I see a lot of questions with titles like "limit question 4" or whatever. Should you number questions? It seems like a poor way of titling questions.

  • 12
    $\begingroup$ Agreed. That is a poor way of titling. Titles should be descriptive. Ideally we want searchable titles, but this is occasionally difficult to arrange. $\endgroup$ – Jyrki Lahtonen Aug 31 '15 at 6:36
  • 21
    $\begingroup$ No, we do not want these. These should be edited away if salvageable, or downvoted and/or closed otherwise. Often, poor titles accompany poorly formatted posts, which are not helpful. Fortunately, it is in users' best interests to give good titles so that others know what they are clicking on. I would like to add that each fall, we get a surge of new users starting the fall semester of classes. The habit of giving good titles is best started early, and should be reinforced both by example and correction. $\endgroup$ – davidlowryduda Aug 31 '15 at 6:56
  • 3
    $\begingroup$ In general I agree with @mixedmath, but when it comes to exercises from very standard texts such as Rudin, including the number in the title might have some merit. In my opinion, this should then be a supplement to a description of the mathematical content of the question. $\endgroup$ – Daniel R Aug 31 '15 at 8:02
  • 9
    $\begingroup$ Comment #4. No, bad titles are bad. And titles that don't let you guess what's inside the questions are bad. Even "Book bla problem 3.21" is a bad title, and should be at least extended by some short description of what the problem contains for those who don't have the book (and not to mention the question's body). $\endgroup$ – Asaf Karagila Aug 31 '15 at 11:25
  • $\begingroup$ @mixedmath That looks much more like an answer than it does a comment. (It might be good to post it as an answer) $\endgroup$ – Milo Brandt Aug 31 '15 at 22:52

Titles like "limit question 4" appear because the site blocks the submission of a question, if a question with identical title already exists. Most likely, the author's intended title was "limit question". They were told they couldn't use that one; adding a number to the end seemed the easiest solution (and if 2 is already taken, just put 3, etc).

Nearly all of those questions are garbage; my Review+ extension uses "title ends with a number" among several heuristic methods to identify downvotable and closable questions.

But if it can be salvaged with an edit, great...

Daniel R.'s comment pertains to a quite different situation: including the number under which the problem appears in some textbook. It's commendable that someone cites their source properly; but the number belongs in the body of the question, not in the title. The title should give readers an idea of whether they would be interested in the question. A reader's thought process is unlikely to be "problem 6.14? No, I don't like that one... I would read if it was 6.26".

In such a case, I normally find a place for the problem number within the question body, and replace the title with something more descriptive.


You must log in to answer this question.