-4

TL;DR of the TL;DR:

Some tags suck. We should probably decide what to do with them as a community. How do we talk about all those tags without it being an unsufferable bother?


TL;DR:

Some users (mainly TREU chatters, myself included) think a couple of tags are irrelevant and pointless, and should be edited out/destroyed altogether, because they're used with no consistency and don't bring any good to the site. However, some other users might like those, so... We need to get people's opinions, so that the community decides what is thrashed and what is kept.

But how to do that is unclear. Chat is an unhandy platform, cluttering Meta with would probably be a bother... So before we talk about the what, we ought to talk about the how.

I repeat, this is not a meta about what tags should be kept or not. This is a meta about how we decide what gets kept or not. It's a question about how we'll manage a meta; a meta meta question, if you wish.


Complete version

The problem

I don't think I'll surprise anyone when I say SFF has some "unclear" tags. I'm talking about non-work, non-franchise, non-media, non-author tags. While some are very popular ( has 1020 questions, 541, etc), there are also some which are only used 5-30 times. So far, everything is factual, it's objective.

Now it gets subjective: some users, myself included, think some of the latter tags:

  • don't really serve any meaningful purpose
  • don't have a clear, universal meaning
  • are irrelevant, just because you stick a "use this tag when talking about ______ in works of SFF" in the wiki doesn't make a good tag for this site;
  • don't really feel required (if has only had 11 questions in 7 years, then it doesn't seem like people need it)
  • so if you don't need it and it doesn't bring anything to the site and it even harms it by bringing inconsistent classification, nuke it from orbit.

Why it should be discussed

On the other hand, the mentioned users are a dozen chatters in The Restaurant at the End of the Universe, who don't represent the community as a whole. Maybe the community is fine with seeing these tags gone, maybe it's not, but in any case, there's not enough activity in chat to qualify that, nor is chat even the place for that to begin with - that's why we have a Meta, am I right?

So far, I didn't see anyone going to Meta/chat about "hey, why was this tag removed? I liked this tag!", perhaps because nobody cares/noticed. Though, discussing whether some tags ought to be killed would settle some things:

  • if there's no discussion and action is taken, stuff will backfire at some point;
  • if there's no discussion and no action is taken, there will be happy and unhappy people, and the unhappy will rant about not being able to act as nothing was discussed;
  • if there's a discussion and a consensus, there will also be happy and unhappy people, but when the unhappy side rants, the happy ones can stuff the Meta consensus in. (I've been told that sounded confrontational; I kept the "sides" unnamed as this would apply regardless of who's happy or not. I wasn't trying to trigger people who want to leave them around, sorry if it felt like it)

How it should be discussed

The workflow is usually simple: one problem → take it to meta → wait and see. Other problem → take it to meta → wait and see.

But in this case, there are many problems, as those tags are unrelated and you may want to keep one but not the other. Concrete stuff time, here's the list of tags that currently serves as the kill-list; by taking a look at its history, you will see that some already got nuked. (includes but not limited to, , , ...)

Some larger ones have been put on "don't touch, ask on Meta" list (here). Includes but not limited to, , , etc.

So now what? How do we handle that discussion?

  1. Does the community even care if these low-attention tags get removed? Or is there a majority of users erring on the side of "meh, let them do it if it pleases them"?
  2. If the community disagrees and wants to discuss it:

    • do we need/want dozens of meta discussions, each about one specific tag? Do we want Meta looking like this (and writing repetitive answers, etc)?

    fake meta homepage riddled with kill-request questions

    And yes, they're all tags on this site

    • do we want one Meta thread with all the burnination requests, one tag per answer, and after X weeks we nuke or not depending on the answer's score? (kind of what was done in Tag Cleanup 2016: Before the burnination)

    • do we want a chat event to discuss this? (no, it's a terrible idea, logistics would be hell and not everyone can access chat, but I wanted a third bullet)

    • ... Other? I'm out of ideas, but people might think of something else.

Post is long already, time for other people's comments/answers/ideas, etc. And once again, this is not a meta about what tags should be kept or not. This is a meta about how we would decide what gets kept or not.


Clarifications, per comments:

  • This is not, and has never been, about mass edits. If a 14-questions tag is to be deleted, that can be done in a week, two weeks, a month... I guess the process would be to push one, two questions at once, once in a while... Nothing "massy".
  • Nobody has to give time to such edits. If four or five hillbillies are willing to do it and clean everything else on the thread while they're at it, the others can sit back and relax.
  • Bumping old questions (as long as it's not invasive, see above) isn't necessarily a bad thing. You get new eyes on the content, so more votes/comments as needed, possibly even new answers, edits... and the folks watching the front page may learn something new.
  • 4
    IMO, there is nothing wrong with the current system. Raise a meta question. You say the burninate tag would clutter meta... So what? Meta is not really that busy. – Skooba Feb 27 at 21:19
  • 1
    @Skooba it's more about people getting bored by 40 repetitive "do we kep this one? Ok, and this one? And that one?" questions – Jenayah Feb 27 at 21:20
  • 1
    @Skooba - but is it really worth it to raise a meta about every tag with 6 questions that's being used inconsistently and is really useless? Where do we draw the line at what needs a meta? – Mithrandir Feb 27 at 21:21
  • 1
    @Jenayah I mean if you see a cluster of tags, save some energy and combine them into one meta post. There are quite a few of those here as well. – Skooba Feb 27 at 21:26
  • @Skooba yep, it's one of the solutions proposed (2nd bullet point of the 3rd bullet list). I think that would be a massive list, but that's also the solution I prefer. Want to post it as an answer so discussion on it can be easier to sort out? – Jenayah Feb 27 at 21:29
  • Discussion about whether or not such tag edits should be done at all has been moved to chat. Please keep comments here relevant to how it should be organised, assuming it should be done at all. – Rand al'Thor Feb 28 at 14:40
3

Just use the normal/current system

If you think a tag should be nuked and can make a strong case for why you think so (i.e. no "cuz itz bad n i dnt lik it") then raise a meta and explain your reasons why you think it should be gone. The answers and voting on them are then the indication on whether or not we should nuke them.

If you're worried about "flooding" meta then group similar tags together where possible.

  • 2
    The need to raise a meta each time also creates an 'annoyance barrier' that prevents frivolous asking – Valorum Mar 4 at 10:49
  • So you're saying that every tag, no matter how small, has to go through meta? – Mithrandir Mar 4 at 14:05
  • 2
    @Mithrandir I think there is some common sense to be had here, if a new tag is created on only one question and it is pointless, it can be up to the individual if they want to nuke it or not. However, you can't really come up with any quantifiable hard value for what thresholds to set on when something doesn't really need to go through meta. Any attempt at setting one will generally end up being too high or too low, the safest bet is to put everything through meta, if you don't you'll end up with a small group of users controlling the tagging system and that will never end well. – TheLethalCarrot Mar 4 at 14:10
  • 1
    TL;DR: It is safer to put everything through meta even if you don't think it needs to because then everyone gets a chance to have a say rather than a small subset of users. – TheLethalCarrot Mar 4 at 14:11
  • 4
    @Mithrandir: The alternative is "TREU gets to make decisions on behalf of SFF as a whole," which is far worse. TREU is "just a chat room" and has no special status here. – Kevin Mar 5 at 16:44
-9

Some tags need to go through a meta process before being removed, and some don't. I'm outlining here a process for determining how it should be decided if a tag needs to go to meta or not.

This is essentially the process we've been using so far in chat, and I think it's a good one. It doesn't allow small groups of people to make decisions about things that would have a large effect on the rest of the community, but does allow for a bit of agility and to not flood meta with dozens of burninate discussions for straightforward cases.

Essentially, small, obviously useless tags can get nuked (but should still have a record, say, in chat or a dedicated meta post that it got nuked), but larger tags or ones that might be important get taken to meta for the larger community to discuss.

Note: All numbers below are non-final; they're there as a rough starting point, and can be adjusted as necessary.


  • Franchise-specific tags should always go to meta.

    The people who care about this (read: chat regulars) aren't always topic experts on everything, so if there comes a tag that might be franchise-specific - or is only being used on questions for a specific franchise - then it should be left or taken to meta so that people who know what they're talking about can weigh in.

  • Tags without wikis, are being used inconsistently, and under 25 questions can be nuked.

    By this I mean no tag wiki (excerpt) at all. Nobody cared enough to write up anything in the wiki; it's ambiguous enough that it's not being used consistently; and it's small, not widely-used, implying that it's not serving a real purpose.

  • Tags with wikis, but not actual guidance, and are being used inconsistently, and under 15 questions can be nuked.

    In this case, someone decided to write something in the excerpt, but it's still got no actual guidance - for instance, someone put in a definition of a term. Since this does imply that it's got at least a bit more interest in it, lowering it to about 15 questions where it's being used inconsistently seems like it would prevent any useful tags from being destroyed.

  • Tags that have usage guidance and are being used inconsistently should be cleaned up and then reevaluated.

    If a tag actually has useful usage guidance, but is being used inconsistently, then the questions with that tag should be cleaned up and retagged. After that's done, look at the tag again: does it serve a purpose? Does it need to be renamed to be less ambiguous?

  • Tags that have guidance and are being used consistently should go to meta.

    This seems rather obvious; if it's being used consistently, has guidance, and seems orderly, then someone wanting to destroy it should go through the proper processes.

  • Tags that don't have guidance and are being used consistently should be given guidance and reevaluated.

    If the tag doesn't have guidance, but there's a clear trend in what it's being used for, then it should be given the appropriate guidance, cleaned up, and then evaluated for its usefulness.

  • Tags that have 7+ watchers should go to meta.

    Having 7+ watchers indicates that enough people find it useful enough to watch, and so we should give people a chance to weigh in before making it disappear.

  • Franchise and work tags shouldn't be nuked.

    This also seems rather self-evident; these are obviously useful and should not be removed.

  • When in doubt, err on the side of leaving the tag, or go to meta.


Basically, any small, harmless decision shouldn't have to go through a long process involving people who don't care. Just do it. Once you reach a certain size, though, you do need some more discussion than just chat.

And, of course, there should always be a record. Perhaps a dedicated meta thread with a list of tags that have been removed, are being removed, and might be removed in the future, simply as a compilation of data?

  • 2
    "being used consistently" / "being used inconsistently" - according to who? – Valorum Feb 27 at 21:53
  • 2
    15 and 25 questions for a tag is a lot, there are lots of work/franchise tags that would be lucky to hit that number. – TheLethalCarrot Feb 28 at 9:57

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .