20
$\begingroup$

According to the tag info of , it is used when questions are about vector calculus. However, it is often tagged with , at questions about convergence/divergence of sequences, series, or improper integrals. Here are such examples:

If there questions were few, then I would edit them directly. However, I don't know what an effect occurs when I edit many questions all together. What can I do to solve the problematic situation?

$\endgroup$
16
  • 2
    $\begingroup$ There is a related post in the Tag management 2017. This is also related: Do we need the “divergence” tag? $\endgroup$ Commented Oct 13, 2018 at 14:45
  • 2
    $\begingroup$ It is recommended not to bump many questions at the same time - see How much bumping is too much? and other related discussions. So in situations like this, a solution might be to "bookmark" the question in some way and do a few of them each time - and return to the list and continue editing later. $\endgroup$ Commented Oct 13, 2018 at 15:25
  • 1
    $\begingroup$ This might be easier if there is a simple way to find such questions and you do not have to compile a list manually - for example, in this specific instance the questions tagged divergence+sequences-and-series are very likely among the questions that need retagging. $\endgroup$ Commented Oct 13, 2018 at 15:25
  • 1
    $\begingroup$ The Stack Exchange staff can remove a tag without bumping any questions. However, this can only be done when a tag is completely removed - not only for a subset of questions. This process is called burnination. $\endgroup$ Commented Oct 13, 2018 at 15:28
  • 1
    $\begingroup$ Seems like it may be hard to prevent this from happening in the future. It's a reasonable mistake. $\endgroup$
    – Alexander Gruber Mod
    Commented Oct 13, 2018 at 16:33
  • 2
    $\begingroup$ @AlexanderGruber Perhaps changing the name of the tag to something more descriptive could help users to notice what the tag is supposed to be used for. (In the thread linked above, the name (divergence-vector-field) was suggested.) $\endgroup$ Commented Oct 13, 2018 at 17:18
  • 1
    $\begingroup$ That sounds good to me. $\endgroup$
    – Alexander Gruber Mod
    Commented Oct 13, 2018 at 18:04
  • 1
    $\begingroup$ Well, it'll be spammy, but I suppose we can just edit in the new divergence tag to the appropriate questions and then synonym the other one with convergence. Might make the front page messy for a bit but this is a common enough tag that I'd say it'd be worth the noise. $\endgroup$
    – Alexander Gruber Mod
    Commented Oct 13, 2018 at 18:07
  • $\begingroup$ @AlexanderGruber I have posted the suggestion to change the name - to see what other users think about this. I am not sure what is (logistically) the best way to do this. I'll leave a few comments in the Tagging chatroom. $\endgroup$ Commented Oct 14, 2018 at 0:45
  • 3
    $\begingroup$ @AlexanderGruber I'd prefer one tag that is called convergence-divergence rather than a synonym. We might end up with the opposite situation of divergence (in the vector calculs sense) being remapped to convergence. $\endgroup$
    – quid
    Commented Oct 14, 2018 at 1:26
  • 1
    $\begingroup$ With only 500 questions with this tag out of our almost million on site, and a perfectly good vector calculus tag, do we even need this tag? $\endgroup$ Commented Oct 16, 2018 at 21:00
  • $\begingroup$ @AlfredYerger Depending on how seriously you mean the suggestion to remove the tag completely, it might be useful to post it as an answer rather than as a comment. (In that way more people will notice it and they can upvode/downvote and discuss in the comments under the answer.) As already mentioned, removal of the tag was suggested here: Do we need the “divergence” tag? The situation at the time was slightly different - the tag-info clarifying the intended use of the tag did not exist back then. $\endgroup$ Commented Oct 17, 2018 at 10:29
  • $\begingroup$ That proves it, mathematicians are boring. If this had been asked on meta.StackOverflow.com, there's no way the title wouldn't have been “divergence of the divergence tag”. $\endgroup$ Commented Oct 19, 2018 at 13:34
  • $\begingroup$ @leftaroundabout You mean something like: Optimizing the tags for maxima and mimima to an extremum? $\endgroup$ Commented Oct 20, 2018 at 8:48
  • $\begingroup$ It's probably not a big deal, but since a new answer was posted, perhaps my answer should be unaccepted while things are sting being discussed. $\endgroup$ Commented Oct 22, 2018 at 18:03

3 Answers 3

18
$\begingroup$

Proposal: Changing the tag name to something more descriptive might make it clearer to the user what the tag is intended for and thus incorrect use will be less likely. (And thus reducing the frequency of this problem in the future.) I would suggest or or something similar.

This is the same suggestion as previously made in the Tag management 2017 thread. (It was suggested to remove the tag completely in the post Do we need the “divergence” tag? - this was before the tag-info was created.)


I will point out that moderators can change the name of a tag without bumping any questions, for details on this see: Can you change the name of a tag? (It is also possible to remove all instances of a tag, this is called burnination.)

However, even if the tag-name is changed, the questions which are tagged incorrectly have to be retagged manually.

However, I don't know what an effect occurs when I edit many questions all together. What can I do to solve the problematic situation?

It is recommended not to bump many questions at the same time - see How much bumping is too much? and other related discussions. So in situations like this, a solution might be to "bookmark" the question in some way and do a few of them each time - and return to the list and continue editing later. This might be easier if there is a simple way to find such questions and you do not have to compile a list manually - for example, in this specific instance the questions tagged divergence+sequences-and-series are very likely among the questions that need retagging. I will add a link to this question (which was asked partly as a reaction to this post): What do you use to mark questions that need editing when you want to avoid excessive bumping?

$\endgroup$
1
6
$\begingroup$

Alternative proposal: Remove this tag.

There are a only approximately 500 questions with this tag, which already refers to only one very specific operation in elementary vector calculus, or calculus on manifolds, for which we also have appropriate tags which are much larger.

In my view, a good tag should carve out an area of mathematics or topic relevant to the community (such as limits-without-l'hopital, a common question we've discussed on meta in the past), while remaining sufficiently broad. Many questions with this tag already contain other relevant tags, and the only ones I can find that do not refer exclusively to calculations in vector calculus. This suggests to me that the divergence tag is often simply something that arises in another problem, perhaps in probability or PDE, where using this tag feels somewhat analogous to using a tag for algebra on a problem involving any formal manipulations.

This would also eliminate the confusion discussed in the other answer, distinguishing vector calculus divergence from divergence in the sense of sequences or series.

$\endgroup$
4
  • 1
    $\begingroup$ Agreed. The questions should use vector-calculus or exterior-derivative instead. $\endgroup$
    – user7530
    Commented Oct 17, 2018 at 16:39
  • 4
    $\begingroup$ I will just point out (as a reaction to the 500 questions remark) that the number of questions should not be very important criterion when considering whether tag is useful or not. There are several tags on this site which I consider useful and have much less question. (Although typically from more advanced topics than this.) For this specific issue - whether we should have a separate tag for divergence (in the sense of vector calculus) - I do not have strong opinion either way. Probably it would be good to hear from users who are active in questions about multivariable calculus. $\endgroup$ Commented Oct 19, 2018 at 11:44
  • $\begingroup$ "only" 500 questions? that sounds like a LOT of questions. $\endgroup$ Commented Oct 20, 2018 at 22:56
  • $\begingroup$ @Santropedro Considering most of them are just fine with 'probability' or 'multivariable calculus' as tags, which each have tens of thousands, I'd say not really. $\endgroup$ Commented Oct 21, 2018 at 3:59
0
$\begingroup$

Non-solution. Nobody seems to have proposed editing the tag description as a solution, which is good, because it doesn't work. Over on [cs.se], we have a bunch of tags for programming languages which basically say, "If your question is about how to write programs in this language, your question is off-topic". We still get "How do I write this program?" questions using these tags. Of course there's a heavy selection bias, here, since people who read and understand the tag description won't post their question.

How to vote on this answer is left as an exercise for the reader. Do you vote up because you agree that editing the tag description won't work? Down for the same reason? Who knows?

$\endgroup$

You must log in to answer this question.

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