I have seen many edit suggestions approved which I consider to be minor edits to old posts (which bump them to the front page), or to actually reduce the quality or readability of the post. Other problems also occur, but these appear to be the most common. As such, I would like to submit for discussion the feature request in the title:

When reviewing suggested edits, cancel rejection and approval votes. More precisely, a suggested edit would only be approved after the difference between the number of approval votes and the number of rejection votes reaches the threshold.

If this feature request is approved, it may be appropriate to also implement the following:

Remove suggested edits which have accumulated a certain number of rejection votes (for example, three rejection votes).

| |
  • 2
    $\begingroup$ Reviewing suggested edits is somewhat biased towards approval, in that rejecting an edit requires to specify a reason, while approving it does not. $\endgroup$ – Stefan Kohl Nov 22 '13 at 14:37
  • 2
    $\begingroup$ @Stefan Kohl: You make a good point. Consider my suggestion then as a manner to introduce a minor bias towards rejection. In fact, I also think we should increase the threshold of required approval votes to three, but I kept that out of this feature request. $\endgroup$ – Ricardo Andrade Nov 22 '13 at 14:39
  • 2
    $\begingroup$ Whether an edit improves LaTeX typesetting is often not easy to judge, since formulas are not MathJax'ed in the review (at least I see them written there sometimes only in source). $\endgroup$ – Stefan Kohl Nov 22 '13 at 14:40
  • 3
    $\begingroup$ @Stefan Kohl: Regarding the difficulty of judging the quality of an edit, and in particular the mathjax typesetting, I usually need to go to the actual post and read it there, and then compare with the suggested edit. Also, many "improvements" to mathjax/TeX typesetting are actually just introducing them where there were none. In several cases, I personally think they are unnecessary and actually decrease the readability of the post, not to mention increase browser issues and slow down rendering. $\endgroup$ – Ricardo Andrade Nov 22 '13 at 14:43
  • 1
    $\begingroup$ Following up on @StefanKohl's remark: a way to see if the new mathjax is alright is to click "improve" as there the preview works in general; one then can "cancel" the improve to get back. Still a bit inconvenient but better than nothing. $\endgroup$ – user9072 Nov 22 '13 at 15:06
  • $\begingroup$ I am confused in the first place about the whole "review suggested edits" thing. On the rare occasions when I've edited a post, the edits took hold immediately --- they never went through a phase of being "suggested". So where are these "suggested" edits coming from? Are they from users without enough reputation to edit directly? $\endgroup$ – Steven Landsburg Nov 22 '13 at 16:35
  • 1
    $\begingroup$ There are some users like myself who use accounts with rep high enough to suggest an edit, but too low to edit directly. When a significant portion of the community has the time to review and apply appropriate standards, the "approved edit" path seems appropriate. Gerhard "Will Become Powerful One Day" Paseman, 2013.11.22 $\endgroup$ – Gerhard Paseman Nov 22 '13 at 16:45
  • 3
    $\begingroup$ @Steven: Exactly. Only users with 2K+ reputation can edit non-CW posts directly. $\endgroup$ – Emil Jeřábek Nov 22 '13 at 16:46
  • 4
    $\begingroup$ And only "trusted users" (20k+) can edit tag-wikis directly. Also, I think every acount's points are high enough to suggest edits; they can even be suggested by non-users/not logged in. $\endgroup$ – user9072 Nov 22 '13 at 17:05
  • 2
    $\begingroup$ I think in order to improve the quality control provided by the suggested edits review queue, it is more important to make the single approval more meaningful than to require more votes to approve an edit. -- Therefore if such change is realistic in any way, I would propose that clicking [Approve] opens a dialog where one needs to tick reasons why the edit improves the post, while rejecting requires only the one click on [Reject]. Also, one person choosing to approve an edit would suffice, and the approver would be listed together with the proposer in the edit history of the post. $\endgroup$ – Stefan Kohl Nov 23 '13 at 14:38
  • $\begingroup$ The last is already implemented (2, not 3)] $\endgroup$ – Abhimanyu Pallavi Sudhir Dec 3 '13 at 8:28
  • 2
    $\begingroup$ Just wanted to let you know that we saw this, there are similar proposals on MSO - each site seems to have different needs and wants when it comes to thresholds, and editing guidelines in general. However, this is something that we really want to be consistent everywhere. I'll follow up here when we've gotten through the discussions. $\endgroup$ – Tim Post Dec 4 '13 at 14:26

You must log in to answer this question.

Browse other questions tagged .