20

A common type of bad suggested edit that we all hate is one that adds random formatting.

At the moment, all suggested edit audits are based on randomly replacing some sequences of words by other vaguely plausible sequences of words. This catches some robo-reviewers but still misses plenty who do a bad job.

Please implement audits that change nothing but make a few words code or bold or italic, in random vaguely-plausible ways.

Improve this question
5
  • 10
    I support this.
    – Doorknob
    Aug 21, 2013 at 18:51
  • 4
    The question here is: are we trying to use suggested edit audits to adjust reviewer behavior or simply to screen out people abusing the system? Right now, suggested edit audits target the latter. Do we want to expand them to start pushing reviewers closer to what the community considers good review behavior?
    – Brad Larson
    Aug 21, 2013 at 19:18
  • 3
    @BradLarson Close/reopen audits do try to adjust reviewer behavior, as witnessed by the frequent disputed-review-audits. Why not do the same for suggested edits?
    – Gilles 'SO- stop being evil'
    Aug 21, 2013 at 19:35
  • 2
    Disputed audits shouldn't really be happening. Some effort should be put into making them more clear-cut, so that we get fewer disputes. Most of the complaints in the disputed audits have been legitimate, from what I've seen.
    – user102937
    Aug 21, 2013 at 21:14
  • This is still very common, possibly from users who disproportionately come from a certain country with over 100M citizens.
    – Nemo
    Apr 8, 2018 at 21:10

1 Answer 1

Reset to default
-4

Audits are meant to catch people who are not paying attention, not to grade auditors on the relative merit of their approvals. The behavior we are trying to change is the behavior of auditors who summarily approve edits without looking at them.

If you can find some way to make the proposed formatting changes indisputably non-sensical (i.e. they should not even be vaguely plausible), I could get behind that (like the ones ben is uǝq describes in his comment below).

Related
Small improvements to the Vandalism / Spam review audits

Improve this answer
16
  • This is not the case for close and reopen audits, so why would it be the case for suggested edits?
    – Gilles 'SO- stop being evil'
    Aug 21, 2013 at 21:00
  • It is the case for all audits.
    – user102937
    Aug 21, 2013 at 21:00
  • No.
    – Gilles 'SO- stop being evil'
    Aug 21, 2013 at 21:04
  • 4
    @Gilles does have a point though. That's all nice in theory, but in practice a failed close vote audit does say "This is community policy. This is how you should vote. Questions like these are fine/should be close". That is not merely a gaming aspect.
    – Bart
    Aug 21, 2013 at 21:05
  • @Bart: It would have to involve something that is unambiguously and unequivocally disallowed by the community. That's not the case in the proposed OP.
    – user102937
    Aug 21, 2013 at 21:06
  • @Gilles: I don't see your point. Audits are there to screen out people who aren't paying attention. That reviewers sometimes change their behavior is a nice side-effect, but it's not the purpose of audits.
    – user102937
    Aug 21, 2013 at 21:07
  • Care to clarify that? Are you suggesting that such issues as code-formatting as a highlighting tool are issues without consensus? I don't really follow.
    – Bart
    Aug 21, 2013 at 21:07
  • @Bart: I am saying that bolding and code formatting are not necessarily bad edits, and community members shouldn't fail such an audit if they are genuinely trying to help.
    – user102937
    Aug 21, 2013 at 21:10
  • We're not talking your average bolding and code-formatting here. But I guess your view of audits is as a fundamental robo-review catcher. Fair enough. Though if not through audits, I don't see what other efficient means we have to educate on some of these issues.
    – Bart
    Aug 21, 2013 at 21:13
  • 4
    @RobertHarvey Shog9 has said several times that close/reopen audits are supposed to align reviewers with community standards, even if that's not their primary purpose. “This question actually makes for a great audit, (…) It's (…) the sort of thing you should learn to recognize when reviewing questions”. What I'm proposing doesn't go that far though: it's only about making sure that the reviewers are paying attention and not accepting bad edits of a certain common type.
    – Gilles 'SO- stop being evil'
    Aug 21, 2013 at 21:14
  • @Gilles: See my edit. The close audits are selected from a pool of questions that are (or rather should be) indisputably on-topic or off-topic. We get audit disputes on those because of failings in the voting system, not because of any weakness in the algorithm that chooses the audit questions. The algorithm itself is sensible, but (in theory, anyway) it does not use borderline questions for audits.
    – user102937
    Aug 21, 2013 at 21:16
  • Anyway, your question would be a better one if you provided an example of what such an edit might look like.
    – user102937
    Aug 21, 2013 at 21:22
  • 6
    for formatting changes that are not vaguely plausible (all require some initial attributes) you could put backticks (or bold etc) every word that is a tag; indent every line beginning th.*x or pl(ea)?(s|z) with four spaces, add dear sirs at the beginning or thank you at the end, change the title to <tag1> - <tag2> issue, change the title to add [solved] and at the same time change the body of the question to add solved - i foo'd the baz (literally), lowecase every i, bold and lowercase every i... etc
    – ben is uǝq backwards
    Aug 21, 2013 at 21:24
  • 1
    @benisuǝqbackwards: I would support any of those as audits. For that matter, I would support changing ben to uǝq as an audit. These are the kinds of things that reviewers are supposed to be stopping.
    – user102937
    Aug 21, 2013 at 21:25
  • I've actually seen all of the above @Robert but I've never seen someone reversing words; something to look out for!
    – ben is uǝq backwards
    Aug 21, 2013 at 21:33

You must log in to answer this question.

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