---

I propose we change the instructions from

The only templates that are acceptable are {{xt}}, {{!xt}}, and {{tq}}; templates such as {{green}} that apply colours to text and are used to highlight examples; and {{collapse top}} and {{collapse bottom}}, used to hide offtopic discussions.

to

The only templates that are acceptable are {{collapse top}} and {{collapse bottom}}, used to hide offtopic discussions, and templates such as {{green}} that apply colours to text and are used to highlight examples.

This is because {{tq}}, {{xt}}, and {{!xt}} all cost four bytes per character enclosed, whereas the colour templates only cost two, and achieve almost the same effect. Removing those three templates will provide a reduction in post-include template size, but little loss of functionality, so I hope this is not controversial. Mike Christie (talk - contribs - library) 11:38, 12 November 2021 (UTC)[reply]

I proposed some replacement text further up to this effect but which encouraged plain markup and also addresses ping templates—do you know offhand how expensive {{u}} or {{ping}} are? Because they're used a lot. ᵹʀᴀᴘᴘʟᴇ ꭗ 12:19, 12 November 2021 (UTC)[reply]

Just tested them: {{u}} has a base cost of only 10 bytes but costs three bytes per enclosed character, meaning that within a FAC that base cost would be 20 and it would cost six per enclosed character. {{ping}} has a base cost of 96 and costs four bytes per enclosed character, so that’s 192 base and eight bytes per character inside a FAC. I’d say these are annoyingly expensive but since they only ever enclose usernames, which are rarely very long, they probably aren’t that big a problem.

The reason I didn’t mention the alternatives that you proposed, for plain text markup, was that there didn’t seem to be immediate consensus for those changes — I was trying to suggest an incremental change that nobody would disagree with. I haven’t decided myself whether I like the plain-text markup idea, but removing xt/!xt/tq seems like an easy step. Mike Christie (talk - contribs - library) 12:33, 12 November 2021 (UTC)[reply]

I'm happy to settle on low-resource alternatives if things like {{green}} are manageable but a note advising to directly notify a user with a wikilink (User:Grapple X, not Grapple X) would shave this down, especially as FACs tend to have multiple pings per review--reviewer pings the nominator with their comments, nominator pings reviewer to show they've responded, back and forth. ᵹʀᴀᴘᴘʟᴇ ꭗ 12:37, 12 November 2021 (UTC)[reply]

I too am worried that the excessive pinging is part of the problem, and think that Grapple X’s note reminding editors to use a direct wikilink instead could help. It might also help remind editors, at the same time, of how often they are pinging people who should be watchlisting the FAC anyway. I read through the entire October FAC archive last night, and was astounded to see how often the FAC Coords are being pinged (as if they don’t read the FACs anyway?). It seems to have become a trend to say “OK, Coords, I have my three supports, please promote”. Three supports is not a guaranteed promote, and how on earth are the Coords managing with all this pinging from people who can’t wait for the next read-through? Also, we should never have to ping the nominator, who should watchlist— it has become customary to use pings throughout Wikipedia, and it might help to get people to rethink that relative to our template limit problem. SandyGeorgia (Talk) 14:09, 12 November 2021 (UTC)[reply]

I would say that perhaps once or twice per nomination the {{@FAC}} was used to request the opportunity to launch another nomination. In most cases that was all it was, not a request to promote. Perhaps if the {{@FAC}} is a problem, it should be removed. It also transpires that some of the co-ords did not even receive the pings, hence the reason in some cases why the ping was repeated. Co-ords themselves have requested leave via this ping to start a new nomination from other co-ords after three supports, an image review and a source review. The Rambling Man (Keep wearing the mask...) 14:15, 12 November 2021 (UTC)[reply]

{{@FAC}} costs 142 bytes, so inside a FAC it costs 284 bytes. Mike Christie (talk - contribs - library) 14:18, 12 November 2021 (UTC)[reply]

Since I became a coordinator I have been surprised at how infrequently the {{@FAC}} template is used. Nominators seem to have an intelligent grasp between necessary queries and notifications and unnecessary pestering. Pending any changes and/or formal decisions can I encourage nominators wishing to communicate with the coordinators to use this function. Gog the Mild (talk) 14:48, 12 November 2021 (UTC)[reply]

It seems to me that these kinds of queries (pinging the co-ords with {{@FAC}} to discuss a second nomination) are exactly suited to a nomination's talk page, so doing so there should be uncontroversial. Using any template there should have no effect and it's the kind of aside not relevant to promotion/archiving which could be shunted off easily. ᵹʀᴀᴘᴘʟᴇ ꭗ 14:53, 12 November 2021 (UTC)[reply]

Agree, if the Coords are not being bothered by excessive pinging, at any rate, these could be placed on the individual FAC’s talk page so as not to consume 284, which is a lot. ( I always thought they were more useful here on FAC talk, where everyone could see them and opine, but a) I had empowered other trusted people to act on those requests in my place, and b) we had far more volume in my day, so I needed those helpers.) SandyGeorgia (Talk) 14:58, 12 November 2021 (UTC)[reply]

There's only been like three since I got added to the ping. It's not bad at all. Hog Farm _Talk 15:04, 12 November 2021 (UTC)[reply]

Well, I have pinged at least two since you became Coord, so you may not be getting all? Or are you referring only to those within FACs, not elsewhere? SandyGeorgia (Talk) 15:20, 12 November 2021 (UTC)[reply]

Only those in FACs. Kinda interesting to check in see "hey, I have 1 new message!". Hog Farm _Talk 15:23, 12 November 2021 (UTC)[reply]

Ah, OK, that’s better, thanks :) SandyGeorgia (Talk) 16:13, 12 November 2021 (UTC)[reply]

Perhaps we should encourage people to ping in their edit summaries instead of in wikitext. Gives the same notification but doesn't use a template (and saves space). Downside: harder to see for others that you have pinged. Alternative: nowiki / remove ping templates when replying. —Kusma (talk) 14:42, 12 November 2021 (UTC)[reply]

I think {{green}} may lose some accessibility compared to {{tq}}, but I'm not an expert. —Kusma (talk) 12:40, 12 November 2021 (UTC)[reply]

I think you're right -- {{green}} doesn't change the font. I know we have one or two colour-blind regulars at FAC; perhaps one of them could comment? Mike Christie (talk - contribs - library) 13:31, 12 November 2021 (UTC)[reply]

Pinging PresN who I've always known to be on-point re: accessibility. ᵹʀᴀᴘᴘʟᴇ ꭗ 13:36, 12 November 2021 (UTC)[reply]

Mike and PresN, would {{highlight}} (which uses yellow) work better with less cost ? SandyGeorgia (Talk) 14:28, 12 November 2021 (UTC)[reply]

{{highlight}} is more expensive than {{green}}; it costs 3 bytes outside FAC and 6 in a FAC per character highlit. Mike Christie (talk - contribs - library) 14:36, 12 November 2021 (UTC)[reply]

OK, thanks! SandyGeorgia (Talk) 14:54, 12 November 2021 (UTC)[reply]

I am on board with whatever version of this gets consensus for now, anticipating it will need to be revisited as it is not enough to address the problem. The problem is lengthy FACs, and the templates within those FACs is only a symptom, and by addressing only one of the symptoms, we will still have load time problems, FAC archive problems, and FACs that don’t necessarily cover all bases, and FACs that are increasingly promoted on only three supports. But. It is crucial to move forward with something to address the problem limiting the page now, so I hope we can come to quick agreement, and I won’t oppose anything we/the rest of us come up with.

My read-through of the past discussions in archives last night indicated that part of the problem is also excessive sectioning of FACs with sub-heads, so unless we also (eventually) deal with that, I am concerned we will quickly be right back here; the sectioning of FACs has come about because of the extreme length and because FACs are now peer reviews rather than Suppoprt, Oppose, followed by rationale with links to extended discussions on talk. But not to hold up this proprosal for that issue. We will need to eventually look at the extreme length and sub-sectioning of FAC pages.

Another area of concern is collapsing; I don’t think it helpful, and don’t think it addresses the underlying issues, rather encourages them by encouraging extremely long reviews which are then masked/hidden. Past discussions in archives indicated collapsing made it harder for Coords to read through a FAC, so we should definitely hear from all four Coords as to whether they view collapsed text as an impediment to their read-through.

All that said, my preference for this text would be to allow either green or yellow (via the highlight template, depending on accessibility), to disallow collapsing, and to remind editors to take care with pinging:

The only template that is acceptable is {{green}} to apply colours to text to highlight examples. User notification templates such as {{ping}} or {{U}} should be replaced with a direct wikilink, such as [[User:Example|Example]]. Offtopic discussions should not be collapsed within templates, but may be moved to the nomination's talk page, supplying a link to extended discussions.

If the Coords say that collapsing doesn’t impede their read-throughs, then letting that go is a compromise I am willing to agree on, although my concern remains that it encourages lengthy FACs rather than summation of whether an article meets criteria, which is the purpose of FAC. SandyGeorgia (Talk) 14:28, 12 November 2021 (UTC)[reply]

Apologies for killing the whole page, and thanks for the fix, Hog Farm; that was a weird problem that Nikkimaria described at WP:VPT yesterday— I’ll follow up there, as it continues to happen. SandyGeorgia (Talk) 14:38, 12 November 2021 (UTC) [reply]

Since banning the collapse templates and template-based pings seems less likely to have quick consensus, could we limit the wording change discussion in this section to just what I proposed, or tweaks that don't change the intention? I deliberately limited the wording in the hope of getting a quick agreement rather than an extended debate. If you want to propose banning template pings or collapses, I'd suggest a separate section for those discussions. That way, anything that has consensus can be implemented without a long discussion. Mike Christie (talk - contribs - library) 14:39, 12 November 2021 (UTC)[reply]

Well, good approach, but there was already a thread about this very topic higher up; sorry for foiling your plan, but we now have two threads. So, we will now have a third proposal, I guess, on the pingie issue?

OK, so, I defer to the Coords on collapsing, but still believe collapsing will lead us further back into the very problem we are trying to solve, and agree with the limited banning of other templates you have proposed as a very temporary fix that will not solve the whole problem, but will get us moving at least. SandyGeorgia (Talk) 15:08, 12 November 2021 (UTC)[reply]

I'll admit to having just fallen into the conventions I see around me after coming back from an extended break so although I recall a time when headings were explicitly not used, I've taken to using them due to their ubiquity (also if I add something new under a header, it falls under that header if I don't make a new one). I'm happy to go back to using semi-colon bolding as was the old convention, which might also remove some of the false positives from the nominations viewer (which seems to count "support" or "oppose" in a heading and in the text below that heading twice). ᵹʀᴀᴘᴘʟᴇ ꭗ 14:35, 12 November 2021 (UTC)[reply]

As an example of how expensive the {{tq}} and {{xt}} can be - this edit and this edit dropped post-include expand size by about 47,000 bytes between the two. Might not seem like that much against 2,000,000, but in comparison I found that Wikipedia:Featured article candidates/I'm Goin' Down/archive1 took up around 32,000 post-include bytes itself. The amount of xt and tq amounted to more than a shortish FAC in just 2 FACs! Those templates are becoming way too expensive for us to continue to use. Hog Farm _Talk 14:39, 12 November 2021 (UTC)[reply]

• Some editors will just ignore all these wails against brackets and templates and carry on using them... no matter what the expense. ——Serial 14:53, 12 November 2021 (UTC)[reply]

  If we can't get people to stick to the rules voluntarily, there's always edit filters or bots I guess. —Kusma (talk) 14:58, 12 November 2021 (UTC)[reply]

  If we make this a rule, we'll probably have to allow the coordinators to remove the template usage if the editor who added it won't. Hog Farm _Talk 15:14, 12 November 2021 (UTC)[reply]

  Serial, I have found that asking them once, and explaining why, does the trick at FAR; all it takes is for someone to routinely patrol and remind those who are doing this. From my read-through of the October FAC archive last night to see where all the green was coming from, it was pretty clear that a lot of this problem could be addressed via conversations with less than five editors. SandyGeorgia (Talk) 15:18, 12 November 2021 (UTC)[reply]

  Just got a feeling that for only semi-regular contributors to the project (as opposed to those of us, you, know, who fire up back-to-back FACs) it might be off-putting--and perhaps harder to remember to follow on that account. Basically, I'm just a little leery about making things more difficult in a volunteer project. I get where y'll coming from; I just think it should be borne in mind that some people might find the things we're talking about a convenience. I agree that, who these people are and how many of the them there are is somewhat of a variable, not to say an unknown quantity  :) ——Serial 15:43, 12 November 2021 (UTC)[reply]

  Yeah, it's a hard balance. We don't want to make it hard, but the page also has to function. And if two FAC's worth of colored text can result in taking up the space as big as several FACs, well, if something needs chucked overboard so the page will run .... Hog Farm _Talk 15:46, 12 November 2021 (UTC)[reply]

  This is why I wanted to be sure to list some simple alternatives as opposed to just saying "don't do this". Rather than simply cutting out the section on a few allowable templates, giving a plain markup alternative should make it much easier to steer usage; carrot first, then stick. ᵹʀᴀᴘᴘʟᴇ ꭗ 15:52, 12 November 2021 (UTC)[reply]

  Keep trying, Grapple; your effort is appreciated! SandyGeorgia (Talk) 16:18, 12 November 2021 (UTC)[reply]

There is a very simple plain alternative that drastically alters formatting without using any templates: just start your new line with a space.

The result looks like this. 

Unfortunately it can't be used inline. But maybe worth considering? —Kusma (talk) 16:44, 12 November 2021 (UTC)[reply]

Wow … that kind of “thinking outside the box” is what we need in this matter! Very hard to write that into instructions, though :) SandyGeorgia (Talk) 16:49, 12 November 2021 (UTC)[reply]

"and templates such as {{green}} that apply colours to text" - Should we be more specific not to use tq, xt, etc? Because they also apply colors, and I don't know how intuitive it will be for people not following this that you can use green to colorize text, but not xt or tq. Hog Farm _Talk 16:02, 12 November 2021 (UTC)[reply]

I initially thought this was a bad idea because we obviously can't list every template that we don't want used, but it's probably worth explicitly naming ones we're finding frequently. I'd suggest if we want to have an exclusion list that the coords just update it without asking for input, based on whatever they are running into and having to remove. Mike Christie (talk - contribs - library) 23:17, 12 November 2021 (UTC)[reply]

Revised proposal

Taking into account Hog Farm's suggestion, here's a revised proposed change. I'll make this a formal proposal section with support/oppose below and a discussion section.

The proposal is to make the following change to {{FAC-instructions}}. Change:

The only templates that are acceptable are {{xt}}, {{!xt}}, and {{tq}}; templates such as {{green}} that apply colours to text and are used to highlight examples; and {{collapse top}} and {{collapse bottom}}, used to hide offtopic discussions.

to

For technical reasons, the only templates that are acceptable are {{collapse top}} and {{collapse bottom}}, used to hide offtopic discussions, and templates such as {{green}} that apply colours to text and are used to highlight examples. Other templates, such as {{done}}, {{not done}}, and {{xt}}, may be removed by any editor.

Supports/Opposes

• Support. As proposer. Mike Christie (talk - contribs - library) 20:11, 14 November 2021 (UTC)[reply]
• Support - seems reasonable as a start, although I'd personally like to hold a conversation about the collapse templates at a future date. Hog Farm _Talk 14:21, 15 November 2021 (UTC)[reply]
• Support per above Eddie891 ^Talk _Work 14:31, 15 November 2021 (UTC)[reply]
• Support. It's a start, at least. I think there's some room for improvement (specifically calling out user link instead of ping templates, for example) but let's see how this affects us going forward and circle back if we need to. ᵹʀᴀᴘᴘʟᴇ ꭗ 14:35, 15 November 2021 (UTC)[reply]
• Support and per HF I would like to discuss removing the collapse templates in a subsequent proposal. Z1720 (talk) 17:12, 15 November 2021 (UTC)[reply]
• Support: I admittedly do not have a strong opinion about this, and I have used these templates in my own reviews, but if this will help the FAC process and those who want to participate in it (as either a nominator, reviewer, or both) even a little, then I will support this. Aoba47 (talk) 22:44, 19 November 2021 (UTC)[reply]
• Oppose because the proposed wording is so garbled that it appears to be proposing the opposite of all that was stated in the discussions that led to it, and what seems to be the intent of the proposal. I said I would not Oppose in general, even though I think this is not addressing the core problems, but the wording proposed is so ambiguous that it will be impossible to enforce and hard for new editors to understand. See my request for clarification in the Further discussion section, which yielded no correction or clarity. Regular editors at FAC may understand the intent of this wording, but it is not likely to be clear to new/future editors (actually, it's not even clear to me) because the wording literally ends up with the opposite of the proposed intent (which is to remove tq and tq2). So, when editors do attempt to follow these instructions and remove unnecessary templates, confusion may ensue, particularly with new editors or reviewers.
  Those editors who are supporting this wording are saying that any editor may remove any other kind of template except the two (only) kinds of templates allowed at FAC (emphasis on only as that is used in the proposal):
  1. {{collapse top}} and {{collapse bottom}},

     but ... "used to hide offtopic discussions"; that is, any time collapse is used for ontopic discussions (which is, does this article meet WP:WIAFA?), those can be removed by any editor. Is this the intent?

  2. templates such as {{green}} that apply colours to text and are used to highlight examples

     so {{highlight}}, {{tq}} and {{tq2}} (all of which apply colour and are used to highlight examples) would be allowed under this second category of only two kinds of templates allowed. Isn't that what we are trying to remove? How will a new editor, not having participated in this discussion, distinguish between {{green}} and {{tq}}, both templates that apply colour to highlight examples?

So, with only two kinds of templates that are allowed, we are left with a long list of templates that can be removed by any editor. This list encompass, basically, the very list this proposal is intended not to disallow, in the Further discussion section-- things like {{U}}, {{endash}}, {{Ping}}, {{Re}}, {{slink}}, {{efn}}, {{cite book}}, {{pb}}, {{Anchor}}, {{reflist-talk}}, "and scores more".

I understand this has happened because Mike has simply used some of the wording that already existed at Template:FAC-instructions, but unless I have entirely missed the point of these lengthy discussions here, the ambiguity and lack of clarity in the proposed wording puts us into a deeper trench of template limits and a deeper well of misunderstanding as new editors attempt to understand how to navigate FAC. I am fairly certain that this proposal does not mean what it says or say what it means, and suggest a reboot to correct the wording to do that. We already have wording in the instructions about not adding subheads to FACs, and that wording was there for very good reasons, but is now completely disregarded. It does us no good to end up with wording no one follows. 'Tis a pity Tony1 is not about to help us fix this, because this wording leaves us tq and tq2 (since they apply colour to highlight examples), and can be interpreted to read that any editor can remove pings and paragraphs breaks. SandyGeorgia (Talk) 22:28, 20 November 2021 (UTC)[reply]

Resolution of the minor wording quibbles at #Adjusting per request for clarification would allow me to remove my Oppose. SandyGeorgia (Talk) 15:05, 27 November 2021 (UTC)[reply]

• Support in light of making definitive objective improvements in the short term rather than spending months debating the minutiae to gain not much. Feels like we have constant filibustering around these parts. The Rambling Man (Keep wearing the mask...) 22:37, 20 November 2021 (UTC)[reply]

  It would be helpful to this page if you worked on not personalizing discussions. (Recognizing this should be placed on your user talk, which you have disallowed.) SandyGeorgia (Talk) 22:50, 20 November 2021 (UTC)[reply]

• Support. Wording could be improved and the proposal could go further but ffs let's make a decision, any decision. We can refine it once we've agreed a starting point. HJ Mitchell | Penny for your thoughts? 06:19, 21 November 2021 (UTC)[reply]
• Oppose. Eh, sorry, but I think that Sandy is right that this rewrite appears to have the opposite effect to that desired. Offtopic discussions don't need to be kept on page and "templates like {{green}}" sounds like it'd be interpreted as allowing things like {{tq}} which are a problem. Jo-Jo Eumerus (talk) 11:33, 21 November 2021 (UTC)[reply]
• Support - we can at least start this off in the right direction. Graham Beards (talk) 11:52, 21 November 2021 (UTC)[reply]
• Support. Strictly an improvement; does not in any way impede further steps. Vanamonde (Talk) 03:38, 22 November 2021 (UTC)[reply]

Further discussion

Please add any further discussion comments here. Mike Christie (talk - contribs - library) 20:11, 14 November 2021 (UTC)[reply]

• Question: Under this proposal, why are off-topic remarks put in a collapsed template instead of moved to the talk page? Wouldn't moving the off-topic comments prevent templates from being used in FACs, and thus save transclusion space? Z1720 (talk) 02:51, 15 November 2021 (UTC)[reply]

  The only thing I'm trying to accomplish with this wording change is to get rid of xt, !xt, and tq. We've had a lot of discussion (above) about various possible changes, but there seemed to me to be little opposition to removing these three. I don't think that removing collapse templates is quite as broadly supported, so I think it would be better to have a separate section for that. Similarly we may want to get rid of {{green}} as well, but again that might not pass so I think would be better as a separate question. Mike Christie (talk - contribs - library) 03:24, 15 November 2021 (UTC)[reply]

  Have we got a quantitative assessment of the impact of removing these? Or is it rearranging deckchairs on the Titanic? For example, if nominators sign each response to a comment, wouldn't that vastly overwhelm any benefit gained by removing these templates? I'm not sure I know the answer. The Rambling Man (Keep wearing the mask...) 08:08, 15 November 2021 (UTC)[reply]

  We do have enough information to indicate that this is a worthwhile change. Search this page for "47,000" and you'll find a post from Hog Farm, who saved that many bytes by removing those templates from just two FACs. If we eventually decide to keep colour templates like {{green}}, and those FACs had used colour templates instead, the savings would have been half that, but that still implies we could save perhaps 50,000 to 100,000 bytes across all FACs. As for signing every response, your signature is 146 bytes and mine is 205, so each instance of a signature would cost that many bytes. (It's not in a template so it doesn't cost double, as tq does.) Yes, that can add up to more than tq/xt/!xt, but I don't think that's a reason not to ban these templates. Mike Christie (talk - contribs - library) 10:06, 15 November 2021 (UTC)[reply]

  Oh no, I'm sure every little helps, but I had no idea that what on the face of it is a reduction of about 143 bytes equates to an actual reduction of 47,000 bytes. The Rambling Man (Keep wearing the mask...) 10:26, 15 November 2021 (UTC)[reply]

  See also this section above, which gives some more details -- part of the problem is that 100 bytes of text in a {{tq}} template costs 400 bytes in WP:FAC. Mike Christie (talk - contribs - library) 10:39, 15 November 2021 (UTC)[reply]

  And is that like a principle of physics which cannot be addressed because it is what it is, or is it something we could somehow improve? Or create a new template which does the same but without that ridiculous overhead? The Rambling Man (Keep wearing the mask...) 10:43, 15 November 2021 (UTC)[reply]

  The best that can be done is to reduce the 4x penalty to 2x, which is what the colour templates such as {{green}} do. That's why I didn't include the colour templates in this suggestion. We may eventually decide to eliminate those too, but since they cost only half as much as xt/!xt/tq that's not as obvious a decision. Mike Christie (talk - contribs - library) 10:47, 15 November 2021 (UTC)[reply]

  (edit conflict) Any template will have the impact of its size multiplied due to the way FAC pages are transcluded again onto WP:FAC, so even a small template ends up several times the size it began. Text formatting templates can really add up because their size is also impacted by the amount of text they're used to format, X many bytes per character is then multiplied when the nomination is transcluded again so it's also more than just the base template size. ᵹʀᴀᴘᴘʟᴇ ꭗ 10:50, 15 November 2021 (UTC)[reply]

  If they're not "templated" but just "marked up", would that incur a similar issue, e.g. if I wrote all my quotes out with code would that be equally profligate? The Rambling Man (Keep wearing the mask...) 11:00, 15 November 2021 (UTC)[reply]

  No that would be better, direct markup doesn't multiply in impact like a template. ᵹʀᴀᴘᴘʟᴇ ꭗ 11:02, 15 November 2021 (UTC)[reply]

  Well that should be an easy win then, tell reviewers to use markup, not templates. The text is presented differently and is still accessible (no colour perception issues) and costs almost nothing to facilitate. The Rambling Man (Keep wearing the mask...) 11:09, 15 November 2021 (UTC)[reply]

  If you subst {{green|fred}}, you get this: <span style="color:green;">fred</span>. In a FAC that would cost 56 bytes -- 52 for the markup and then 1x [correction: 2x, because of the double transclusion] the number of characters in the span tags. Tq doesn't work that way; it references a css file, so it's a lot messier to achieve with tags. I think some users would find that a bit harder to deal with in the editing window, so again I would say that should be a separate question since not everyone might agree with that -- after all, if we do decide to require markup, not templates, it still means banning tq/xt/!xt. Mike Christie (talk - contribs - library) 11:14, 15 November 2021 (UTC)[reply]

• Per WP:SIG: Signing your posts... both for the article and non-article namespaces, is required and facilitates discussion by identifying the author of a particular comment. Other users can then navigate to a talk page and address their comments to the specific, relevant user(s). Discussion is an important part of collaborative editing, because it helps all users to understand the progress and evolution of a work. ——Serial 10:37, 15 November 2021 (UTC)[reply]

  Yes? The Rambling Man (Keep wearing the mask...) 10:39, 15 November 2021 (UTC)[reply]

  Wut. ——Serial 10:50, 15 November 2021 (UTC)[reply]

  Do you have a template in your signature? -- Guerillero ^{Parlez Moi} 12:25, 15 November 2021 (UTC)[reply]

I substed a copy of every FAC into a sandbox earlier today, and calculated how much each template cost towards the page limit. Here's what I found -- I did an exact count for tq, because it's so large, but just estimated the others. These costs include the doubling/quadrupling effect of the two levels of transclusion. A reminder: the limit for transcluded text is about two million -- 2,097,152, to be exact.

• {{tq}}: 317 transclusions, with a total cost of 156,672 bytes (and this is after Hog Farm took out a 47,000 in cost by getting rid of scores of tq uses)
• {{U}}: 106 instances, cost 5,000
• {{endash}}: 89 instances, cost 300
• {{xt}}: 79 instances, cost 39,000
• {{Ping}}: 75 instances, cost 11,000
• {{Re}}: 39 instances, cost 5,000
• {{slink}}: 25 instances, cost 2,800

Other than an instance of {{efn}} and one of {{cite book}}, nothing else seems likely to cost more than a thousand or so bytes. There was only one collapse pair. There were about fifty other templates beyond those listed here -- {{pb}}, {{Anchor}}, {{reflist-talk}}, and scores more. If someone would like to spotcheck my numbers (probably worth doing) the sandbox is here.

The tq and xt templates together cost about 200,000 bytes. Replacing them all with {{green}} or another colour, would save about half of that, but if we think 100,000 is still too expensive for colour-coding the text, we may consider banning that too. The pings and collapses aren't worth banning just for the sake of their size; I know some people don't like collapsed text (I'm not a fan, myself) so perhaps someone will propose we ban that too, but a ban couldn't be justified just on the basis of the include size problem. Mike Christie (talk - contribs - library) 01:33, 16 November 2021 (UTC)[reply]

Wow. Tq seems to be a real problem, there. Kusma's outside-the-box idea of putting a space before text seems more inviting by the minute. Jo-Jo Eumerus (talk) 11:37, 16 November 2021 (UTC)[reply]

I may have found a way around the issue, at least for {{green}}, and perhaps for {{tq}} and its cousins too. The collapse templates used to be a huge problem for FAC because they took all the collapsed text inside the template as an argument. That was fixed by creating {{collapse top}} and {{collapse bottom}}, which simply place the appropriate tags at the ends of the text -- the downside for a user is you have to put two templates in, but it makes the collapse templates very cheap in our terms. I tried doing the same thing for {{green}}: see User:Mike Christie/greent and User:Mike Christie/espan. The first is essentially a "green start" template, and the second is a "green end" template (or end anything that creates but doesn't finish a span tag). If you put text between these two templates like so: {{User:Mike Christie/greent}}This is some text that should be green.{{User:Mike Christie/espan}}, the text will turn green, but the include cost is only the size of the templates themselves. It won't matter how much text you put between them.

We still need to ban xt etc., but this might mean we could bring back an alternative. We might be able to create an {{xtstart}} and {{xtend}} that could be used like xt. I would want to ask about this at WP:VPT first, because I could see some problems that might arise -- for example, it would make it very easy for users to mistakenly turn a whole page green. All they'd need to do is forget to put in the end span tag. Mike Christie (talk - contribs - library) 12:06, 16 November 2021 (UTC)[reply]

I mainly use quote marks and/or italics which are very cheap. (t · c) buidhe 12:12, 16 November 2021 (UTC)[reply]

Yes, this might be too much trouble to go to for something which isn't strictly necessary anyway. Mike Christie (talk - contribs - library) 12:15, 16 November 2021 (UTC)[reply]

• Just wanted to add that templates can be useful where there are a lot of review comments and you need to highlight a few that haven't been addressed. Otherwise I agree that they should be deprecated and editors guided to use "<insert highlighted text here>" or some alternative normal text formatting like italics etc. I used to use templates like xt routinely in my early days, but haven't for years except in exceptional cases for a few I need to highlight. Peacemaker67 (click to talk to me) 23:14, 20 November 2021 (UTC)[reply]

Request for clarification

Mike, your proposed text (bold emphasis mine) says:

… the only templates that are acceptable are {{collapse top}} and {{collapse bottom}}, used to hide offtopic discussions, …

Per the bolded text, I am reading this to mean that collapsing is NOT used for on-topic (relevant to WIAFA and the review) discussions. But some of the Support declarations seem to indicate that other editors may be reading this differently (that is, the collapsing will continue for resolved review discussions, which others want to discuss further–I agree if that is the case). If we are indeed collapsing only off-topic, I have the same concern others have raised (why not just move to talk)? If we are instead allowing collapsing of review content, that is a different discussion. Confused as to intent here, please clarify. Also, the context of the word only may be clear to us relative to the rest of this discussion, but could confuse future/new participants; you have given above a long list of other templates that are still in use and will likely to continue to be used (endash, ping, etc), so some clarification of the scope of the wording may be in order. All templates other than those listed are not being discouraged, so only could mislead or confuse new participants. SandyGeorgia (Talk) 14:56, 17 November 2021 (UTC)[reply]

The change I'm proposing is made with no prejudice to changing further wording -- it just seems to me that {{tq}} is eating enough space that we need to get rid of it. The "offtopic discussions" wording is only in this proposal because it's in the same sentence with the proposed change. And I agree that we probably need more discussion on the use of other templates. Mike Christie (talk - contribs - library) 01:19, 18 November 2021 (UTC)[reply]

So I think we all may (or may not) know what we mean relative to this discussionn, but I’m not sure the wording says we think it says, so I’m not sure what we’re !voting on (we want to get rid of tq et al, but what are we actually saying about collapse and only, and what will new participants make of it? SandyGeorgia (Talk) 01:27, 18 November 2021 (UTC)[reply]

Well, I don't think we'll have enough space in the header to ever discuss the entire nuances of template use. Maybe a brief synopsis there and a link to something like a Wikipedia:Featured article candidates/Allowable templates? It feels like we're stuck trying to create rules/instructions for a complex and nuances topic and trying to fit it into the cramped space of the header. If we do end up creating a list of allowable templates, it would be best to handle each of the major classes separately. I've become convinced that this talk page struggles to handle complex discussions for a variety of reasons. Hog Farm _Talk 22:42, 20 November 2021 (UTC)[reply]

Not a great wordsmith, but if all we are trying to do is remove tq and tq2, why don't we just say that? (Of course, all of this begs the question of shuffling the chairs on the deck of the Titanic, and would be immensely simpler if we actually addressed the core problems, rather than try to fiddle this template problem.) SandyGeorgia (Talk) 22:46, 20 November 2021 (UTC)[reply]

At least how I understand this, I'm seeing getting rid of those templates as just a start. I'd still like to see collapse gone, a consensus formed as to what can/should be the on the FAC talk page, and several other changes, but I'm really worried that if any of the long-term fixes are going to be made, it's going to need to be in incremental improvements, as several things need to be done, but it's difficult to keep discussion here focused. Hog Farm _Talk 22:51, 20 November 2021 (UTC)[reply]

Yes, I think this is just a start and we do need further discussion. (I plan on closing this tomorrow if there are no more substantive comments and no opposes, unless someone else closes it first.). Sandy, re “why don’t we just say that”, that’s what I meant to do — all the wording does is remove those. I quoted more than that in the proposed change to make it clear what the change would be but I can see that might have been taken to mean I supported the rest of the wording. Mike Christie (talk - contribs - library) 22:56, 20 November 2021 (UTC)[reply]

I recognize it's not easy, and I recognize the problem is because you kept some of the current wording, but we're not there, and are going to add confusion to new participants with the proposed wording. SandyGeorgia (Talk) 22:59, 20 November 2021 (UTC)[reply]

After giving it another reading, I can see where the confusion Sandy is referring could come from. I don't know how to fix the wording and make it clearer exactly what would and would not be okay. {{Green}} is less expensive than {{xt}}, and the wording is trying to steer users away from the more expensive ones, but it's not clear because in practice tq and green do the same thing. A bit of a puzzle - I think there's case to get rid of the color templates, but not sure that an entire ban on all, including the ones like green, would get consensus. Yet we're too limited on space in the instructions to get into all the details. Hog Farm _Talk 23:05, 20 November 2021 (UTC)[reply]

What if we just outright stopped with templates, and left a note in the instruction how to manually the colour span formatting achieved by {{green}}? Specifically calling it out with something like "templates should not be used to format text, but markup like <span style="color: green;">color formatting</span> may be used in their place". ᵹʀᴀᴘᴘʟᴇ ꭗ 23:11, 20 November 2021 (UTC)[reply]

Substing {{green}} does the same thing and is a bit easier to remember. Tq/xt make a bit of a mess of the edit window when subst'ed. HJ Mitchell | Penny for your thoughts? 23:18, 20 November 2021 (UTC)[reply]

Then if it's easier to grok we could simply state that only {{green}} is permitted and must be substed, if the net result is the same. ᵹʀᴀᴘᴘʟᴇ ꭗ 23:20, 20 November 2021 (UTC)[reply]

re Hog Farm on “incremental improvements”. Yes, and we generally all agree that the immediate incremental movement is to get rid of tq and tq2, right? Yet that is not what the proposal does, wrong? In fact, it does the exact opposite by allowing templates that use colour to highlight examples (which describes tq and tq2). It would be optimal to fix this before it goes live, right? SandyGeorgia (Talk) 22:58, 20 November 2021 (UTC)[reply]

No, tq changes both font and colour (and technically could do more than that since it refers to a style sheet). To me, it’s a problem that right now coords can’t go through and remove, or request the removal of, tq. Changing this wording would allow that. I suppose some editors may not realize that tq does more than change colour but the coords can disabuse them of that. And if they change to using {{green}}, which we may also ban eventually, that’s a huge improvement over tq. Mike Christie (talk - contribs - library) 23:05, 20 November 2021 (UTC)[reply]

But Mike, that is a nuance that is lost on new editors. We need to be more clear. We know what we intend but this wording doesn't get us there and will further confuse the very editors we need to reach. It reads the opposite of what is intended. Let's fix it now. I am heading out for the evening, and having an early Thanksgiving tomorrow. I hope you can find a way to fix this, which means you can ignore my Oppose if that is done ... SandyGeorgia (Talk) 23:26, 20 November 2021 (UTC)[reply]

I believe the text is still an improvement since it lets the coords remove tq, but since you’ve opposed I’m not going to close this; I’ll wait for a coord to close it. I feel that if we were to change the wording now it would invalidate the support votes, and it would be better to close this discussion and move on to the next. I do have ideas about what could be done next, but I don’t want to raise those ideas in this discussion — it’s so hard to get change around here that I think we should take what we can get. Mike Christie (talk - contribs - library) 23:42, 20 November 2021 (UTC)[reply]

The wording does not give the Coords the power to remove tqs. It gives everyone else the power to remove almost everything except tqs. I understand you not wanting to invalidate the Support votes already there, but what on earth is being supported, when the wording is exactly the opposite of the intended effect? There is no urgency; FAC is unlikely to hit template limits in archives again until next October, so why put through something with wording that is obviously not what you intended, hurriedly, yet can confuse new (and even existing) participants? Right now, the wording says that I (as “any editor”) can go through and remove pings. Or that I, as “any editor” can go through and remove anything that starts with double brackets (is a template) except in two limited cases: collapsing off-topic content only, or highlighting text in colour. SandyGeorgia (Talk) 02:09, 21 November 2021 (UTC)[reply]

Adjusting per request for clarification

Mike Christie, if I am understanding your proposal correctly, I believe the adjustments below would solve the ambiguities mentioned. And, unless I am misunderstanding your intent, it is likely that pinging those already supporting would allow us to get this done. I hope these few adjustments (below) reflect your intent, while removing the ambiguities. (Strikes to indicate removal, bold to indicate addition.) Regards, SandyGeorgia (Talk) 14:22, 27 November 2021 (UTC)[reply]

PS, because this is necessary for technical reasons, yet goes against WP:TALK guidelines (to never alter another editor’s post), it seems we should really work to get it right, and have a broad consensus, rather than a hurried adjustment. We don’t want to end up in dispute resolution over changing another editor’s posts, but the proposed wording could have us removing things like citation templates and paragraph breaks, while leaving ambiguity about the very thing we intend to remove (tq and tq2). SandyGeorgia (Talk) 14:41, 27 November 2021 (UTC)[reply]

Current as of Oct 31	Revised proposal	Adjusted per request for clarification
Please do not use graphics or templates on FAC nomination pages. Graphics such as  Done and  Not done slow down the page load time, and complex templates can lead to errors in the FAC archives. The only templates that are acceptable are {{xt}}, {{!xt}}, and {{tq}}; templates such as {{green}} that apply colours to text and are used to highlight examples; and {{collapse top}} and {{collapse bottom}}, used to hide offtopic discussions.	Please do not use graphics or templates on FAC nomination pages. Graphics such as  Done and  Not done slow down the page load time, and complex templates can lead to errors in the FAC archives. For technical reasons, the only templates that are acceptable are {{collapse top}} and {{collapse bottom}}, used to hide offtopic discussions, and templates such as {{green}} that apply colours to text and are used to highlight examples. Other templates, such as {{done}}, {{not done}}, and {{xt}}, may be removed by any editor.	Do not use graphics or complex templates on FAC nomination pages. Graphics such as  Done and  Not done slow down the page load time, and complex templates can lead to errors in the FAC archives. For technical reasons, templates that are acceptable are {{collapse top}} and {{collapse bottom}}, used to hide offtopic discussions, and templates such as {{green}} that are used to highlight examples without altering fonts. Other templates, such as {{done}}, {{not done}}, {{tq}}, {{tq2}} and {{xt}}, may be removed by any editor.

Sandy, I think you and I agree on the overall goal here. I also agree with you that giving the coords the ability to remove {{tq}} et al. was one of the goals, and if I were to write the proposed text again I think I'd change the given example from xt to tq. However, I feel that since tq is no longer named as protected, the wording already discussed does enable them to remove it. So I think I would still rather let this go through, as it appears it probably will, and then tweak it. Part of the reason is that I'm not sure I would agree with the other apparently minor changes you've made -- for example, you've added "complex" to "Do not use graphics or complex templates"; I don't think "complex" is well-defined, and in any case it's the PEIS that matters, not the complexity. And altering fonts, per se, is not the issue; there would be no harm in altering fonts if it could be done without the PEIS cost. I can think of other ways to address the underlying issue -- just as we have {{collapse top}} and {{collapse bottom}}, we could have {{green top}} and {{tq top}} which would cut 90% of the cost of those templates at a stroke. So we might end up re-adding some form of the troublesome templates to the permitted list. Or we might split the page in two -- "Newer FACs" and "Older FACs" separately transcluded, with just a list at the main FAC page. Or we might change the instructions to say that no template that includes quoted text inside the body of the template is permitted, since that's mostly the issue, and would be a form of words that would immediately allow the fifty other templates I found in the test I ran. Or someone might come up with another issue. So I think we should let this pass and then look at next steps. As I said above, we have such a hard time making progress I don't want to derail what seems to me to be an important change. I was even hesitant about posting the list of possibilities in this very paragraph, for fear it would start another conversation about possible changes to the wording and derail the current proposal. Mike Christie (talk - contribs - library) 16:01, 27 November 2021 (UTC)[reply]

I added complex as it was a word you used, and in trying to distinguish that when you say only, you don't really mean only (we're not intending to delete paragraph breaks, citation templates, and the like, nor are we intending to remove templates that only add color); there could be a better way to distinguish that. But my bigger concern is that if you really intend to give this ability only to Coords, it should not be saying any editor, as we don't want "any editor" to end up in a messy dispute if they follow these instructions and alter someone else's post. And I really don't think if we just fix these things now, that any of the editors who supported wouldn't immediately support again, along with me removing my oppose (win-win). SandyGeorgia (Talk) 17:24, 27 November 2021 (UTC)[reply]

Close request

@FAC coordinators: : could one of the coordinators close this? Mike Christie (talk - contribs - library) 12:58, 15 December 2021 (UTC)[reply]

@FAC coordinators: : Pinging again. If y'all would prefer I ask for a third-party closer, let me know. Mike Christie (talk - contribs - library) 13:07, 19 December 2021 (UTC)[reply]