000 seconds
Lua memory usage	3.55 MB/50 MB

CPU time usage	2.276 seconds
Real time usage	2.362 seconds
Preprocessor visited node count	4858/1000000
Preprocessor generated node count	19687/1500000
Post-expand include size	305965/2048000 bytes
Template argument size	2472/2048000 bytes
Highest expansion depth	11/40
Expensive parser function count	3/500
Lua time usage	0.828/10.000 seconds
Lua memory usage	3.55 MB/50 MB

CPU time usage	3.324 seconds
Real time usage	3.409 seconds
Preprocessor visited node count	4858/1000000
Preprocessor generated node count	19687/1500000
Post-expand include size	305965/2048000 bytes
Template argument size	2472/2048000 bytes
Highest expansion depth	11/40
Expensive parser function count	3/500
Lua time usage	0.808/10.000 seconds
Lua memory usage	3.55 MB/50 MB

CPU time usage	1.920 seconds
Real time usage	1.983 seconds
Preprocessor visited node count	4858/1000000
Preprocessor generated node count	19687/1500000
Post-expand include size	305965/2048000 bytes
Template argument size	2472/2048000 bytes
Highest expansion depth	11/40
Expensive parser function count	3/500
Lua time usage	0.640/10.000 seconds
Lua memory usage	3.55 MB/50 MB

CPU time usage	2.468 seconds
Real time usage	2.550 seconds
Preprocessor visited node count	4858/1000000
Preprocessor generated node count	19687/1500000
Post-expand include size	305965/2048000 bytes
Template argument size	2472/2048000 bytes
Highest expansion depth	11/40
Expensive parser function count	3/500
Lua time usage	0.818/10.000 seconds
Lua memory usage	3.55 MB/50 MB

I did those measurements at different times, replacing cite web with either variant. I'll keep these around for a while in case others want to test, as well.

So, I am considering whether the increase in memory and time is worth the added benefits. If no, it should remain a separate template with basically no long-term usage, but if yes, it should be merged. What do you guys think? —PC-XT+ 07:44, 22 March 2014 (UTC)[reply]
If it were to be added to the existing template(s) that should be discussed at their talk page, not here. And it has been already, Help talk:Citation Style 1/Archive 5#Should autofix more cites, and there was no consensus to add it.--JohnBlackburnewordsdeeds 11:19, 22 March 2014 (UTC)[reply]
  • Merge parts later after consensus: Recent talks have raised ongoing discussions about the amount of autofixing to use, but the extra processing time only occurs when invalid parameters are present, handled in isolated parts of 2 Lua functions of the main Lua module. The extra memory is equivalent to using another large Lua-based template, and is reduced when no invalid parameters are present (minimum +0.3 MB), so that is neglible, and templates are not rejected when they increase Lua memory usage by 20%. -Wikid77 20:52, 22 March 2014 (UTC)[reply]
  • To be clear, I'm not going to try !voting merge in this discussion, because that's not an option. I will look for the proper place for further discussion. Wikid77's answers are adequate to keep me from !voting to delete. If there is still no consensus to merge, I still think it should be kept as a sandbox for now, pending further discussion on merging parts of this, unless userfying would be better for some reason. —PC-XT+ 03:32, 24 March 2014 (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). No further edits should be made to this section.