I see two comments of Robin Chapman on the answer Kernel of the tangent map split into many more comments. Am I the only one to see this? Thanks for checking.

  • $\begingroup$ I see the same thing. $\endgroup$
    – Blue
    Sep 11 '10 at 12:39
  • $\begingroup$ @Day Late Don : Thanks for your answer! $\endgroup$ Sep 11 '10 at 12:45
  • 4
    $\begingroup$ Robin Chapman explains here: meta.math.stackexchange.com/questions/779/… $\endgroup$
    – user856
    Sep 11 '10 at 12:47
  • 9
    $\begingroup$ Wow. This is the most annoying 'feaure' so far... $\endgroup$ Sep 12 '10 at 18:21
  • 1
    $\begingroup$ Why do you keep retagging it as "bug" if it's apparently not a bug? $\endgroup$
    – Rasmus
    Oct 16 '10 at 21:58
  • 6
    $\begingroup$ @Rasmus: I think to make the point that we think it is a bug, or at least, a surprising and undesirable change. As I see it, we have basically no agency in getting the SE people to roll back this undesirable change to our site. In turn, it seems fair for them not to stop us from expressing our dissatisfaction with it. $\endgroup$ Oct 17 '10 at 4:53
  • 3
    $\begingroup$ Shouldn't this dissatisfaction be expressed on meta.stackexchange? And is it really such a huge problem? I mean, 99% of the users or more cope with it quite well. $\endgroup$
    – Rasmus
    Oct 17 '10 at 11:02
  • 2
    $\begingroup$ Rasmus, have you done a survey to justify your figure of 99% "coping with it quite well"? $\endgroup$ Oct 17 '10 at 12:13
  • 3
    $\begingroup$ @Robin: Does reading a certain amount of comment threads qualify as survey? $\endgroup$
    – Rasmus
    Oct 17 '10 at 18:02
  • 4
    $\begingroup$ @Rasmus: No it doesn't. For instance, my comments may appear fine, but I have to often struggle quite a bit with editing comments (sometimes have to type them in another editor and paste them here, out of fear of posting accidentally). So I'm most certainly not included among users who "cope with it quite well", but you couldn't tell from reading comment threads. More importantly, why should anyone have to cope with something like this? The benefits are minuscule compared to the hardships. $\endgroup$ Oct 17 '10 at 20:42
  • $\begingroup$ Thank you, ShreevatsaR. Apparently, I was too hasty. For what it's worth, I like the new feature -- but could cope as well without it. =) $\endgroup$
    – Rasmus
    Oct 17 '10 at 22:29
  • $\begingroup$ @T..: If you want to write to Jeff, add a comment with "@Jeff". Revisions to the post won't be shown in a user's inbox. $\endgroup$
    – kennytm
    Oct 21 '10 at 15:56
  • $\begingroup$ @KennyTM: how would Jeff receive it if he did not post an answer or comment in the thread? This is why I wrote it as above: he seems to read discussions tagged [bug]. $\endgroup$
    – T..
    Oct 21 '10 at 17:22
  • $\begingroup$ @T..: meta.stackexchange.com/questions/36052/… $\endgroup$
    – kennytm
    Oct 21 '10 at 17:23
  • 8
    $\begingroup$ @Jeff Atwood: the math.SE moderators are now wasting their time fixing this problem instance by instance, by manually splicing together comments that are disrupted by the new feature. If it looks like a bug, and walks like a bug... [and by the way, thanks for the upgrade to the @comment feature allowing this message to be sent to your inbox.] $\endgroup$
    – T..
    Oct 21 '10 at 18:10

Another problem with this "feature" is that comments are posted while still being edited.

This leads to problems when other users quickly comment on the half-baked comment, causing awkward out-of-sequence effects in the comment threads as replies address the half-baked material, which is then deleted or edited or reposted in version 2.0 after the comments discussing it have already appeared. These time machine causality effects make some of the comment threads hard to understand for readers not involved in the exchange of remarks.

A genuine click on the POST THIS COMMENT button should be required before comments are broadcast.

  • 3
    $\begingroup$ Thanks for these words of common sense; I fear that common sense may not prevail, alas. $\endgroup$ Oct 16 '10 at 19:50

There has already been feedback against this new "feature" whereby Pressing Enter in comment box unexpectedly submits form. Despite many comments there showing how this "feature" is undesirable, there is no reply at all from the Stack Exchange people, and the latest edit by Jeff Atwood on the question here shows it's not as if they're unaware of this either. Chalk this up as another instance of Jeff Atwood autocracy, I guess. I find this annoying as well, and don't know what to do. :-)


It is likely because Robin pressed Enter which causes the comment to be submitted. I have merged the comments manually.

  • $\begingroup$ Thank you very much! $\endgroup$ Sep 11 '10 at 13:15
  • 7
    $\begingroup$ So this "feature" either (1) consumes moderator time, or (2) makes it difficult for one of the biggest contributors to the site to fully use comments (e.g., having to express everything within a few words or have it mangled). $\endgroup$
    – T..
    Oct 17 '10 at 2:14

I blame me. OK, I didn't make the actual code changes, but I did hint more than a few times that enter could submit comments.

Either way, Jeff is aware that it is causing problems so maybe the old behaviour will revert before too long. In the interim, I suggest (to borrow a meme from "chat"):

/blame @marc
  • 5
    $\begingroup$ Perhaps you could advocate a remedy to this act of deliberate sabotage? $\endgroup$ Nov 6 '10 at 7:57
  • $\begingroup$ Marc, thanks. There's no reply from Jeff, so maybe you can say something about what's going on? $\endgroup$ Nov 26 '10 at 5:52
  • $\begingroup$ @ShreevatsaR at the moment? nothing as far as I know $\endgroup$ Nov 27 '10 at 20:38
  • $\begingroup$ @robin do you recognize Marc? Apparently he was once a student of yours. $\endgroup$ Dec 8 '10 at 10:17
  • $\begingroup$ I am testing whether $\endgroup$ Jul 13 '11 at 14:47
  • $\begingroup$ the behavior has been reverted. It seems it has not. $\endgroup$ Jul 13 '11 at 14:48

You must log in to answer this question.

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