1

I'm trying to figure out what are good practises for editing questions, and since I don't think there are rules you can always follow, I'll try giving specific cases.

For example, this question's title Dereferencing a structure pointer as a diffrent structure's pointer in C

Now, I'm thinking, after reading the body of the question, that the title should be more like this:

Will my program's output change if I use a different compiler (I'm dereferencing a structure pointer as a different structure's pointer in C)

Would that be a correct edit? And in general, are edits in that spirit (adding details to the title to make the body less surprising)?

9
  • 3
    Your proposed replacement title seems overly wrong... you don't need all of the details in the title. (That said, I don't think the current title is particularly good...)
    – Wooble
    Commented Jan 10, 2014 at 12:19
  • OK, that's the discussion that I was looking for. So what is a good title for this question? A good fit, so to say.
    – sashoalm
    Commented Jan 10, 2014 at 12:21
  • 7
    Without any knowledge of the subject (and without having looked at the question), it seems as though you should just switch it around: "Does dereferencing a structure pointer as a different structure's pointer alter output if I change compiler?" Commented Jan 10, 2014 at 12:34
  • +1 to @ben's suggestion, that looks like a good title.
    – Stijn
    Commented Jan 10, 2014 at 12:37
  • @Stijn Keep in mind that he asks it about his particular code, not in general (if you read the body). I mean, how do we know if he asks it for any code, or for his code in particular, that's why I wrote "Will my program's output" instead of "Will a program's output".
    – sashoalm
    Commented Jan 10, 2014 at 12:39
  • @satuon I also don't have any knowledge on the subject, but if that title isn't a good fit I'd say it's not the essence of the question. But I'm not sure.
    – Stijn
    Commented Jan 10, 2014 at 12:42
  • IMO the question itself is not really good. The poster tried to provide a simple sample but somehow failed to make his point. When creating an SSCE the essence of the problem should be kept intact, and somehow this has not happened here, AFAIK. This particular code doesn't really look like something that somebody would write. More like an example to demonstrate something like inheritance.
    – Devolus
    Commented Jan 10, 2014 at 13:06
  • @benisuǝqbackwards Your suggestion looks good, I changed the title according to it.
    – sashoalm
    Commented Jan 10, 2014 at 14:25

1 Answer 1

3

There are a couple of common problems I see with question titles:

  1. The title is not descriptive; it doesn't describe the problem the user is having.
    Variation: the title is what the user typed into Google to try and find an answer to his question.

  2. The title is "How do I do this thing," but the real question is "Why doesn't this work?" This frustrates Google searchers, who are expecting a "how to" article, but encounter a highly-localized troubleshooting question instead.

For the first category of questions, if the question body is lucid enough where I can summarize the actual question being asked in 150 characters or less (or whatever the character limit is) I'll do that.

For the second category, I change the title to a summary of the specific problem being encountered by the OP. This is the situation you've encountered in the question you linked.

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