Would be useful if you could see in the question, somewhere, the number of downvotes it received, and the number of upvotes it received.

Some questions are downvoted 1 time and upvoted 1 time, then appears the score 0. But no one knows (only the op) that the question was downvoted and upvoted.

Maybe a feature like that would be useful to classify better the quality of the question.

  • They don't want you to see that because they fear you could be suggested by the exact numbert. This is only for high rep users (over 1k) and it's by design. – Danubian Sailor Nov 29 '13 at 13:13
  • @ŁukaszL. as far as I know that's not fear. It's more mundane thing - aggregated votes are easily cached along the post itself. recalculating up / down is costly. Caching 2 values more is, in a way, costly too (in updates cost). cheapest (cpu-wise) is to limit the number of people who would see it and then recalculate if they really want. – Mołot Nov 29 '13 at 14:29
  • @Mołot then it would be limited to registered users to prevent bots from reading that data... there surely is more behind – Danubian Sailor Nov 29 '13 at 15:07

You can see that information when you have enough reputation (1k) for it.

It is a costly operation and that is why it is only available to a limited user base.

| improve this answer | |
  • oh, Thanks for the info, i did not know that, by the way, can you explain me why you need 1k rep to see that? "Costly operation" ? – Paulo Roberto Rosa Nov 29 '13 at 12:48
  • 1
    @PauloRoberto he already did: "It is a costly operation". It means higher database load than just displaying aggregated number. – Mołot Nov 29 '13 at 12:49
  • @PauloRoberto - um. Because it is costly. To limit the number of people using the feature, so it doesn't hammer the servers. – Oded Nov 29 '13 at 12:50
  • ok server performance cost. Now i understand that, thanks. – Paulo Roberto Rosa Nov 29 '13 at 12:51
  • I don't believe doing group-by count on index costs so much more that single count, that it must affect the UI – Danubian Sailor Nov 29 '13 at 13:13
  • @ŁukaszL.: The votes are cached but the statistic must be fetched AFAIK. – juergen d Nov 29 '13 at 13:13
  • 1
    @juergend but it's not an argument, because both can or can not be cached, depending on design. – Danubian Sailor Nov 29 '13 at 13:15
  • 2
    @ŁukaszL. - doing a group by millions of times is more costly than not doing it. – Oded Nov 29 '13 at 13:19
  • 1
    If you go by the SEDE schema @ŁukaszL. it's not the same at all. The total score is denormalised into the Posts table so it's just there and available. Votes are in a separate table so you're adding an additional index range scan (up and down), group by and join for every click. – ben is uǝq backwards Nov 29 '13 at 13:23

You must log in to answer this question.

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