While reviewing the flags on a Programmers post I found this:

Bad Twitter bot

You can clearly see that the post was flagged 5 minutes before the post was Tweeted. While in this case the post remains open, it might not have done and there would have been (another) closed question Tweeted to the world.

Posts with close votes aren't Tweeted so can the same logic be applied to posts with flags?

Yannis raises an interesting point in the comments - perhaps the number of helpful flags raised by the flagger should be taken into account as well. This would mean that a flag from someone with 300 helpful flags will stop the Tweet, while a flag from someone with only 10 won't.

Of course, once the flags have been cleared and if the post is still open then it should be eligible for Tweeting once again.

  • 18
    But then the twitter bot will have nothing to tweet about on P.SE... :) Mar 8, 2012 at 13:28
  • 4
    @MarvinPinto I tend to agree with you, any single flag short circuiting a tweet is probably not what we want. But there is this very nice thingy that gives some of sort credibility to flags, if only I could remember how it was called...
    – yannis
    Mar 8, 2012 at 13:35
  • @YannisRizos But in all seriousness, we really can't expect mods to clear flags quickly enough for the twitter bot to matter. By that, I mean by the time the bot is finally allowed to tweet something, the question won't be as new anymore -- if that sort of thing matters at all (just spitballing). Mar 8, 2012 at 13:40
  • @MarvinPinto Yes you really can't expect us to clear flags so quickly (hear that 10k+ users?). There is a very interesting thing though, both our Marshals don't have the vote to close privilege, I'd say having 500 helpful flags is good enough to short circuit a tweet.
    – yannis
    Mar 8, 2012 at 13:43
  • 5
    Perhaps the whole automated tweeting thing needs to be looked at as a whole per a discussion on the Programmers Meta. I don't think questions that aren't representative of a community should be tweeted, so is there a better way to introduce human review of questions that are scheduled for being tweeted? Mar 8, 2012 at 14:05
  • @ThomasOwens - I agree with that, but this will do as an interim measure.
    – ChrisF Mod
    Mar 8, 2012 at 14:05
  • @ChrisF Yes, anything better than what we have now, that's for sure. Mar 8, 2012 at 14:11
  • But the flaggings might not be genuine.
    – neverMind9
    Mar 20, 2019 at 3:38


Browse other questions tagged .