This happens once or twice every week with little variation:

enter image description here

Let's go through the situation quickly:

  1. A new user has his first attempt at improving the content of an SE post. He puts up a decent attempt, and adds an edit summary highlighting what he did. So far, so good.
  2. But, they add mathjax to the title, which is bad for us for several reasons discussed in the past. Since this is the only thing they did, we have to reject this suggested edit, which was rightfully the case here.
  3. However, the problem here are the "Reasons" that have been given for rejection:
    • one is a canned response that correctly describes the problem ("actively harm readability"). But, how do we expect the new user to know if their suggested edit to the title was harming readability?
    • the second is a response that more specifically addresses the problem, but still doesn't give the user a reason as to why MathJax is not allowed in the title.

Both of the reasons are likely to not help the user completely understand what's the matter here, and they are likely to repeat the error again, or give up suggesting edits entirely facing dejection on their first attempt. Not everyone knows about Meta, and where to find the information about "mathjax in titles".

Of course, I am not suggesting that suggested edit reviewers should comment about "No MathJax in titles" and link to a meta post. That would take much more time and is too difficult to put in practice. The return would be not worth the time.

However, I wonder if there could be a better way to reject such edits. Since this requirement (No MathJax in titles) is unique to our community, I wonder if a custom reject reason could be implemented - that would say something along the lines of:

The edit is an attempt at improvement, but we avoid MathJax in the title. Please see - chemistry.meta.stackexchange.com/q/261 - for details.

(sitting at 140 chars, maximum in the custom close reason is 150)

I think this would easily allows reviewers to reject the edit while also giving the user a complete understanding of where they went wrong.

Is this doable? Or is there another way that achieves the same motive? Thoughts?

PS: Within an hour of my submitting this meta post, the same user did the same thing to the title again. Here you can you see that the custom close reason looks much neater:

enter image description here

Of course, the problem at hand is how to execute this custom close reason efficiently.

  • 1
    $\begingroup$ While the custom box says "Causes Harm," it's the best place to give a custom rejection reason, even if the edit isn't truly "harmful." I would paste in your message there. $\endgroup$ – jonsca Mar 20 '18 at 1:51
  • $\begingroup$ @jonsca So, moderators can't add more reasons to the Suggested Edits Reject box than there already are? (similar to how they can change the close vote reasons?) $\endgroup$ – Gaurang Tandon Mar 20 '18 at 2:25
  • 2
    $\begingroup$ No, those are standardized across SE, as far as I know. $\endgroup$ – jonsca Mar 20 '18 at 2:31
  • $\begingroup$ Oh, well that's sad :( (PS: I had to modify my custom message slightly to fit 150chars which I just realized above ----^) $\endgroup$ – Gaurang Tandon Mar 20 '18 at 2:38

Whenever I review the suggested edits, what I do before accepting or rejecting is to look at all the constructive edits, destructive edits, and missed out edits (certain things the editor did not fix).

If I see more constructive edits than destructive and missed out edits combined, I will approve and edit to fix the remains of his edits. I'll also ping him/her in the post's comments and tell them what mistakes they made in the edit. Once I get a response from the editor, I will promptly delete my comment and flag the editor's response as No Longer Needed.

If there are more destructive and/or missed edits, I will reject the edit (or rollback if already approved) and leave a comment telling them why the edit harmed the post (I would leave an edit summary if rollbacked).

Now, in your situation, you say that the only mistake that the editor made was to add $\LaTeX$ in the titles, and all other edits to the body of the post were constructive. I feel rejecting the edit was a bit too harsh, considering that adding $\LaTeX$ to the titles itself is mostly done with good intentions, and rarely maliciously. It would have been better to approve the edit, and edit the title yourself. This way, the editor gets both credit for his edit, and also learns not to add $\LaTeX$ to the titles.

  • $\begingroup$ Thanks for saving my post from being Roomba-ed! Though I disagree with your last paragraph. "It would have been better to approve the edit, and edit the title yourself." No, I disagree. There a lot of ways in which people suggest "good intentioned edits", one of which is when they add more information to the author's post than the author intended. We reject such edits as "deviates from the author's intent". So, in this case of MathJax also, the correct path seems to be to reject the suggested edit. What say? $\endgroup$ – Gaurang Tandon Apr 16 '18 at 0:53
  • $\begingroup$ @GaurangTandon Roomba has less power on meta sites, for what I'd say are fairly obvious reasons. $\endgroup$ – orthocresol Apr 16 '18 at 23:30
  • $\begingroup$ @ortho ah, didn't know that. $\endgroup$ – Gaurang Tandon Apr 17 '18 at 0:34

You must log in to answer this question.

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