There have been a number of questions in the Close part of Review lately which were basically asking for help creating an algorithm to do some mundane task (see here, here, here for example). I wonder if some of these people could be helped by migrating them over to https://softwareengineering.stackexchange.com/. Even if you don't think these questions in particular should go there, I'll bet that those who are okay with migration in general would be okay with the option to migrate to the programmers website. Hence my question: are others in favor of adding this feature? If so, can someone make a request to the appropriate powers?

  • $\begingroup$ Just be careful when choosing which programming website to migrate to. The first of the cited questions would be answered with "Why don't you do your own homework yourself?" on every math. or programming related forum I'm a part of. Also I am as reluctant to allow myself to judge what level a not completely obvious programming question appearing on MO is as to allow the programmers judge whether they should move a math. question from their site to MSE or to MO. $\endgroup$ – fedja Aug 30 '13 at 18:47
  • $\begingroup$ I removed the discussion tag since it didn't make much sense. Posting a feature-request is the right way to make a request to the appropriate powers. $\endgroup$ – François G. Dorais Aug 30 '13 at 20:20
  • $\begingroup$ Somewhat related meta.mathoverflow.net/questions/572/… a request for a path to cstheory.SE though discussion suggested maybe rather we should have cs.SE (the more general one). And for the first of the question actually this might be the right site (though I am not really sure). I do not know programmers.SE well but I really doubt we get any reasonable number of question that fit there; maybe SO but Programmers seems not needed to me. David White do you know programmers.SE well? Perhaps I am wrong. $\endgroup$ – user9072 Aug 30 '13 at 21:24
  • 8
    $\begingroup$ You shouldn't suggest migration to sites you are not familiar with. The scope of Software Engineering is not what you think it is. This is one of the reasons opening user migration paths is taken seriously. I think to open a migration path you need approval from both sides, not just the migrating site. Even on cstheory we don't get that many programming questions and I don't think MO will be much different in that regard. $\endgroup$ – Kaveh Aug 31 '13 at 5:23

Yes, it's possible, but as Anna Lear explained in an earlier answer, there are some requirements. There needs to be a clear pattern of questions to be migrated and the migrated posts need to have low rejection rate on the target site. In the mean time, if you need to have a post migrated, just flag the moderators and explain the situation. Moderators can migrate any new post to any site in the network.

  • $\begingroup$ Okay, thanks for the answer. I don't think there is enough of a pattern yet for the programming migration but my gut says we're headed that way. I guess time will tell. $\endgroup$ – David White Aug 30 '13 at 20:48

I don't really like our setup at the moment. Anna said we need to establish precedents of migrations before the open migration paths, but for now only moderators can actually do these migrations. Because we're actually a long established site where moderators area relatively rarely involved in closing questions, we're sort of stuck.

It would be helpful, I guess, if posts that should be migrated get flagged, even if they are already closed. We can reopen then migrate them.

  • $\begingroup$ I think to create a culture of things getting flagged for migration would be important, regardless, since as explained in the post linked by François G. Dorais there is an upper-limit of four (other, ie excluding meta) sites with open migration paths. At the moment we have two (math and stats) but say some migrations to tex, academia, mathematica, scicomp, cstheory, cs and a couple others could be useful on ocassion; a lot more than four. [At the moment, some of the sites are still in beta so here is another reason for no path, but since this is about long-term.] $\endgroup$ – user9072 Aug 31 '13 at 10:00
  • $\begingroup$ @quid: They will increase if there is a need, but I think we need to train users to know when & how to migrate before giving them a longer list. $\endgroup$ – François G. Dorais Aug 31 '13 at 11:44
  • $\begingroup$ @FrançoisG.Dorais (or also Scott Morrison): personally, I somewhat changed my mind about a longish list (from the time of the first discussion). I doubt it will ever work well in practice (except for math.SE) as migrations to sites other than math.SE seem too rare and also due to the feelings of some against migration. Flagging for moderators seems more viable to me. What is your opinion: would it be good/bad/neutral for somebody to go over recently closed posts and to flag things they consider as good candidates for migration? $\endgroup$ – user9072 Aug 31 '13 at 13:40
  • 1
    $\begingroup$ @quid: If done carefully this would help identify potential target sites. Be very careful to make sure the questions are actually suitable for the target site because rejected posts will negatively impact the possibility of opening a migration path. $\endgroup$ – François G. Dorais Sep 1 '13 at 2:20
  • 3
    $\begingroup$ That's what needs to happen in a nutshell. If you migrate 10 questions a month to Stack Overflow and only 2 or so are rejected in the same period, then a migration path from you to SO would likely make sense. We just need some statistical assurance that questions too programmy for MO would be well received on SO in order to be certain that we're providing a better experience for users by setting up the path. If most of them were rejected (closed once being migrated as anything other but a duplicate) then we're not really improving the experience of the question author, which is the goal. $\endgroup$ – Tim Post Sep 3 '13 at 15:37

You must log in to answer this question.

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