« first day (1708 days earlier)      last day (3395 days later) » 

5:04 AM
!mustache help
@AstroCB Command format: !mustache[ 1-5] directURLToImage
@AlexisKing trivia: there are over a million users on Stack Overflow alone that qualify for deletion but haven't been deleted due to the throttle. And there are under 4 million registered users on Stack Overflow.
5:06 AM
sounds like the throttle is too strict
ya think?
Every week, we delete 1000 of them. Every week, there are a few more than that who qualify...
@Shog9 Doesn't SE like having those inflated user stats, though?
@Shog9 Then why hasn't that throttle been lifted some?
I guess it doesn't look that good to have a lot of 1 rep accounts.
Why don't you just delete all 4,000,000? Then you'll know the ones that sign back up really want to stick around.
5:09 AM
@AstroCB I don't know that tons of dead accounts really helps any
Is it really that much more work than just changing a number somewhere?
@AlexisKing well... There's probably a reason for the throttle. Like, it takes too long to delete 1000 users or something.
Of course, we could just do it every day.
@Shog9 How could it possibly take very long to delete 1000 users?
Or it used to take too long, and now it's just a number that everybody's afraid to touch.
5:10 AM
Shoulda used a sans-serif font for the headstones
Are you saying that user deletion is somehow an extremely expensive operation?
@JasonC ...and mostly forgot about until someone brings it up
@Shog9 That's a quarter of the registered SO user base, though.
@AlexisKing beats me. We already had a massive backlog when I joined the company.
@SmokeDetector tpu
5:11 AM
@AstroCB Blacklisted user and registered question as true positive: added title to the Bayesian doctype 'bad'.
Meh, fair enough.
@SmokeDetector who the heck casts a close vote?
Zombie accounts.
It's entirely possible that the original idea was to slowly chip away at the backlog, and then... that stopped happening at some point and no one noticed.
5:12 AM
@Braiam I see more "I'm voting to close this question because it is spam" than you'd think.
Oh, also, @Shog9, I posted this a couple days ago, and I'd be curious to hear a word or two from you on it. Am I just wrong? (And when I say "a word or two", I'm being pretty honest... I don't expect you to write out one of your typically thorough explanations.) meta.stackoverflow.com/q/289458/465378
I like being in top something % of users. Don't delete the zombies.
Just tell me if I'm being stupid.
@AlexisKing your question is too broad.
That few enough words?
@Shog9 Maybe it was a safety of some sort in case things went crazy.
5:13 AM
also plausible
@Shog9 You are an evil human being. :)
@pizza but the price per slice of pizza is higher the less number of pizzas available
[ SmokeDetector ] Bad keyword in title: Revitol Anti Aging Cream by lintania wioz on meta.stackexchange.com
yes xkcdBot, well done
5:14 AM
@SmokeDetector tpu
@AstroCB Blacklisted user and registered question as true positive: added title to the Bayesian doctype 'bad'.
@Shog9 Okay, here's what I'd really like to hear your opinion on. I'll make it a yes/no question. Do you agree with me that the "too broad" close reason is abused, if not on a similar scale as "too localized" and friends, at least in similar ways?
@AlexisKing yes
@AlexisKing Every close reason is abused.
Okay. Neat. I'm not being a complete idiot.
5:16 AM
@AlexisKing You may still be being a complete idiot.
MCVE can be used for almost anything.
Too Localized, Goldilocks, Too Broad.
@tchrist MCVE – not enough code, Goldilocks, MCVE – too much code.
@AstroCB I think "too broad" lets people close things that they personally don't want to answer, but someone else could answer very well.
And I'm not convinced that's enough to make a question off-topic.
@AlexisKing I agree.
5:17 AM
@AlexisKing I'm voting to close this question because puppies.
@AlexisKing it's not. But that was never the intent. Plenty of on-topic questions are simply asking for more than we can - or should - provide.
Where this gets dicey is when the question is easily answerable, but someone votes to close anyway because they don't feel the asker deserves an answer.
@Shog9 But what are we gaining from closing those questions rather than leaving them possibly unanswered?
Believe me, I'm all for closing things. If anything, I'm a CV-nazi.
"I want a jQuery code for making the screen slough off like in Doom transition screens"
@AlexisKing Higher signal-to-noise in search results.
That's not too broad.
5:19 AM
@Shog9 That's too broad.
@AlexisKing personally I use too-broad when I think that the question is asking us to either provide an entire solution from vague requirements. or they are asking something like 'What is Java?'
"Too broad" is used all the time where "lacks minimal understanding" would have been, because you can argue that it would be "too broad" to tell the asker everything they'd need to know before being able to solve the problem.
Still gonna piss someone off.
You don't need Too Broad for Madey Uppy reasons when you can make up a new one whenever you please.
A: How to flag a question for protection

Braiam if I want to vote against the deletion, what action can/should i take ? Fix the question. That is always, always the first action. Can't fix the question? Try to re-ask the question in a way that fits Stack Overflow guidelines and import over the content. That sometimes means separating...

5:19 AM
@Shog9 I think it needs to be reworded. Really, that both too broad and too localized.
I'm not going to repeat it again
@AlexisKing it's neither
@AstroCB Too Basic / Needs to Show Research
[ SmokeDetector ] Nested quote blocks in body: who to canvert free Pdf to Excel by juldu nuls on superuser.com
@tchrist You could technically use MCVE there, too.
5:20 AM
@Shog9 It's a mix of both. It's too broad because the OP didn't explain a specific problem, just asked for some vague concept as a question. It's too localized because it's unlikely that many other people will need that specific thing in the future.
@AlexisKing no, here's the real reason it would be closed: the asker didn't search
@Shog9 On the other hand, questions like "How do JavaScript closures work?" is broad, but not too broad.
Because it's answerable and applicable to other people.
I'd imagine it would just be closed as "unclear" by people who didn't know what that transition effect from Doom was. It wasn't a good example question.
@AlexisKing A lot of questions like that have old canonical dupes.
5:22 AM
@AstroCB Right, they do, but new canonical dupes aren't being created because they get closed as too broad.
I just came up with that off the top of my head because I was thinking about Doom earlier. Yes, I'm old. Also, anything you can think of, someone wrote it in JavaScript.
@Shog9 Anyway, I'm not really looking to argue this to death right now, I'm just positing the idea that "too broad" is doing as much harm as good in its current form.
@Shog9 that seems to inflate someone JS ego
Shog maybe you'll be the only other person besides me in here to appreciate this: somethingawful.com/news/pirated-1999-files
5:24 AM
!mustache help
@Shog9 DOOM isn't old. ADVENT is old. Maybe PONG.
@tchrist your face is old.
yes, PONG is old too
> 2. Gin & Juice cover by Phish (actually by System of a Down) (actually Ben Folds).mp3 (Actually the original by Snoop Doggy Dogg)
Ah, napster
[ SmokeDetector ] Offensive answer detected: which way should I handle a little boy who likes girls toys! by Huns on parenting.stackexchange.com
5:27 AM
> Limewire Pro Enterprise Edition Cracked - Perfect for Businesses Fully Legit Includes Elf Bowling.executable
@AlexisKing "When should I use present tense? What does a compiler do? What are the rules for commas? Where should I put my braces?"
@Shog9 archie
@tchrist I'd argue that the second question is actually answerable. Numbers 1 and 3 are too broad because they're not well-defined. Number 4 is primarily opinion-based.
[ SmokeDetector ] Bad keyword in title: Skin care tips for women oily skin by cynthia sisto on meta.stackexchange.com
5:29 AM
@tchrist What about quommas?
@tchrist stop reading WP
@JasonC Yo momma's what again?
@JasonC I'm going to continue this here because I don't think the comments are a good place to do this.
@tchrist That's right.
@Braiam I was on ARPANET in 1983. How 'bout you?
5:30 AM
@tchrist I completely forgot that existed.
@JasonC Basically, just what I was saying with Shog—new canonical duplicates aren't being created anymore because all of them are considered "too broad" while they could really be left open with no harm done. If they go unanswered, then oh well, but if people actually want answers to that question, someone will come along and write a fantastic answer, and then they'll get a ton of rep for it. The site works! What's the problem here?
@tchrist I don't have recollections of that time, heck I don't have recollections of 2010!?
The trouble is when people ask things like "how do I make webzone?" that things need to be closed for being too broad because such an answer cannot exist.
@tchrist And that weird gopher phase.
@AlexisKing just accept that every good question was already asked and answered
except lisp
5:32 AM
We shouldn't close questions just because we're too lazy to write a good answer ourselves even if one could theoretically be written.
@Braiam There are new languages coming out all the time. :)
But we do.
@AlexisKing Well you could go write the answers.
@tchrist We do what? Close them? Yes. That's my point. We do that but we shouldn't.
@JasonC I can't write the answers if the questions are closed. :)
You could do a self-answered question, too, if you think you have good canonical duplicate source material.
You could make a case and vote to reopen.
@JasonC Yes, but that shouldn't be the only way to ask those kinds of questions.
5:33 AM
(╯°□°)╯︵ ┻━┻
@JasonC Yeah, but then the system isn't working right to begin with.
Sure it is.
@AlexisKing no, we just close crappy question that happens to be too broad
CV easy, answer hard.
@JasonC Okay, picture this. Give me a second to type this out.
5:34 AM
if you ask a good question that is almost on the fence of being too broad...
The community wants those questions to be closed, and they are. Therefore the system works. Your best bet is to do what you are doing now and post on meta to try and change the general attitude such that those questions don't get closed.
@JasonC Saying the fact that people close questions means that those should be closed is a terrible argument.
@JasonC Tell me how that works out for you.
5:35 AM
Look what happened with too localized and lacks minimal understanding.
@tchrist How what works out for me?
case point:
A: Why doesn't [CSS feature] work in [browser] but works in others?

bjb568Though it is not always the case, one of the most common reasons why a property like transition or animation works on some browsers and not others is because of vendor prefixes. What are vendor prefixes? At the time version 4 of Firefox was introduced, the CSS transition module specification wa...

@Roombatron5000 Burn with bleach, hydrofluoric acid, then fire, then a meteor strike
@AlexisKing Yes. Those close reasons were removed.
@JasonC And yet you're claiming that the current ones are infallible? o.O
5:36 AM
Unless that's not the "what happened" you were referring to?
@AlexisKing I never said that, or implied that.
@JasonC The "what happened" I referred to is that they were abused.
Questions were closed that shouldn't have been closed.
gratuitous graph of questions closed, by reason, over the past 20 weeks:
Sure, maybe 19/20 of them should have been closed.
5:37 AM
@AlexisKing And subsequently they went away because, as a community, we didn't like the abuse. Therefore the system worked. So as I said, keep bringing up the point on meta. If the culture shifts such that the community does not like the "too broad" reasoning any more, it will be removed. Don't make me talk in circles. :P
@Shog9 I thought we were optimizing for pearls, not sand?
A title for your discussion: "Are we eating our seed corn?"
@Shog9 Maybe I misinterpreted your point, but the point I was making is that good "too broad" questions might be very rare, but closing them forecloses on fantastic answers.
@JasonC is this what you call a tavern brawl?
@AlexisKing Then answer.
5:39 AM
@James No. This is why I hate text. I'm totally trying to sound reasonable.
@tchrist You're not looking at the long term. Let me write the example I've been trying to, but I keep getting caught up in the discussion. :P
@JasonC No, not implying that you aren't being reasonable, just seems like everyone is talking with everyone about the same thing that no one will agree on
Concrete, specific, actual examples are key.
@James Oh. No, that's just what you call a bunch of engineers in a chat room.
5:41 AM
@AlexisKing so the idea of "optimizing for pearls" is that you got all the sand you want. All the sand you'll ever need. You don't get more pearls by throwing more sand at the same # of oysters. Someone has a brilliant answer, you want them to have the opportunity to write it... But you don't increase the chances of getting that answer by just dumping more and more sand on 'em.
Anyway, let's say a user asks a question that's pretty broad. Maybe something like "How does pattern-matching work in Scala?" That's an answerable question, but it's really hard to make a good answer quickly. It takes a lot of work. People aren't going to answer it immediately.

But people are going to want an answer to that question. It gets upvoted. Maybe it sits there for *months* without an answer. Perhaps someone puts a +200 bounty on it. It gets featured. Some Scala pro takes the time to sit down and write out a 1500 word answer that explains it *perfectly*. That user gets a ton of up
@James I edited right as you caught me.
@tchrist @JasonC @Shog9 ^ That's sort of what I'm referring to.
anyone remembers the namespace for the css3 draft?
w3.org page is in the way
@Shog9 Yeah. I think I'm worried that we're closing all the questions that could have really big pearls.
5:42 AM
the big problem we have right now isn't a lack of sand, a lack of opportunity. If anything, it's that there's so much sand that we've buried the oysters.
@AlexisKing That's how it does work now on some questions, and not on others.
@AlexisKing Necromancer can be awarded more than once.
@JasonC It's a lot rarer than it used to be (partially because the site quality has massively declined as a whole, of course).
@tchrist she is saying that we are closing them almost instantly
Yeah, that. ^
5:44 AM
Kevin's been working on something... It's pretty cool, he's gonna post about it tomorrow... But the rough idea is, you give folks a way of finding their way to other questions they might be interested in answering. Not by shoving a big pile of Java questions in their face as soon as they answer some obscure algorithm question with [java] on it, but by actually doing a proper search using terms in the current question and then picking the top few results.
[ SmokeDetector ] Bad keyword in title: Revitol Anti Aging Cream by myrtl stev on meta.stackexchange.com
Instanly gone
@Shog9 should just borrow wisdom of Jeff:
A: At what level does "do X for me" become off-topic?

Jeff AtwoodI certainly defer to the others here who are active participants -- but our general philosophy is to heavily favor answerers. We feel that the world is awash in questions, but not answers. Answers are the real unit of work in any Q&A system. Therefore, the only logical thing to do is to maximiz...

@Shog9 That sounds neat, but it doesn't help if some of the coolest questions to answer are already closed.
Stop dumping sand on people; find that perfect little bit of mother-of-pearl and give it to the oyster.
5:45 AM
@AlexisKing All right. Consider this: If, in general, the question "How does pattern-matching work in Scala" was closed, but truly had value, what stops said pro from making a self-answered question? I would imagine that a Scala pro who understood what was valuable would self-motivate to add that question to the SO database.
operative phrase: If this means aggressively closing unworthy or uninteresting questions, so be it.
@Shog9 I don't care about the sand. I do care about the mother-of-pearl that the community deems is off-topic.
In other words, if it should exist, it will.
@AlexisKing no, it doesn't. But again, the working theory is that any worthwhile question will be asked again. Has been asked again, numerous times. Hasn't yet gotten a worthwhile answer yet maybe, but not for the lack of asking.
@JasonC I don't think that's true.
5:46 AM
So what you could do is start encouraging people to create canonical self-answered duplicate sources for questions they believe have value, even ones that haven't been asked yet.
@JasonC A Scala pro who already gets pattern-matching isn't going to really feel motivated to write out a massive post about it simply because they want to share that knowledge. But if there's a question with 25 upvotes and a 200 rep bounty, then sure, maybe.
E.g.: Hey everybody, think of the best question you possibly can about the field you're the most interested in. Search for it here. If it hasn't been asked yet, add it. A push to increase canonical questions.
@AlexisKing But if the question truly has value, then a Scala pro will know it, and will invest in writing it for future rep gains. If they don't see that value, then perhaps it's not a great long term question to begin with.
@JasonC Sometimes it's very non-obvious to a pro what the community really needs. I know this firsthand.
Part of the point of Stack Overflow is that good questions can rise to the top because people look for them and vote on them.
@AlexisKing Ok then, do you have a specific example of a valuable question that has popped up on more than one occasion but has been closed? (Note that I associate "being asked more than once" with "high value to the community")
5:48 AM
@AlexisKing My most highly upvoted question/answer is a self answer question I did because I was sick of not having a resource to point to when answering similar types of questions.
@James Then how do some people rack up dozens of Necromancer badges? I've got 34 myself and lots of folks have many more.
So, now and then, I like to remind myself that Stack Overflow isn't 100% crap by opening up the 10K tools and looking at the highest-voted answers for [time period]
@James s/My/Most self answered questions/
@JasonC I ran into them all the time when I monitored the [java] tag (especially with Java 8), but I stopped doing that after I hit 10k. Maybe I'll start doing that again to add fuel to the discussion.
@AlexisKing Find an example.
5:50 AM
@JasonC I will. Not tonight, but I will.
Here's a question asked and answered just a couple of days ago:
A: What are the actual uses of ES6 WeakMap?

Benjamin GruenbaumFundamentally WeakMaps provide a way to extend objects from the outside without interfering with garbage collection. Whenever you want to extend an object but can't because it is sealed - or from an external source - a WeakMap can be applied. A WeakMap is a map (dictionary) where the keys are w...

You have to go look for those old questions.
At first glance, that's pretty broad.
@Shog9 It helps that ES6 is a new technology. People are much more trigger-happy on technologies like Java, even if someone explicitly points out "hey, I want a solution with Java 8 now that it has all these cool new features".
Once you find that example, I propose asking and self-answering it rather than spending the time posting on meta about it. It's hard to find those gold mines, you're lucky when you do.
5:51 AM
@JasonC I think they'd be easier to find if people left them open and upvoted them instead of immediately crying out "too broad, too broad!"
@AlexisKing you have a 9 days window, why haven't you found one?
@Braiam What do you mean a 9 days window?
@AlexisKing They should be just as easy to find. Closed questions are readily available in search results.
[ SmokeDetector ] Bad keyword in body, Blacklisted website: Bio Diamond Herbal Anti-aging Skincare For Today's by ewasken730 on meta.stackexchange.com
Q: Need help for including a username and a password and simplifying

HodhodI'm making a FTP Client that needs a username and a password. My problem is that I don't know how to include them in the program. I want it continue only if the username and password are correct. Futhermore I would like to ask if my program could be simplified more

5:53 AM
@JasonC I think I haven't emphasized a part of my argument enough, but I should.
At the risk of meta-effecting a close vote stampede on my pet question/answer this is the question I was talking about http://stackoverflow.com/q/24257449/2357233
from the title, yes it's waaaaaaay to broad, but I gave a comprehensive through answer
@AlexisKing No. You should pause for a day or two and find some examples.
@JasonC And I will, but hear me out.
mm... @Shog9 I'm supposed to see the number of pending reviews (close queue) on a tag page even if I don't have 3k?
(This is relevant.)
A lot of the time, these questions aren't terribly obvious to people who have a comprehensive understanding of the technology in question. Things like pattern-matching are so darn basic that people frequently aren't going to realize that, oh yeah, this could make a good answer.
5:55 AM
@James That wouldn't have struck me as too broad, actually.
@Braiam no
oh cool, it takes me to LQRQ
So leaving those questions open and letting them accumulate upvotes is actually informative—it tells people what the community actually needs.
This is not always obvious.
@JasonC Really? huh, I had to fight people to let it stay
@AlexisKing Right. But if people are active on the site they will recognize that the question was asked many times, that's the signal.
5:56 AM
Oh, just expressing my annoyance (frustration?) at a too common phenomena of wanting to do something but it isn't possible for no good reason other than "we didn't do in a way that supports that". — Michael 7 mins ago
meh, you were right @pizza and whoever else guessed
@JasonC There would be a lot more people writing great answers if being hyperactive on the site and wading through loads of crap wasn't a prerequisite for finding these questions.
Remember, you and I are not the average members of SO.
@Shog9 Those are all answers, though, right?
@AlexisKing Sorry, but you're going to have to do a lot of research to back up the claim that "being hyperactive on the site and wading through loads of crap" is truly the main reason why people aren't writing great answers. That's way too hypothetical.
@Braiam I don't get it
@AlexisKing right. Those are the pearls.
@James would you get it if I tell you that I suck at regexes?
@JasonC What? People have left SO all the time. I didn't think this was under discussion. Where was that awesome image on that one meta post talking about the three kinds of SO users?
High-rep users who have basically given up is the group I have in mind right now.
@AlexisKing Yes, many people have left SO.
A: Why is Stack Overflow so negative of late?

Mysticial Why is Stack Overflow so negative of late? I hate to pile on, but I couldn't resist summarizing the problem. And I apologize ahead of time for any potentially offensive language that I might use. Basically there are 4 camps of users on Stack Overflow: The "caretakers" who want to keep the...

5:59 AM
@JasonC Many people who could write great answers. I think the primary reason for leaving is the sense that finding good questions to answer is too much of a chore.
Reel this back in. Go find some examples.
@Shog9 Yes, that's it. :)

« first day (1708 days earlier)      last day (3395 days later) »