16
$\begingroup$

If you take a look at this particular revision history, I made an edit to my answer at 05:41:36Z, which is just under 3 minutes after the original timestamp of posting that answer at 05:38:43Z. This revision was noted despite my edit being made under 5 minutes. Normally a modification isn't recorded (or, at least, not shown) until an edit is made after 5 minutes has passed. This is the whole purpose of the grace period.

Has the system changed recently? If this change was indeed made, perhaps its intention is to encourage users like me to be more careful with typing up their original posts.

$\endgroup$

1 Answer 1

Reset to default
20
$\begingroup$

The five-minute grace period is not gone, but it has been recently changed. Information can be found in Shog9's answer on MSE. In this particular instance what happened was

  • [a] comment has been added to the post since the previous revision by anyone other than the editor.

This ends the five-minute grace period, meaning a new edit will result in a new revision of the post.

The intention for this particular condition seems to be to make sure that comments don't seem completely idiotic when made during the grace period of the post which is then itself quickly edited: we can find out which revision of the post was commented on.

$\endgroup$
18
  • 3
    $\begingroup$ While this makes sense to apply to questions, when an answer has been posted, which the Meta.SE proposal suggests (aside: please don't use MSE for Meta.SE, it is confusing :-)), it is completely idiotic (IMO) to apply the same to an answer when a comment has been posted. In almost all cases, the conversation will go "What do you mean?", "Oh was editing, sorry", delete comments. $\endgroup$
    – Aryabhata
    Commented Mar 26, 2015 at 7:22
  • 1
    $\begingroup$ Now if comments could not be edited after someone has replied to them, that would be great. $\endgroup$
    – Najib Idrissi
    Commented Mar 26, 2015 at 7:22
  • 3
    $\begingroup$ @Aryabhata I see these changes as making it easier for the site mods. In terms of answers, if a "Didn't you mean...?" comment is made followed by a ninja edit, the comment probably should be flagged as obsolete, but without the revision history it may not be entirely obvious to us that something was changed, or that the comment is now obsolete or otherwise unnecessary. ("MSE" is fairly standard network-wide for meta.SE. I've personally used "math.SE" to refer to our site well before the introduction of meta.SE. Context generally makes it clear which site is being referred to.) $\endgroup$
    – user642796 Mod
    Commented Mar 26, 2015 at 8:51
  • 9
    $\begingroup$ As someone who had been annoyed by seeing nonsense or wrong answers turn into apparently plagiarized answers within the grace period and with timestamps misleading about what actually happened, I can see why this could be a good feature for answers. $\endgroup$
    – Jonas Meyer
    Commented Mar 26, 2015 at 11:15
  • $\begingroup$ @JonasMeyer: For that one would pretty much have to always save the initial draft of every answer/post. It's a feature that's been requested on MSE, but, alas, it is (status-declined). $\endgroup$
    – user642796 Mod
    Commented Mar 26, 2015 at 17:32
  • 1
    $\begingroup$ @ArthurFischer: How will it help? Say I made a few non-ninja edits to my answer. How will you tell the difference without actually opening up the revision history? You would still need to read the current version of the answer, wouldn't you? And if the comment was within 5 minutes of the answer being posted (and no revision history), you can probably delete the comment without looking, anyway. Anyway, I haven't been a mod, so have no clue how it is. I make a lot of edits in the first 5 minutes. If someone posts a comment, now those edits will show up in the history and the active page etc. $\endgroup$
    – Aryabhata
    Commented Mar 26, 2015 at 19:14
  • $\begingroup$ @JonasMeyer: Is plagiarism really a huge problem? And is this really required? (i.e. without this you cannot catch them?). The initial draft save + 5 minutes grace would be much better solution IMO. $\endgroup$
    – Aryabhata
    Commented Mar 26, 2015 at 19:15
  • 1
    $\begingroup$ @Aryabhata: For this particular type of plagiarism, I only really know of one user, who seemed to do it a lot. I have seen it happen, but I don't know what you mean about "catch them". If there is no record, I might be wasting my time in pursuing it. Now, a comment on an initially useless post might help. Yes, initial draft saving would be good, probably better, but I have seen instances where this new behavior would have made suspect actions clearer. While it may not be common that it is required, I do not understand how this new behavior would be a problem. $\endgroup$
    – Jonas Meyer
    Commented Mar 26, 2015 at 19:31
  • 1
    $\begingroup$ @Aryabhata: In particular, why would it matter that it shows in the active page? It would have already been bumped less than 5 minutes prior. Why is it a problem that the edits would show up in the history? $\endgroup$
    – Jonas Meyer
    Commented Mar 26, 2015 at 19:33
  • $\begingroup$ @JonasMeyer: The edits would be too trivial to merit a revision (the reason for having the 5 minute grace in the first place!). As to how many times a post getting bumped to the top of the list for 5 minutes, is an annoyance, which people tolerate differently. Imagine 10 different 5 minute quick edits happening at the same time. Anyway. $\endgroup$
    – Aryabhata
    Commented Mar 26, 2015 at 20:04
  • $\begingroup$ @Aryabhata: I guess I would need data on how many edits this would effect to decide if I agree. I am thinking that it wouldn't affect most answers. $\endgroup$
    – Jonas Meyer
    Commented Mar 26, 2015 at 20:29
  • 2
    $\begingroup$ @JonasMeyer: I almost always get the latex wrong on the first try :-) Usually takes me a few edits to fix everything. You can probably get (old) data on the proposal which suggested the 5 minute grace in the first place. It might not be relevant now, though. Anyway, since the change has already been made, I guess we can just wait and see... $\endgroup$
    – Aryabhata
    Commented Mar 26, 2015 at 22:09
  • $\begingroup$ @Aryabhata: But do you typically get fast comments about your LaTeX being wrong? $\endgroup$
    – Jonas Meyer
    Commented Mar 26, 2015 at 22:49
  • 1
    $\begingroup$ @Aryabhata suppose somebody comments "In the second equation shouldn't the inequality be the other way round?" OP edits to change the inequality. The comment is flagged as obsolete. If mod has a revision to see if there was an edit, it is immediate the comment is obsolete. If they do not have a revision and and edit might or might not have happened then they cannot know if the comment was taken into account or not. (They might be able to decide if the remark is [still] relevant but this is a lot more difficult and possibly impossible.) $\endgroup$
    – quid
    Commented Mar 26, 2015 at 23:54
  • 3
    $\begingroup$ @Aryabhata No I do not trust a single user flagging a comment as obsolete (else why not autodelete). Whoever might have thought the comment is not relevant [anymore] could have flagged it. It is not unlikely it was the author of the answer, but they might also have gotten it wrong, too [else why not auto-delte]. And in any case your parenthetical remark is a big "if." Even if the author chooses to ignore the comment indefinitely, it could remain a relevant piece of information hinting at an error in the answer that is all but obsolete (even if the author might not like it). $\endgroup$
    – quid
    Commented Mar 27, 2015 at 10:50

You must log in to answer this question.

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