Well, I hope y'all enjoyed your happy hour... now get back to being miserable.

Morning...

And I thought I wasn't late to being Happy...

:(

@JourneymanGeek Congratulations on earning 10,000 rep this year!

cv-pls meta.stackexchange.com/questions/311075/…

@SonictheInclusiveHedgehog oh thanks

chaos-emerald-pls meta.stackexchange.com/questions/306201/… (@Shadow @rene)

→ 12 messages moved to Chimney

@rene If you're cleaning up SD messages, can't you clean up the corresponding commands also?

@πάνταῥεῖ uh ... I'm a robot ... I click a link offered by a userscript ... blame the userscript ...

Could be probably imroved, yes.

@πάνταῥεῖ it would work if you reply to the smokey reports instead of the sd ... shortcut. Then I still need to remember to tick replies

I'll take that into account for future.

@πάνταῥεῖ this is what I do when I move messages: i.stack.imgur.com/iKjkl.png

I See, explicit command prefixes would to that job as well ...

Could blame JasonC but we've not seen him for ages ;p

→ 7 messages moved to Chimney

let's see if this fix is worth keeping. I saw some weird behavior earlier where clicking on the clean link opens a new page. Strangely enough I can't repro that any more.

[ SmokeDetector | MS ] Blacklisted website in answer: Export Exchange Public folders to a network drive, sharepoint, or a PST by jameswatt on serverfault.com

@SmokeDetector tpu-

sd why

@πάνταῥεῖ [:7015146] Body - Position 3240-3248: erectile

@SmokeDetector fp

@πάνταῥεῖ Registered answer as false positive.

sd puff

> youtube.com/watch?v=fNLhxKpfCnA

> youtube.com/watch?v=IXdNnw99-Ic

> youtube.com/watch?v=vi7cuAjArRs

I am here, yup.

[ SmokeDetector | MS ] Bad keyword with email in answer, pattern-matching email in answer: Unknown email adress logged into my Gmail. by frank on webapps.SE

@SmokeDetector tpu-

Hello all! I made a userscript for keyboard shortcuts in StackExchange comments. Most of the SE power users are here, so I thought you might be interested to have a first look :) All feedback is welcome. Thanks!

"Most of the SE power users are here ..." Are they?

@πάνταῥεῖ well, maybe not most, but certainly many ;)

Your script looks neat anyways ...

Could be a good question at SE Code Review

@πάνταῥεῖ thanks!

@πάνταῥεῖ maybe I miscommunicated, but I wasn't hoping to get the code reviewed though. I was instead hoping for beta-users of the userscript for this initial release.

I'm avoiding to install user scripts as hell personally.

sure, no problem

Should I flag NAA reviews that I believe were decided incorrectly?

Not to get it undeleted, but because I don't think the reviewers have read the relevant discussions on meta about what exactly constitutes NAA.

@Shog9 Do you have an opinion?

Just to get the users warned?

I guess so

I've always flagged suggested edits that were unambiguously bad and made it through the queue, and my flags generally get accepted.

But I guess there's some controversy over what constitutes NAA (even though the "official" position is pretty clear) so maybe the mods don't want to deal with it.

@Radiodef what do you mean by "NAA review"?

LQP review of an answer that was clearly flagged as NAA.

Here's the review I was thinking of flagging, for reference stackoverflow.com/review/low-quality-posts/19977277

What do you mean by "clearly flagged as NAA"?

It could have been flagged as very low quality too, I guess, but the comments suggest the users deleted it for being "not an answer".

So what's the problem here?

You want the answer undeleted?

I think the users who voted to delete haven't read the guidelines on how to respond to use these kinds of flags, such as described here: meta.stackoverflow.com/questions/265552/…

To be clear, I don't know if the users should get in trouble for this kind of thing or not. That's why I'm asking about it.

Ok, you are probably right, since... Who reads stuff on meta? But... That's not a problem. An answer getting deleted when it should not is a problem.

Would it be productive to ask a question on meta about this so you or somebody else can give a more substantial answer?

Depends. You should identify a problem first

I don't know JavaScript well enough to say whether the deleted answer was actually useful. My only problem is that it appears to be an answer, given the guidelines I've read in the past. If it doesn't really matter that people in this queue don't follow those guidelines to a T.

I meant: If it doesn't really matter that people in this queue don't follow those guidelines to a T, then I guess I would have no reason to flag this sort of thing.

So I guess my question here really has two parts: first, does it matter that answers are deleted via the queue which are an answer using strict guidelines from meta, and (if so) should those answers be flagged?

Mumbai, mumble ...

@Radiodef NAA is one of the rarest cases ...

@Radiodef so, first off: the queue doesn't tell anyone which flags were raised. There's stuff in the queue with NO flags on it; the system just decided it looks crappy.

Yes, you can guess... But the queue doesn't ask you to guess. It asks if there's a problem with the post.

So if there's a problem with the post, the post will probably get deleted, regardless of what flag was raised.

I am not really a friend of those review queue mechanics.

The critical question here is: was this post problematic enough to warrant deletion?

So am I just wrong for believing that cases like the one I linked to shouldn't be deleted (as a general rule)?

I do know something about JavaScript. The answer is pretty close to worthless, but might help if the asker has very specific needs. Which, I doubt is the case, but I restored it anyway for that slim chance.

The comments suggesting it should be a comment are the real problem here.

What I thought is that (in a nutshell) NAA means it's e.g. a comment on another post or a question and that VLQ means it's e.g. unintelligible.

Review deleted it as low quality, which it is, and also happens to be the name of the queue that deleted it.

And that neither flag should be used for answers that are e.g. wrong or otherwise not very good.

I'm..

Probably less hung up on that than some folks might be

meh

if I see something that's technically an answer but blatantly and dangerously incorrect sometimes I'll just vtd in the queue

you're not supposed to but still

Review is, in general, rather more strict than the moderators. But in this regard, moderators will often decline flags that require topic-specific knowledge, out of practical necessity.

@Magisch jury nullification...

The judge gives you instructions on how to interpret the law, but... You don't actually have to follow those instructions

In general you should though

audits and all

but dangerously incorrect barely-answers don't get selected as known good audits usually

That reminds me, there seem to be no audits on smaller sites

or, if there are, I haven't encountered any in a good 100 reviews

Not enough data for audits on most sites

The vast majority of LQP audits that I get seem to be spam

A feature to hand select some would be nice then

but dev time...

Considering the complaints I see about audits...

@Magisch The "Fliegen-S(h)ites" trolling off quickly :-P .....

@Catija most of those are because audits are system selected and sometimes you get edge cases, iirc

I don't think anyone's ever complained about a suggested edit audit

@Magisch oh yes they have

@Magisch They do so, a lot ....

Most sites don't have thousands and thousands of review queue items.

Wasn't one of the recent fixes that had to be done relate to the improve edit option on a suggested edit audit?

@Magisch SE dev time "stinks off" OP conscious time a lot. Let's say a rate of 1/million.

@Catija there is a fix for that pending

It's complicated

You're talking about this one: meta.stackoverflow.com/a/368118/4548692

?

I was looking for it but that was the closest I found.

@Catija oh, that's not suggested edit audits, that's first post / late answer audits. We never enabled support for suggested edits as an action, so... That just behaved oddly if you had < 2k

Ah. That was the one I was thinking of, misremembered it being about suggested edits.

There's another old bug with Improve, but not related to audits.

Right now, Improve changes get mixed in with the suggested edit if you try to roll back. So if you edit just part of a question as your suggestion, and the reviewer touches other parts... You get the combination as the only thing you can roll back to.

Ah, so it should create two separate edits, then.

I spent a good bit of the last week trying to figure out how that should work. The suggested edit part is straightforward, but then there are a bunch of cases where two or more edits actually got stored simultaneously or very close to onr another, and it's pretty important to behave consistently there too.

Like... Normal edits. Not suggested.

I do recognize that sometimes the OP and an editor will muck up each other's edits somehow. I've had that happen... it looks like the OP is reverting an editor's fixes but they're really just submitting their own and incidentally rolling back the other at the same time.

There's some especially weird stuff if you go back to the old days when Community edited duplicate links into the post itself. Lots and lots of collisions with actual human editors.

I guess in code they have you check out a version of it and then merge it back in so if edits were made in the interim, you can combine the two... but that'd probably be a bit cumbersome to do in what's likely an uncommon situation.

Jarrod remarked, wistfully, "I wish we'd just built this on git or something"

Oh, that reminds me... I finally got to see what the ToC banner looks like for posts that are in review and got some screenshots for the README to add.

That would be complicated to use for people who don't know git, I would think.

Not... Like you have to push from the command line to post a question, but for back end storage.

Although a CLI for SO would be pretty sweet

@Catija Two little words describe this scenario: "edit conflict"