00:00:51 -!- Bike has joined.
00:01:22 <Bike> they did an interview with the tsareav's dad
00:01:36 <Bike> said they were stupid bastards
00:01:59 <olsner> the people interviewing or his sons?
00:03:15 -!- Bike_ has joined.
00:03:48 <Bike_> the dad said whoever did the bombings were bastards
00:03:54 <kmc> i heard they interviewed their uncle
00:03:59 <Bike_> their uncle went on a little rant about it and how chechens weren't like that
00:04:01 <kmc> "why do you think they did this?" "because they're losers"
00:04:11 -!- Bike has quit (Disconnected by services).
00:04:12 -!- Bike_ has changed nick to Bike.
00:04:32 <Bike> cutting to the core of the terrorist mythos
00:09:28 -!- benuphoenix has changed nick to benuphoenix|away.
00:10:54 <kmc> https://tahoe-lafs.org/pipermail/tahoe-dev/2013-April/008192.htmlnice
00:10:58 <kmc> https://tahoe-lafs.org/pipermail/tahoe-dev/2013-April/008192.html nice
00:11:57 -!- Bike_ has joined.
00:12:30 <elliott> that research project is cool
00:13:00 -!- Bike has quit (Ping timeout: 252 seconds).
00:14:46 -!- Nisstyre-laptop has quit (Quit: Leaving).
00:14:49 -!- Bike_ has changed nick to Bike.
00:15:02 <Sgeo> So, I guess there is a very good reason to verify that incoming data meets constraints that only happen to exist because the data won't contain weird things... if a service not under our control is buggy and would respond badly to weirdness that it should respond sanely to
00:15:36 <Sgeo> Trusting external services to be competent at security is a bad idea
00:18:38 <elliott> i like slept through _why's day
00:18:47 <elliott> http://whytheluckystiff.net/ ok
00:19:13 <Bike> <img src="Moving-picture-of-the-words-get-psyched.gif" />
00:19:50 <elliott> http://www.scribd.com/doc/136875051/-why-s-complete-printer-spool-as-one-book
00:26:20 <nooodl> elliott: holy shit i can't wait to read all of this
00:26:26 <nooodl> it's so _why!!! at a glance
00:28:39 <Sgeo> I should stop talking about work
00:29:18 <Fiora> kmc: that sounds really cool, is there like, a link to the presentation?
00:35:25 -!- carado has joined.
00:35:31 -!- carado_ has quit (Ping timeout: 246 seconds).
00:37:07 <kmc> not that i have
00:46:27 <Sgeo> Is OWASP well-liked? Are any of its ideas flawed?
00:47:46 <Sgeo> Suspect in custody.
00:48:34 <kmc> Sgeo: i've found their wiki to be useful; it does vary a lot in quality though
00:49:18 -!- Phantom_Hoover has quit (Remote host closed the connection).
00:52:46 <Sgeo> "Common web application platforms such as Java Struts/J2EE, Ruby on Rails, and PHP can theoretically prevent developers from introducing most classes of vulnerability in the first place. However, the current state of the framework industry is more driven by features than by security; any conflict between the two is usually decided in favor of adding features and ease of use, as opposed to difficult-to-use security enhancements. Some framework
00:52:46 <Sgeo> s even have built-in vulnerabilities out of the box!"
00:52:56 <Sgeo> I am beginning to hate everyone
00:53:05 <Sgeo> Well, no, I was much angrier this morning
00:53:30 <Bike> wait why are you hating people
00:55:43 <pikhq_> It's generally good policy.
00:56:35 <Sgeo> Because an external widely-used-by-lots-of-organizations competent-seeming web service has a blatant security issue
00:57:13 <Lumpio-> PHP prevents people from making security holes
01:04:54 -!- Nisstyre-laptop has joined.
01:19:16 <kmc> hi shachaf
01:21:22 -!- carado has quit (Ping timeout: 246 seconds).
01:21:32 <shachaf> WE DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, WARRANTIES OF TITLE, NON-INFRINGEMENT, MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, QUALITY, ACCURACY, TIMELINESS, COMPLETENESS, QUIET ENJOYMENT, NON-ENCUMBRANCES, NO LIENS, AND SYSTEM INTEGRATION
01:21:44 <shachaf> I really wanted a warrany of quiet enjoyment. :-(
01:23:11 -!- mnoqy has joined.
01:23:46 -!- mnoqy has changed nick to monqy.
01:24:01 -!- monqy has changed nick to mnoqy.
01:24:29 <shachaf> did monqy have any good messages
01:24:52 <shachaf> maybe next time, mnoqy. maybe next time
01:25:01 <olsner> maybe not next time either
01:25:14 <shachaf> maybe not next time, olsner. maybe not next time
01:25:57 <olsner> maybe never, shachaf. maybe never
01:26:47 -!- Nisstyre-laptop has quit (Quit: Leaving).
01:29:55 <Bike> huh apparently that's a real thing
01:30:19 <Bike> covenant of quiet enjoyment
01:33:48 -!- WeThePeople has joined.
01:40:07 -!- Bike has quit (Ping timeout: 264 seconds).
01:41:36 -!- Bike has joined.
02:01:06 <Sgeo> https://www.owasp.org/index.php/Don%E2%80%99t_trust_services
02:05:55 -!- nooodl has quit (Ping timeout: 258 seconds).
02:36:44 -!- doesthiswork has quit (Ping timeout: 256 seconds).
02:46:46 -!- WeThePeople has quit (Quit: Leaving).
02:50:48 -!- heroux has quit (Ping timeout: 276 seconds).
02:51:27 <Sgeo> https://www.owasp.org/index.php/How_to_write_insecure_code
02:51:52 -!- heroux has joined.
03:15:12 -!- sirdancealo2 has quit (*.net *.split).
03:15:12 -!- mroman_ has quit (*.net *.split).
03:18:00 -!- Nisstyre-laptop has joined.
03:20:12 -!- sirdancealo2 has joined.
03:20:12 -!- mroman_ has joined.
03:49:20 <shachaf> kmc: are you going to become a cryptographer
04:17:25 <shachaf> kmc: Did you know C++ supports try-catch for constructors?
04:18:00 <shachaf> Foo() try : x(...) { ... } catch(...) { ... }
04:25:26 <Jafet> Did you know you can use a try-block anywhere
04:25:38 <Jafet> It doesn't have to be a constructor
04:26:10 <shachaf> The usual try syntax doesn't involve an initializer list.
04:26:13 <Jafet> void f() try {} catch {}
04:27:37 <Jafet> Sadly this isn't legal??? []()try{}catch(...){}()
04:28:00 <Jafet> Second class citizens
04:34:25 <kmc> shachaf: wow
04:34:27 <kmc> did not know
04:37:58 <shachaf> Jafet: do you know everything about c++
04:38:30 <Jafet> Nobody knows everything about C++. I'm a nobody.
04:40:39 <coppro> function-try-blocks are pointless on regular functions
04:40:52 <coppro> but serve a distinct (if questionable) purpose for constructors
05:12:12 -!- heroux has quit (Ping timeout: 258 seconds).
05:14:05 -!- heroux has joined.
05:20:58 -!- Bike_ has joined.
05:21:11 -!- FreeFull_ has joined.
05:25:28 -!- Bike has quit (*.net *.split).
05:25:29 -!- FreeFull has quit (*.net *.split).
05:25:29 -!- hogeyui has quit (*.net *.split).
05:34:03 -!- heroux has quit (Ping timeout: 258 seconds).
05:35:07 -!- hogeyui has joined.
05:35:32 -!- heroux has joined.
06:14:09 -!- Sgeo has quit (Remote host closed the connection).
06:19:40 -!- Sgeo has joined.
06:31:49 -!- doesthiswork has joined.
06:33:38 -!- ogrom has joined.
07:12:25 <oklofok> can you be reached in http://imgur.com/W8gUtNi?
07:13:19 -!- Bike_ has changed nick to Bike.
07:15:43 <oklofok> "<FireFly> I think what that game is missing is that it isn't taking place on a torus" one thing i'd like to do is to have a game where the play area is some fucked up manifold that changes its global form
07:15:50 <oklofok> but i dunno how that should work
07:16:55 -!- c00kiemon5ter has joined.
07:40:22 -!- mnoqy has quit (Quit: hello).
07:43:59 <oklofok> fizzie: it seems i have about 150000 points. i think i have to make some changes to the game.
07:45:52 -!- c00kiemon5ter has left.
08:02:34 -!- Bike has quit (Quit: leaving).
08:25:13 -!- Zerker has joined.
08:37:42 -!- ThatOtherPerson has joined.
08:55:12 -!- c00kiemon5ter has joined.
09:02:53 -!- epicmonkey has joined.
09:14:46 -!- FreeFull_ has changed nick to FreeFull.
09:16:15 -!- ais523 has joined.
09:17:17 <ais523> I think I hit a new record for how undebuggable a bug can be
09:17:24 <ais523> basically, I was bounds-checking against the wrong array
09:17:33 <ais523> and thus any attempt to try values near the start of the array worked
09:21:19 -!- ogrom has quit (Quit: Left).
09:22:39 <ais523> it was eventually found via the time-honoured method of staring at the code without running it
09:23:19 <Zerker> As opposed to bisection search?
09:26:22 -!- carado has joined.
09:29:39 <ais523> Zerker: on what? the array?
09:29:58 <ais523> if I'd thought "perhaps the first half of the array is being treated differently" I'd have found the bug much faster
09:30:04 <ais523> but it's not the sort of thing I tend to assume
09:30:32 <ais523> that'd involve trying to work out what the correct values were halfway through the code
09:31:35 <Zerker> Which is nontrivial even for one specific set of input values that has been proved not to work?
09:32:04 <Jafet> http://en.wikipedia.org/wiki/List_of_sets_of_four_countries_that_border_one_another
09:32:17 <Jafet> I think most of my programs have been written for at most one input
09:34:05 <ThatOtherPerson> My favorite debugging method is "hit random buttons until it works"
09:34:58 <fizzie> oklofok: I have 184319 points.
09:34:58 <FreeFull> With code I write, I usually know what the bug is as soon as I see it
09:35:23 <fizzie> I think imma close the tab now.
09:37:11 <ais523> afk for a length of time that will likely be between half an hour and an hour
09:37:59 -!- Phantom_Hoover has joined.
09:43:09 -!- carado_ has joined.
09:50:40 -!- doesthiswork has quit (Quit: Leaving.).
09:52:38 -!- Zerker has quit (Quit: Colloquy for iPad - Timeout (10 minutes)).
09:54:51 -!- Zerker has joined.
09:59:22 -!- ThatOtherPerson has quit (Quit: Page closed).
10:07:02 -!- nooodl has joined.
10:39:33 -!- xitrix has joined.
10:50:01 <ais523> a bit more than an hour, it seems
10:56:23 -!- Zerker has quit (Quit: Colloquy for iPad - Timeout (10 minutes)).
10:57:01 -!- Zerker has joined.
11:35:20 -!- nooodl has quit (Ping timeout: 256 seconds).
11:39:02 -!- carado has quit (Quit: Leaving).
12:06:29 -!- Zerker has quit (Quit: Colloquy for iPad - Timeout (10 minutes)).
12:15:32 -!- nooodl has joined.
12:18:20 -!- sebbu has quit (Read error: Connection reset by peer).
12:18:47 -!- sebbu has joined.
12:19:25 -!- sebbu has quit (Changing host).
12:19:26 -!- sebbu has joined.
12:29:21 -!- Koen_ has joined.
12:42:25 -!- btiffin has joined.
12:46:41 <btiffin> Thinking about a superhero eso. Should it be rigid, datatypes by hero, and action verbs by superpower? Or loosey goosey, code mixed as appropriate? Could be fun, detailing the Spiderman datatype, errr, non-trademarked-human-with-spider-like-powers (tm) linklist type
12:48:22 <Phantom_Hoover> suggn. drop the idea if it's just "a superhero esolang"
12:48:38 <Phantom_Hoover> themed languages rarely turn out well except as one-off jokes
12:51:25 <btiffin> Sound advice. Accepted. :-)
12:55:22 -!- Taneb has joined.
12:57:00 <btiffin> Ahh, but my first, cbrain, is. That's what I get for being new ;-)
12:58:14 <Taneb> We all know whose fault that is
12:59:02 <Taneb> The blame lies with my maths teachers, of course
12:59:06 <Taneb> I have homework to do
12:59:25 <ais523> btiffin: thematic languages are OK as long as they tie the theme to the semantics somehow
13:02:46 <Taneb> What's that blog that's the best for linking people who say web dev sucks because PHP
13:03:44 <Taneb> I think this is either a kmc or an elliott question
13:04:28 <btiffin> ais523: I predict superhero the language will remain an idea, pondered and not much else, as per Phantom_Hoover's reality check.
13:30:55 <Taneb> ...why the hell does str_rot13() exist in the PHP standard library
13:31:06 <Taneb> Does that even do what it looks like it does
13:33:48 -!- copumpkin has quit (Ping timeout: 252 seconds).
13:34:24 -!- copumpkin has joined.
13:34:44 -!- Phantom_Hoover has quit (Ping timeout: 258 seconds).
13:39:23 <fizzie> "[str_rot13] Changelog: Version 4.3.0: The behaviour of this function was fixed. Before this fix, the str was also modified, as if it was passed by reference."
13:39:27 <fizzie> Nice that it had a bug too.
13:41:00 -!- ThatOtherPerson has joined.
13:42:02 <fizzie> But, I mean, the standard library has things like "hebrev — Convert logical Hebrew text to visual text" and "convert_cyr_string — Convert from one Cyrillic character set to another". One should not be surprised.
13:43:57 <fizzie> (From what I can tell, convert_cyr_string is like mb_convert_encoding except it has one-character special names for the few encodings -- k for koi8-r, w for windows-1251, i for iso-8859-5, a or d for x-cp866 and m for x-mac-cyrillic -- it supports.)
13:58:42 -!- nooodl has quit (Ping timeout: 256 seconds).
14:03:14 -!- ion has quit (Quit: Thanks, btrfs. BUG: unable to handle kernel NULL pointer dereference at 00000008).
14:10:11 -!- ion has joined.
14:20:58 -!- ThatOtherPerson has quit (Quit: Leaving).
14:22:17 -!- Zerker has joined.
14:30:23 -!- ion has quit (Quit: btrfs kernel code dereferenced NULL again. Rebooting to recover.).
14:32:50 -!- Zerker has quit (Quit: Colloquy for iPad - Timeout (10 minutes)).
14:33:12 <Taneb> I remember when I was little and lived in Australia there was a cartoon I used to watch that had a koala who used to say "Extra-ordinary!"
14:34:38 <Taneb> Blinky Bill, a googling reveals
14:36:29 -!- epicmonkey has quit (Ping timeout: 255 seconds).
14:37:04 -!- ion has joined.
14:42:17 -!- Zerker has joined.
14:46:46 <Sgeo> I did play with REBOL a little, but not much
14:54:15 -!- Zerker has quit (Quit: Colloquy for iPad - Timeout (10 minutes)).
14:58:56 -!- ThatOtherPerson has joined.
15:17:27 -!- nooodl has joined.
15:20:31 <Taneb> In my head I just started comparing PHP to the English language
15:24:08 <Sgeo> That... is the most awesome comparison
15:25:22 <Taneb> It's a mess that people claim is easy but really isn't. It's a god-awful mix of half a dozen other languages, combining features from each that weren't meant to be mixed
15:25:41 <Taneb> It is inconsistent and has dozens of stupid, arbitrary rules
15:28:37 <Taneb> But it's still really commonplace, it's the de facto language because it is the de facto language
15:34:35 <TeruFSX> does english have a bizarre unusual name for some common feature
15:35:04 <TeruFSX> if so it is a perfect comparison
15:39:27 -!- WeThePeople has joined.
15:45:28 <mroman_> "English is a god-awful mix of half a dozen other languages, combining features from each that weren't meant to be mixed and it's inconsistent especially when it comes to spelling.
15:45:36 <mroman_> Nevertheless, everybody uses it because they got nukes."
15:46:05 -!- Phantom_Hoover has joined.
15:46:08 <Taneb> mroman_, I think everyone started using it when the two most powerful countries were the British Empire and the US
15:46:09 <mroman_> where they = native tongues.
15:46:38 <mroman_> it's ruining other languages all over the world as well.
15:47:22 <Taneb> Phantom_Hoover, I started it, so it really is
15:47:29 <mroman_> this conservation smells bad.
15:47:59 <Taneb> Basically, English and PHP are kinda similar
15:48:15 <mroman_> They suck but everybody has to use it.
15:48:43 <mroman_> It's probably already to late to push esperanto
15:49:09 <mroman_> If NK really had nukes, they could push it
15:49:16 <mroman_> not sure if they want to though.
15:49:21 <Taneb> Esperanto wasn't that great if you weren't from western Europe, I seem to recal
15:49:38 <olsner> NK would use nukes to force the rest of the world to learn esperanto?
15:50:19 <olsner> Phantom_Hoover: no idea
15:51:01 <mroman_> Taneb: But asian languages probably all are very different from european languages.
15:52:26 <mroman_> Phantom_Hoover: Isn't it "Hey look at my nukes, can I haz free candy"?
15:56:07 <Taneb> I'd advocate lojban, but I don't really want to.
15:56:58 <mroman_> Some language based on logic.
15:58:48 <mroman_> I don't get it why they even wasted time invening lojban.
15:58:58 <mroman_> all previous languages failed
15:59:17 <mroman_> why would they have thought that lojban wouldn't fail too
16:01:07 <mroman_> you can copyright a language.
16:03:18 <mroman_> And I highly doubt that laws will be written in Lojban.
16:03:43 <mroman_> They actually want the laws to be interpretable to weird extents.
16:04:47 <Taneb> ThatOtherPerson, how naive
16:05:02 <mroman_> The judge is allowed to reinterpret the law as he pleases
16:05:33 <mroman_> which means that nobody really knows what's lawful and what not.
16:06:00 <mroman_> which is why consulting a lawyer is usually useless
16:06:15 <Sgeo> iirc, if you take two Lojban words and combine them in a certain way, you need to also know what that combined word means, not just the individual parts
16:06:18 <mroman_> because he can't tell you what you want him to tell you
16:06:45 <mroman_> there's a 50% chance he tells you "well, that depends on how the judge interprets the law"
16:07:43 <mroman_> there's no jury in switzerland.
16:08:37 <Phantom_Hoover> i mean it's basically the same here except the jury does the interpreting
16:08:38 <mroman_> e.g when making a contract
16:08:52 <mroman_> there is a "template" in the law for contracts
16:09:05 <mroman_> which states what is what when not otherwise defined in the contract
16:09:11 <Phantom_Hoover> but anyone saying 'lawyers are useless, don't bother with them' makes me think they're running their mouth off
16:09:21 <mroman_> some of the things in that "template" can be overwritten in the contract
16:09:23 <ThatOtherPerson> Really, what good would a programming language where anything you say is "up to interpretation" be?
16:09:34 <mroman_> And you would think the law tells you which can be overwritten or not
16:09:45 <ThatOtherPerson> (at which point someone will probably give me a link to a language like that -_-)
16:09:46 <Phantom_Hoover> "everything should basically work like computers, it'd be perfect"
16:09:49 <mroman_> the judge decides that as he pleases.
16:10:11 <mroman_> Phantom_Hoover: They are not useless.
16:10:18 <mroman_> but they can't answer all the questions.
16:10:22 -!- carado_ has quit (Ping timeout: 246 seconds).
16:10:31 <mroman_> You can't tell them a scenario and expect them to tell you whether it is legal or not.
16:10:31 <Fiora> ThatOtherPerson: I think there's a joke about C and undefined behavior here XP
16:10:41 <mroman_> because they don't know it
16:11:33 <mroman_> There are absolutely no objective criterias for when something falls under copyright or not
16:11:53 <mroman_> you effectively don't know if something falls under copyright law or not UNTIL a judge said "yes" or "no".
16:12:00 -!- carado_ has joined.
16:12:17 <Phantom_Hoover> you can still guess at how the judge will interpret it
16:12:28 <Phantom_Hoover> and i daresay for most matters a lawyer knows better than you
16:12:28 <mroman_> But guessing is not knowing.
16:12:48 <ThatOtherPerson> Really, it reminds me of how a professor told Isaac Asimov "Just because you wrote it, what makes you think you have the slightest idea what it's about?"
16:12:51 <mroman_> The objective of your laywer is to get the judge to interpret the law in a certain way.
16:13:02 <mroman_> but you can't guarentee that.
16:13:11 <ThatOtherPerson> Now I can't get the picture of the Python interpreter saying that to someone out of my mind
16:14:36 <mroman_> It's only murder if you brutally kill someone.
16:14:43 <elliott> what the fuck is everyone talking about and why...
16:14:46 <mroman_> but what "brutally kill someone" means depends on the judge ;)
16:15:41 <mroman_> If you kill someone by accident it's not murder.
16:16:14 <mroman_> it's not enough @intentional killing
16:16:36 <ThatOtherPerson> elliott: We're all having separate monologues that are only slightly related
16:16:49 -!- ais523 has quit.
16:17:31 <mroman_> it's the judge who says if you killed someone intentional or not.
16:18:51 <mroman_> murder has to be intentional AND extreme
16:19:15 <mroman_> I mean, I'm pretty sure it's the same in the US
16:19:20 <mroman_> except the jury has more say?
16:19:27 <mroman_> If the law were clear you would not need a jury.
16:19:49 <elliott> are you sure there's not some kind of language barrier here... like if i make a detailed plan to kill someone and then go up to them and poison them or whatever i am going to get arrested for murder
16:19:55 <elliott> but none of that is very brutal
16:20:00 <mroman_> So your lawyer has to persuade the jury to make a certain decision?
16:20:06 <ThatOtherPerson> mroman_: the jury isn't so much for interpreting the law as for saying whether or not someone actually did the crime
16:20:22 <mroman_> Thet it's something else :)
16:20:30 <elliott> "If the law were clear you would not need a jury." <-- is this meant to be an argument against interpretation? surely you can see the disadvantages of a totally unambiguous law that covers every single case and is rigorously adhered to
16:20:42 -!- Bike has joined.
16:21:03 <mroman_> elliott: I can see the disadvantages, yes.
16:21:33 <mroman_> the ambigous law version has the disadvantage that you can't make a 100% statement whether something is legal or not.
16:21:46 <mroman_> until a judge decided on legal or not
16:21:58 <mroman_> in which case other judge later will usually refer to the previous ruling.
16:22:12 <mroman_> but they don't have to, of course.
16:22:54 <mroman_> As a citizen I want to know if somethin is legal or not
16:24:43 <mroman_> but it turns out that you don't have a chance in most cases
16:24:58 <mroman_> even if you consult a lawyer in advance.
16:25:13 <mroman_> except for cristal clear cases
16:25:27 <mroman_> Like: "Do I have to pay my taxes?"
16:27:10 <mroman_> It was about if you meet your friends to do sports
16:27:24 <mroman_> and then some kids gather and play with you
16:27:39 <mroman_> are you liable for what happens to those kids?
16:27:49 <mroman_> in terms of "parental watch"
16:27:54 <mroman_> or whatever that term is in english.
16:28:20 <mroman_> that's actually a simple question which occurs daily.
16:28:28 <mroman_> but he could not give me a "yes" or "no" answer.
16:28:57 <mroman_> and that's the day I completely lost trust in the legal system.
16:29:27 <mroman_> esentially every day you live in a legal gray area and you have no clue what you have to do and what not.
16:29:46 <Bike> is this a poem
16:30:17 <mroman_> because I'm very frustrated about the legal system :)
16:30:36 <mroman_> I'm supposed to follow the law
16:30:50 <mroman_> but the law can't tell me for sure if I'm following it or not.
16:30:56 <coppro> I SUPPORT THE FOLLOWING: ---
16:31:19 <Bike> --- is pretty rad
16:31:25 <mroman_> That feels like the state is mocking me.
16:31:44 <Bike> the state is mocking you, personally
16:31:52 <Bike> and snickering with its buds
16:32:02 <mroman_> and doesn't what me to know so he can beat me up whenever he wants to.
16:32:45 <Phantom_Hoover> the state is generally immensely concerned with persecuting random programmers
16:32:55 <elliott> let's face it you have 0 risk of actually getting beaten up by the state so why do you care
16:33:06 <mroman_> elliott: The risk of that is > 0.
16:33:19 <Phantom_Hoover> and the legal system is set up this way for that express purpose, not because it's actually quite a sensible way of building it
16:33:42 <mroman_> If there's a risk I must be afraid everyday of it happening.
16:34:08 <elliott> right now there's a risk i will get so sick of this that i destroy the internet entirely
16:34:18 <mroman_> But I don't care about the internet.
16:34:24 <mroman_> So I don't worry about that.
16:34:50 <Taneb> mroman_ fought the law and the law won
16:35:08 <elliott> how many kinds of drunk or high are you anyway
16:35:08 -!- kmc has set topic: Bible camp's cancelled. | | I am making the world better by CONFUSING THE TIME CUBES | Underhanded C Contest: http://underhanded.xcott.com/?page_id=5 | http://codu.org/logs/_esoteric/.
16:35:22 <Bike> a blank topic thing? what is this
16:35:39 -!- kmc has set topic: Bible camp's cancelled. | I am making the world better by CONFUSING THE TIME CUBES | Underhanded C Contest: http://underhanded.xcott.com/?page_id=5 | http://codu.org/logs/_esoteric/.
16:35:52 <mroman_> my body doesn't process alcohol.
16:35:55 <Bike> i am the topic lawyer
16:35:57 <kmc> Bike: there were actually 50 zero width invisible multiocular Os in that space
16:36:00 <mroman_> and I don't do drugs because they are definetely illegal.
16:36:20 <Bike> "drugs" in general aren't illegal, what the hell
16:36:27 <mroman_> If I were to posses drugs I would be afraid every second that the police will find them.
16:36:36 <mroman_> I would literally be shaking.
16:36:39 <Fiora> I have ibuprofen in my bathroom!
16:36:56 <Bike> FIORA THIS IS THE POLICE. GIVE UP NOW. WE KNOW YOU DEAL WITH HEADACHES
16:36:58 <Phantom_Hoover> mroman_ you should possibly See Someone about your crippling anxiety
16:37:10 <elliott> Fiora: arrested for crimes against livers??
16:37:22 <mroman_> But turns out they can't help you.
16:37:35 <mroman_> They can drug you, but that doesn't fix the actual problem.
16:37:59 <pikhq_> I have prescription drugs!
16:38:24 <coppro> pikhq_: yes, some presecription drugs help fuck
16:38:24 <mroman_> also they don't actually want to help you.
16:38:49 <pikhq_> coppro: 'Fraid this one has an occasional side effect of hindering fuck. Alas.
16:38:51 <Fiora> yes I have prescription drugs too @_@ I even have schedule 4 drugs
16:39:13 <Fiora> one of the least regulated on the list? <.<
16:39:19 <elliott> 4 is the wimpiest one right
16:39:27 <elliott> oh apparently there is a 5
16:39:30 <Fiora> um, I think 5 is? I'm not sure
16:39:46 <mroman_> Drugs you have to schedule daily?
16:39:48 <Fiora> categories of restricted drugs in the US
16:39:57 <elliott> cannabis is schedule 1 and cocaine is 2 or something right
16:40:01 <Fiora> http://en.wikipedia.org/wiki/Controlled_Substances_Act#Schedules_of_controlled_substances
16:40:11 <elliott> slightly badder than cocaine
16:40:31 <Fiora> e.g. codeine/cocaine/morphine/oxycodone are II
16:40:39 <Phantom_Hoover> (why are drug classificiations always school-themed? there's a conspiracy here i'll bet)
16:40:55 <Bike> haaaa why are so many of the schedule Is non-habit-forming hallucinogenics
16:40:57 <elliott> ok yes cannabis is schedule I
16:41:07 <elliott> Bike: because they're bad!!!
16:41:10 -!- DHeadshot has joined.
16:41:11 <pikhq_> Bike: Hallucinating is baaaad.
16:41:15 <Fiora> Bike: I think "schedule I" is "things with no legitimate use"
16:41:17 <mroman_> There are probably harder drugs
16:41:20 <Fiora> not specifically "it's worse"
16:41:29 <elliott> Bike: stick to safer stuff like cocaine and PCP
16:41:36 <Fiora> of course marijuana being on there is dumb but there's some logic to it
16:41:43 <Fiora> Phantom_Hoover: topical anesthetic, apparently?
16:41:47 <Bike> LSD has sometimes been used for alcoholism
16:41:52 <Bike> whatever, i guess, HIPPIES
16:42:02 <elliott> heroin is a "painkiller"!!
16:42:03 <pikhq_> Fiora: Things with no legitimate use, a high potential abuse, and a lack of any safety without medical intervention.
16:42:16 <pikhq_> Making honestly a lot of schedule I ridiculous.
16:42:22 <Fiora> yeah, it's kind of silly
16:42:55 <Bike> and it doesn't have alcohol, k
16:42:56 <pikhq_> In practice schedule I is "drugs that are 'scary'"
16:43:07 <pikhq_> Bike: Alcohol actually has a legit medical use.
16:43:34 * Fiora imagines mouthwashes as contolled substances?
16:43:36 <elliott> forgetting things; drowning sorrows
16:43:43 <pikhq_> Bike: A treatment for methanol poisoning is application of ethanol.
16:43:47 <Bike> well yeah, i mean like, drinking alcohol
16:43:57 <Bike> not rubbing alcohol n stuff
16:44:00 <pikhq_> By being drunk, you prevent the methanol from being processed in the liver.
16:44:12 <pikhq_> And so instead you piss it out.
16:44:17 <elliott> GHB, a general anaesthetic and treatment for narcolepsy-cataplexy and alcohol withdrawal with minimal side-effects
16:44:21 <mroman_> I like how I can turn a language discussion into a drug discussion.
16:44:24 <pikhq_> Well, except you are going to have a *hell* of a hangover.
16:44:41 <mroman_> I'd probably be a very bad politician.
16:44:44 <Bike> sorry fiora you're not really sick
16:44:56 <Fiora> elliott: if I had to guess a lot of the list seems to be very old and politics means it never gets updated/fixed
16:45:04 <Fiora> because then they'd have to admit they were wrong about an evil drug or something?
16:45:42 <mroman_> If it were a superpower I couldn't have it anayway.
16:45:44 <pikhq_> Fiora: Some people seem to assume it's on the list because it's evil, and getting it off the list is tantamount to legalizing rape.
16:45:45 <elliott> something about that old story of cannabis illegalisation being driven by the fact that it makes the black man™ think he's as good as the white man™
16:45:49 <fizzie> The schedulers are kind of international: http://en.wikipedia.org/wiki/Single_Convention_on_Narcotic_Drugs
16:45:56 <Fiora> GHB is also a date-rape drug.
16:45:57 <Bike> nutt sack, what
16:45:57 <fizzie> Like, our law refers to those lists in that thing too.
16:46:05 <fizzie> Schedules, not schedulers.
16:46:12 <fizzie> I suppose process schedulers are international too?
16:46:51 -!- Ekib has joined.
16:46:56 <Phantom_Hoover> Bike, the head Science Guy What Tells The Government About Drugs (the eponymous Nutt) basically said the drug policy was a load of bullshit
16:47:00 <elliott> if it treats narcolepsy wouldn't it be like the opposite of a date-rape drug... (n.b. i have no idea how narcolepsy or date-rape drugs really work)
16:47:12 <HackEgo> Ekib: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page. (For the other kind of esoterica, try #esoteric on irc.dal.net.)
16:47:14 <Fiora> elliott: maybe in small quantities or something? but modafinil is way better for narcolepsy
16:47:24 <Bike> damn for a second i thought you were going to say everything i did backwards
16:47:24 <Ekib> sdrawkcab did i gnihtyreve yas ot gniog erew uoy thguoht i dnoces a rof nmad
16:47:31 <Fiora> so it seems like it doesn't really have any uses besides assholes drugging people
16:47:33 <elliott> ok who is setting up all the bots to reverse people
16:47:37 <elliott> with reversed versions of their name
16:47:39 <Phantom_Hoover> whereupon the government demonstrated their commitment to a rational, evidence-based drugs policy founded on sound medical advice by sacking him
16:47:41 <elliott> it got old the second time
16:47:47 -!- Ekib has quit (Remote host closed the connection).
16:48:03 <elliott> precaution: make sure your name is a palindrome
16:48:25 <kmc> onlinehome-server.com sounds like a domain used to control botnets
16:48:26 <fizzie> Incrementing me by one makes me a palindrome.
16:48:26 -!- Phantom_Hoover has changed nick to Phantom_motnahP.
16:48:30 <Bike> http://en.wikipedia.org/wiki/File:Rational_scale_to_assess_the_harm_of_drugs_(mean_physical_harm_and_mean_dependence).svg haha welp
16:48:49 <kmc> didn't he get fired for that
16:49:03 <Bike> «In January 2009 he published in the Journal of Psychopharmacology an editorial ('Equasy – An overlooked addiction with implications for the current debate on drug harms') in which the risks associated with horse riding (1 serious adverse event every ~350 exposures) were compared to those of taking ecstasy (1 serious adverse event every ~10,000 exposures)»
16:49:31 <fizzie> Oh, I remember that horse thing.
16:49:53 <Bike> "He was asked to go because he cannot be both a government adviser and a campaigner against government policy. [...] As for his comments about horse riding being more dangerous than ecstasy, which you quote with such reverence, it is of course a political rather than a scientific point." hoo boy
16:50:02 <Phantom_motnahP> i think nutt also made the point that if you have a guest and you offer them a bowl of nuts and a bowl of ecstasy it's far safer for them to go for the ecstasy
16:50:14 <elliott> the primary risk of ecstasy is drinking too much water or something right, or is that a myth
16:50:21 <kmc> elliott: think that's a myth
16:50:27 <kmc> in fact that myth is probably harmful
16:50:32 <elliott> because it makes people not drink water?
16:50:32 <kmc> because way more people are hurt due to not drinking enough water
16:50:47 <elliott> can always count on kmc for my #accuratedruginformation
16:54:49 <kmc> happy 4/20 nerds
16:55:29 <Koen_> do I want to know what's special about 4/20?
16:55:55 <Bike> cannabis day hooray
16:56:13 <pikhq_> Except in Colorado, where it's healthy recreational drug day.
16:56:34 <Taneb> My favourite ice cream man got arrested for drug dealing last year
16:56:41 <elliott> Taneb: is this the ukip one
16:56:51 <fizzie> Taneb: You seem to have bad luck with ice cream.
16:56:52 <Bike> no it's the UKIP one sorry
16:56:54 <Taneb> This was the one by the high school
16:57:00 <Taneb> Not the one by the first school
16:57:03 <elliott> did he sell weed ice cream
16:57:26 <Taneb> He did sell cocaine ten pence mix ups, though
16:57:53 <kmc> more than ten pence I expect
16:58:11 <elliott> he got arrested for bankruptcy
16:58:45 <elliott> kmc: what's the price of cocaine #accuratedruginformation
16:59:06 <kmc> pikhq_: i can only imagine the crowd at CU Boulder right now
16:59:22 <Fiora> kmc: let me steal a joke a friend told me yesterday
16:59:25 <kmc> i mean even before it was legalized, this was a school where the authorities sent an official message warning students with respiratory problems to stay away from the quad
16:59:32 <Fiora> this is a special day, because it's celebrated by a well known, popular fantasy race
17:00:08 <elliott> i groaned in recognition solely because i've seen someone make that joke before
17:00:16 <kmc> <______<;[->++<]
17:00:24 <elliott> one of those love/hate days w/ the internet
17:00:36 <kmc> Phantom_monaD
17:00:37 <elliott> Phantom_motnahP: i thought you were reconciling with the elves
17:00:54 <kmc> Fiora: that's me
17:01:27 <kmc> elliott: i don't know how much cocaine costs
17:01:35 <kmc> i've never bought any cocaine or 'done' any cocaine
17:02:02 <elliott> kmc: sounds very "square", are you not "hip" to the "kids"
17:02:17 <elliott> i am "peer pressuring" you right now
17:02:28 <kmc> pier pressure
17:02:31 <Fiora> maybe with enough pressure, he might
17:02:34 <Taneb> kmc, apparently cocaine is really expensive
17:02:49 <Taneb> why didn't I think of that
17:03:14 <elliott> wow Phantom_motnahP you didn't tell me how hard it was to stop yourself revealing that your pun doesn't actually exist when you did it
17:03:16 <Bike> pier pressure is a serious phenomenon, it's hard to keep those things going in the presence of wave motion
17:03:55 <kmc> in the movie Mr T's Be Somebody or Be Somebody's Fool, there's a segment about peer pressure which takes place on a pier
17:04:11 <kmc> the protagonist is a small kid sitting on a bench
17:04:25 <kmc> and various 'no-good teenagers' come up to him and try to literally shove alcohol and cigarettes in his mouth
17:04:56 <kmc> and Mr. T stands about 4 feet away with his arms crossed shaking his head sternly
17:05:03 <Phantom_motnahP> elliott, i still really want to know how oerjan managed to get the mackerel one
17:05:04 <kmc> but doesn't actully intervene of course
17:05:42 -!- AnotherTest has joined.
17:05:47 <elliott> kmc: this actually happened to me irl
17:05:58 <Bike> "The production drew strongly on new wave and R&B culture of the mid-1980s to appeal to children to respect elders, avoid peer pressure, and build self-confidence."
17:06:10 <Bike> Roots - Mr. T says "Ya can't know where you're going if ya don't know where you're from" and explains the symbolism of his gold chains.
17:06:30 <Bike> Recouping - When a kid trips on the sidewalk, "Dr. T" shows how one can preserve their dignity after an "absoludicrous" mistake by playing it off as a break dance move.
17:06:50 <Bike> Mr. T's Tale - Mr. T tells his version of Romeo and Juliet and gives a pro-reading message.
17:06:57 -!- Zerker has joined.
17:09:26 -!- WeThePeople has quit (Quit: Leaving).
17:12:11 <kmc> ah Romeo and Juliet, the inspiring story of a three day relationship between a 13 year old and a 17 year old that ended in 6 deaths
17:13:00 <Bike> imagine if most teen romances ended in six deaths
17:13:19 <Phantom_motnahP> i hate it because i had to spend two fucking years of english classes 'analysing' it
17:13:38 <Fiora> I can't take it seriously anymore because
17:13:41 <Fiora> I can only think of the DiCaprio movie
17:13:48 <fizzie> Two years of analysing Mr. T, oh, what a fate.
17:13:50 <Bike> imo, Much Ado > R&J
17:14:03 <pikhq_> It's one of Shakespeare's more boring plays. And fucking idiots think of it as being a romantic tale.
17:14:12 <kmc> i haven't read much shaxpear
17:14:27 <Taneb> I like Othello and the Comedy of Errors
17:14:37 <Taneb> And Richard II and Richard III
17:14:40 <Bike> shakespeare in film is just great overall fiora, you know there's an adaptation of Hamlet involving the "Denmark Corporation"
17:14:57 <elliott> i can't believe we literally talked about this weeks ago.
17:15:03 <elliott> should i say the same things about hamlet
17:15:05 <Bike> IT WAS A GOOD ADAPTATION OKAY
17:15:12 <elliott> no it's not just your line!
17:15:52 <Bike> it's hamlet In Modern Times, man
17:15:57 <pikhq_> kmc: He's a decent playwright that a lot of people misunderstand. Badly.
17:16:11 <pikhq_> First secret to Shakespeare: about half of it is dick jokes.
17:16:55 <Fiora> was it kimchee or shachaf or someone who wanted more astrophysics
17:16:58 <Fiora> http://arxiv.org/pdf/1106.0923v2.pdf
17:17:10 -!- Zerker has quit (Quit: Colloquy for iPad - Timeout (10 minutes)).
17:20:04 <Taneb> I'm gonna go get ready to go out
17:20:16 <Taneb> elliott, avoid the golf club if you want the universe to stay intact
17:20:27 <Taneb> Also where is the golf club
17:20:55 <Taneb> Hexham golf club, not Tyne Green golf club
17:21:23 <Taneb> Oh god, it's right on the other side of town
17:24:07 <coppro> Fiora: the gun says 'SWORD 9mm'
17:24:24 <Bike> http://www.harkavagrant.com/index.php?id=238 shakespeare
17:24:27 <Fiora> you probalby remmber better than me -_-
17:24:34 <Taneb> coppro, there were two guns, I think
17:24:37 <Bike> excellent use of metaphor and... pathetic fallacy?
17:24:48 <Taneb> At the beginning they used a SWORD 9mm
17:24:58 <Taneb> But Lord Capulet (I think) had a LONGSWORD
17:24:58 <coppro> also this channel is wonderful
17:25:20 <Bike> lord capulet don't fuck around
17:25:26 <Fiora> I'm just remembering the moment where the guy is like BRING ME MY LONGSWORD
17:25:34 <Fiora> and he gets a gun made by Sword
17:25:37 <Fiora> and it is just the best
17:25:55 <Fiora> I think we watched it in class, I'm not sure
17:26:00 <Fiora> in like, middle school
17:26:21 <Bike> best part of that movie was mercutio though
17:26:24 <coppro> elliott: I just said it is
17:26:31 <Fiora> "The debris produces a small but massive accretion
17:26:40 <Fiora> (this sounds so contradictory out of context)
17:26:40 <Bike> how shakespearian
17:26:57 <coppro> Fiora: if you know what "massive" means, it doesn't
17:27:12 <Fiora> yeah, I know <.< the context is "it's white dwarf degenerate matter, so yes it's small and massive"
17:27:25 -!- Zerker has joined.
17:35:37 -!- Bike_ has joined.
17:38:17 -!- Taneb has quit (Quit: Leaving).
17:38:37 -!- Bike has quit (Ping timeout: 256 seconds).
17:38:54 -!- ogrom has joined.
17:40:08 -!- Bike_ has changed nick to Bike.
17:41:02 -!- DHeadshot has quit (Read error: Connection reset by peer).
17:41:08 -!- DH____ has joined.
17:43:47 -!- Zerker has quit (Quit: Colloquy for iPad - Timeout (10 minutes)).
17:44:06 -!- DH____ has quit (Read error: Connection reset by peer).
17:44:28 -!- DHeadshot has joined.
18:25:28 <Sgeo> I think drinking hot coffee everyday broke my ability to swallow. It feels like I can only gulp now
18:29:58 <kmc> can you quaff
18:31:24 <kmc> ThatOtherPerson: do something useful
18:31:53 <ThatOtherPerson> perhaps they could also tell me something to do in a useful manner
18:32:19 <kmc> contribute to Mosh: https://github.com/keithw/mosh/
18:37:29 <AnotherTest> although I probably don't have the required time/will to get into the project
18:37:35 <elliott> kmc is getting paid so much right now
18:42:37 <Sgeo> Would Mosh work well or poorly for Nethack/other roguelike servers?
18:42:44 <Sgeo> Especially due to the prediction stuff
18:42:55 -!- DHeadshot has quit (Ping timeout: 258 seconds).
18:43:17 <kmc> probably well
18:43:26 <elliott> prediction wouldn't do anything
18:43:33 <Bike> a well-paid spambot
18:43:36 <Sgeo> Also, why UDP?
18:43:53 <Bike> it's supposed to work on mobile connections isn't it
18:43:55 <elliott> it's effectively doing the work of tcp in some ways
18:43:55 <kmc> prediction is pretty conservative by default; it will only echo characters if it's seen a character echoed on the same line
18:44:00 <Sgeo> Because keystrokes and c... ah
18:44:03 <elliott> except it's not based on sending a stream
18:44:07 <elliott> it's based on synchronising a state
18:44:08 <Bike> often not practical to establish a "real" "connection"
18:44:33 <kmc> yes several reasons for UDP: a) roaming, b) making your own flow control that's better suited to the interactive traffic c) synchronizing states rather than sending an in-order reliable stream
18:44:44 <Sgeo> Maybe that portion of the protocol could be useful elsewhere?
18:44:45 <AnotherTest> but... UDP can cause a lot of implementation hassle :D
18:44:59 <kmc> if a packet is dropped and the terminal has updated in the meantime, the server can send the new state; it doesn't need to retransmit the old packet exactly
18:45:04 <Sgeo> Although, I guess state synch is domain specific
18:45:08 <kmc> Sgeo: yes, KeithW has already applied it to video conferencing as well
18:45:16 <kmc> the state sync protocol is reasonably generic
18:45:28 <kmc> it can synchronize any object type with serializable diffs, basically
18:45:34 <kmc> you can join #mosh and talk to him about it more
18:45:45 <kmc> not yet :/
18:45:49 <kmc> you should add IPv6 support
18:45:57 <Sgeo> Serializable diffs and unique IDs for each state?
18:46:01 <kmc> something like that
18:46:07 <kmc> there's a paper linked to from mosh.mit.edu
18:46:09 <Sgeo> Because otherwise how would it know which diffs to use
18:46:10 <AnotherTest> kmc: Maybe I should, maybe I shouldn't bother though
18:46:12 <kmc> which lays out the protocol somewhat
18:46:44 <Sgeo> Also, I can't remember that book on security that kmc recommended
18:46:45 <AnotherTest> software written in 2013 should do IPv6 I think tho
18:46:55 <kmc> AnotherTest: great, we look forward to receiving your patch
18:47:15 <kmc> Sgeo: The Tangled Web?
18:47:19 <kmc> it's about web security
18:47:21 <AnotherTest> kmc: I said maybe, which probably means "unlikely" in this particular case
18:47:22 -!- mnoqy has joined.
18:47:57 <kmc> AnotherTest: sorry to be sarcastic, it's just that we all agree IPv6 is a high priority, but nobody is working on Mosh full time atm
18:48:05 <kmc> KeithW is a grad student with lots of grad studenty work
18:48:07 <kmc> i'm a lazy bum
18:48:10 <kmc> i might start contributing more though
18:48:31 <Sgeo> "With coverage extending as far as planned HTML5 features, The Tangled Web will help you create secure web applications that stand the test of time."
18:48:32 <kmc> AnotherTest: in particular he believes (and I'm inclined to agree) that we can't really claim IPv6 support unless we also support roaming between v4 and v6
18:48:35 <Sgeo> Book sounds a bit old?
18:48:44 <kmc> because roaming is a core part of the Mosh experience
18:49:09 <kmc> and that in turn means that mosh needs to be aware of multiple server IPs and switch between them if traffic isn't getting through
18:49:16 <kmc> right now Mosh doesn't even notice that it's roamed
18:49:24 <kmc> it's just always sending UDP packets and sometimes they get through
18:49:29 <AnotherTest> kmc: do you allow public-key authentication with anything else than RSA?
18:49:38 <Sgeo> Can't roam during the SSH handshake, right?
18:49:43 <kmc> AnotherTest: all authentication and public-key crypto is handled by SSH
18:50:10 <kmc> AnotherTest: the 'mosh' wrapper script uses ssh to launch a mosh-server on the remote host, which prints a symmetric crypto key, and then it launches a mosh-client locally with that key
18:50:20 <kmc> from there they communicate using 128-bit AES in OCB mode
18:50:31 <kmc> this is great because it means the Mosh crypto story is super simple
18:50:37 <kmc> and all your strange SSH authentication methods still work
18:50:43 <kmc> AnotherTest: hm?
18:50:47 -!- Bike has quit (Read error: Connection reset by peer).
18:50:47 -!- Bike_ has joined.
18:51:02 <kmc> no it's a totally different protocol
18:51:03 <Sgeo> Also, writing NetHack in the language where every program is NetHack is super simple.
18:51:03 <AnotherTest> kmc: I thought this was like an extension of SSH or somethign
18:51:12 <kmc> http://mosh.mit.edu/
18:51:34 <Sgeo> Surely the complexities of ssh in some sense add to mosh complexity?
18:51:47 <kmc> AnotherTest: the 'mosh' wrapper script uses ssh to launch a mosh-server on the remote host, which prints a symmetric crypto key, and then it launches a mosh-client locally with that key
18:51:48 -!- Bike_ has changed nick to Bike.
18:51:58 <kmc> then the SSH connection terminates and mosh-server and mosh-client talk to each other using the Mosh protocol
18:52:24 <AnotherTest> there is a mosh client running on the server?
18:52:26 -!- doesthiswork has joined.
18:52:38 <kmc> Sgeo: yeah, but people already trust SSH, and keeping it secure is someone else's responsibility
18:52:43 <elliott> Sgeo: ssh is only used for authentication/handshakey stuff
18:52:47 <kmc> AnotherTest: no, there's a mosh-server on the server, and a mosh-client on the client
18:52:53 <elliott> have you tried reading the page
18:52:53 <kmc> you know hat
18:52:55 <kmc> you know what
18:52:57 <kmc> why don't you read the page
18:53:09 <kmc> or at least pay attention to the things people are saying in channel
18:53:21 <kmc> it spells out pretty explicitly at the top what the advantages over SSH are
18:53:38 <elliott> it's ok kmc. i'm using mosh... right now
18:54:14 -!- Bike_ has joined.
18:54:20 <Bike_> does anyone know chinese? you should traslate this two thousand page book for me because victorians can just go fuck themselves
18:54:25 -!- Bike has quit (Disconnected by services).
18:54:28 -!- Bike_ has changed nick to Bike.
18:54:33 <Sgeo> Is there any support for decoupling mosh from ssh and using something else for auth+handshake?
18:54:36 <kmc> another advantage of delegating connection setup to SSH is that there's no persistent mosh daemon that needs to be installed and run as root
18:54:56 <Sgeo> There's a persistent ssh daemon that needs to be installed and run as root
18:54:59 <Bike> imo fuck victorians. fuck england
18:54:59 <kmc> instead you have one mosh-server process for each connection, running as that user, and it could just be running out of their homedir
18:55:12 <kmc> Sgeo: right. but people already have that.
18:55:21 <kmc> and would want to keep it even if they install mosh
18:55:26 <kmc> because SSH does some things Mosh doesn't
18:55:47 <kmc> Sgeo: there are examples further down the page of how to launch mosh-server and mosh-client manually
18:55:51 <elliott> well you need root if you want to log in as users
18:56:15 <elliott> probably not that useful now that people whitelist ssh users
18:56:21 <elliott> but running a bunch of servers would sort of suck too
18:56:21 <kmc> i know somebody wrote a wrapper to launch mosh over HTTPS, and someone else wrote one for SMS :)
18:56:24 <AnotherTest> so it uses SSH for the authentication, and the rest is a different protocol?
18:56:26 <elliott> you'd need... a supervisor running as root
18:56:30 <elliott> that starts up all the little servers
18:56:38 <elliott> unless everyone just put it in their crontab or something which uhhhh
18:56:41 <pikhq_> Technically it could use something else for authentication.
18:56:59 <pikhq_> But SSH is already there.
18:57:12 <Sgeo> "But you'll need working UDP."
18:57:32 <Sgeo> I mean, in the sense of people not having access to working UDP
18:57:38 <Bike> are there systems that don't have working udp
18:57:47 <pikhq_> Some particularly bad NATs don't pass it through sanely.
18:57:50 <kmc> if your server is behind a firewall then you need to forward a port for it
18:57:55 <kmc> one port per concurrent Mosh connection
18:58:16 <pikhq_> But those NATs break a lot of stuff.
18:58:36 <AnotherTest> You probably need root access to forward ports though
18:59:01 <Deewiant> AFAICT eduroam at our university blocks UDP completely
18:59:20 <Sgeo> Deewiant, ! How do you live without BZFlag? (iirc BZFlag uses UDP but I might be wrong)
18:59:33 <pikhq_> A lot of games use UDP.
18:59:48 <kmc> yeah, some really restrictive public WiFi doesn't allow UDP
18:59:50 <pikhq_> Blocking UDP is a sign of incompetence.
18:59:51 <Deewiant> I've never heard of BZFlag so I guess the answer is "without worrying about it"
19:00:05 <kmc> the only traffic which can reliably make it across the Internet unmolested is 443/tcp
19:00:09 <Deewiant> eduroam isn't public (although at the campus I guess it practically is)
19:00:09 <kmc> and this is why we can't have nice things
19:00:26 <Sgeo> Deewiant, fantastic multiplayer game
19:00:31 <pikhq_> Which is why you should run SSH on 443/tcp.
19:00:50 -!- copumpkin has quit (Ping timeout: 252 seconds).
19:00:52 <Sgeo> (Ok, maybe fantastic is overstating it)
19:00:54 <Deewiant> I noticed this when I noticed my laptop's clock being off by a minute or so
19:01:22 -!- ThatOtherPerson has quit (Quit: Leaving).
19:02:23 <kmc> that's what http://www.rutschle.net/tech/sslh.shtml and https://github.com/stealth/sshttp are for :)
19:02:56 <AnotherTest> kmc: how does the "If the client changes IP addresses, the server will begin sending to the client on the new IP address within a few seconds." part work?
19:03:13 -!- copumpkin has joined.
19:03:16 <kmc> AnotherTest: the client is always sending packets to the server
19:03:29 <kmc> if the server gets an authenticated packet from a new IP, it will start responding to that IP instead
19:03:36 <kmc> the client can roam but the server can't
19:03:45 <kmc> (but we have plans to add some degree of server roaming)
19:04:11 <kmc> note that the client doesn't even notice it's roamed
19:04:17 <kmc> which is nice as far as making the design simple and robust
19:04:45 <kmc> for example it works just as well if a router upstream of you has roamed
19:04:52 <Sgeo> "The heartbeats allow Mosh to inform the user when it hasn't heard from the server in a while (unlike SSH, where users may be unaware of a dropped connection until they try to type)."
19:06:34 <Sgeo> "SO 2022 locking escapes
19:06:34 <Sgeo> Only Mosh will never get stuck in hieroglyphs when a nasty program writes to the terminal."
19:06:35 <fizzie> Deewiant: Tried to go check inside, if it says anything official about eduroam: "Aalto University´s web services will have an extensive maintenance downtime on 19.-21.4.2013. This is because the university´s data center is moving. We apologise for the inconvenience."
19:06:40 <fizzie> (Also note: ´ in place of '.)
19:06:43 <Sgeo> How does that locking stuff even happen
19:07:08 <Deewiant> fizzie: How do you access inside from outside
19:07:22 <fizzie> Deewiant: By... typing "inside.aalto.fi" to a browser?
19:07:50 -!- Lumpio- has quit (Ping timeout: 256 seconds).
19:07:57 <kmc> Sgeo: there are terminal control codes to tell the terminal to switch to a different character set
19:08:02 <Deewiant> It might've been yesterday that I tried it
19:08:08 <AnotherTest> kmc: but it doesn't roam when you change devices of course. Could you add this if you public-key cryptography instead of AES OCB?
19:08:09 <Deewiant> "The certificate is only valid for the following names:vhosts.aalto.fi , halli.tkk.fi , hse-halli.tkk.fi , itp.aalto.fi , itp.hse.fi , kuore.org.aalto.fi , mapple.aalto.fi"
19:08:35 <fizzie> Deewiant: It worked fine last time I used it from outside, probably a week or three ago.
19:08:59 <kmc> AnotherTest: maybe, but the simple solution is to use screen or tmux on the server (which provides a lot of things mosh doesn't)
19:09:04 <Deewiant> I've probably done that approximately never which is why I just assumed it's inside-only
19:09:38 <kmc> it wouldn't be "instead of AES" so much as "in addition to"; almost every application of public-key crypto uses it to establish a symmetric key
19:09:45 <kmc> and public key crypto is more complicated to begin with
19:10:01 <kmc> so it would be a huge increase in the complexity of the security-critical codebase
19:10:06 <fizzie> Deewiant: It does require the (Shibboleth?) login from outside, can't recall if that was the case from inside.
19:11:03 <AnotherTest> kmc: Yeah. I guess you could do this too if you redid the key-agreement
19:11:36 <kmc> another thing people want is the ability to reattach to a mosh-server if the corresponding mosh-client dies (e.g. if your laptop reboots)
19:11:47 <kmc> and that's also not possible with mosh itself, and is also solved easily enoguh with screen / tmux
19:12:00 <AnotherTest> Well, that would fall under the same thing as device switching I guess
19:12:05 <kmc> but you do have to manually kill the abandoned mosh-server which is a pain
19:12:24 <AnotherTest> If every mosh client had a static public key I guess it might be possible
19:12:27 <kmc> or you do something like 'mosh host -- screen -dR foo' in which case the new mosh connection will make the old mosh-server exit
19:12:42 <AnotherTest> as you could use something like DH or FHMQV to agree on a new key
19:12:53 <elliott> mosh + dtach handles this perfectly well btw
19:14:16 <kmc> Sgeo: the locking control codes are obsolete because UTF-8 provides a way to print any character in a stateless way
19:14:21 <kmc> but some programs still use them
19:15:50 <kmc> for ncurses programs you sometimes need to set NCURSES_NO_UTF8_ACS=1
19:16:05 -!- AnotherTest has quit (Read error: Operation timed out).
19:20:15 -!- copumpkin has quit (Ping timeout: 252 seconds).
19:23:53 -!- copumpkin has joined.
19:31:03 -!- Lumpio- has joined.
19:31:08 -!- Lumpio- has quit (Client Quit).
19:31:37 -!- Lumpio- has joined.
19:39:43 -!- epicmonkey has joined.
19:45:12 <fizzie> https://dl.dropboxusercontent.com/u/113389132/Misc/20130420-fungot.flac
19:47:13 <fizzie> But would not explaining it ruin it?
19:49:00 <kmc> i guess one problem with convincing someone to drop out of college and join your startup is that you then feel really guilty firing them even if they're a shit employee
19:49:34 <fizzie> Hey, where *is* fungot, anyway?
19:49:52 <Bike> kmc: how many startups have you founded. like twenty
19:49:53 -!- fungot has joined.
19:50:08 * Lymia feeds fungot more hemlock
19:50:09 <fungot> Lymia: last year for a company of approximately 2 employees. the way i'm seeing the " fnord"
19:50:27 <c00kiemon5ter> fizzie, fungot is existence transformed into an irc connection giving life into a presense named fungot
19:50:28 <fungot> c00kiemon5ter: evaluating ( display ' ( " .ss". scheme-mode) auto-mode-alist)))
19:50:56 <Lymia> fungot knows Lisp o.o
19:50:56 <fungot> Lymia: nm, it works, so i was trying to
19:55:03 <fizzie> fungot.flac is what you get if you render fungot's source code (in DejaVu Sans Mono, rotated 90 degrees) into a 513x6592 pixel bitmap, then consider that image as a magnitude spectrum (for the [0, 11.025] kHz range, with 1024-sample windows and 75% overlap) and resynthesize a time-domain signal with that spectrum and arbitrary phase, as per (Griffin et al., 1984).
19:55:03 <fungot> fizzie: to address the limitations of your compiler. i don't
19:55:48 <fizzie> The "beat" is probably lines of text.
19:58:37 -!- ogrom has quit (Quit: Left).
19:59:34 <olsner> it's a lot less dissonant than I expected
20:00:22 <Bike> fungot should get signed with virgin babylon
20:00:22 <fungot> Bike: in all seriousness, but still
20:01:20 -!- rapido has joined.
20:01:40 <fizzie> The hiss in the background is because my implementation of (Griffin et al., 1984) had some... issues -- http://sprunge.us/PXGX -- when I tried to feed it a spectrum with that many real zeros in it. (So I just added a -17 dB (relative) or so background white noise.)
20:03:45 -!- xitrix has quit (Ping timeout: 252 seconds).
20:06:58 <fizzie> (The part around 50-55s in is the Underload interpreter.)
20:10:24 <rapido> topic? esoteric programming with multiset alternative evaluation model: {1,2} + {3,2:4} => {4,3:5,2:6}
20:11:36 -!- Bike_ has joined.
20:14:14 <rapido> {{1,2} {3,4}} => {1,2,3,4}
20:14:42 -!- Bike has quit (Ping timeout: 276 seconds).
20:15:15 <rapido> oops: {{1,2},{3,4}} => {1,2,3,4}
20:15:59 <rapido> {{1,2} {3,4}} => {1 3,1 4,2 3,2 4}
20:16:10 -!- Bike has joined.
20:17:03 <rapido> i understand the model but interpreting expressions by hand is tricky
20:19:09 -!- Bike_ has quit (Ping timeout: 252 seconds).
20:22:16 <rapido> what does the following expression evaluate to?
20:22:23 <rapido> ({1,2} + {2:3,4}) * {5,2:6} ?
20:25:20 <rapido> {2:20,3:25,4:24,7:30,2:36}
20:27:06 <Bike> i have no idea what you're talking about
20:27:23 <rapido> Bike: good! so it's pretty esoteric
20:28:24 <rapido> i'm talking about a esoteric evaluation model
20:29:12 -!- c00kiemon5ter has left.
20:29:21 <rapido> that yields 'alternative' values as opposed to single values
20:30:41 <Bike> if i had to do that in the way that seems natural to me i'd say {3:20,12:36}
20:30:43 <rapido> reference: prolog, generators, etc.
20:31:17 <rapido> Bike: interesting, what's natural to you?
20:31:33 -!- epicmonkey has quit (Ping timeout: 248 seconds).
20:31:37 <Bike> well, i guess i'm taking the sets as ordered pairs
20:31:38 <kmc> you can write a multiset monad in Haskell can't you
20:31:40 <Bike> so probably not that great
20:32:12 -!- ion has quit (Quit: Rebooting the server once more due to btrfs issues).
20:32:20 <mnoqy> Logic(T) might be satisfactory??
20:32:35 <kmc> Bike: i haven't founded any startups and don't think i'd want to
20:32:41 <mnoqy> list "almost works"
20:32:59 <kmc> it involves a bunch of stress and a bunch of non-technical problems and you don't get paid
20:33:37 <rapido> ({1,2} + {2:3,4}) * {5,2:6} => {2:4,3:5,6} * {5,2:6} => {2:20,3:25,4:24,7:30,2:36}
20:33:42 <kmc> i thought i might like being Employee #1 at a startup but I couldn't get along with Employee #2 and they decided to keep him instead of me out of I think guilt / obligation
20:34:30 <kmc> so i dunno
20:34:59 <Bike> rapido: x:y is supposed to be a range, yes?
20:35:16 -!- ion has joined.
20:36:02 <rapido> so {2:3} + {3:4} => {6:7}
20:36:46 <rapido> Bike: may be you can suggest a better notation (without using *)
20:37:23 <Bike> nah i get it now
20:37:33 <Bike> just ignore me
20:38:54 <kmc> well you can see a multiset as a function from element → count
20:39:43 <kmc> so you could write like {5↦1, 2↦6}
20:39:45 <rapido> kmc: true, but i like to prefix the count to the element (syntactically)
20:39:59 <kmc> i recommend not special-casing the case where count = 1, i think that produces confusion
20:40:22 -!- Phantom__Hoover has joined.
20:41:06 <kmc> anyway if you implement this as a monad in Haskell then you can use all kinds of existing generic code to play around with it
20:41:54 <kmc> i guess it's hard to make a Monad instance, though, because you need equality on elements, and probably ordering or some hash function if you want efficiency
20:42:01 <kmc> so it's the usual "restricted monad" problem
20:44:10 -!- Phantom_motnahP has quit (Ping timeout: 256 seconds).
20:46:45 -!- rapido has quit (Quit: rapido).
20:47:53 -!- Phantom__Hoover has quit (Ping timeout: 240 seconds).
20:48:11 -!- rapido has joined.
20:48:46 <rapido> kmc: equality and ordering is indeed very important
20:49:14 <rapido> every multset (of multisets) must be ordered
20:51:12 <rapido> counts can be fractional
20:51:30 <rapido> to make it more interesting
20:52:30 <rapido> more generically: counts need not be numbers
20:52:55 <rapido> counts need just to be objects that adhere to the arithmetical laws
20:53:55 <rapido> http://github.com/rapido/spread/blob/master/A%20new%20look%20at%20multimaps.pdf?raw=true
21:00:17 <rapido> counts as 'mathematical' objects does make things more interesting
21:00:54 -!- Phantom__Hoover has joined.
21:05:58 -!- DHeadshot has joined.
21:09:07 -!- Sgeo has quit (Remote host closed the connection).
21:09:11 -!- Phantom__Hoover has quit (Ping timeout: 245 seconds).
21:11:39 -!- Sgeo has joined.
21:13:59 -!- rapido has quit (Quit: rapido).
21:19:42 -!- rapido has joined.
21:26:36 -!- conehead has joined.
21:34:17 -!- rapido has quit (Quit: rapido).
21:52:08 -!- Phantom__Hoover has joined.
21:56:29 -!- Phantom__Hoover has quit (Ping timeout: 252 seconds).
22:08:31 -!- conehead has quit (Quit: Computer has gone to sleep.).
22:19:58 -!- Taneb has joined.
22:25:20 -!- augur has quit (Remote host closed the connection).
22:37:02 <Taneb> I need something to do to cheer me up
22:37:17 <kmc> masturbation?
22:37:31 <kmc> tv watching?
22:37:50 <kmc> applied cryptography puzzles?
22:37:54 <Bike> norwegian-murdering?
22:38:15 <Taneb> Bike, can't be bothered right now
22:38:36 <coppro> Taneb: helping out on NetHack 4?
22:39:14 <kmc> http://www.matasano.com/articles/crypto-challenges/
22:41:11 <Taneb> kmc, I've asked to be in
22:41:31 <Taneb> How long should the wait be, roughly?
22:42:28 <kmc> i think less than a day
22:42:40 <kmc> but i think they're getting a lot more responses now than before
22:43:05 <coppro> Taneb: you should help populate the NH4 bug tracker
22:43:16 <Taneb> coppro, but I suck at NetHack!
22:43:27 <coppro> we have several hundred bugs in need of filing
22:44:05 <Bike> are they hilarious as NH3's listed bugs
22:44:44 <coppro> Bike: we need to file all those bugs in the new bug tracker
22:51:46 -!- btiffin has left.
23:04:04 -!- carado_ has quit (Ping timeout: 246 seconds).
23:16:54 -!- carado_ has joined.
23:22:12 -!- doesthiswork has quit (Quit: Leaving.).
23:26:02 <Bike> http://en.wikipedia.org/wiki/The_Flyting_of_Dumbar_and_Kennedie why do people not like scotland, again?
23:28:39 -!- Taneb has quit (Quit: Leaving).
23:29:37 -!- lahwran has changed nick to lahwran-.
23:29:39 -!- lahwran- has changed nick to lahwran.
23:38:38 <tswett> I'm trying to think of a historical figure that it would make sense to name the universe after.
23:38:57 <tswett> I'm thinking maybe Aristotle or Newton.
23:40:13 <Bike> aristarchus of samos
23:41:08 <Bike> er, Madhava of Sangamagrama.
23:42:04 <Bike> or Kolmogorov, just because he's Kolmogorov
23:45:52 <tswett> I'm just going with Aristotle.
23:45:59 <Bike> boooooriiiiing
23:46:08 <Bike> dude thought women had less teeth than men
23:46:22 <tswett> Yeah, but he wrote books about literally* everything.
23:46:38 <tswett> And a significant amount of what he wrote was correct.
23:46:59 <Bike> so did everyone educated, back when there wasn't much to write about
23:49:13 <Koen_> tswett: do you mean he wrote about *literacy* everything?
23:49:27 <tswett> He wrote about ludicrously everything.
23:52:49 -!- augur has joined.
23:52:49 -!- augur has quit (Remote host closed the connection).
23:52:58 -!- augur has joined.