« first day (2155 days earlier)      last day (2862 days later) » 
01:00 - 20:0020:00 - 00:00

8:20 PM
@TIPS where? MSE? Might still be the privacy change mega ultra inbox item. It can fool even veterans...
@TIPS k
 
@ShadowWizard That thing, this world, we all, everything that lives forever . . .
is not a report.
 
if having the association bonus makes you a veteran then I'm never using that phrase again
 
@Quill Chatting with another veteran makes you a veteran.
Now, as always, the question follows as: Who's the first veteran?
 
Trivia: the highest median time-to-review for suggested edits on any graduated site is Expression Engine, with 212 minutes.
 
@Shog9 That . . . is median?!
 
8:25 PM
Interesting to note: Language Learning has 76 meta posts while Retrocomputing only has 41.
 
The lowest is Math, with 8 minutes
 
@Quill More than half of which aren't really solving the real problems.
Or maybe they are.
Been some time since I last checked LL.
@Shog9 This is expected.
 
Stack Overflow is 27 minutes
 
Chemistry would be short too.
 
18 minutes
 
8:27 PM
I guessed.
 
The fastest beta site is SpaceExploration, with 24 minutes
 
We edit pretty fast, since there are usually many people jumping to edit a post, even though we don't have many editors.
 
The slowest is Patents, with 1672 minutes
 
And well, seeing someone else submit an edit earlier and the system lying to your face isn't ideal.
 
but Patents is kinda special; the slowest normal beta site is Homebrew with a median time-to-review of 1539.5 minutes
 
8:28 PM
@Shog9 This is too much still.
Wait, how are those calculated?
 
edit approve/reject time - edit creation time
 
Aha.
 
or do you mean how is median calculated?
 
@Shog9 Both
First got answered
So I guess being kinda in the middle is typical and good for Chem.
 
I haven't handled a mod flag in six days... and there's only been two since then anyway
 
four mods probably wasn't the best idea
 
Thanks anonymous
@Quill I can haz your diamond?
 
uh no
 
You're in a very very very young beta. What else did you expect?
 
@Quill yup this is the official definition
34
A: Which badges do not cause a notification for experienced users?

balphaSilver and gold badges are always considered veteran-worthy, so you'll always be notified of those. On real child metas (that excludes Meta Stack Exchange), you're always considered a veteran; on all other sites, you're considered a veteran if you have received the association bonus, i.e. you ha...

 
8:31 PM
@Quill just suspend someone, as practise?
 
> you're considered a veteran if you have received the association bonus, i.e. you have 200 rep somewhere.
 
Code Review has a median review time of 22 minutes
 
@rene I'm ready CC @Quill
@ShadowWizard Who says we can't tweak the official definition to be something less lame?
 
Jun 15 at 7:40, by Quill
otherwise I'll just ask bjb to sign up
 
:D
 
8:33 PM
Jokes aside, don't push it @Quill. You could've said the same thing if it were three mods, because a single person can handle two flags in a week. That's not the point.
The site's not gonna stay like this.
What you should be very wary of, I, as someone with some experience on a language sites think, is being flooded with ELL questions.
It might not even take as long as graduation, and happen in the beta days.
 
also bears. Always be wary of bears.
 
"Language Learning, yeah, cool name, let's go get my thesis proofread there."
 
so like Math Overflow for linguists?
 
@Shog9 The brown ones, or the white ones?
Wait, the PURPLE ONES?!
@Quill ELU.
 
the ones that live in Quill's trees and drop on your head, killing you
also the microscopic ones that live in water. Those things are fierce
 
8:36 PM
I was half way through the same joke....
 
I haven't checked Linguistics closely, but I think they're fine-ish with such questions, since, you know, this is language and stupid sentences can bear some interesting linguistic discussions to themselves.
 
I don't use the video feature of my oneboxer very often, but when I do, damn it's good
 
Regarding our discussion on bears . . .
 
If you would have told me before that my best scoring meta.so post would be about a UI design, I would have laughed at you ...
 
And now you're laughing at me?
 
8:42 PM
so does this mean you're not laughing at us now?
 
Smiling
 
I'm thinking 5 pending edits should suffice on graduated sites, and probably 20 on beta sites to allow them to still function.
22
A: Raising rep by flooding the suggested edit queue

Shog9bluefeet brought this up on a call earlier today, and we got to talking about the various problems we're still facing with suggested edits: Scores of tiny edits take time away from reviewers and fill the queue From a reputation-earning perspective, making comprehensive edits is counter-product...

 
@Shog9 Oh God. Thank you
No really. THANK YOU
This is already pushed or what, right?
 
@TIPS naw, still hashing out details of how it should work (/ if it should work)
 
@Shog9 20 is too low a bar. I don't want to review 10 crap edits from a user.
 
8:48 PM
Site that would be most hurt by this: Islam, where one heroic editor has single-handedly fixed awful titles on most of the posts.
 
Well, we could arrange it in a way so it wouldn't stop mass-edits that are good.
 
sure, just approve them
The lesson there is that Islam needs more active reviewers
 
Well, first of all, 20 suggested edits won't look much in, say, Math.SE, but they would in Chem.SE.
So a single number for all graduated sites isn't gonna work.
 
that's why I'm thinking 5 for graduated
 
We never get 20 suggested edits a day, unless it's a mass-editing event that I (or some other user that knows what they're doing) lead.
Oh wait
 
8:50 PM
5 is not much of a burden on Math, but normally any 5 edits would get approved before piling up on Math because Math reviews very quickly.
 
Scrolling up
20 on betas seems too much, but the sites I'm active on are no longer in beta, so I don't have a say there.
15 or 10 would be saner.
 
On, say, Emacs, 20 edits is a lot but edits don't get reviewed very quickly so it's not as difficult to pile up 20 edits.
 
Well, hmm, how many active betas are there that don't review fast?
In the silent betas, I don't think anyone would go on to edit 20 posts, and usually a mod arrives to approve/reject faster than most reviewers do.
Heck, even in those betas the problem is almost nonexistent, so we could rule the problems that might come with the rate-limit in those ones.
 
@TIPS lots? 100-300 minutes median review time for quite a few.
 
That's not a lot. For a beta.
 
8:55 PM
Thing is, for quieter sites you do want to allow for the possibility that an editor might come in, make a bunch of edits over the course of an hour lunch, and then leave... And reviewers aren't gonna show up for another 8 hours because of timezones or whatever.
Sites that are only getting a couple questions per day often have this sort of pattern among users
 
Usually even in betas close to graduation, you get some random people come to review and review fastness is determined by how often they click that queue.
 
I think we wanna reserve the ability to tweak this limit per-site, just because there are always weird exceptions... But first, I'm looking for sane defaults.
 
I.e. most betas have only fewer than 10 active people in moderation.
So unless something hangs in the queue for more than a day, it's not a problem.
 
20 also has the advantage of being the normal limit for reviews per day, so 2 reviewers can clear the queue on a small site with only one editor (which is not uncommon either)
 
which is usually the case.
Mhm.
So, which is it going to be? Pending edit count or unapproved/rejected edit count?
 
8:58 PM
@TIPS same thing
oh, you mean pending plus rejected.
 
No, I meant which of the two between "has 5 reviews in the queue" and "has two reviews rejected in the past 15 minutes"?
Or other options coming from a combination of the two, yeah.
 
the latter without the former doesn't do much
I gotta run some numbers on the combo though
 
Well, this comment is very spot on:
Been playing with the query. I see a lot of users whose editing work needed some guidance. I've often rejected things with a custom reason, hoping their next edit would be something I could Approve. Implementing this form of throttling would help a lot. — S.L. Barth 2 days ago
I sometimes see the user in a spree, and the edits are substantial but leave bad problems with the post.
I wanna make sure that if I reject, they'll get the memo.
Usually, when I reject, they don't even check, and continue with their behavior, which is annoying and once even made me ping a mod in chat.
 
yeah... Thing is, this is a tiny % of editors doing this. And some of them are doing it well... I just wanna smooth out the process for everyone, since right now it's too easy for folks to shoot themselves in the foot and annoy everyone, which then reflects badly on the majority of editors who aren't doing this.
 
So the "rejection"-related reason is in.
That's good enough by me.
I guess it hurts the brain more to think of the exact best number. Maybe push this and smoothen or roughen it up according to feedback?
 
9:02 PM
I'm really not sure why people think it's good to fix a small typo in the title and change nothing else. I suppose that's rep farming...
 
or the typo just annoys you
@TIPS I figure, got one chance to make it not-annoying and then forever after to make it ideal.
 
Well, the people I'm talking about submit like 20 edits at a time...
And sometimes, if they change other stuff, it actually makes stuff worse!
 
@hichris123 Sometimes it's just honest mistakes though.
 
It's the easiest way to farm rep if you know nothing about the topic
My first few hundred rep points on SO were from edit +2s
 
I guess not much in SO's cases, because more users and the urban city law thingy, and also there aren't many editing guidelines and tweaks and finer points and stuff.
 
9:05 PM
the flip side of that is that Stack Overflow needs more copy-editing, basic typo fixes and so on
 
This whole meta question is a shame. Because the C++ community don't want their little snowflake to be closed they are imposing to the rest of the community their views, because if consensus is reached that those kind of questions are to be closed, they will not get away with it. — Braiam 3 hours ago
 
But in a site like Chem, with such heavy citation, tagging and mathjax uses and all, it's next to impossible to get a perfect edit done in a medium or long post.
 
Overstatement of the year right here ^^.
@Shog9 Usually the posts this occurs on are either complete trash or need some love though. The former shouldn't be touched, the latter need someone to care for the post.
 
@hichris123 Go Braiam
 
Whee, that question was closed again. Now for another reopen cycle...
 
9:07 PM
Woohoo
Go SO.
Promise me an enjoyable drama.
 
Trivia: 57% of all posts containing "jabascript" are positively scored
reopened as a counter to hyperbole
 
Well, I suppose that is a fix of some sorts...
I pretty much agree with this comment:
Your argument about "too broad" does not make any sense. You quoted the portion that reads "good answers would be too long for this format", implying that this rule is being violated. But the existing answer puts the lie to that implication. It fits within the character limit and works well with "this format". So this is just another case where someone wants Stack Overflow to become a "debug my code" site, rather than a world-class resource for programmers, a body of expert knowledge about practical programming problems. Use your brain as something other than a pattern matcher. — Cody Gray 10 hours ago
 
@Shog9 Haha tahnks in advantage 😁 😏😅
 
@hichris123 isn't pattern matching the fundamental principle of human learning?
@Shog9 I was gonna make a Star Wars joke, but everything I came up with was Lukewarm
 
I suppose you could say that but Cody's referring to "It looks like x, therefore it must be y."
 
9:18 PM
I get the feeling Yakk doesn't understand what Documentation will be like once merged into SO
 
I don't think anyone does right now... the way it looks now will probably change when it becomes public.
 
@Quill so is breathing, but breathing alone doesn't qualify you to determine what is on/off topic on SO.
@hichris123 let's go with, "An unstoppable juggernaut, rolling over everything in its path as it marches, Sherman-like, toward the ocean of knowledge leaving naught but a trail of burned southerners in its wake".
 
That's... sounds... interesting.
 
@Shog9 with some of the comments people make about SO, you'd think that breathing would teach them all they need to know
on that note, I had an interesting discussion with Pops about that the other day
 
That's what you get for being in the middle East. :(
 
10:04 PM
I just remembered I didn't respond to an email from a colleague I got nearly a year ago about how to implement a feature I built
oh well
 
 
1 hour later…
11:05 PM
[ . . . ]
 
user315433
11:39 PM
@Shog9 EE motivates the introduction of 3rd category (after beta and graduated), shouldn't-have-graduated.
 
technically that's a fifth category
 
user315433
@hichris123 (1) Saw "JavaCript" in a title; (2) suggested an edit; (3) Saw 38 more of those and went on...
 
01:00 - 20:0020:00 - 00:00

« first day (2155 days earlier)      last day (2862 days later) »