« first day (1870 days earlier)      last day (3142 days later) » 

4:00 PM
Guyses I think there's some potential bug with spoiler markdown/
 
>! 1. Haha
2. Haha
Doesn't work.
>! 1. Haha
>! 2. Hahaha
 
@UnitedStatesOfAmerica Code for what?
 
The API code
After you login and approve the app
 
Also doesn't work because it's not properly indented as numbered lists should be.
 
4:01 PM
I'm obviously not understanding. What code do you need to input each time?
 
!!/random
@UnitedStatesOfAmerica 0.660626117726
@Andy access code
 
So, should I report it?
 
@UnitedStatesOfAmerica It doesn't change. Register the app and keep the code it generates. Your token stays the same.
 
Ok, will do when I can
!!/random 1 10
 
@Andy you think that's a bug?
 
4:07 PM
!!/random 1 10
@UnitedStatesOfAmerica 9
!!/random 1 10000
@UnitedStatesOfAmerica 4014
 
y u no do teh testingz in sandbox?
 
> This is a photocopier. The photocopier is your friend.
 
Falls asleep
 
> That's not the title page
> Press the plunger to get your change
 
4:08 PM
@Pro you think that was a bug in the editor?
 
Editor?
 
8 mins ago, by inɒzɘmɒЯ.A.M
>! 1. Haha
2. Haha
 
Ah.
 
8 mins ago, by inɒzɘmɒЯ.A.M
>! 1. Haha
>! 2. Hahaha
 
let me check
 
4:09 PM
None of these two work.
 
The second works for me, assuming that you want to display both things in a spoiler.
 
@inɒzɘmɒЯ.A.M No. It shouldn't change. When you authorize an app to act on your behalf it provides you with a token. You pass that token through the app when you do something.
 
[ SmokeDetector ] Few unique characters in answer: Is rust from steel rainwater barrel harmful? by user567 on gardening.stackexchange.com
 
If it changed each time, you've have to reauthorize it every time you did something
 
4:11 PM
Or wait, do you want to make it appear like an ordered list, @inɒzɘmɒЯ.A.M?
 
@Andy Didn't you mean this to be for @Uni?
@ProgramFOX Hmm.
I mean yes.
 
@inɒzɘmɒЯ.A.M Well in that case, then it doesn't work for me either.
 
@inɒzɘmɒЯ.A.M You asked if I thought it was a bug
 
[ SmokeDetector ] All-caps title: LAN TO WAN CONNECTION TWO ROUTERS by DarkNet99 on superuser.com
 
@Andy It's not one?
 
4:12 PM
>! <ol><li>abc</li><li>def</li></ol>
 
No.
 
Let me reread what you said.
 
@inɒzɘmɒЯ.A.M ^ Ugly workaround.
 
@ProgramFOX HTML is always ugly.
 
1. Developer creates app
2. User authorizes app to act on their behalf (receive a token)
3. User uses app to do *thing*. Passes token to app
4. App uses passed token to do *thing*.
5. API sees token belongs to user and does performs action under user's name
 
4:15 PM
@Andy Uh, inɒzɘmɒЯ.A.M was talking about the editor bug in his ping, not about the API stuff.
 
Well, now I'm all confused. Carry on. I'm going to go eat lunch or something
 
@bjb568 Chemistry?
 
Half a letter grade (it's midterms about now) for not showing work on one assignment. I got all the questions right on that assignment…
@inɒzɘmɒЯ.A.M yorp
 
4:22 PM
@bjb568 You've got a free tutor in the Tavern if you ever didn't understand what humans study.
0
Q: Spoiler text doesn't work with numbered lists - Intended behavior? Workaround?

inɒzɘmɒЯ.A.MThe spoiler markdown is sometimes cough messed up. Today's edition: Spoiler doesn't work with numbered lists. >! 1. Foo 2. Bar 3. Baz renders like ! 1. Foo 2. Bar 3. Baz I could instead do >! 1. Foo >! 2. Bar >! 3. Baz but that also turns into Sheesh, it's ugly! I cou...

4 days to the lunatic badge. Yay!
 
@inɒzɘmɒЯ.A.M I understand what humans study.
 
@bjb568 Even humans don't understand what humans study.
 
4:48 PM
@SPArchaeologist fair enough, guess it's not possible to make everyone happy :)
@SPArchaeologist hmm... sounds like something that you don't like... thanks for the tip!
@inɒzɘmɒЯ.A.M no spoilers in chat afaik
chat got very limited markdown :/
 
@ShadowWizard Not chat.
Real site.
 
5:03 PM
[ SmokeDetector ] Link at end of body: The blood vessels then that how by cornato tom on superuser.com
 
sd trueu-
 
5:33 PM
[ SmokeDetector ] Numbers-only title: 1 4 10 11 19 21 25 29 36 by Darren on math.stackexchange.com
 
5:49 PM
sd why
 
@DragonLordtheFiery [:3974886] Body - Position 37-58: trainingstationinc.co
 
sd - falseu-
 
[ SmokeDetector ] All-caps title: CANNOT START UNITY by Paul Delage on askubuntu.com
 
6:30 PM
@SmokeDetector Smokey's going to give me the Copy Editor badge shortly after giving me the Marshal badge...
 
6:48 PM
@SmokeDetector tpu-
Where is everybody....? Smokey, I feel lonely!
 
I'm always here...
 
@rene That's good to know :-)
 
7:05 PM
!!/tea
 
@S.L.Barth brews a cup of lemon tea for @S.L.Barth
 
I'm starting to understand Uni's pain... I want spam to flag and it just won't appear!
 
Gahd it's the 3rd week of a tech club and we haven't started the projects that we'll be presenting, but we're enduring a presentation now on presenting skills.
The club leader (student) is crazy
 
7:20 PM
 
Is there currently a proposal on meta about having close votes scale with reputation the way delete votes do?
@Undo Long ago flagged.
 
@tchrist I know ;)
I'm hoping I can get an inheritance from him
 
7:51 PM
New close options:
 
For the republicans
 
@tchrist In favor! But for which questions should I use "none"?
 
user259867
@tchrist Shog occasionally says he's in favour of that happening. In this chat, for example.
 
I recall.
 
user259867
Aug 18 at 21:27, by Shog9
@durron597 I think 24 votes/day is reasonably generous as a baseline. Rather than artificially spiking this on sites like SO, that limit should scale with rep - so at 10K you'd have 38, 20K you'd have 58, 800K you'd have 1618...
 
8:05 PM
(removed)
 
False positive.
 
@NormalHuman Done.
1
Q: Proposal to make close votes scale with rep

tchristThere have been several proposals for how to handle the ever-mounting close-vote queue. Dupehammer has been especially successful, and there have been several proposals since them for awarding more close votes one way or the other, including one to give extra close votes, only accessible via /re...

 
okay folks, I need some fake information
 
@tchrist your get 50 close votes on SO, and 40 reviewtasks per day due to the cv queue size
 
@rene Oh right.
Fixing.
I didn't mean to try to pull in the 40-vs-50 thing.
 
8:19 PM
@Undo use this image
 
@bjb568 Wonderful
 
@tchrist The problem I have with that is some people (Roombatron/@CRABOLO for one) are the top CV'ers but don't have a ton of rep.
 
@hichris123 why is that a problem?
 
I was wondering that myself.
 
typical, you ask for an explanation and they run...
 
8:26 PM
Someone needs to breathe more and read slower, alas.
And maybe be nice.
 
@rene That the proposal would give more close votes to the people who don't close vote a ton but have a ton of rep than the people who close vote a ton & don't have a ton of rep.
@rene I do have other stuff to do. ;)
 
@hichris123 I know, I am just teasing you...
 
I know, I know. Flowers have a nice sense of humor.
 
@hichris123 Is that a typical small site problem? I don't expect much imbalance on SO.
 
Eh, not sure.
Although... remembering back, I liked this proposal of yours @tchrist:
18
A: Give extra close votes, only accessible via /review

tchrist⁢        ⁢        ⁢        ⁢     Double-Badger Superpowers ⁢                                                 Photo credit: Chris Noble at The Wildwood Trust TL;DR: Give holders of CVRQ Steward + silver tag-badge holders double-weighted CVs in that badged tag Give holders of CVRQ Steward +...

 
8:38 PM
The problem is that it's complicated.
Too much dev time.
For too little gain.
However.
I really like the idea of privs that derive from accomplishment not just popularity or longevity alone.
 
The think I like though is that it avoids the "give more close votes" problem and gives people a more powerful close vote.
(sorry if my thoughts are all over the map here... they kinda are)
 
when I say log n, what the heck I'm saying?
what is the base?
 
@Braiam 10?
 
@Undo dunno, Sturges didn't say
 
18
Q: When log is written without a base, is the equation normally referring to log base 10 or natural log?

john smithFor example, this question presents the equation $$\omega(n) < \frac{\log n}{\log \log n} + 1.4573 \frac{\log n}{(\log \log n)^{2}},$$ but I'm not entirely sure if this is referring to log base $10$ or the natural logarithm.

 
@Bart as always, there's always a SE question for it... and a whole world that doesn't search for it
 
:D
 
[ SmokeDetector ] Few unique characters in answer, repeating characters in answer: How to find villages? by rgr on gaming.stackexchange.com
 
sd trueu-
 
9:00 PM
Let me ask this as constructively as possible. Programmers gets only 25 questions per day; why then is gnat so tetchy about close votes there? With 24, everybody gets to vote on very nearly every question that come in.
 
@tchrist Some people are zealous
 
Was that jealous? :)
I have another datapoint that may help explain it.
The top closing sites are Programmers, The Workplace, Network Engineering, Christianity, Patents, and ELU in that order.
 
@tchrist part of it is cleanups.
 
Notice that that looks very little like the order for the top sites in terms of most anything else.
 
The other part...
 
9:03 PM
@tchrist historical + migration reasons
 
So they are having trouble marshalling the troops to clean out the old stables full of crap questions that are a poor fit for the site?
 
I... guess? Not sure though.
 
Programmers is the top closing site.
It doesn't really matter why.
This may be connected to his touchiness. I don't know.
I do know that I blow through my close votes on ELU more often than I don't, but I only get 24 there. However, I almost never do so through the review queue.
That said, I'm the top CVRQer there. But I seldom use it much.
Without discipline, all 24 CVs are gone within the first hour UTC.
 
the eternal figth for supremacy between closing incoming crap and historical crap, and the anoyance when you have to think too much before using your votes, so you can deal when you see one
 
Part of that all three of our most active moderators are gone to bed by then, and the stuff they would just unilaterally close needs five of us that late in the day.
The other part is that India and China are getting up soon. See the first part.
 
9:11 PM
tl;dr tchrist moderator for ELU
 
no no no
I couldn't close as much then.
Already I have to be careful because of Mjölnir.
I want the Community to work on this, not one person.
Look at my lovely reps!!!!
2¹²
You’ll never see that again.
 
@tchrist mm.... I'm confused
Community or the community?
 
Sorry.
Not Community♦, but the Community at large.
@Braiam Could you please do me a small favor? Could you tell me what the CVRQ looks like on U&L right now?
I realize it will look different for you, but it's a general figure.
 
@tchrist two questions, 1 dupe and another unclear
 
Never mind, I looked. Forgot I could.
Yeah, very low.
 
9:19 PM
well, that's because we normally close stuff fast
 
You get a lot of questions.
About a third more a day than ELU.
So ~80 instead of ~60.
But we close far more as a percentage than you do, and we have a close queue thats hundreds deep.
Aw gee, even Math has only 59 in its queue.
 
more engaged users, mayhap? Also, two of the most active users are mods too
 
Does this show 158 in the CVQR for you?
@Braiam Now that #2 makes a difference.
 
today on UL: Questions Closed: 14 Questions Asked: 113 Close percentage: 12.39 %
@tchrist 222
 
@Braiam AHAH! That difference shows the ones I’ve already voted on. :(
 
9:24 PM
last 7 days: Questions Closed: 95 Questions Asked: 651 Close percentage: 14.59 %
 
Where are you pulling those data from so quickly/conveniently?
Time travel.
Lemme check ours.
 
man, I tip my hats off of whoever implemented (gave the idea of) that
 
Over Closed Asked Percent
 1d     12     65 18.46 %
 2d     51    129 39.52 %
 7d    174    435 40.00 %
14d    363    915 39.67 %
30d    769   2036 37.77 %
90d   2455   6284 39.07 %
 
for comparisions, today on AU: Questions Closed: 41Questions Asked: 193Close percentage: 21.24 %
 
So it takes us two days to approach the real mean. That’s ok.
 
9:30 PM
last 7 days: Questions Closed: 308Questions Asked: 1419Close percentage: 21.71 %
 
Only 22%? You guy aren’t trying hard enough! :)
 
90 days stats are slow!
 
Fancy that.
You know what, I bet it takes us two days to close things because we only have 24 votes and there are twice that number of questions to close.
Because we don't have enough close voters.
Again.
 
UL 90 days: Questions Closed: 1143 Questions Asked: 8002 Close percentage: 14.28 %
 
1
Q: Let's have some better leaderboards for review tasks

durron597The problem @tchrist recently asked this question: Proposal to make close votes scale with rep While I completely agree with the sentiment, I wholeheartedly disagree with the solution. As one of the more active close voters on Stack Overflow and a top five close voter on Programmers.SE (in the...

 
9:33 PM
Yes, very low.
 
still waiting AU's
 
@durron597 What is your proposal?
 
@tchrist Programmers is the top closing site because we have a really dedicated userbase that works really hard and uses all their votes every single day
 
[ SmokeDetector ] Bad keyword in answer, offensive answer detected: how to block proxy websites to enter tumblr by wetpussyxxxwetpussy on stackoverflow.com
 
@tchrist Whoops, forgot a sentence.
@SmokeDetector tpu-
 
9:36 PM
@durron597 Don’t bury your lede.
 
@SmokeDetector cv-pls too
 
[ SmokeDetector ] Offensive answer detected, blacklisted user: How to add a where clause in a MySQL Insert statement? by wetpussyxxxwetpussy on stackoverflow.com
 
sd trueu-
sd false-
Not whitelisting the user, though.
 
@durron597 Do you believe that Review Queue leagues would somehow help the people who run out of close votes?
Or are you trying to encourage more people to top off?
 
[ SmokeDetector ] Manually reported answer: How to open a PDF file in the Kindle app for Windows? by ayush on superuser.com
 
I've also protected the question. Six answers are deleted, and a seventh will be removed shortly.
 
@DragonLordtheFiery Are the others of the same ilk?
Maybe you should leave it unprotected if it does a good job of smoking out spammers!
 
There's a bunch of "thanks!", "me too!", and spam answers.
 
Me too deserves protecting against. Persistent spam bait, well, this might be a feature not a bug. :)
Just a thought.
 
10:01 PM
@tchrist I would hope that it could work as an incentive, yes
 
Incentives are good.
 
@tchrist I'm trying to encourage more people to top off
What I don't want is for SO staff to say "okay, we gave people more close votes, lets move on to other problems"
 
@DragonLordtheFiery Your protection racket is still in its infancy, Grasshopper: 3833380 :)
 
:-)
I've protected quite a few questions since I earned the privilege, but I don't protect questions without a good reason.
 
@tchrist Be careful, google said "105" results for mine SO protections but in reality it was only 62
 
10:07 PM
> Our systems have detected unusual traffic from your computer network. This page checks to see if it's really you sending the requests, and not a robot. Why did this happen?
 
Complex queries most often trigger this CAPTCHA.
 
That’s just SE’s way of telling you that your Windoze box has been inducted into the DDOS zombie army.
 
Especially if you use really complex parameters like numeric ranges (10..100) or operators such as site:.
Numeric ranges can cause Google to spend several seconds processing your query.
 
10:32 PM
@durron597 I agree that getting more people to vote really is the best scalable solution. I wonder though, if the problem is that once you've run the close-vote review queue, you have frustratingly few close votes to spend outside of the queue, would it also work to have "extra" votes that can only be spent on the queue, or must be first spent there or something?
 
10:56 PM
[ SmokeDetector ] Few unique characters in answer: Sanitizing/Escaping user input and output by user5348397 on stackoverflow.com
 
sd trueu-
 
user259867
11:17 PM
Spam? money.stackexchange.com/a/53966 it does disclose affiliation, but on the other hand is nothing but an ad.
 
@NormalHuman Flagged.
Pretty obvious when you have nearly 600 spam flags under your belt.
 
I feel it is spam.
@DragonLordtheFiery I would be surprised if he had fewer than you.
 
@NormalHuman has 1390 flags on Mathematics and 1079 on Meta Stack Exchange. Can't access the spam flag counts, though; these are total flag counts, which are public.
Then again, he's one of the maintainers of SmokeDetector and so he probably does have more spam flags.
My focus has been on Super User, where we get all sorts of tricky spam that often escapes less experienced reviewers.
As of now, I have 2299 helpful flags on Super User, of which 581 are for spam.
 
@DragonLordtheFiery Check SO. ;) I think the Normal Human knows decently well what to flag, just wanted a sanity-check.
 
2300 flags. That's an awful lot of flags for the rep he has.
 
11:26 PM
He’s just understated and modest. He has 2300 on SO, 1267 on Drupal, 555 on AskUbuntu, and plenty more elsewhere. I’ll let you extrapolate.
 
@tchrist Yeah, and that means I'm not modest. That's my fault. Apologies to anyone insulted.
 
We know you're working on it, Krait.
 
I've only just started using SmokeDetector on a regular basis to augment my existing cleanup work on Super User.
 
Isn't Smokey in your chatroom?
He’s in ours.
 
Tried running an instance of it myself, didn't work out.
Well, I'll just keep on flagging. :-)
 
user259867
 
@DragonLordtheFiery I’m pretty sure ELU has a Smokey presence because @Undo runs Smokey on chat.stackexchange.com, just like here.
@NormalHuman Sigh. Have you ever hung out with MJD in real life?
 
@NormalHuman Answer is already gone.
 
I have; I like him, and I find him sometimes hilarious and sometimes brilliant. But he can definitely set people off when he wants to, and sometimes when he doesn't.
Not that there is any excuse for abusive vandalism.
 
user259867
No interactions outside of meta.math, where we sometimes agree and sometimes don't. We don't answer same kind of questions on the main site.
 
user259867
He put together the MathJax tutorial, the top-voted post on meta math (1K votes by now). And started the collection of words that should not be in titles which I expanded and Shog used in a warning pop-up.
 
user259867
11:45 PM
Depressing to click on those and find that the current number of appearances is about twice what it was at the time of writing.
 
Do we need to blacklist help just as problem is?
 
ELU has more than 5,000 questions with “correct” in the title, of which like 56% are closed or deleted.
Hm, maybe that's not quite right. I would have to recheck Jon's post and figures. But it's horrible.
 
user259867
Help is partially blacklisted on SO; not in every context. There are no hard blocks based on title keywords outside of SO.
 
user259867
Except CJK characters
 
11:49 PM
SO blocked “problems” for like a 21% close-rate, whereas we have “CORRECT” at 51% and “GRAMMA” at 56%.
And we don’t block those.
This happens because there is no upward bound on the number of sentences that non-native speakers can post and ask us to tell them whether or not it is grammatical correct. And they do.
We could also have a popup like people putting regex in the titles get, rather than a true block.
That might do more towards getting folks to post better questions.
I don't know how many sites have what popups.
 

« first day (1870 days earlier)      last day (3142 days later) »