7
$\begingroup$

We have a few questions where it seemed like cs.SE might be their natural home. For example,

and I imagine we'll have more as time goes on.

We have a migration policy that is very conservative: for the most part we merely close and suggest another venue. The question here is:

Should we have a different policy for cs.SE, because we can fairly claim to understand the forum a lot better, and a number of us are active there (including all three cstheory mods). Specifically, should mods be free to migrate to cs.SE and not merely close and suggest reposting ?

| |
$\endgroup$
  • 2
    $\begingroup$ For the record, I'm in favour of migrating the third question you mention. $\endgroup$ – Dave Clarke Mar 27 '12 at 15:10
12
$\begingroup$

As one of the moderators, I'm in favour of migrating questions that we see fit. I think the policy should be different for the reasons you mention: we understand the forums a lot better and we are active in both. This will avoid the problems we had a few months ago where strange questions were migrated from StackOverflow, simply because the moderators there had no feeling for what CSTheory was all about.

I think that this would be good for both communities, especially cs.SE, because it will be more positive for the OPs being directed to the more appropriate site. Closing a question is always harsh, especially for new users who are unaware of the level of CSTheory.

In one of the questions, there was some doubt, so I asked for a majority vote before migrating. This too could be a part of the policy we apply at CSTheory.

| |
$\endgroup$
  • 3
    $\begingroup$ +1 for the 'harshness' issue. $\endgroup$ – Suresh Venkat Mar 27 '12 at 14:58
  • 5
    $\begingroup$ I am pro migration, but please keep in mind that we are supposed to migrate only good questions. One or two migrated questions were initially not well posed irrc. In such cases, closing and asking them to rephrase for cs.SE might be better. We can always overhaul the question there, though. $\endgroup$ – Raphael Mar 28 '12 at 21:22
  • 1
    $\begingroup$ The good questions point is noted. $\endgroup$ – Dave Clarke Mar 29 '12 at 8:31
8
$\begingroup$

The current practice is to migrate an off-topic question only when it is completely clear that it is suitable on the destination site. You are saying that sometimes it is completely clear that a question is suitable on cs.stackexchange.com. I do not see any need for a different policy here.

| |
$\endgroup$
2
$\begingroup$

I think this migration path works fine. I have one request, though: Please delete comments that will be obsolete after the migration ("This is offtopic here", "Maybe we should migrate that to cs.SE?", ...) before hitting the button. Thanks!

| |
$\endgroup$
  • 4
    $\begingroup$ No, the comments should not be deleted on the originating site, because they are the record for why the question was migrated. The system should at least add a comment saying something like “the question was migrated from cstheory.stackexchange.com to cs.stackexchange.com” automatically so that people know which comments are before migration and which are after migration, but this is not currently implemented. Personally I do not use “here” and always say either “cstheory.stackexchange.com” or “cstheory” instead, but it is cumbersome. $\endgroup$ – Tsuyoshi Ito Apr 2 '12 at 20:36
  • 3
    $\begingroup$ I think this will stop being an issue when Computer Science gets its own moderators who can remove those obsolete comments. This is one of the reasons that I think having a co-moderator between the two sites would be quite helpful. $\endgroup$ – Kaveh Apr 2 '12 at 23:54
  • $\begingroup$ Agreed, the problem should vanish soon. $\endgroup$ – Raphael Apr 3 '12 at 5:49
0
$\begingroup$

The problem with migrating can be seen in this question : Is there some other way to describe a formal language other than grammars? The OP did not "migrate" with his question, making a future not answered question, eventhough there are plenty of answers.

I think it would be better to direct the OP toward the cs.SE site, and let him ask his question again than migrating. I am not sure it is optimal for a beta-site to have questions with no accepted answers.

| |
$\endgroup$
  • 3
    $\begingroup$ That the OP doesn't migrate is a problem, though the fact that an answer is not accepted is less problematic, IMHO. In this case, the OP has not logged on since yesterday, so perhaps he has not yet seen that the question has been migrated. $\endgroup$ – Dave Clarke Mar 28 '12 at 10:41
  • $\begingroup$ The fact that the answer is not accepted is less problematic in our case, but is it also the case for a site that is on beta? Doesn't it need a certain percentage of question with accepted answers to get out of beta? What if everyone were to do this? $\endgroup$ – Gopi Mar 28 '12 at 11:20
  • 2
    $\begingroup$ I'm not sure that it does matter if the question has no accepted answers. What matters more is the number of answers and the number of unanswered questions. I do agree that it is problematic, but I think for people who really care about their questions, it won't be a problem. $\endgroup$ – Dave Clarke Mar 28 '12 at 11:22
  • 1
    $\begingroup$ If a question is migrated the OP gets a message telling him so. As long as he comes back to check on his question, he'll find it. $\endgroup$ – Peter Taylor Mar 29 '12 at 13:04
  • $\begingroup$ @Gopi, AFAIR, a question with an upvoted answer is considered answered by SE software even if no answer is accepted by the OP. $\endgroup$ – Kaveh Mar 29 '12 at 18:11
  • 2
    $\begingroup$ The user has since migrated and even accepted an answer. I'd count that as win for migration. $\endgroup$ – Raphael Mar 30 '12 at 12:30

You must log in to answer this question.

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