7
$\begingroup$

The question sandbox is now a well used tool on world building and some users have now posted multiple questions there. As the sandbox treats users posts as answers this has led to users conceivably having many answers to one question. Is there a limit on the number of answers a user can post on a question and if so how can we avoid this limit in order to allow active sandbox users to continue posting on the sandbox?

$\endgroup$
  • $\begingroup$ I don't think there's a limit, but I haven't done research. There's also a recent proposal here somewhere to restart the sandbox periodically when it gets unwieldy. (Sorry, no time to find it now but maybe this comment will help.) $\endgroup$ – Monica Cellio Aug 18 '16 at 22:36
  • $\begingroup$ Don't forget that as you graduate your question to the main site, you are requested to add a link in the aggregate CW answer, then self-delete your sandbox answer. $\endgroup$ – cobaltduck Aug 19 '16 at 12:19
8
$\begingroup$

While I could not find the limit of answer per user on the same question, I've found that there are no limit to the total number of answers to a question.

But there is apparently a limit on the number of posts (not counting comments) for the site:
2 147 483 647 ($2^{31} - 1$). I doubt we are going to reach that.

Source: Is there a maximum number of answers to one question?

Edit: assuming that we still have an average of 4.7 answers per question, like during the beta phase, with 7295 questions, we should have about 41 581,5 posts just on the main site after almost 2 years (yes there is a half answer somewhere, plenty in fact). At this speed, it will take us 103 290 years to reach the limit.

$\endgroup$
  • 2
    $\begingroup$ Oh whew; at first I thought you said 2 million. :-) $\endgroup$ – Monica Cellio Aug 19 '16 at 3:21
  • $\begingroup$ I took the liberty to comma-ize the number to avoid the potentially unpredictable word wrapping issue, as well as add the derivation. $\endgroup$ – a CVn Aug 19 '16 at 9:19
  • $\begingroup$ @MichaelKjörling Actually the original message was with commas so I decided to get rid of them to avoid confusion, not the opposite. $\endgroup$ – Vincent Aug 19 '16 at 13:59
  • $\begingroup$ Well, this way works, too. $\endgroup$ – a CVn Aug 19 '16 at 14:05
  • $\begingroup$ I think I see a dare here. I wonder if we can convince every user to provide a decent answer on every non-close-worthy question... $\endgroup$ – Frostfyre Aug 19 '16 at 16:11
  • $\begingroup$ If a site reaches that is it frozen or closed or what? $\endgroup$ – Bellerophon Aug 19 '16 at 21:15
  • 1
    $\begingroup$ @MichaelKjörling: You could of course simply write the number in MathJax with \,. $\endgroup$ – celtschk Aug 20 '16 at 14:23
  • $\begingroup$ @Bellerephon by the time that limit is reached it will be feasible to simply up the limit to $2^{64}-1$. Its already possible, but not done to save a bit disk space on the index since its unnecessary. $\endgroup$ – Polygnome Aug 22 '16 at 9:59
  • $\begingroup$ @Polygnome That would be $2^{63} - 1$, actually - remember the bit for signing (SQL Server indices are signed ints). Still, a large number. And by the time that's exhausted, 256-bit integers may well be commonplace. $\endgroup$ – ArtOfCode Aug 22 '16 at 16:59
  • $\begingroup$ @ArtOfCode Nothing prevents you from using the negative indices as well ;) Its just not done that often in practice. $\endgroup$ – Polygnome Aug 22 '16 at 17:41
  • $\begingroup$ @Polygnome Sure, you can use them, but even including them it's only $2^{63}-1$ - you never use one bit because it's used for signing. $\endgroup$ – ArtOfCode Aug 22 '16 at 18:04
  • 1
    $\begingroup$ @ArtOfCode $2⋅2^{63}=2^{64}$ Utilizing negative numbers gives you the one bit loss back... because the information is not "lost" for the sign, you use it. $\endgroup$ – Polygnome Aug 22 '16 at 19:13

You must log in to answer this question.

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