Duck test:
If it looks like a duck, swims like a duck, and quacks like a duck, then it probably is a duck.

In this case:
If it behaves like the Original Poster, then it probably is the Original Poster.


Sometimes a new user asks a question using unregistered account. After a while some substantial edit suggestion from another user appears. Nicknames are at least similar and this second user behaves like the Original Poster (answers comments, adds relevant information, narrows the scope etc.). Let's assume the second account is registered.

I can see this suggested edit in my review queue.

My point of view

I'm almost certain there is the same real person behind both accounts, still I have no evidence. All I can see is the second account "quacks like the OP".

I have no doubt I should give them some useful feedback. Last time it happened my comment under the question was like:

I believe you tried to edit the question. See I accidentally created two accounts; how do I merge them? Merge the accounts, then you will be able to edit freely because you will own the question.

I tend to give the user some time (I neither accept nor reject, nor skip the suggested edit). After a while, if the accounts are not merged (yet), I reject the edit, unless it's already accepted or rejected. I can mark "this edit deviates from the original intent of the post" as a reason.

If I'm wrong and the accounts are merged eventually, the user can edit again (as the question owner this time).


This last time such edit was accepted. It so happened I disclosed my rejection in comments and two high-rep users (+10k, +25k) argued that "accept" was the right thing to do.

I'm not sure because this is what we can read about merging accounts [emphasis mine]:

After you contact us, the Stack Exchange Team will reach out to verify that you own both accounts. If we can confirm your ownership, we will initiate a merge.

From this I understand some verification and confirmation is required to tell there's one and the same real person behind multiple accounts. Yet it seems to me some users think it's enough to "quack like the OP".

The accepted edit did make the question better, I admit. Should it matter? With this attitude we can take whole bunch of questions that are broad, lack some information etc. and make them "better" by making up and adding new details. We don't do this. Instead we do reject edits for the following reason, if it applies:

Even edits that must make drastic changes should strive to preserve the goals of the post's owner.



  1. What verification and confirmation is required to merge accounts? (especially unregistered with registered one).
  2. Do users with enough reputation see something (e-mail? IP?) that allows them to reasonably tell apart the genuine Original Poster with another account from an impostor?
  3. I can't see anything like it (do I miss something?). Should I perform a duck test then? Should I accept edits from users that "quack like the OP"? What is the right thing to do?

Secondary (out of curiosity; these questions may be left unanswered):

  1. When accounts A and B are merged, will B's pending edit to A's post be automatically accepted? or will it still occupy the review queue? (This question suggests the latter but it's old; some automation may have been added.)
  2. Can moderators merge accounts without explicit merge request? Do they?

2 Answers 2


We can't merge accounts any more. We basically use a duck test + a quick look at whatever PII we have, and pass it up to a CM. There's nothing that a regular user can do. In many cases I actually suggest OP request a merge themselves just to ensure they're aware of it. There's been cases of folks creating what seemed like a endless stream of new unregistered accounts on other sites.

Duck test is good enough for edits. We can always roll back if your duck is actually a mutant possum.

I don't actually know the effect of pending edits on account merges, so just approve if it seems appropriate.

  • A mutant possum. I cannot help but laughing loudly in my office when I read that part. Aside from that, I'm curious to why that functionality is removed from diamond moderator? It is a very useful utility and there's no potential misuse. Abuse should be obvious enough, because merges should be rare.
    – Vylix
    Commented Oct 12, 2017 at 9:20
  • 4
    Cause if we mess up, fixing it is a pain it seems. And it happened a few times.
    – Journeyman Geek Mod
    Commented Oct 12, 2017 at 9:22
  • I would have thought that it would be automated to the point that nothing would happen until the user had confirmed it by accessing both links emailed to the separate accounts. No? Commented Oct 14, 2017 at 21:18
  • No. Apparently not how it works.
    – Journeyman Geek Mod
    Commented Oct 14, 2017 at 23:45
  • 1
    I doubt anyone's ever going to mistake a possum for a duck. A platypus on the other hand... ;)
    – jmbpiano
    Commented Oct 19, 2017 at 21:19

The phrase "the goals of the post's owner" can be understood in many ways (as can "strive to"), and if there's any definitive official guidance on what it's supposed to mean, it doesn't seem to be very well publicized. So different people can and do draw the line at different places. That said, in this particular situation, I'd very much suggest interpreting in narrowly: the main goal of someone asking a question is getting their problem solved, and any edit that helps them achieve that goal — or at least isn't likely to hinder it — is good and should be approved.

In particular, let's assume that the question in its original form is vague and unanswerable or off-topic and about to get closed, and the edit makes it answerable and on-topic. Let's consider the possible cases:

  1. The editor is the OP. In this case, the edit is clearly appropriate.
  2. The editor is adding information provided by the OP elsewhere, e.g. in comments. Again, this is appropriate.
  3. The editor isn't the OP, but has correctly guessed what the OP meant to ask, e.g. because they have the exact same problem and can provide the missing information that the OP failed to include in their original question. If we knew for sure this was the case, there would also be no reason not to approve the edit.
  4. The editor isn't the OP, and the issue they describe in their edited version of the question is not, despite their superficial similarity, actually the same as the OP's.

The last case is the only one where the edit could be considered to be contrary to the OP's goals, since answers to the edited question would not necessarily solve the OP's problem. However:

  • If the real OP is still active on the site, they will be notified of the edit and can revert it. Indeed, if they're paying any attention to their question at all, they should still notice that it has been edited, and can revert the edit (or re-ask their question with more details) if they want.
  • If the OP isn't paying any attention to their question, then they won't know if anyone answers it, either. So they really won't have lost anything.
  • In any case, if the original question lacked the information necessary to properly answer it, then the OP wouldn't have received an answer to their original problem anyway. So, again, even if the additional information provided in the edit doesn't exactly match the OP's situation, approving the edit still won't actually harm the OP.

So, regardless of whether the editor is actually the OP or not, in this particular case (where the original unedited question is off-topic or unanswerable) there really isn't any reason not to approve the edit. The worst case scenario is that the OP later comes back, notes that their question has been edited and answered in a way that doesn't actually help them, and feels annoyed — but even then, they won't have actually lost anything.

That said, to reduce the chance of potential confusion and annoyance if that turns out to be the case, I'd suggest leaving a brief comment noting that you've approved the edit, and explaining why. While you're at it, you can also suggest that, if the editor is indeed the OP, they should merge their accounts, e.g. like this:

I've approved an edit to this question [link to review item here] suggested by @user123456, since it adds essential information that may actually make this question answerable. If that's you, [question author], you may want to merge your accounts. If not, and if the edits by user123456 don't describe your actual problem, you may revert them yourself. However, in that case, please do provide the requested information yourself so that your question can be answered. Alternatively, you may also ask a new question with the additional information.

Note: AFAIK, the @-ping will work if the suggested edit has already been approved when you post the comment. Otherwise only the OP will be notified of the comment. However, if the edit does eventually get approved, the editor will at least get a notification about the +2 rep linking to the question they edited.

Also, just to be clear, I'm not suggesting that going around editing other people's unclear questions by guessing the missing information is generally a good idea. But even if that's what someone might have done, it's not really a reason to reject an edit that turns an otherwise bad question into a good one. Especially since, as the duck test suggests, most likely the editor is the OP.

You must log in to answer this question.

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