Sub-tags and over-tags

In group theory there is a global tag, group-theory. There are also a number of sub-tags, such as finite-groups, abelian-groups, topological-groups, geometric-group-theory, etc. I am sure there are similar sub-tags for other subjects.

Would it not be a good idea to put something on these tags to "persuade" question askers to also use the group-theory tag (or at least the abstract-algebra tag, although this is a bit general, but maybe that's just my opinion)? It seems a number of questions which are straight out of an introductory course do not tag group-theory, but virtually identical ones do.

EDIT: As suggested in the comments, I have done this to (most of*) the group-theory related tags. However, it would be nice if this was done uniformly across the board...

*I didn't touch the (group-cohomology) tag as it needs a tag in the first place and I do not really know enough about it to write one! It would be useful if someone more knowledgeable about this than me wrote it...

I am not sure if this is necessarily a good idea.

1. Considering that only 5 tags are allowed per question, asking users to observe hierarchies may not be worth our while. For example: would you ask a user who tagged lie-groups to also include topological-groups, smooth-manifolds, group-theory, and differential-topology just because the "hierarchy" calls for it?

2. And why isolate group-theory? If you are going to ask users to consider using "parent tags", all inheritances should be observed. Why isn't general-topology a parent tag for topological-groups when group-theory is?

3. In fact, we have previously established tag wikis which calls for exactly the opposite of what you are proposing.

4. Note that there are character limits to tag wiki excerpts, and line breaks are not observed. Is this request (for users to use both the parent and child tags) really important enough to be placed in the limited space?

5. And if we were to add such a request, I think we should have something beyond just "Consider using..." A few more words would be better. Something like "For questions focusing on the group theoretical aspects of topological-groups, consider using also the [group-theory] tag" would be much more informative and avoid points 1 and 2 above.

• For point two, I isolated group theory because it is the subject I know most about. I should have made that more clear in the question - the question was meant to be about the general senario, but using the example of group theory. – user1729 Nov 20 '12 at 11:52
• For point 1 basically my point is that tags which are used by people sitting a first course in group theory, such as (finite-groups) or (abelian-groups), which do not require someone with an in-depth knowledge of the subject really should be accompanied by the (group-theory) tag (perhaps other people use SE differently, but I tend to just go to my favourite tag and nowhere else, and so I miss these questions). Lie groups is a subject unto itself, and anyone using such a tag can probably work it all out for themselves. Perhaps they should be asked to "consider using any other appropriate tags"? – user1729 Nov 20 '12 at 11:54
• For point four, the excerpt is meant to be just that: an excerpt. You only need one line, maybe two - if you are using a tag you just need to know that you have the correct tag, you do not need to read an in-depth discussion or a complete definition. That is what the wiki proper is for. – user1729 Nov 20 '12 at 11:57
• I agree with your point five, but only for the more "advanced" tags, and not (finite-groups) etc. – user1729 Nov 20 '12 at 11:58
• (...and I haven't said anything about point three because I would rather have a discussion than cite precedent...unless that precedent was discussed before?) – user1729 Nov 20 '12 at 11:59
• Roger on your first comment. About second comment: you explicitly mentioned "topological groups" and "geometric group theory" in the OP; neither of those are encountered by people sitting a first course in group theory. If you limit to only extremely specific cases like "finite groups" or "abelian groups", I agree with your third comment. [That is to say, I agree with your fourth comment. But I don't think such requests belong at all on the more "advanced tags".] – Willie Wong Nov 20 '12 at 12:00
• (About my second comment) Oh, I am sorry! I seem to have no read over my question before replying to your post! What I said in my response to your point one is my proper opinion; what I said in the question should be edited to reflect this! – user1729 Nov 20 '12 at 12:03
• I do not understand the second half of your comment: "If you limit to only extremely specific cases like "finite groups" or "abelian groups", I agree with your third comment. [That is to say, I agree with your fourth comment. But I don't think such requests belong at all on the more "advanced tags".]" Third comment? Fourth comment?...... – user1729 Nov 20 '12 at 12:06
• A propos my point three, the edit history to the galois-theory and field-theory tags indicated that I was the one responsible for creating the tag wikis. (I had thought otherwise though....:-p). There may have been a discussion held before I was even actively participating in the site (see here), though in the early days we were not always so clear about documenting everything on Meta. Sometimes things are just hashed on in comments of the relevant questions. – Willie Wong Nov 20 '12 at 12:08
• @user1729: when I wrote this your fourth comment hadn't appeared. After I hit the enter key I noticed that the second half of my comment is essentially what your wrote in your fourth comment. So I added a parenthetical to express that fact. Sorry about the confusion. – Willie Wong Nov 20 '12 at 12:10
• Can I try and clarify your opinion? (I do not know what that it a question - I suppose I am asking you if you want to reply to what follows...): You agree with me on the basic tags - that they should be accompanied by their over-tag. However, you feel that the more advanced tags can stand on their own and encouraging people to use other tags waste space in the excerpt. This wasting of space is important here because the more advanced tags need a more in-depth excerpt. – user1729 Nov 20 '12 at 12:13
• Almost: it is not just that it wastes space. Many of the more advanced tags lie at the intersection of many fields, and it is not even clear which tags (if any) should be considered its "over/super/parent tag". Moreover, one needs to consider how a tag is used. If I were looking for a question on finite-groups, I will consider searching both (finite-groups) and (group-theory). But if I were looking for a question about geometric group theory, I will never expect to find anything useful under (group-theory). – Willie Wong Nov 20 '12 at 12:17
• – Willie Wong Nov 20 '12 at 12:17
• Unfortunately, chat does not work on my computer. Now, you say that if you were looking for a question about geometric group theory then you will never expect to find anything useful under (group-theory). This may be the case, but it should not be, because questions about geometric group theory are answering questions about groups! If you look at the questions with this tag then the vast majority of them should be tagged (group-theory) too. (If you are doing research in group theory you are, broadly speaking, either doing rep. theory or geometric group theory, or studying top. groups...) – user1729 Nov 20 '12 at 12:41
• Ah, chat doesn't work for you. But can you read the transcripts? If not I'll copy and paste what I wrote afterward here. I think that will address your previous comment. – Willie Wong Nov 20 '12 at 12:51

I would like another solution: make some kind of tag hierarchy but if somebody wants to see all post with a particular tag, it also gets to see the question of all related sub-tags,

In the example above if a question is tagged lie-group it also shows up when I check on , and other more general tags.

• +1: It's a cool idea, but -1 because that's not really the way tags typically work (elsewhere too, not just StackExchange), and it would be confusing to get used to. So, net +/- 0. – apnorton Aug 23 '13 at 19:59