@Rob Without creating an account on MS, the user will not be permitted to edit or delete comments. One of the substantial points of creating an account on MS which is authenticated with the user's SE account is for us to verify that the login on MS is, in fact, that SE user.
We're not going to allow what is in effect random individuals to edit/delete comments on MS. Without an authenticated account, we can't differentiate some random person visiting the site from the actual SE user.
We allow any user with SD privileges to create comments on MS, even if they don't have an MS account. Primarily, this will be moderators. In large part the responses which moderators give to SD reports are valuable and useful to keep with the report as a comment on MS. There are, of course, times when moderators joke around. It happens; it's noise, but to toss out all comments by moderators just so that a few/moderate number of joke replies are not kept is not reasonable.
Ultimately, we have a process in place such that any user who has an account on MS can request removal of individual comments which they see and don't find useful to the post/project. That's really good enough (if we were appropriately attributing the comments, even just as text within the comment on MS, as that's required by the CC BY-SA license).
It would be possible for us to record the chat ID of the user who created the comment and associate previously recorded comments with a subsequently created user. From a programmatic standpoint, that would require changes in SD and MS, along with adjusting the post_comments table in the MS database.
That seems like a substantial amount of work for relatively little gain, which could be handled in other ways for the rare cases where someone cares substantially (e.g. after creating an account, the user copying the comment text into a new comment and requesting the old one being deleted).
Again, I definitely agree that we should be providing attribution to the correct user, even if that's just text with a link within the normal MS comment text. That's something which we certainly should change. Note that the comment would still be "owned" by SmokeDetector, as the user that actually created the comment. I have created a GitHub issue for this.
As to the more substantially involved cases of recording the chat ID and allowing subsequent creation of an MS account to take ownership of the comment, please feel free to create an issue in the MS GitHub repository, as that's where we'd need to track such a feature request.
@Rob Oh, I'm so sorry, I didn't realize I was causing trouble by trying to answer your question. I'll make sure not to do that again unless you explicitly ping me. Also, thanks for showing me a new term "shoulder surfing", I didn't know it. I guess that sort of does apply to me; I read through the transcript/conversations to learn valuable stuff (but without the malicious intent :)).
@cigien Re: That post I flagged as R/A (sorry about the late response - I've been 'elsewhere' for a while) ... as far as I can tell, it's gibberish in any language, so R/A is appropriate.
... it was ultimately dealt with by a helmet-wearing moderator, so I guess I'm safe (for now).
@AdrianMole Hmm, I'm not sure about that. It really looks like it could be a natural language. I think NAA is appropriate for SD feedback on that one. As far as a flag on the site goes, sure, I could see r/a flags being accepted on that. Though I'd be surprised if the spam penalty were applied. Thanks for clarifying.
Also, what do you mean you were "elsewhere"? You know the rules, once you're on SE, you can't leave ;)
@cigien I put it through Google Translate when I first saw it (in Review) ... and Google came up with the same gibberish. And Gibberish is worthy of R/A, so I flagged it as such. Maybe Google was wrong?
Well, in 1972, I was doing better(?) things than learning programming languages ...
... but OP needed to put that in their post.
... as it stands the post is worthy of a gibberish flag. I'll need to look more carefully as to exactly how the Orang Helmet handled it ... but my red flag has been marked "helpful".
.. Orang ... "Orange"? Nor sure about that. Freudian slip?
