I posted a question and tagged it "infinite-sequences". A high-reputation editor added a more general tag: ca.analysis-and-odes. Now, in principle, having a more general tag is fine---but this tag is the wrong one. In particular, none of the topics mentioned under "tag info" have much to do with infinite sequences. I think that the right general tag is "real-analysis". However, I don't want to start an editing war with a high-rep user.
6$\begingroup$ As a rule, every question should have one of the top-level tags with an Arxiv prefix. $\endgroup$– Emil JeřábekJan 29, 2014 at 15:04
5$\begingroup$ Add the real-analysis tag, but leave the others as well. $\endgroup$– Andrés E. CaicedoJan 29, 2014 at 15:08
5$\begingroup$ To expand on @EmilJeřábek: If you were to submit a paper about this to the arXiv which category(s) from the ones given here arxiv.org/archive/math would you choose? You should at least use the corresponding tag (and possibly additional ones). $\endgroup$– user9072Jan 29, 2014 at 15:49
1$\begingroup$ Dear @John Bentin, I do apologize if you think my edit was inadequate. While I did not say so before, you are free to undo my edit or change the tags as you see fit. I will stand by my edit, but it is your question. By the way, I also apologize for accidentally cutting out a letter from your question, which I did not notice until now. $\endgroup$– Ricardo AndradeJan 29, 2014 at 20:16
1$\begingroup$ As long as I am here, I would like to also subscribe to the ideas presented in the comments above by Emil Jeřábek and quid, and in the answer given below by François Dorais: the top level tags (the ones with two letter prefixes like 'at.algebraic-topology') are useful for classification and search purposes, and also for increasing visibility of the questions to the right people. I will agree this set of top level tags adapted from the arXiv has several issues, but it has mostly become the norm on MathOverflow. $\endgroup$– Ricardo AndradeJan 29, 2014 at 20:22
$\begingroup$ @RicardoAndrade: Apology accepted. Thank you. The replies to my question support your tag, and I accept it now. But I will add mine, as recommended. $\endgroup$– John BentinJan 30, 2014 at 8:54
1$\begingroup$ I would hope that on MathOverflow most disagreements do not end in war! Surely making a comment, as per @quid's answer below, is more often than not going to result in a positive and constructive discussion that will resolve any difficulties. (Evidence for this is provided by Ricardo's gracious response above.) $\endgroup$– Nick GillFeb 4, 2014 at 10:24
$\begingroup$ Somewhat related: What are top level tags, and why should I try to use them? and Why are MO tags formatted as they are? $\endgroup$– Martin SleziakJan 19, 2018 at 6:49
It's tempting to think that the only role of tags is for classification purposes. However, tags play another important role which is to increase a question's visibility to certain users. There is a balance between the two roles:
- Don't use an irrelevant tag just to attract specific users.
- Don't just use highly specialized tags with little or no following.
In this case, the question is hard to classify; ca.analysis-and-odes is not completely irrelevant and real-analysis seems fine too. They both have large following. Since the question is not close to the five tag limit, so why not have both?
I would like to answer this question in two parts:
For the general question what to do if you disagree with an edit to your post. First, it is possible to comment-notify editors of a post (despite them not being suggested in autocompletion), and thus you could ask them regarding the edit and perhaps you find common grounds. Yet, second, I think on MO it is by and large common understanding that the owner should have in most cases the last word regarding their posts, and thus in case of actual conflict you should feel free to revert.
For the particular matter. I think the tag perhaps appearing 'wrong' could be a misunderstanding of tagging-conventions of MO, which at the moment (due to software-upgrade) are still not that visibly explained as they used to be. As said by Emil Jeřábek it is strongly encouraged to use at least one top-level tag for each (mathematical) question. These are the tags with two-letter prefixes that correspond to the math subject categories of arXiv, so this list http://arxiv.org/archive/math The one used for retagging your question ca.analysis-and-odes seems the relatively most fitting to me (except one goes for the utmost general gm.general-mathematics). In particular, on MO this tag (ca) is more or less the catch all (real) analysis tag. Note for instance that classical-analysis is a synonym of it. In some sense real-analysis is more or less a subtag of ca. (One can be of the opinion this top-level system is not optimal, but it is current standard, and the retagging seems in line with it, while the original tagging was in fact not.)
4$\begingroup$ Historical note: Not only is classical-analysis a synonym of ca.analysis-and-odes, the original name of the tag was
ca.classical-analysis. The arXiv category is "classical analysis and ODEs", which is unfortunately too long. The reason for the change was that ODEs had no obvious major tag and people tended to (erroneously) use ap.analysis-of-pdes for that purpose. $\endgroup$ Jan 29, 2014 at 23:59
1$\begingroup$ I'll just add an update to @FrançoisG.Dorais' comment above: After the limit for tag names was increased to 35 characters, MO moderators renamed this tag to (ca.classical-analysis-and-odes). $\endgroup$ Aug 3, 2018 at 7:00