« first day (2326 days earlier)      last day (2686 days later) » 

user315433
12:00 AM
It's the Unhappy Hour courtesy of MIT admissions.
 
@Shog9 It's just supposed to be.
 
What a failure.
 
Wait, so, websocket client error events really give no way to get details about the error?
I guess I see the rationale... w3.org/TR/websockets/#concept-websocket-close-fail... but that's irritating.
Especially because the error details actually do appear in the dev console, like "ERR_NAME_NOT_RESOLVED", which kind of hilariously defeats the purpose of W3's choice there.
 
user315433
@Dr.YSG This is not strange at all. The error message in the developer tools is meant for the user of the browser so it can contain sensitive information. OTOH some random JavaScript code is generally not trusted. Otherwise you could write a worm (like a port scanner or a DDoS script) and just spread it via some random ad network. — jpc Jan 12 '15 at 14:03
 
Or does it, I guess it really just doesn't want the info to go to scripts.
Oh yeah like that.
Well whatever this seems super easy and convenient so far, I can take missing error info.
 
user315433
12:09 AM
> This W3C spec reads like a text on gnostic mysticism pic.twitter.com/CpIYdwrLRP -- Jaiden Mispy at 12:41 PM - 11 Nov 2016
 
user315433
user image
2
 
Lol
 
 
1 hour later…
user202362
1:24 AM
 
user202362
 
user202362
4:05 AM
 
8:14 AM
@zaq just for fun... are they talking about something related to Tor or other similar stuff?
 
9:12 AM
I think any question posted on meta should come with a banner which says "This is Meta, please delete your question" @rene ;)
 
9:34 AM
@JasonC but but... I heard that if you get strong enough, you will be just given the opportunity to continue train with a strange Egyptian-like cat in the afterlife.
 
 
2 hours later…
11:06 AM
Not quite sure if this is the right place to ask, but searching for "@editor" on meta ignores the "@"
However, does using "@editor" in a comment notify a suggested editor please?
 
11:23 AM
@Bart Ha, I try to be helpful, right? Doesn't hurt anybody to bring in some Dutch bluntness ...
@JamesWebster no, it doesn't. You can ping the user which edit was approved and/or the dupe hammer close voter. Those don't auto-complete though, so you have to type the username yourself.
 
Thanks @rene. This was just for an invalid suggested edit on one of my posts, I wanted to provide a comment with my rejection, but in the end, I settled for just the "other" box
Thanks for the link. I did check on meta, but the search box just ignored my "@"
 
Yeah, that is a known limitation of search
 
I thought I had read that @editor was supported, but from that question it seems I'm wrong.
"Are there any special keywords for notifying users?

No. "
 
:D
we are not a social network ... so pinging annoying as much users as possible is not a core function
 
12:24 PM
@JamesWebster editor yes but user whose edit was rejected no.
 
@ShadowWizard, Ah I had read it somewhere. Thanks.
 
1:05 PM
you try to qualify to be hired? ;)
 
1:38 PM
Am I the only person with problems with the hamburger menu on mobile chat? It's like there's some offset between where I touch and what menu item gets selected. And I definitely can't select "starred".
 
2:14 PM
status-no-repro
 
3:06 PM
@ShadowWizard Weird. I've actually had problems with that menu working right consistently since the new chat interface way back when.
 
user315433
3:39 PM
Good morning.
 
[ SmokeDetector | MS ] Manually reported answer: My boyfriend's dog hate me... help! Warning! Long post by henderneth on pets.stackexchange.com
 
 
1 hour later…
5:00 PM
Today's xkcd comic (Negativity):
 
user315433
5:28 PM
An easy way to block a lawn is to bury it under a few feet of snow.
 
user315433
5:46 PM
8
Q: In the spirit of the holidays

Larian LeQuellaI've noticed a few threads have become a little heated as of late. Since the days are getting shorter up here in the northern hemisphere in preparation for all the solstice themed holidays, and a few folks may be getting a bit testy, I am kindly asking that before posting, you count to 10, take a...

 
user315433
Nice post, now featured on Skeptics.
 
Can you block a lawn like you block a hat? Question for either G&L or Crafts
@JasonC what phone / what browser?
 
@Shog9 Kyocera Brigadier using the uh... browser.
 
@JasonC ok, what OS version?
 
Android Chrome I guess
@Shog9 Both 4.4 and 5.1.
I updated it "recently".
 
5:52 PM
My guess would be slightly miscalibrated touchscreen, but who knows
 
Nah it's spot on. I actually tested that. The issue is unique to SE's mobile chat hamburger menu.
SE sites do other odd things on my phone too that no other sites do.
 
@JasonC strange. I have not observed this.
 
Like they seem to kill the autocorrect a lot (chat does this, ask question does this, comments though are usually OK). Also for some reason on most SE sites the usual hold-press text edit features like copy and paste aren't available. It's weird.
 
you sure you're using Chrome?
 
I'm guessing I am, since it has an incognito tab feature. Hold on let me find an about box.
It is Chrome 54
 
5:56 PM
ok. I'm on 55, but that's not old and I don't think chat has changed significantly since I was on 54
strange
 
Yeah I dunno I just kind of live with it.
The system has beaten me into submission, I've just accepted that no phones do anything correctly, and so I never complain. It's a sad state of affairs. Welcome to the future.
 
From here things can only get better ...
 
Heh, I'm not so sure about that...
 
6:12 PM
I'm an optimist ...
 
Didn't know you were an eye doctor
 
laugh track
 
+bluefeet THAT'S AN OPTOMETRIST, MORON.
Oh I mean @bluefeet, sorry, I thought this was YouTube.
 
Sure, OK. I can be a boat as well. Whatever.
 
I guess Michigan just recently legalized mass self-driving car tests on public roads.
 
6:21 PM
Is that the same car Uber was using in SF or LA?
 
Not sure. Were they using Bolts?
Ah they were using Volvos.
 
@JasonC you guys are too cranky.
 
I'm working on a web application project, what do you expect?
 
browser glitches?
 
Lol, probably.
I'm putting all my faith in Bootstrap here.
 
6:31 PM
So you can blame Bootstrap?
 
@rene Hey, you use git a lot, yeah? Is doing development on the master branch then occasionally merging it into a release branch a reasonable strategy?
Well, because I have no idea what I'm doing, and Bootstrap buttons look nice.
 
@JasonC Yeah, that would mimic SVN un-stable trunk
with a small team that should work
 
Ok, and then, if I have a production system, that's always on release, as long as the local repo is on release, I should be able to just git pull origin release to update it with no other git weirdness?
It took me a while to figure out how to merge master into release on the command line, but the "network" graph on gitlab didn't turn out looking how I expected it to look, although it seems to work.
Yeah it's a super small team.
 
@JasonC yeah, that should be the expected thing to happen.
 
Cool thanks.
I really need to rethink my workflow here, I'm trying to come up with this software auto-update system, for these Pi-based devices with no real UI, that involves the least amount of work on my part and it's a nightmare. It works great on the user side. But on the dev side it's gonna get awful real fast.
I think I got myself into a bit of a haste-makes-waste situation.
 
6:47 PM
@JasonC do those Pi's do git pulls as well?
 
6:58 PM
@rene No.
I've got an update server, it does git pulls from release.
When it pulls an update it runs a little build/sanitize script to remove source that shouldn't go to the Pis, .git dirs, etc. Packs it up into an archive, then serves it on request.
 
OK.
 
It also sticks a file containing the git hash of that release in the archive. The Pi's read that file to know what version of the SW they have. They use that in their update server queries so the server knows if a new update needs to be served or not.
That's about it.
But for some reason it's getting sticky on my end. I think it's because I just am not so great with git.
The other thing that's making it confusing is I've got all the code for this project in the git repo, including a bunch of stuff that doesn't affect the Pis (although they are the only thing that needs auto-updates).
So the "release" branch is actually more of a "new sw for pi" branch.
Because other stuff like the production server, if I put it in the release branch every time I update it, it triggers all the devices in the field to receive unnecessary updates, because it changes the release hash even though the pi sw itself didn't change. So it's messy.
I either need to separate the Pi stuff into its own repo, or more clearly define what my branches are for. There is no good "production update server release" branch, for example.
 
Maybe move the Pi stuff to their own repo?
 
Yeah that's kind of what I'm starting to think.
I'm also questioning if git is the most appropriate "back-end" for version management and deployment here. I feel like it's great for the job and saving me work. E.g. the hashes are great, I don't have to version the sw explicitly. But I wonder if it's making things more difficult.
And maybe instead of the update server updating from git it should just be manual deployment of Pi builds to the update server by the devs.
 
you can have two remotes on your local repo
 
7:07 PM
OR I could use explicit version numbers instead of commit hashes. But some of my devs are really lazy, with one-time short term jobs on the project, I don't trust anybody to do things properly.
I don't quite understand that concept but don't explain it, one thing on my todo list today is to sit down and go through a full git primer. I've been winging it but I'm hitting knowledge barriers now.
 
yeah, I wouldn't rely on manual process if you run the risk of updating field deployments
 
That's also the reason why I don't want to use master for releases. Some of the people on this project just won't be able to get their head around working in a git branch.
 
user315433
FR: penalize WB questions on HNQ to prevent them from overtaking the list.
 
You know I think the other problem is that I've done too much work in too short of a time period, and a lot of it is first-time-for-me stuff, and it's confusing even in my head. I think the system might not actually be so bad. But I really need to take a step back, document everything properly, and then re-assess. I'm in too deep right now.
 
user315433
> You may say I'm paid to copy and paste code from Stack Overflow. I'm actually paid to know WHAT code to copy and paste from Stack Overflow. -- Dandy Alexander at 10:57 AM - 16 Dec 2016
 
user315433
7:18 PM
 
user315433
I'm in a good company in SO league...
 
user315433
Wonder what's the tiebreaker there, is it overall rep or username?
 
@zaq I think the primary key, id on the user table.
 
user315433
The blog post Accept Your Own Answers linkes to an earlier one, Why Can't I Accept My Own Answer? but the link is 404.
 
7:42 PM
@zaq created an PR
 
user315433
Thanks. There's also a broken Uservoice link there ("not alone" in the first line), but I doubt anything can be done about that.
 
Nope, The wayback machine doesn't have it either
this is the post from Shog9 to close UserVoice: meta.stackexchange.com/a/127898/158100
Unless there was some migration I guess that content no longer is present.
 
8:24 PM
@rene mine's been sitting for over a year now :(
o_0 all the colleges are sending their template emails today for some reason.
Hello, bjb568, you should totally finish the common app for $SCHOOLNAME.
 
@bjb568 you got some attenion ... but the designers might disagree. I have yet to meet one that easy accepts any critique ...
 
But they said nothing. Bah.
 
user315433
@bjb568 Make sure that your essays are undergirded and infused with ethics and integrity.
 
I have no idea what that means, but alright.
 
8:52 PM
@zaq it's probably still there, somewhere, but UV's search is awful
and of course they don't really let Google index properly
(and they've changed IDs and URL schemes so many times...)
I used to try updating links now and then, but... It's a hopeless task.
 
@Shog9 maybe do then what I suggested in the comment on the PR? Just link to your meta answer about UV being closed?
for all links to UV still in the blog ...
 
maybe
I don't do anything with the blog anymore
And it's probably migrating to another platform soon, so I'm reluctant to bother with improvements
 
Another? Because one rewrite wasn't enough?
 
I used to spend a ton of time tweaking formatting and links and most of that got trashed in the last migration, so unless it gets super-convenient at some point I'd rather just make sure critical info is on meta somewhere.
 
user315433
@hichris123 New marketing people want new marketing tools.
 
8:56 PM
pretty much, yeah
 
The blog is the intern project? With every new intern the blog migrates
 
which is fine; they actually use it, so they should have tools that work for 'em
meta sites work decently well for announcements and the occasional essay these days, which is all I ever used the blog for anyway
and it's sooooo much less busywork to write a meta post than it is to write a blog post
 
SE seems like it's picking up the tooling side of technical debt: too many projects, too many things, too much stuff to deal with.
 
user315433
 
it was about 30 people when I joined, and now it's about 300. You can force 300 people to standardize on tooling, but it's usually not a good idea unless they're all doing the same thing.
 
8:59 PM
I mean in more than that sense, though. Like how many blogs (and how many different blogging platforms) is SE running right now? How many projects are running concurrently?
 
user315433
At least there's still some stable output (on Internet Archive and elsewhere) and it's not porn.
 
user315433
@hichris123 They are all at stackoverflow.blog
 
@zaq Plus the WordPress ones that are still left...
 
user315433
Yeah, subtract those in near future, and you have progress: fewer platforms.
 
user315433
Well, and he mod newsleer hing.
 
9:03 PM
yeah, that's migrating to something else too
not 100% sure what yet
 
@zaq Did someone steal your ts?
 
user315433
Yay, I broke a key on my 3.5 year old Chromebook! Guess which one.
 
escape
 
user315433
Geing myself a gif for holidays.
 
't' just seems too obvious
 
user315433
9:05 PM
I suppose I can still type by having "t" in the clipboard.
 
Nov 11 '14 at 19:11, by Mooseman
user image
That seems like a good holiday gif
 
sure ... but can we keep the conversation going so it scrolls out of my viewport?
 
Aaawww ... :(
 
user315433
I have a chart with SE employee count somewhere...
 
9:09 PM
you need to watch that gif for a couple of hours to prepare for Happy Hour
 
I'll be drunk then before happy hour starts
 
that's the idea
 
I can do that
 
user315433
 
user315433
9:12 PM
Joel says it's 300 now, but 290 on the company page.
 
that time period is not long enough
 
@Shog9 oh, that is nice.
 
user315433
:( I should have begun scraping that page earlier.
 
user315433
Maybe I don't need a new Chromebook. As long as only one key cap is missing, I know what it is.
 
user315433
Google search brought up www.replacementlaptopkeys.com as the top result. Spammers have succeeded... :(
 
9:35 PM
slightly inaccurate graph based on account records:
 
user315433
Big Careers expansion in 2013...
 
user315433
There was a post announcing 10 new employees, and soon after that they were no longer announced
 
user315433
Stack Overflow en español announced community ads, but they are still in beta... what gives?
 
user315433
It's not clear what exactly graduation means for lang-so sites, but I supposed ads were a part of that.
 
9:53 PM
Ok, so I figured out that my class rank (which is on my transcript but which I can't easily see) is bad. What's important is that I'm not in the top 10% of my class, which basically disqualifies me from getting admitted anywhere respectable. The reason for this is
1) I will only have had 3 years of high school (2 that are shown and put in the class rank) and my freshman year grades were bad (explained by an essay).
2) Class rank was probably calculated before I sent my high school the college transcript, so dual enrollment classes aren't weighted.
This is probably a huge factor in getting rejected from MIT. 98% of accepts are in the top 10%.
So should I send them an email describing circumstances and ask for advice on applying to other schools which would have the same problem?
 
user315433
They are admissions, not admission counselors... I'd ask at your school.
 
Ha, my school won't do anything.
The admission people are pretty nice. Some of them call themselves admission counselors.
I think there's also a different email to contact students who are sorta in charge of some part of the process.
 
user315433
Anyway, if your class rank is bad, and is important for admissions (which it is), the way to go would be to improve the rank.
 
Well, duh. But class rank is calculated not including this year.
I'm probably just dead'd.
 
user315433
10:08 PM
It's weird the school won't just tell you what your class rank is, given that they report it to universities about you.
 
user315433
You should be able to know whatever it is that they tell others.
 
user315433
10:20 PM
!!/hats
 
@zaq WE LOVE HATS! Winter Bash will begin in 2 days, 1 hour, 39 minutes, and 30 seconds.
 
Hey, @Shog9, quick question - are moderators able to make questions with score < -8 show up on the front page of a meta site?
 
user315433
No, this is automatic.
 
@E.P. sure. They can upvote them to -7.
 
user315433
Only if the score is -8 exactly.
 
10:25 PM
or they have friends with votes
 
@Shog9 thing is, the downvotes express disagreement, but just because we disagree with it doesn't mean that it should be hidden from view
 
hang on, I wrote about this somewhere
 
I've never seen the case for the hiding on meta, but I would hope for a case-by-case mechanism to turn it off when the conversation is important for the site even if the score is very low
 
user315433
34
Q: Meta sites' front pages should not hide questions with very low scores

kennytmOn Meta SO, downvoting merely means "disagree." This convention carried over to per-site metas. But questions with very low scores — e.g. this question on Meta Math SE with net score of -5 — are automatically hidden from the front page. The fact that a valid discussion seems to be deleted causes ...

 
@zaq yeah, and I agree with Steven Jeuris's answer
the accepted-answer-status-completed status is bogus on that one, the original request was watered way down
 
user315433
10:28 PM
I think the current system is fine, there is an infinite supply of stupid ideas that hiding pushes out of the way.
 
user315433
A mod can make a post featured regardless of the score, which brings way more visibility.
 
user315433
(If the discussion is really that important)
 
I can't find it. Anyway, the gist was that at some point you're just pillorying a question by keeping it visible, and that doesn't help anyone.
 
@Shog9 it's certainly pillorying if you make it featured
 
so, who did that?
 
10:30 PM
(as in zaq's suggestion)
but hiding it smells of censorship as much as keeping it looks like pillorying
 
right. So funny story, I'm pretty sure that didn't work for a long long time, until we had some company announcement that got massively downvoted while featured.
 
it's a case by case thing at best
 
so, you want stuff not-censored, don't downvote it
 
@Shog9 since when does "downvote" on meta mean "censor"?
 
answer, "Hey, buddy, I think this is a terrible terrible idea but I want everyone on the site to see it so I'm gonna upvote it instead. But just so we're clear, this post is worse than Hitler"
 
user315433
10:31 PM
@E.P. Hiding it automatically is not censorship. But you want a mod to personally decide which of -8 posts are shown and which are not... Step 2: accusations that mods decide on this unfairly.
 
it means "I disagree", not "I disagree, and I think you should shut up and no one should ever be able to listen to you"
 
The flip side of that is... There's only n slots on the home page.
 
@Shog9 yeah, but this is on a pretty slow meta site
 
user315433
The few who want to see heavily downvoted posts know how to find them.
 
If you consider not showing something there to be censorship, then we're censoring the vast majority of questions on most metas.
 
10:33 PM
it's not the main front page, that'd be silly
 
So, we censor. Ideally, we censor the stuff that everyone hates and will never be implemented
But, I guess we could hide the good ideas randomly instead
 
@Shog9 No. This question just had an answer, but you couldn't tell it from watching the meta front page
 
so?
 
@Shog9 so, the answer is an important part of the conversation we need to have
it's not a pleasant conversation
but it's good for the site
and it doesn't help anyone if it just drops out of everyone's radar
 
Then upvote it
 
user315433
10:35 PM
Won't help, I just downvoted the question further.
 
@Shog9 I am happy to upvote the answer
 
but not the question
why not?
you think it's an important conversation
but you won't upvote it?
 
@Shog9 because I disagree with everything in the question
 
user315433
This discussion may be important, but I don't want it to be framed by that question by that user.
 
yeah...
 
user315433
10:36 PM
So, -1 to the question, and I hope you'll have that discussion elsewhere.
 
user315433
Ideally, you want to end up with a thread you will be using for reference in the future. That one isn't it.
 
@zaq how would that look like?
 
About six years ago, we had a user who showed up on meta and managed to get a bunch of folks mad at him
 
that I'd really like to know
 
I wrote, of his participation at the time...
> He is obviously trolling, and has been since he first showed up here.

What's funny is that he's doing so by doggedly supporting the same ideas that some of his most active enemies supported in the past: SO as a social network / MMORPG, down-votes as unacceptably negative, etc.

I don't really care. I'm going to down-vote and argue against the same lousy ideas regardless of who proposes them. If anything, his attitude makes this easier...
sometimes, the worst thing you can do for an idea is to let the folks who agree with it support it.
 
10:38 PM
@Shog9 for clarity, I'm talking about this thread.
 
Also, @E.P., read this:
179
A: Can we talk about the voting culture here on Meta?

Shog9So I've been kinda chewing on this since your email, because... It's one of those simple questions that defies a simple answer. You're not the only person to raise this concern, of course: several respected users, moderators and even co-workers have expressed dismay upon seeing reasonable questio...

@E.P. ...why am I not surprised...
 
@Shog9 indeed.
 
user315433
@E.P. You think there is an important issue to discuss calmly, so put up a post that is conducive to that. Preferably not this week or next week, there is no rush.
 
@zaq Frankly, the discussion this time has been as calm and reasoned as I could reasonably hope for
 
@E.P. then bookmark it so you can close other questions as duplicates of it when they're less calm and reasoned.
 
10:41 PM
@Shog9 That I can do, but it doesn't help the problem that it's still hidden from view when I think it shouldn't.
Anyways, arguing here won't do much for anyone.
I just wanted to know if the software mechanism exists or not.
 
@E.P. on the one hand, I kinda get your point
on the other, Physics has enough drama, as do most sites.
Folks who want to get in on it can find it
but I've gotten some really tired emails from folks in the past who just wanted to talk about physics, and were tired of having to argue about Physics instead
 
@Shog9 and I'm really tired of having the same old discussion in comments and in chat about how some answers are really wrong, and of having those discussions degenerate to shouting matches.
 
Just downvote and move on
 
@Shog9 I don't think ignoring this problem solves it.
 
user315433
 
10:48 PM
yeah, you and Shog9 are wrong in this instance. I'm going to stop arguing now.
 
11:12 PM
See, this is what happens when you don't have a jumping fox on-screen
 
user315433
11:33 PM
> Every time someone on a stack overflow-based forum posts a Let Me Google That For You link I wanna just slap the dogshit outta them -- Bloody Honey at 3:17 PM - 16 Dec 2016
 
user315433
Is this still a thing?
 
user315433
Is there an auto-nuking regex for lmgtfy?
 
user315433
Time for BigQuery, I guess
 
user315433
Heh, a lmgtfy comment on an answer that is itself a link to Google, to a question that is OT/rec.
 
user315433
11:40 PM
But no autonuke. :(
 
user315433
BigQuery doesn't have magically linkifying columns...
 
user315433
Anyway, there are only 49 comments with lmgtfy.com in them, so I don't know what they guy was talking about. Unless it was another SE site, because the ban isn't network wide for some reason (??)
 
user315433
For example, this is 2015:
 

« first day (2326 days earlier)      last day (2686 days later) »