11
$\begingroup$

Recently, I gave a solution to a complex analysis problem that avoided the standard complex analytic tools, but relied on a property of the modified Bessel function of the first kind.

I bet this happens quite often: someone proposes a question, tagging it accordingly to the area of mathematics in which the problem is expected to belong, and a solution comes from a completely different area.

For such situations, I was wondering about extending the tag-system also to answers, in order to underline certain peculiar connections between different branches of mathematics.

What do you think about it?

$\endgroup$
3
  • 1
    $\begingroup$ This is (to some extent) related: meta.math.stackexchange.com/questions/2612/… $\endgroup$
    – Martin Sleziak
    Aug 29, 2014 at 21:39
  • 1
    $\begingroup$ There are a couple old feature requests on MSE (back when it was MSO) that are (status-declined). $\endgroup$
    – user642796
    Aug 30, 2014 at 5:18
  • 2
    $\begingroup$ I've felt the need for this too. There are times in which a question is about (tag1) and I give an answer whose nature is related to (tag2) and (tag2) has little to do with (tag1) and the question itself, at least at first sight. $\endgroup$
    – Git Gud
    Aug 30, 2014 at 9:25

2 Answers 2

Reset to default
5
$\begingroup$

Add the tags to the question

This will ensure that people who are looking for applications/uses of your technique in your answer can see questions that it can be applied to. People who are looking for a solution to a problem via your method might look at a question and see that the tags do not cover therefore they create a new post to solve your way. If you add the tags to the question these people can that the question covers what they wanted.

The only disadvantage of this is that people may become confused. If people are searching for questions related to Bessel Functions and they stumble across complex analysis they may reconsider retagging or misunderstanding. I would propose adding a footnote to the question explaining that there is a solution with Bessel functions however this may be more hassle than it is worth.

Overall if there is a good way to deal with my previous point then this may be a plausible way of dealing with things like this effectively.

$\endgroup$
4
  • 3
    $\begingroup$ I almost entirely disagree with this answer. According to the Help Center, »A tag... describes the topic of the question« and from the Tags page »A tag... categorizes your question with other, similar questions.« (all emphasis mine) That one can use techniques from (foo) to answer a question about (bar) does not make (foo) a topic of the question. If somehow solving the question inherently brings up the techniques of (foo), there might be a case for adding the tag. $\endgroup$
    – user642796
    Aug 31, 2014 at 13:02
  • $\begingroup$ @ArthurFischer I agree that the help center outlines tags in this way however are there any examples of discrepancies related to tagging questions as (foo)? As far as I can see this does not cause any harm. $\endgroup$
    – Ali Caglayan
    Aug 31, 2014 at 18:50
  • 2
    $\begingroup$ Are there discrepancies? Almost certainly! The site is currently getting over 350 new questions every day (which will drastically increase soon), and we have an exceedingly small cohort of users engaged in the re-tagging of questions. However most of the discrepancies are due to the ignorance of (usually new) users, not to an (IMHO) abuse of tagging. I would also argue that making relevant questions harder to find does hurt the site. $\endgroup$
    – user642796
    Aug 31, 2014 at 19:29
  • $\begingroup$ @ArthurFischer There is a dead end then. If you consider that tagging questions unstrictly as such harms the site then implementing tagged answers as a feature would be the only other option. As these are status-declined the future looks bleak. $\endgroup$
    – Ali Caglayan
    Aug 31, 2014 at 19:43
15
$\begingroup$

In principle, answer tags could fit into the system, but I can see SE's viewpoint that the added complexity would not pay off. Most users are confused with question tags already.

Adding marginally-related tags to questions is a double-edged sword: tags are also used to ignore questions. If a question is tagged , those who have neither interest nor expertise in these functions (e.g., me) will not see it, missing out on a nice complex analysis problem (which does not need Bessel's functions at all, as the elegant solution by Daniel Fischer demonstrates).

Those interested in how Bessel functions are used in complex analysis can search just for that. I also think that to actually "underline certain peculiar connections between different branches of mathematics" one needs something like a blog post, not a tag.

Finally, given the state of the site's questions (tags, titles, formatting, content) it seems to me that we are trying too hard to polish fingerprints off some window while the window next to it is broken.

$\endgroup$

You must log in to answer this question.

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