# Should we eliminate [soft-question]?

Given "The Death of Meta Tags", should we eliminate the [soft-question] tag? Near as I can tell, there are three types of questions tagged with [soft-question]:

1. those that are also tagged [big-list] (more than half)
2. those that are valid questions, but for which [soft-question] does not add any meaning
3. those that are (or should be) closed

There are a few question that are only tagged [soft-question], which also needs to be fixed.

• No! Soft-question is a tag for people who don't want to read nonmathematical questions. I don't want to have to block out every single off-topic (read: philosophy of mathematics, mathematics-education, history-of-mathematics, pedagogy, etc.) tag individually. Soft questions are specifically those that are not asking a mathematical question. Please keep it.
– 97832123
Aug 8, 2010 at 19:32
• @Isaac: What reasons from the linked blog post do you agree with, and feel should apply here? Aug 9, 2010 at 2:54
• The meaning of soft-question is completely unclear. For example, @97832123 implied (I think) that all math ed/pedagogy questions are soft questions, and I disagree. big-list covers the majority of soft-question and is much easier to define. To me, soft-question is no different than subjective in the blog post. Aug 9, 2010 at 6:49
• The idea of soft-questions is explained in the MO FAQ. You should take a look there.
– 97832123
Aug 9, 2010 at 21:27
• @97832123: At a quick glance, I didn't see anything there, nor did a search for "soft-question" find anything--could you please point me to the specific spot in the FAQ? Aug 9, 2010 at 21:45
• @Isaac: There are some sentences on the topic in ¶2 of "what kind of questions should I ask here." The Community Wiki section could also be seen as relevant. Aug 10, 2010 at 3:36

I'd suggest

• removing soft-question from big-list-questions — they don't give any new information there
• creating a meta thread to gather links on posts that [probably] should be closed
• leaving (at least for now) soft-question on other questions — it's useful e.g. for adding soft-question in ignored (or interesting) tags
• and we should try to leave no questions tagged only soft-question
• I agree wholeheartedly with all four points. [big-list] is very nearly a subset of [soft-question]: of the 11 questions now marked [big-list] but not [soft-question], only three are genuinely asking for many examples fitting sharply defined criteria. Aug 9, 2010 at 12:28

A single soft-question tag is useful for people who want to see all soft questions, or who want to filter out all soft questions. Although I don't know about the first category, 97832123's comment confirms that people in the second group exist.

The main reasons I can think of for not having such a tag are:
It is too vague. It incorporates many different topics: pedagogy, book recommendations, math software, and gives them all the same tag. I think that this is not a problem because any such categories that grow large enough will get their own tag. See the 14 questions tagged books.

It adds little information about the question. To which I would say "it adds some information." If you see a question tagged algebraic-geometry on the front page with a fairly unhelpful title, whether or not it also contains a soft-question tag may guide whether you click it or not. In this case, I think the tag is doing its job.

It uses up limited space. Questions only get 5 tags total. If I spend 3 on soft-question, big-list, and books, that only leaves two tags for real content! If I want to ask for a list of books for teaching calculus in a physics context, then what do I do?
This I think is a real problem, and a possible reason to consider restructuring the tags. If we can think of some synonym set that would achieve this, though, I'd rather support that than removing soft-question.

• (Re: This I think is a real problem) well, maybe it will be real problem at some point, but for now I don't see any problem with soft-question tag Aug 9, 2010 at 7:33
• It adds little information about the question - Few tags are genuinely informative, since most users of the site will be able to infer most tags. Rather they are navigational aids. I value the [big-list] tag more than the [soft-question] tag, but [soft-question] is still of value. Aug 9, 2010 at 12:31
• the proper solution is to disallow soft questions, honestly. Aug 18, 2011 at 2:16
• @Jeff Atwood: Alas the M.SE community here is different than SO in many ways see meta.math.stackexchange.com/questions/17483/… You ended up creating programmers.SE for soft-questions that would have been asked on SO. Also, there's now ME.SE & academia.SE where some, but not all math-related soft-questions can go.
– Fizz
Feb 27, 2015 at 9:07

This seems to be a perennial question here... So, I'd like to throw in my couple of cents:

1) Soft questions aren't necessarily bad, even though some think there's too many of them. Although this may seem like just a pronouncement ex cathedra, you'll see in a moment why I think they're not bad...

And as a more concrete proposal,

2) The tag should probably be changed to suggest using one of , , , and possibly other more specific tags dealing with certain kinds of soft-questions. This is how is handled (as a tag) and it seems a good way of dealing with overly broad categories. Whoever wants to filter out soft questions can probably also filter out [some of] these more specific (soft-question) tags. Basically, my point is that discouraging the use an overly broad category is a good idea, but having this tag on a question that's currently not easily fit into any other "soft" category is probably better than not having it around.

(Honestly it would probably be even better if tags could be explicitly organized in a tree or DAG like categories can be organized in Wikipedia/Mediawiki, but I guess we won't see that happen with the SE software.)