We have had multiple instances of the following process recently.

  1. An on-topic question is migrated to us (cs.SE).
  2. We close as "unclear" because it lacks detail and/or effort, telling the user that the question may be reopened when they improve the question (standard operating procedure).
    The question is put "on hold".
  3. The user can not edit because the question is now locked and opens a new question.

This is obviously bad behaviour: a (new) user trying to conform with site policy get a firm stick between the legs. Also, the migrating site gets told "please don't migrate stuff like this" even though the question could be salvaged.

I found that rejected migrations are locked automatically for some reasons (1, 2) that are not invalid. I think that only migrated questions closed as "offtopic" should be rejected, or we should be able to choose whether to reject, but I guess the situation in these regards is [by-design].

As a compromise, I propose that migrations be not rejected until their status changes from "on hold" to "closed" (or it is deleted). That leaves the asker some time to improve their question and does not send wrong signals to the migrating site prematurely.

This still sends bad signals if the user does not improve their post, but I guess we'll have to take that hit unless we only reject migrations if they are closed as offtopic.

  • 4
    Really, if the question is unclear and needs closed, we shouldn't be migrating it in the first place. – animuson Feb 2 '14 at 17:39
  • 5
    @animuson: That's impossible to decide (in practice) because policies differ. For instance, Mathematics routinely answers homework dumps while Computer Science does not. (Note that the new state "on hold" was specifically introduced to encourage users to improve their unsuited questions.) – Raphael Feb 2 '14 at 17:48

We have a clear policy of "not migrating crap." Had the user asked the question on the correct site, it would have been evaluated on its merits there.

Questions should be closed without migration when they are not quality questions, based on the standards for question quality on the site where the question was asked. If people are migrating questions without considering question quality, they're doing it wrong.

Because the mental deficiency that causes the posting of the question to the wrong site is often the same one that causes the question to be a poor-quality one, it is rare that I actually migrate questions. I see migration as an escape hatch for very good questions that just happen to be posted to the wrong site, not a means of directing traffic.

Migrating marginal questions causes other problems; the OP will often repost their question on the correct site before we can get their original migrated, and now you have Two Problems.

In other words, I don't think we should change the way the system works just because some people don't know how to migrate correctly.

  • Thanks for explaining how things should work, but you did not address my points. Restating my comment on the question: policies differ. A question that may be offtopic but look okay on, say, Software Engineering or Mathematics may still be put on hold on Computer Science. We can't expect people to know/understand policy on other sites, so they should migrate from their PoV. Furthermore, assuming migration was a mistake (how can it be if the user was happy to improve their question?), why punish a) the asking user and b) the target site for something people on the migrating site skrewed up? – Raphael Feb 2 '14 at 18:31
  • That's exactly the point I'm trying to convey. Because users who migrate questions can't be expected to understand all of the variations in question quality norms on all the target sites, they should simply close low-quality questions, not migrate them only to have them closed on the target site. Remember, the OP can always repost their question to the correct site. Botched migrations are far more disruptive than simply having the OP repost, and question quality standards really shouldn't differ all that much from site to site. – Robert Harvey Feb 2 '14 at 18:35
  • 1
    "and question quality standards really shouldn't differ all that much from site to site" -- well, they do. As long as you assume they did not we won't be able to agree. (For instance, on Computer Science we close perfectly clear and ontopic questions if they don't show effort (e.g. homework dumps). Users on other sites don't know that and maybe can't even tell the difference.) – Raphael Feb 2 '14 at 18:48
  • So what do your proposed new rules buy us? Can't the OP just repost? I am resisting this because I see it as "Oh, the target site will just deal with any quality problems the question may have," and I'm not generally in favor of that, for reasons that I've already explained. What prevents you as a mod from simply clearing the migration history on the target site to unlock the post? – Robert Harvey Feb 2 '14 at 18:51
  • BTW, Stack Overflow and Programmers both treat homework dumps and code dumps in exactly the same fashion as Computer Science does. – Robert Harvey Feb 2 '14 at 18:56
  • 1
    The rules buy us usability. Reposting is of course possible, but why not use the tools we have? Note that following my proposal the target site can still reject crap migrations (just with a time delay). – Raphael Feb 2 '14 at 19:07
  • 2
    I did not know we could fix the issue by clearing migration history; now that's an answer to the problem, thanks! (However, if it's a mod-only feature it does not scale well. For the volume I have encountered, it's a good solution nevertheless.) – Raphael Feb 2 '14 at 19:08

You must log in to answer this question.

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