8
$\begingroup$

Basically I was looking around the queue and found this suggested edit: https://stats.stackexchange.com/review/suggested-edits/117203

It became obvious to me the user was having difficulties with the account(s), so he posted an update with another account as an edit.

The suggested edit was approved (not by me, I skipped it). But I can see reasons to actually reject it. The user should be asked to merge the accounts first. We should treat the edit as if it was another user.

Why? Because it becomes too simple to vandalize posts doing this. I've never seen it, but it became obvious to me after seeing people posting updates with multiple accounts how easy it would be to pretend you're another person and suggest an update, and how it would easily pass in the queue.

I've seen it reported in meta before, so I'm posting this thread as : should these edits be accepted before accounts are merged?

I want to know it so I can act accordingly in the review queue.

New example: Develop Prior from Previous Studies using Bayesian Updating: Poisson-Gamma Example

$\endgroup$
3
  • 3
    $\begingroup$ It is important to fix unclear questions as fast as possible, so yes, I'd say such edits should be approved. I have never seen an attempt to vandalize a post by suggesting a resonably-looking edit (and unreasonable edits will not pass the queue anyway). $\endgroup$
    – amoeba
    Commented Sep 20, 2016 at 15:08
  • 1
    $\begingroup$ @amoeba, perhaps this comment of yours could be converted into an answer? $\endgroup$
    – Andre Silva
    Commented Sep 20, 2016 at 15:14
  • $\begingroup$ @amoba there aren't bounds to the extent an edit is considered legitimate duplicate-account updating. A person could simply change the meaning or focus of the question without the OP consent and it would pass in the queue. $\endgroup$
    – Firebug
    Commented Sep 20, 2016 at 16:40

2 Answers 2

Reset to default
8
$\begingroup$

I agree with @amoeba: it's important that authors have the ability to update or improve their question as quickly as possible. Otherwise it may be closed as unclear, or answers that do get posted will fail to address the author's main concern. We have many ugly, hard-to-follow threads that are a nightmare for future readers, resulting from substantial updates or clarifications appearing after some answers have already been given. It is certainly worth trying to prevent this.

I frequently approve such edits. But I also post a note saying:

Please register & merge your accounts (you can find information on how to do this in the My Account section of our help center). Then you will be able to edit & comment on your own question.

Copy-and-pastable text:

    Please register &/or merge your accounts (you can find information on how to do 
    this in the **My Account** section of our [help]). Then you will be able to 
    edit & comment on your own question.

Clues that this is what is going on:

  • The question is recent,
  • Both the original account and the editing account are new (or the edit is anonymous, which indicates the author has logged out - I would post a slightly different comment in that case, suggesting that the author should log in),
  • The edit contains a clarification, or adds some information that only the original author would be expected to have.

It strikes me as remarkably unlikely that a new question with fewer than a dozen views not only will be stumbled on by a random visitor, but also that they should have mistakenly attempted to edit and improve it.

If I ever get a confused comment from a surprised poster wondering why somebody else has edited their question and reslanted it with their own additional information, then I will know I've made a wrong call, but it hasn't happened to me yet!

$\endgroup$
5
  • 2
    $\begingroup$ Thanks for the input, sounds like that's the consensus then. I'll start to do it like that. $\endgroup$
    – Firebug
    Commented Sep 20, 2016 at 17:10
  • 1
    $\begingroup$ (+1) Editing questions on editing in Meta raises metaphysical questions. $\endgroup$
    – Nick Cox
    Commented Sep 20, 2016 at 18:08
  • $\begingroup$ @NickCox Thanks for the copy-editing, that is satisfyingly clearer. You might enjoy casting your editorial eye sometime over the list of stock comments (mostly inherited from gung, though with several other contributors) that I borrowed this comment from. $\endgroup$
    – Silverfish
    Commented Sep 20, 2016 at 18:55
  • $\begingroup$ Thanks; they seem very well worded to me. I suspect @gung has refined them more than once. $\endgroup$
    – Nick Cox
    Commented Sep 20, 2016 at 19:07
  • $\begingroup$ @NickCox I appreciate your time, thanks. $\endgroup$
    – Silverfish
    Commented Sep 20, 2016 at 19:08
4
$\begingroup$

(Take this as personal opinion speaking as an ordinary user)

If it's an edit you'd accept if it was from a different user - an improvement without changing the intent too much - you could accept it whether or not you were sure it was the original poster.

However, if it actually is the original poster suggesting an edit from a second account, I regard the merging of accounts as generally more important to fix first -- this is so we're not running around after the same person doing it again 20 minutes later and again an hour after that. If we do what they want (accept the edit) but rely on them to go to the trouble of merging when they can get things done by making us run around instead, why would they?

[Sure there are well-intentioned users who will do their best, but many people are lazier if they can be.]

So I'd suggest using some judgement -- while it's good to get questions improved quickly, it's also important avoid the OP continuing to suggest a number of further edits from the second account (or a third...) instead of merging.

$\endgroup$
9
  • $\begingroup$ This is an interesting perspective. Do you think such frequent-editing-without-account-merging it is a real danger? $\endgroup$
    – amoeba
    Commented Sep 21, 2016 at 13:42
  • $\begingroup$ @amoeba the combined concerns of that issue and of the account never being merged between them worry me more than the possibility that the edit is not applied as quickly as possible. $\endgroup$
    – Glen_b Mod
    Commented Sep 21, 2016 at 15:47
  • $\begingroup$ Do we have any evidence of people continuing to operate without merging their accounts? (You may well be more party to that information than me. I've generally seen people either merge their accounts in the end, or they seem to disappear entirely once the edit goes through. But I'm not seeing the full picture of course.) Also, do we have evidence on holding up a user's edit encouraging them to register/merge properly? (On their end, they see the edited version with a "pending" notice after they write the edit. And I wonder how many people check to see comments on edit rejection.) $\endgroup$
    – Silverfish
    Commented Sep 21, 2016 at 16:11
  • $\begingroup$ Previous comment looks rather critical but I'm basically curious what the workflow looks like from the point of view of a user who doesn't go through the account merge process and whose edit gets rejected. Presumably once the edit is rejected, the "pending" notice is removed and the question goes back to looking like it did before to them. Do they get any particular notification of the rejection? $\endgroup$
    – Silverfish
    Commented Sep 21, 2016 at 16:13
  • $\begingroup$ It's been a while but I believe suggested edit rejections are notified. [On the previous comment -- I don't mind the critical nature of the question ... that's kind of what discussion is for. As long as it's reasonably polite (and it was) there's no problem. To address the question you raised there, I don't think that it's particularly common, but it does still concern me] $\endgroup$
    – Glen_b Mod
    Commented Sep 21, 2016 at 16:23
  • $\begingroup$ I have a feeling, but I'm not 100% sure, that rejected edit suggestions might not be notified. But I might be misremembering something that happened to me. $\endgroup$
    – Silverfish
    Commented Sep 23, 2016 at 20:29
  • $\begingroup$ @Silverfish Your memory is more likely to be accurate than mine. $\endgroup$
    – Glen_b Mod
    Commented Sep 23, 2016 at 23:36
  • $\begingroup$ Ahh, things have changed anyway: meta.stackexchange.com/questions/120624/… - but still not given as a notification. $\endgroup$
    – Silverfish
    Commented Sep 23, 2016 at 23:42
  • 1
    $\begingroup$ Incidentally the problem here is for people who have not yet mastered our system of accounts, and some of them are presumably making a new account each time they visit or attempt to interact. So I guess it doesn't matter to them whether the rejection is notified or not - they won't see it either way! $\endgroup$
    – Silverfish
    Commented Sep 23, 2016 at 23:44

You must log in to answer this question.

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