7
\$\begingroup\$

I often prefer to improve other people' answers instead of writing one of my own (unless mine would be considerably different). To be entirely honest, I do this for petty selfish reasons - to show off that I can, to avoid having to explain the answer myself, and maybe to gain some reputation.

The typical behaviour I observe then is they either suggest to me I should post a separate answer (which I, being lazy, refuse) or they credit my username with the number of characters I saved. That's why sometimes I only give clues, to make the author come up with the improvements alone, and to avoid complicating things.

What would be a good rule of thumb to tell if an improvement is worth posting as a separate answer? From the PPCG community point of view, is an unexplained shorter solution preferable to an explained longer one? Is credit necessary? Wouldn't it be better to upvote the comment instead of @-mentioning in the answer?

\$\endgroup\$
  • \$\begingroup\$ How does improving others' answers gain you reputation? You mean "street cred" rather than SE rep? Upvotes on comments does not increase your SE rep. \$\endgroup\$ – Adám Feb 20 '18 at 21:47
  • \$\begingroup\$ For what it's worth, I quite liked it when you have me a clue in the comments (even if I didn't solve it on my own) \$\endgroup\$ – H.PWiz Feb 20 '18 at 21:53
  • 3
    \$\begingroup\$ @Adám I didn't even realise it doesn't increase SE rep ... Anyway, I can always post an alphabet triangle challenge if I run low on rep :) \$\endgroup\$ – ngn Feb 20 '18 at 21:57
  • \$\begingroup\$ @H.PWiz glad to hear that :) but some may not like it because it means spending extra time to think, apart from time for re-explaining the solution \$\endgroup\$ – ngn Feb 20 '18 at 22:00
  • \$\begingroup\$ @ngn You may actually run low on rep. You lost almost a tenth of your rep this week! \$\endgroup\$ – Adám Feb 20 '18 at 22:10
  • \$\begingroup\$ @Adám great, there's still enough fuel for nine full-bonus challenges :) \$\endgroup\$ – ngn Feb 22 '18 at 11:42
  • \$\begingroup\$ I delete my improvement suggestion comments after they are obsolete, so it makes sense from my perspective that the change is recorded in the edit or answer. If it's not credited, it might get confusing whether the edit clashed at the same time with the comment or the comment had a direct effect. \$\endgroup\$ – Unihedron Feb 28 '18 at 0:36
  • \$\begingroup\$ @Unihedron If you delete your comment, it won't be clear what part of the code you improved. It probably doesn't matter all that much, but I do take pleasure in showing off golfing tricks. \$\endgroup\$ – ngn Feb 28 '18 at 1:02
  • 1
    \$\begingroup\$ About explanations: my personal opinion is that an answer with an explanation is always better than an answer without one, and I rarely upvote answers that have no explanation. Why? Because I want to appreciate your cleverness. Unless your submission is a pretty straightforward Python/C/JavaScript/QBasic program, I generally won't be able to understand what you did unless you explain it to me. That said, it doesn't have to be super in-depth, just a quick summary of the algorithm and any funky syntax or interesting golfing tricks you used. \$\endgroup\$ – DLosc Feb 28 '18 at 2:56
19
\$\begingroup\$

You are free to choose whether to comment or post your own answer

There is absolutely no threshold upon submitting your own solutions, you are free to choose whether to leave a comment or post your own answer. Yeah, even duplicate answers are allowed! I personally comment:

  • If the improvement(s) is/are relatively small.
  • If I would not have been able devise my own answer to be shorter than the original one, without seeing it first.

I personally answer separately:

  • If I arrived at my solution independently, and if it is not identical to another answer that predates mine.
  • (Sometimes) If the other user tells me that I should post it separately.
  • If I feel like my improvement(s) is/are significant enough to warrant a separate answer.

When it comes to giving credit, I think it is necessary. Morally speaking, you should mention those that helped improving your post, regardless of how significant their suggestions were.

Giving clues instead of spoiling is also completely fine. That might also incentivize competition, which is definitely in the spirit of PPCG.

From the PPCG community point of view, is an unexplained shorter solution preferable to an explained longer one?

Of course, explanations are encouraged, but not strictly necessary. No solution is preferable over another one, as long as both meet our rules.

\$\endgroup\$
  • \$\begingroup\$ Another possible reason to comment/new answer: if you use a different algorithm, or change the existing algorithm substantially. \$\endgroup\$ – Rɪᴋᴇʀ Mar 3 '18 at 16:38

You must log in to answer this question.

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