7

As we can see for example here some people happen to post votes based on the username of last editor, not on the content of the post. That should not happen.

Please introduce review audits on Drupal Answers to test if people really pay attention and don't vote because they just expect that this user couldn't make a post OK.

8
  • 1
    Agreed, this would be a good idea
    – Clive Mod
    Apr 17 '14 at 9:34
  • I would be interested in hearing how you go about implementing this. How are your auditors going to tell the difference between someone who rejected a reopen request "based on the username of last editor", and someone who rejected it because, after careful review, found that the edit didn't fix the problem? I can assure that you can't find this out by analyzing voting patterns (I have checked, based upon the presumption that your good self is a very careful and diligent reviewer, and I am a lazy one). Apr 17 '14 at 9:51
  • 2
    @FreeRadical for implementation - it's already there. It is a matter of turning it on. It is turned on on Stack Overflow and various sites in SE network.
    – Mołot
    Apr 17 '14 at 10:10
  • Well, something named "review audits" exists. We probably have different opinions about whether the implementation matches the moniker (see my answer). However, I doubt that this will game will produce more diligent reviews. The most likely outcome of introducing this is that we will have fewer reviews. (Since reviews diverging from majority opinion are penalised, the optimum strategy is not to review if one is not sure about what the majority thinks.) Oct 17 '14 at 11:16
  • @FreeRadical I posted it after some really bad reviews got accepted. For example wrapping list in code blocks, or putting bolds all over. Less reviewers is actually a good thing if it means such edits are not accepted.
    – Mołot
    Oct 17 '14 at 11:19
  • @Mołot. But at least you've achieved your goal (fewer reviewers). I've more or less stopped reviewing as a result of your snide remark about introducing "Review audits" to stop people like me from reviewing (you link explictly to a comment made by me in this feature request). (I have never accepted any of those you correctly call "Bad Edits" tho'.) Oct 17 '14 at 11:56
  • @FreeRadical I posted it after someone openly admitted he reviews based on usernames, not on actual post content. Was it you? Maybe, don't know, don't care, and still thinks reviewers that don't want to read what's actually in post should get scared away from reviews, no matter who they are.
    – Mołot
    Oct 17 '14 at 12:00
  • @Mołot, your last admission about why you posted (as a direct response to me) it true. You even included a link to my statement in your feature request. Please do not pretend this feature request has anything to do with accepting bad edits "wrapping list in code blocks, or putting bolds all over" (this was no longer a problem in April). This was done by a single reviewer that was not me five months earlier. Your attempts to associate my statement in April 2014 to bad reviews made by someone else in December 2013 is a both unnecessary and untruthful. Oct 17 '14 at 23:36
1

While I've been typing, I see that there now is a link to What are review tests audits and how do they work, which I now understand is what you're talking about.

First, I would not call this an "review audit". At least not if one believes that the definition of "audit" is:

planned and documented activity performed by qualified personnel to determine by investigation, examination, or evaluation of objective evidence, the adequacy and compliance with established procedures, or applicable documents, and the effectiveness of implementation. (Source: Wikipedia.)

Like most things on SE, the described feature is a gamification device, where the objective of the game is to match, as close as possible, a known outcome.

To take the case of Reopen votes, the description says that this will test the gamers ability to correctly call:

a question which the system has determined should be open or closed based on various criteria (highly-voted with no close votes, quickly closed with no answers, etc).

There is at least three ways to play this game:

  1. One can carefully review the question, and try to decide whether it should stay closed or be re-opened, based upon one's best objective analysis.

  2. One can look for patterns in the phrasing of the question known to trigger a knee jerk response from a sizeable portion of the community (present company not excepted) and try to match that pattern.

  3. One can press "Skip" whenever one faces a review that actually requires careful analysis (as opposed to a knee jerk response).

I'll leave it as an exercise to the reader to work out which of the three strategies outlined above that shall yield the highest benefit/cost to the individual game player, given that the objective of this game is to avoid suspension of privileges.

Personally, I think that the optimum individual gaming strategy leads to the least desirable outcome for the DA community.

0

You must log in to answer this question.

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