24
$\begingroup$

I recently came across a problem statement question about a block matrix inverse. The original author did not include context, and the question got put on hold.

I personally know a lot of situations where matrix inverses like this occur, as I regularly encounter matrices of this form in my research. They are quite interesting (to me at least).

Is it appropriate or innapropriate for me to improve the question by adding my own version of context, considering that the author of the question has supplied none?

$\endgroup$
  • 2
    $\begingroup$ I don't see any problem with this so long as you frame the question in a light where OP comes across as genuinely wanting to learn. So often a question is just, "I have this problem. [Something something] I don't know where to start!" These are certainly going to be put on hold and downvoted very quickly, but I think you can add context in the sense of making the question not look like a desperate plea for help. You could even add a "thought or two" so long as you don't destroy the integrity of the post by adding a great deal of information that is not OP's. That's just my opinion. $\endgroup$ – Daniel W. Farlow Jul 29 '15 at 17:32
  • 1
    $\begingroup$ Some related discussion in chat: chat.stackexchange.com/rooms/2165/conversation/… $\endgroup$ – Martin Sleziak Jul 29 '15 at 17:37
  • 4
    $\begingroup$ You could consider asking your own question which links to the original as inspiration. $\endgroup$ – Peter Woolfitt Jul 29 '15 at 19:08
  • 12
    $\begingroup$ @PeterWoolfitt Well I already know the answer (and indeed supplied an answer before the question got put on hold). The motivation for editing and trying to get it repoened is more in line with the philosophy that stackexchange sites should not just help question askers, but also function as a repository of good question answer pairs for future people. $\endgroup$ – Nick Alger Jul 29 '15 at 22:22
  • $\begingroup$ This answer explicitly mentions in connection with old posts that: "They can be edited to bring them into a more appropriate shape". However, it is possible that the views of the site have shifted since then and maybe today most users would prefer posting a new question. $\endgroup$ – Martin Sleziak Apr 15 '18 at 6:30
  • $\begingroup$ Here is another answer which mentions editing a question as an option. However a lot has changed on the site since those answers were posts (in 2014 and 2015). Perhaps at some point this issue has to be discussed again. $\endgroup$ – Martin Sleziak May 1 '18 at 14:40
  • $\begingroup$ An instance of this was discussed here. That was to some extent about two fresh moderators playing with their new toys, but IMO some good points relevant to this discussion were also raised. $\endgroup$ – Jyrki Lahtonen Dec 19 '18 at 8:47
19
$\begingroup$

To edit posts to improve them is appropriate. However, one should make sure to respect the author's intent.

Moreover, I recommend to focus such activities on questions that one actually finds worthwhile.

$\endgroup$
4
$\begingroup$

I'd been advised about this before, and been told that an appropriate thing to do would be to ask a new question: your question, in your voice with your context (so it won't be closed). Then flag your question for moderation attention, requesting that the other question be merged with yours. This way any good answers to the old question won't be overlooked or lost just because the old question was closed for missing context.

The idea is that editing the other question by another user is "putting words in their mouth".

$\endgroup$
3
$\begingroup$

But how sure can you be that the context you supply is anywhere close to the asker's context?

If you're absolutely certain of what the original context is, then maybe it's appropriate to supply it. But if you're any bit less than absolutely certain, then I think it is inappropriate for you to supply your own context.

I think the best thing to do in such cases is to add a comment prompting for context. "Did you mean to refer to positive integers only?" "Are the triangles supposed to be equilateral?" "Do the numbers have to be distinct?" etc.

$\endgroup$
  • 7
    $\begingroup$ Doesn't the questioner receive notice of the edit? If so, then they are free to revise the added context if need be. In either case, the question is improved and any good answers are better protected against deletion, $\endgroup$ – Bill Dubuque Aug 4 '15 at 22:02
  • $\begingroup$ @Bill you'd know that better than me. $\endgroup$ – James47 Aug 4 '15 at 22:04
  • 1
    $\begingroup$ I also like to respect the original author's intent as much as possible, but if you combine this with all the other rules here, one is forced to reach an unfortunate conclusion -- Whenever a potentially good question is asked with no context, that question becomes off-limits from math.stackexchange forever. Realistically the original author is never going to edit it, and if someone else asked the same question with their context, it would be closed as a duplicate. $\endgroup$ – Nick Alger Aug 5 '15 at 4:28
  • $\begingroup$ Speaking only for myself as someone who occasionally reviews suggested edits, I'd be inclined to reject an edit that assumes too much on the original author, but maybe that's a short-sighted way of looking at it (and the way the review queues are structured, short-sightedness is encouraged). Of course in your case, Nick, your edits wouldn't come up in the suggested edits review queue. I don't know if I would be so pessimistic about the OP editing, I suppose there's a bit of a judgement call here: a user with a reputation score of, say, 150, probably cares enough about the points to react to dv $\endgroup$ – Robert Soupe Aug 6 '15 at 4:32
0
$\begingroup$

Presumably your answer already explained your own version of the context of the question. At least, if I were answering a question that I found interesting despite OP's lack of context, I would put something in my answer to describe the context in which I found the question worth answering.

Someone searching for that context might not find the original question, but they would likely find your answer, which would lead them to the desired question-answer pair.

If the question is on hold and you think there is a good possibility for someone to usefully add yet another answer, you could (with at least 3000 reputation) vote to reopen the question.

A comment on the original question itself explaining an interesting context for the question might encourage other people to give useful answers if they have any (if the question is reopened).

$\endgroup$

You must log in to answer this question.

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