Go to this answer on a question


On that post click on "add/show X more comments" then immediately click "add/show X more comments" on the post directly above it.

A box pops up "You may only fetch comments every 2 seconds (click box to dismiss)"

When you click the box to dismiss, you discover that the "Add comment" box has appeared, but the additional comments that you tried to view are nowhere to be found.

You have to reload the page to find them.


I've reproduced it on both FF and IE 7 on my Windows Vista box


Reproduced (FF3.5, Windows XP SP2). Oddly enough, the first set of comments failed to load with that message, while the second set loaded successfully.

Regardless, it left me with no choice but to reload the entire page in order to view both sets of comments.


The comments all loaded fine for me despite seeing the box.

Opera 10.00 Beta Build 1551
Platform Win32
System Windows XP

  • Do you have any other Browsers on your box? Maybe I ought switch to Opera... – devinb Jul 14 '09 at 21:38

I can confirm the bug on Chrome / Vista SP2


I reproduced this in Google Chrome, Win32, WinXP SP3. None of the additional comments for the second clicked link appeared after dismissing the box and I had to reload the page to get them.

Also reproduced in IE8 - same system.

However, I'm not sure this is really a bug. The warning tells you what happened and you can refresh the page to get the comments. Is this really something we want the guys spending time on fixing?

  • 1
    At the very least append a (refresh the page to fetch them) message. – Vinko Vrsalovic Jul 14 '09 at 21:30
  • @Jeff, I wouldn't consider it a high-priority bug, but a bug nonetheless – devinb Jul 14 '09 at 21:37
  • @devinb: I guess it could be a bug but I can also see how it could be by design, which would make it an oversight and therefore a feature request. For the dev team to decide. But I agree, it's not high-priority. :) – Jeff Yates Jul 14 '09 at 21:39
  • @Jeff, bad design is a bug. =P – devinb Jul 16 '09 at 13:45
  • @devinb: That's probably a matter of opinion. If it is working as designed, that's not a bug. But from the perspective of a user, it may be perceived that way. – Jeff Yates Jul 16 '09 at 17:51
  • @Jeff: If I design a slot machine without a 'pull arm', it would not matter if it was implemented perfectly to spec. It's still a bug. It does not matter what stage in development the bug was created (bugs can exist before you reach the implementation stage) they are still bugs. At this point, I'm not arguing about this specific bug, I'm just saying that in general a bad design is a bug even if it was perfectly executed by the coder. – devinb Jul 16 '09 at 18:49
  • but, (+1) for sticking up for your opinions in a respectful, well articulated and thoughtful manner. – devinb Jul 16 '09 at 18:52

As for what the replacement behavior should be, I'd recommend keeping track on the client (the browser) how many seconds have passed since the last fetch, and queue the request so that it won't be sent until, say, three seconds have passed since the last fetch.

To the user it would just look as if the comment is taking a bit longer than usual to load, and the server won't get the overhead of having to receive the request and then deny it.

  • Well, the expected behavior would be the same as the "you cannot upvote more than once every X seconds" or "you cannot comment more than once every X seconds" it would just ignore your request, and you'd be allowed to do it when the timeout has expired. – devinb Jul 24 '09 at 18:07
  • Well, that's the expected behavior right now, yes, but I proposed the alternative behavior because it's simple to implement (since it's only JavaScript), it takes away an extra operation from the server (most of the time, the server would still need to stop comments being fetched too often to prevent abuse), and it's more intuitive to the user (clicking once is better than clicking twice.) – Blixt Jul 24 '09 at 18:11

You must log in to answer this question.

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