I suggested in a comment here that is a meta-tag and perhaps should be deleted.

The blog entry on meta-tags quoted this:

The reason meta-tags are a problem is that they do not describe the content of the question. They describe some other aspect of the question, like the author’s skill level, or the author’s motivation for asking it, or generally what “kind” of question it is (poll, how-to, etc.).

Meta-tags are actually a subset of a larger problem that I usually call dependent tags. These are tags that don’t say anything by themselves – you can’t tell what the question is about unless they’re paired with some other tag (or several of them). These tags are a problem because people don’t realize this and will often use that as the question’s only tag.

I submit that doesn’t tell you anything about the content of a question but rather just what “kind” of question it is, and is therefore a meta-tag.

However, I’m not entirely convinced that all meta-tags are necessarily detrimental, so I’d like to open the discussion:


Full disclosure up front: I personally "live" in that tag. It's one of the tags I have marked as favorite, and it's one of the tags I have used to "promote" our site by creating this tag set over at StackExchange, which has brought quite a lot of traffic our way on quite a few occasions, and seems to be rather popular in general (it's still on page 2 after all this time).

I totally see how this tag can be regarded as meta. However, I don't think that it is any more meta than . And it's actually more useful than certain non-meta-tags. The tag doesn't tell you anything about the content of a question and it doesn't tell you anything about what “kind” of question it is. (Is it about usage? spelling? etymology? pronunciation?) With the tag , you at least know the latter. In fact, this tag is exceptionally clearly defined, it tells you everything about the question except for what word the OP is actually looking for, but that's why he's asking the question in the first place. I don't recall anyone ever using it by mistake.

In short: I vote to keep it, but I will live with any decision the community takes on this.

Edit: Also, I think we should pause for a second and ask ourselves, why do we have tags in the first place? What purpose do they serve? They are not just decoration. They are there for navigation. They are there so that questions can be sorted into meaningful categories, which can then be easily browsed, followed, or ignored. I just can't imagine anyone following and ignoring , or vice versa. But I know of at least one person (myself) who follows the ; I see that at least two other people follow the aforementioned tag set on StackExchange, which includes (and ); and I know of at least one user who has recently commented here on meta that he doesn't care about this type of questions, and would rather ignore them. The tag is clearly doing more good than harm.

Edit 2, in response to comments: yes, I am convinced that can work as the only tag on a question. I have seen it do just that. And in fact, if it were deleted, then what else would you tag these questions with? In that regard, it is no different than or . Here, let's just break down what questions we have on this site:

  • if your question is about etymology, then it's, well,
  • if it's about dialects, it's
  • if it's about, dare I say it, grammaticality, that's (I hope we have settled on that one)
  • if you don't know how to spell a word, that's
  • if you're not sure how to pronounce it, that's
  • if you have two (or more) words and are not sure which one to use, that's
  • if you are not sure where a word goes in a sentence, that's
  • if you have a word and are not sure about its meaning, that's
  • and if you have the meaning but don't know how to express it in a single word, then it's, umm, what? ? Well, sure, sometimes it is, but in any case, part-of-speech tags are secondary, additional, completely orthogonal tags that could be applied to all of the above questions as well. In fact, it's the POS tags that won't work on their own — and note how nobody takes that to mean that they are somehow "meta".

can, and does, work as the only tag on the question. Sometimes you are not sure if you're looking for an adjective, an adverb, a participle, or a noun. You just have that vague idea floating in your head, and you want to pack it into a single word, and then build a sentence around that word accordingly, be it a noun or a verb. If that tag got deleted, people would (have to) resort to tagging their questions with something less appropriate, such as , which would water down that tag, or , or , or some other nonsense. Or they will just keep re-inventing as or something. Now that would do more harm than good.

  • 1
    I think you make some good points about tag utility. It is not obvious to me at all that meta-tags are necessarily harmful nor that certain popular non-meta-tags are useful. The blog entry on meta-tags doesn't give clear guidance on deciding what is and is not a meta-tag nor does it give a general explanation of why meta-tags are harmful. – nohat Feb 1 '11 at 21:22
  • @nohat uh, did you read the same blog post I did? the bedrock guidance is CAN THIS TAG BE USED AS THE ONLY TAG ON A QUESTION? If the answer to that is "no" then it is a meta tag. – Jeff Atwood Feb 2 '11 at 2:36
  • @Jeff, it says “if the tag can’t work as the only tag on a question, it’s probably a meta-tag” and “if the tag commonly means different things to different people, it’s probably a meta-tag.” Is that the definition of a meta-tag? A tag that couldn’t work as the only tag on a question and/or a tag that means different things to different people? What do either of those things have to do with being meta? Sounds to me like signs that you might have a meta-tag, not a clear definition of the characteristics of meta-tags. – nohat Feb 2 '11 at 3:18
  • @nohat this is just semantics; those are strong indicators, very strong. You can read "probably" as EXTREMELY LIKELY TO BE. I thought about editing the blog post to say this, but.. semantics. – Jeff Atwood Feb 2 '11 at 4:01
  • 6
    @Jeff Semantics is exactly what we’re talking about—the meanings of words of phrases. But you say it dismissively, as though the meaning of what you write were some irrelevant detail. To me, there is a qualitative difference between an indicative test (however strong) and a definitive one. If it looks like a duck and it quacks like a duck, it’s probably a duck, but looking like a duck and quacking like a duck do not define what a duck is. – nohat Feb 2 '11 at 4:47
  • @Jeff: please see my edit. Thank you! – RegDwigнt Feb 2 '11 at 8:55
  • @reg I think this looks good, though [grammaticality] is almost as tortured as [acceptability] and I still wonder about [usage] – Jeff Atwood Feb 2 '11 at 9:07
  • 3
    @Jeff The concept of grammaticality is not as flexible and relative and wishy-washy as you think it is. It is a useful working concept with a clear, precise definition which is widely understood, agreed upon, and it is used in an everyday way by linguists around the world. – nohat Feb 2 '11 at 17:10

I almost completely agree with Reg, except I come down much harder on the "no, it's not actually meta at all" side. I believe that "single-word-request" does work just fine as the only tag on a question, and makes it very clear that the questioner is essentially asking for a reverse dictionary lookup.


Thinking a little bit more about this, the guideline for meta tags is still the same:

Can this tag be used as the only tag on a question? If the answer to that is "no" then it is a meta tag.

I think here on English that some of the tags we are discussing, like , are more akin to the [java] tag on Stack Overflow -- it works as the only tag, and when you click on that tag, you know you're getting a list of Java language questions.

In other words, it is unlike the [poll] tag (meta: this question is a poll about.. uh, what?), but very much like the [java] tag (not meta: the content of this question is in the Java language).

  • 1
    So [usage] is a meta-tag or it isn't? What about [acceptability]/[grammaticality]? What about [single-word-request]? I appreciate that your thoughts on the subject are evolving, but now I'm confused on where you stand. – nohat Feb 2 '11 at 3:21
  • Maybe another meta-tag test could be “does this tag mean ‘This question is a [tag]’”. In that case, [usage], [acceptability], and [grammaticality] would be tags, but [single-word-request] is a meta-tag. – nohat Feb 2 '11 at 3:24
  • 1
    @nohat definition of meta: it's not about the question but about some other aspect of it, e.g. "this is a poll!" versus "this is about hotdogs!". In other words tagging a question [poll] tells you nothing about what the topic of the poll is. I'm not sure why this is so confusing to you. – Jeff Atwood Feb 2 '11 at 3:49
  • @nohat the subtlety here is this -- a question tagged [poll] is not a meta tag on a site about polls. See: [usage], etc. – Jeff Atwood Feb 2 '11 at 3:54
  • 1
    @Jeff I was confused on where you stand on the meta-ness of [acceptability], [usage], and [single-word-request]. Still am. It sounds like you are saying they are not meta tags, but that is different from what you were saying before. Hence my confusion. – nohat Feb 2 '11 at 4:43
  • @nohat I think acceptability is pretty marginal since it is so.. relative, and the word itself is just hard to parse and easy to misunderstand. But I support it as a synonym of [usage]. – Jeff Atwood Feb 2 '11 at 5:32

You must log in to answer this question.

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