What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 122 Stack Exchange communities.

(relates to my other question can't upvote edited post)

Suppose this sequence of events occurs:

  1. User X posts an answer (revision #0).
  2. I read revision #0 of user X's answer.
  3. User X edits the answer to revision #1.
  4. I decide to EVALUATE (= upvote/downvote/remove vote/accept, I don't know what you guys call it) user X's answer. But I'm still looking at revision #0 of the answer.

When the client-side part of SO sends an EVALUATE request to the server, does it send the revision # of the displayed answer to the server? If so, could the server send back info to warn the user before accepting the EVALUATE request?

At present there's no indication on the client-side that an answer has been updated. I would have to reload the page to check if the answer I'm voting on is the latest revision.

share|improve this question
add comment

2 Answers

This seems unlikely to happen very often. If you read an answer and want to vote on it ten minutes later, reloading the page seems pretty reasonable

share|improve this answer
    
I disagree. From time to time, I will read an answer, then go try it, then come back and vote. The burden is on me to reload the page before I vote... which may be reasonable, but it's another source of errors. –  Jason S Mar 7 '11 at 18:05
    
...and it may be only a minute later, not ten minutes. –  Jason S Mar 7 '11 at 18:08
    
@Jason Between you checking an answer "from time to time", and people editing their answers from time to time, the odds that they'll exactly overlap and mess you up are really small –  Michael Mrozek Mar 7 '11 at 18:39
    
Well, you might think differently if you had been involved in that discussion for my C++ array initializer question. There were a lot of changes happening very fast. One person had to get me to clarify which part of his post I had been commenting on: while I was composing my comment, he had edited the post. There were at least 3 instances of concurrent editing during the first 20 minutes after I posted my question. –  Jason S Mar 7 '11 at 19:19
    
"Between you checking an answer "from time to time"" -- and why do I have to do this anyway? Why can't the server and client help me do this, instead of me having to reload the page every 60 seconds? –  Jason S Mar 7 '11 at 19:20
add comment

While I agree this could be a problem in a few specialized cases, I don't think it's a big enough deal to 'fix.'

Right now, if an answer/question has been edited after you have 'evaluated' it, you are free to change your evaluation. This is meant to combat a very similar problem that you describe.

share|improve this answer
    
-1, because your 2nd paragraph has a flaw, which was illustrated by my other post (meta.stackexchange.com/questions/82047/cant-upvote-edited-post/…). I upvoted, then the user edited their post, then I removed the upvote, then I saw that the user had edited their post (which had happened before I removed the upvote), then I went to re-upvote, and could not. –  Jason S Mar 7 '11 at 18:07
    
@Jason, so the scenario was - upvote, edit, remove upvote, edit, upvote? –  jjnguy Mar 7 '11 at 18:10
    
The scenario was the sequence him:post#0, me:upvote#0, him:edit#1, me:remove-upvote#0, me:read#1, me:upvote#1 -- where the problem is that because I hadn't refreshed the page, SO server interprets my remove-upvote request as applying to revision #1 (thus blocking me from changing my vote later), even though I was judging the post on revision #0. –  Jason S Mar 7 '11 at 18:20
1  
@Jason, I feel that this scenario would be so rare that it isn't necessary to fix. You should simply edit the post yourself (trivial edit to 'unlock' your vote), and then change your evaluation accordingly. –  jjnguy Mar 7 '11 at 18:26
add comment

You must log in to answer this question.

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