00:13:08 -!- ter2 has quit (Ping timeout: 255 seconds).
00:35:53 -!- oerjan has quit (Quit: Nite).
00:37:58 -!- copumpkin has joined.
00:44:16 -!- tertu has joined.
00:45:58 -!- ^v has joined.
00:47:32 -!- ter2 has joined.
00:47:32 -!- tertu has quit (Disconnected by services).
00:50:34 -!- tromp has quit (Ping timeout: 240 seconds).
00:51:06 -!- tertu3 has joined.
00:53:51 -!- ter2 has quit (Ping timeout: 250 seconds).
01:02:55 -!- ter2 has joined.
01:05:45 -!- tertu3 has quit (Ping timeout: 252 seconds).
01:06:05 -!- tromp has joined.
01:06:21 -!- tertu has joined.
01:08:19 -!- ter2 has quit (Ping timeout: 240 seconds).
01:32:14 -!- shikhout has quit (Remote host closed the connection).
01:33:47 -!- ter2 has joined.
01:33:47 -!- tertu has quit (Disconnected by services).
01:41:48 -!- ter2 has quit (Read error: Connection reset by peer).
01:42:12 -!- ter2 has joined.
01:46:39 -!- tertu3 has joined.
01:47:38 -!- Phantom__Hoover has quit (Ping timeout: 240 seconds).
01:49:38 -!- ter2 has quit (Ping timeout: 240 seconds).
01:57:38 -!- tertu3 has quit (Ping timeout: 240 seconds).
02:00:55 -!- atslash has quit (Quit: This computer has gone to sleep).
02:01:31 -!- atslash has joined.
02:06:42 -!- atslash has quit (Quit: This computer has gone to sleep).
02:09:35 -!- tromp has quit (Remote host closed the connection).
02:10:11 -!- tromp has joined.
02:10:19 -!- tertu3 has joined.
02:14:14 -!- tromp has quit (Ping timeout: 240 seconds).
02:42:27 -!- Sorella has quit (Quit: It is tiem!).
02:48:13 <kmc> what's lpc
02:59:35 <Sgeo> bitly is actually encouraging people to change passwords, right?
02:59:49 <Sgeo> Gmail filed the email under spam
03:00:44 <monotone> The notice is on bitly's actual site header, if you're paranoid.
03:01:07 <Sgeo> I'm more worried about Gmail filing it under spam and thus possibly screwing people over
03:01:20 <Sgeo> For anyone who might not check or believe their spam folder
03:01:45 <monotone> Welcome to security notifications in the 21st century...
03:11:03 -!- tromp has joined.
03:11:31 -!- conehead_ has quit (Quit: Computer has gone to sleep).
03:28:28 -!- hk3380 has quit (Ping timeout: 265 seconds).
03:32:41 -!- tertu3 has quit (Ping timeout: 264 seconds).
03:37:20 -!- tertu3 has joined.
03:44:06 -!- ter2 has joined.
03:45:54 -!- tertu3 has quit (Ping timeout: 252 seconds).
03:46:52 -!- tromp has quit (Remote host closed the connection).
03:54:31 -!- tromp has joined.
04:01:10 <kmc> "The exclusive or of two different Gold codes from the same set is another Gold code in some phase."
04:02:00 -!- tromp has quit (Remote host closed the connection).
04:02:32 -!- tromp has joined.
04:03:55 <Sgeo> Why do I have codes?
04:05:07 <Sgeo> My last name is Gold
04:06:34 -!- tromp has quit (Ping timeout: 240 seconds).
04:08:55 <Sgeo> To Randall Munroe: If you're going to reinvent the web, there's a high chance you're more likely to do it sanely than the current web
04:08:56 <kmc> i guess that's why, then
04:12:27 <monotone> Until someone figures out something else that they want to bolt on top and it gets broken again.
04:53:14 -!- password2 has joined.
04:56:24 -!- password2 has quit (Max SendQ exceeded).
04:57:00 -!- password2 has joined.
05:07:34 <HackEgo> bypacoin sodateditcoin poricoin treencoin singercoin bfccoin dateftcoin selfcoin bupercoin frizattcoin shercoin sallcoin bituffcoin reignencoin homocoin muecoin hcriecoin lammcoin prolanshakcoin varcoin
05:30:33 <FireFly> fungot: today's wisdom, please
05:30:33 <fungot> FireFly: and fnord only has guile, i proclaim it sucks. they fixed my favorite bug, or so
05:33:59 <fungot> shachaf: i just clicked on page 3 saying ' refer to diagram' which turns up quite a lot
05:34:31 <shachaf> you know what they say, fungot
05:34:31 <fungot> shachaf: they're all secret pervs.) will that be the main proponents of lolcode, and neither riastradh nor i have lots
05:34:40 <fungot> Available: agora alice c64 ct darwin discworld enron europarl ff7 fisher fungot homestuck ic irc* iwcs jargon lovecraft nethack oots pa qwantz sms speeches ss wp youtube
05:35:31 <myname> why is there a difference between site and side in english? as a german it confuses me pretty often
05:38:58 <FireFly> site in the 'website' sense?
05:39:23 <monotone> They're two etymologically different words that just happen to sound similar.
05:39:49 <FireFly> I just realised they're the same in swedish too, but from the unrelated sv:sida = en:page
05:52:05 <myname> it's both "seite" here
05:52:53 -!- password2 has quit (Ping timeout: 255 seconds).
05:57:26 -!- MindlessDrone has joined.
06:03:25 <lambdabot> Title: Creative, Search & Social | Connected Digital Marketing - WebSide
06:03:30 <lambdabot> Title: WebSight Design: Bay Area Web Design, Development & Hosting
06:03:36 <lambdabot> http://www.youtube.com/user/webseid
06:04:16 <lambdabot> http://de.wikipedia.org/wiki/Webseite
06:05:27 <Jafet> (also, Website /= Webseite.)
06:05:36 -!- ^v has quit (Quit: http://i.imgur.com/Akc6r.gif).
06:22:16 -!- FreeFull has quit.
06:39:07 -!- ter2 has quit (Ping timeout: 276 seconds).
07:18:50 -!- Bike has quit (Ping timeout: 255 seconds).
07:19:18 <FireFly> http://www.cnet.com/news/ancient-d20-die-emerges-from-the-ashes-of-time/
07:26:04 -!- MindlessDrone has quit (Quit: MindlessDrone).
07:31:09 -!- slereah_ has joined.
07:53:25 -!- atslash has joined.
07:58:26 <mroman_> Somebody wants to reinvent the web?
07:58:51 <slereah_> Let's just buy some copper wire and some routers
08:01:05 <FireFly> Didn't Russia say they were going to do approximately that?
08:01:37 <slereah_> They used to have their own network back in the USSR days
08:01:48 <slereah_> Also in France we had the minitel~
08:02:00 <slereah_> The earliest network for the public
08:27:00 -!- Patashu has joined.
08:28:49 -!- quintopia has joined.
08:34:51 <mroman_> Is there yet a 2D language where you have to modell "data flow"?
08:34:58 <mroman_> in addition to instruction flow
08:35:15 <mroman_> i.e. data is stored on the 2D grid itself
08:35:59 <mroman_> (without having a data pointer that points to some cell in the grid)
08:36:18 <mroman_> You'd actually have to move data around.
08:36:55 <Jafet> Wireworld is one that's actually designed to resemble circuits.
08:37:15 <mroman_> http://codepad.org/9f6lWtqu
08:37:28 <mroman_> DULR change directions of the data stream
08:37:40 <mroman_> dulr take data from their "opposite" point and send them
08:37:55 <mroman_> i.e. r takes the data from the cell to the left and sends it rightwards
08:38:05 <mroman_> until it arrives at another data cell
08:38:35 <mroman_> (the above code should calculate a=a+2*r, with a initialized to 2)
08:39:57 <mroman_> http://codepad.org/H1IXEaXj <- like that
08:40:04 <Jafet> Also, circuit layouts are 2D computer languages
08:40:06 <mroman_> there's no interpreter yet. Just sketched out some things in my hand
08:40:28 <mroman_> the result from the addition is moved to the 0 above the *
08:41:23 <mroman_> 2d computer languages with timing constraints ;)
08:53:53 -!- Tod-Autojoined has joined.
08:54:54 -!- TodPunk has quit (Read error: Connection reset by peer).
09:11:23 -!- drdanmaku has quit (Quit: Connection closed for inactivity).
09:20:20 -!- atslash has quit (Ping timeout: 255 seconds).
09:26:53 <FireFly> I don't remember if I ever beat all the Rubicon levels
09:27:10 <FireFly> Or whatever the java-applet game based on RUBE was called
09:29:04 -!- atslash has joined.
09:40:07 -!- edwardk has joined.
09:40:34 -!- atslash has quit (Quit: This computer has gone to sleep).
09:45:09 -!- Patashu_ has joined.
09:45:09 -!- Patashu has quit (Disconnected by services).
09:52:11 -!- sign has quit (Ping timeout: 252 seconds).
10:11:08 -!- ais523 has joined.
10:12:47 -!- boily has joined.
10:17:26 -!- oerjan has joined.
10:28:24 <oerjan> @tell kmc <kmc> what's lpc <-- the interpreted C-like OO language in which lpmuds are mostly programmed.
10:28:46 <oerjan> https://en.wikipedia.org/wiki/LPC_(programming_language)
10:29:16 <myname> http://upload.wikimedia.org/wikipedia/commons/c/c9/Chaos_magic_ritual_involving_videoconferencing.JPG wat
10:34:36 -!- shikhin has joined.
10:48:45 <oerjan> <FireFly> http://www.cnet.com/news/ancient-d20-die-emerges-from-the-ashes-of-time/ <-- argh _two_ tall, space-stealing bars
10:50:55 <oerjan> i am mystified that web designers don't understand how annoying that is on laptop screens
10:51:26 <oerjan> (admittedly, many probably do, which is why i don't see it _everywhere_)
10:53:06 <ais523> oerjan: I've actually been criticised for /not/ doing that on a website
10:53:14 <myname> since when are laptop screens considered different from others?
10:53:19 <ais523> I wanted to post a rant about how that should be a client-side setting, not a server-side setting
10:53:26 <ais523> but I didn't, because I didn't think anyone would care
10:53:30 -!- hk3380 has joined.
10:53:39 <oerjan> myname: they are wider than they are (not very) tall
10:58:25 <oerjan> it may be connected to my absent-minded-ness; when things disappear off screen too fast i forget the context of what i'm reading.
10:58:58 <oerjan> and go into ocpd up-down arrow mode.
10:59:48 -!- Darkgamma has joined.
11:00:08 <oerjan> add to that that i've started to zoom text to strain my eyes less...
11:00:48 -!- atslash has joined.
11:00:50 <oerjan> ...that would be the day the bot is gone.
11:01:02 <boily> `welcome Darkgamma
11:01:03 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: welcome: not found
11:01:13 <oerjan> (Gregor is of course always missing)
11:01:19 <boily> the welcomes are dead!
11:01:21 <ais523> boily: all the welcomes got deleted a while ago
11:01:25 <oerjan> oh bother. just a minute...
11:01:33 <boily> ais523: weren't some restored?
11:01:44 <ais523> Darkgamma: Welcome to the international channel for esolang discussion, development and deployment! Our wiki is available at http://esolangs.org.
11:01:53 <ais523> umm, gah, remembering the welcome message is hard
11:02:06 <ais523> except the first sentence, that was in the topic for /years/, in various variations
11:02:09 <Darkgamma> ais523: yeah, I'm aware of the wiki
11:02:11 <HackEgo> patching file welcome \ patching file r13elcome \ patching file relcome \ patching file rwelcome \ patching file welcome \ patching file welcome13
11:02:17 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: relcome: not found
11:02:28 <oerjan> sorry, undo is also broken :(
11:02:30 <HackEgo> 98076 \ a \ app.sh \ bdsmreclist \ bin \ canary \ cat \ complaints \ :-D \ dog \ etc \ factor \ fb \ fb.c \ head \ hello \ hello.c \ ibin \ index.html \ interps \ lib \ paste \ pref \ prefs \ quines \ quotes \ r13elcome \ relcome \ rwelcome \ share \ src \ test \ Test \ Test.hi \ Test.hs \ UNPA \ welcome \ welcome \ welcome13 \ Wierd \ wisdom \ wi
11:02:41 <ais523> why are the welcomes in the root dir, not bin?
11:02:54 <ais523> anyway, I preferred it when this channel was about esolangs
11:02:57 <oerjan> ais523: because `undo is broken
11:02:58 <ais523> rather than about the welcomebot
11:03:03 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: /hackenv/bin/relcome: Permission denied \ /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: /hackenv/bin/relcome: cannot execute: Permission denied
11:03:17 <oerjan> `run chmod +x bin/*elcom*
11:03:22 <HackEgo> 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.)
11:03:58 <boily> the balance of the Universe was Restored. now I can go to work.
11:04:09 -!- boily has quit (Quit: don't ask about the chickens.).
11:05:00 <ais523> I actually had an esolang idea a couple of days ago
11:05:11 <ais523> I was working on a program, and writing a bunch of comments to explain why it was doing things
11:05:25 <ais523> Darkgamma: that's to do with your client
11:05:29 <ais523> IRC doesn't have a link syntax
11:05:36 <ais523> so the client tries to parse the link, and some of them get it wrong
11:05:47 <ais523> it's OK, you can have a bunch of conversations in parallel in IRC
11:06:00 <ais523> so after a while you get used to having a bunch of conversations in parallel, sometimes in the same channel, or the same people, or even both
11:06:35 <ais523> anyway, the standard rule is that the source code should say the "what", but it doesn't say the "why", so you need comments to explain it to other developers
11:06:44 <ais523> and my idea was: why don't we explain why the code is doing what it does to the computer?
11:07:14 <ais523> so that if we have an explanation saying "we're doing this so that foo", and it doesn't cause foo, we get a warning or an error
11:07:24 <ais523> then I got a little stuck trying to figure out how you explain a reason to a computer
11:07:35 <oerjan> sheesh, i thought the whole point of enclosing the link in <> was to _decrease_ the chance of clients parsing it wrong
11:07:36 <ais523> my current idea is that you make statements about the future
11:07:52 <mroman_> the only thing really disturbing me is that scrolling on that website is laggy
11:08:01 <ais523> oerjan: blame the qwebirc people, I guess
11:08:07 <oerjan> mroman_: let me guess, you don't see them because no enable javascript?
11:08:25 <ais523> Darkgamma: it's not really like COME FROM
11:08:32 <ais523> it's more like, you make a prediction about the future
11:08:41 <ais523> the compiler tries to prove it, if it can, great
11:08:42 <mroman_> oerjan: I have IE with JavaScript enabled
11:08:47 <ais523> if it can find a counterexample, the program errors out
11:08:58 <ais523> if it can't do either, it puts in runtime checks, and complains if the statement ever appears to be false
11:10:05 <mroman_> 1900 times something I guess
11:10:12 <int-e> forever love; while (love) {}
11:10:20 <ais523> Darkgamma: I don't, really, it's quite a new idea, and one that might not ever be properly fleshed out
11:10:21 <mroman_> If I'd surf with 800x600 things'd probably look different
11:10:25 <ais523> I have enough vaporware languages already
11:10:25 <oerjan> mroman_: they are bars on _top_ and _bottom_
11:10:41 <oerjan> mroman_: i surf with non-maximized window + zoom
11:10:45 <ais523> but innovative esolang ideas are rare, so I like to share them with the channel in case they inspire other people
11:11:22 <oerjan> mroman_: you are looking at http://www.cnet.com/news/ancient-d20-die-emerges-from-the-ashes-of-time/ right?
11:11:31 <Darkgamma> ais523, hm I don't know how it'd end up compiled but doing it in an interpreter seems not extremely difficult
11:11:39 <mroman_> Or are you refering to the ad bar below the menu bar?
11:12:11 <ais523> Darkgamma: yeah, if you're merely making statements like "I do this so that there won't be a divide by zero exception"
11:12:15 <ais523> and then there's a divide by zero exception
11:12:19 <ais523> then that line of code probably isn't doing its job
11:12:31 <mroman_> If I surf with a non-maximized window with 200% zoom those bars take almost 90% of the window
11:13:02 <mroman_> but with normal window size and 100% zoom it doesn't disturb me at all
11:13:09 <oerjan> mroman_: menu at top, menu at bottom. except ... it's suddenly stopped doing that D:
11:13:20 <mroman_> the site is still laggy when scrolling
11:13:32 <oerjan> now it's just the one hovering on top
11:13:40 <oerjan> a while ago it was only bottom.
11:13:48 <oerjan> so i guess the site is just screwy.
11:13:56 <mroman_> I also hate websites that hide the scrollbar on the right
11:14:02 <mroman_> How am I supposed to scroll then
11:14:41 <mroman_> I like scrolling with the left mouse button
11:15:08 <oerjan> ais523: are you reinventing design by contract?
11:15:24 <ais523> there's a difference between a precondition/postcondition, and a reason
11:15:34 <ais523> I'm sort-of inventing asserts, except time-displaced
11:15:42 <ais523> like, the innovation isn't assert(foo);
11:16:06 <ais523> but assert(foo will be 5 even after unrelated function X returns);
11:17:28 <oerjan> mroman_: hm normally on sites with hidden scrollbar it reappears when i move the pointer
11:17:34 <Darkgamma> do you assert that foo will be 5 after a specific function or after any general function
11:17:57 <HackEgo> Agent “Iä” Smith is an alien with a strange allergy to avian body covering, which he is trying to retroactively prevent from ever evolving. On the 3rd of March, he's lawful good.
11:18:10 <mroman_> letting websites style the scrollbar was a dump idea anyway *i think*
11:18:15 <ais523> oerjan: the wisdom database isn't particularly elucidating, really
11:18:37 <mroman_> I hope they can style the "back" and "forward" button in HTML5
11:18:43 <mroman_> because that would make _so much_ sense
11:18:46 <oerjan> 't use @nick addressing here, and lambdabot keeps us honest
11:19:58 <oerjan> ais523: that's just what an alien would say
11:20:13 <mroman_> I also don't get why browsers don't display a scrollbar on pages where you wouldn't need to scroll
11:20:19 <mroman_> because it makes the window some pixels wider
11:20:27 <mroman_> and when there's enough content, the window is suddenly smaller
11:20:44 <mroman_> and designers freak out because now everything is a little bit displaced when you compare it to a page with less content
11:20:49 <mroman_> so they do stupid things to fix that
11:21:03 <mroman_> like inline scrollbars and shitt
11:21:05 <oerjan> do you have an extraneous "don't" in that first sentence
11:21:16 <mroman_> javascript emulated scrolling or whatever it's called
11:21:21 <ais523> mroman_: can't you just do overflow-y:scroll, or whatever it is?
11:21:49 <mroman_> but then you have a scroll bar sort of in the webpage itself
11:22:52 <mroman_> the best thing you can do is have your menu bars "fixed"
11:22:56 <mroman_> so they don't move while scrolling
11:23:19 <mroman_> and then make some dummy div to extend the webpage to 101% height so that a scrollbar on the right is always there
11:24:16 <mroman_> and some quirks I'm pretty sure are just done to be able to place more ads .
11:25:17 * oerjan suddenly imagines a web browser which is made for ignoring all that enterprisey stuff. it'll be called Lenin, and its tagline should be "Because capitalists should not control the Web."
11:25:55 <mroman_> http://news.slashdot.org/story/14/05/12/2013236/lectures-arent-just-boring-theyre-ineffective-too-study-finds o_O
11:26:04 <oerjan> it'll probably be judged illegal in the US if anyone tried that, though.
11:27:05 <oerjan> mroman_: are you actually surprised at this finding tdnh
11:27:16 <mroman_> But I don't listen to lecturers
11:27:24 <mroman_> And I don't like to participate
11:27:42 <mroman_> I just sit in class for the sake of being physically present
11:27:57 <Taneb> mroman_, I've almost stopped even that
11:28:03 <mroman_> It's not like they're teaching anything new
11:28:25 <mroman_> So far this semester I've only had As
11:28:28 <oerjan> can i offer "teaching methods that teach all people identically aren't effective", then?
11:28:57 <oerjan> (i'm afraid i have no evidence, only foaming prejudice)
11:29:05 <mroman_> Some Exam about Lisp, Prolog, Java Bytecode and Functional Programming
11:29:30 <mroman_> One question just had some Java Bytecode and you had to interpret it like a freaking jvm
11:29:54 <ais523> oerjan: surely they're effective for some subset of people, and ineffective for others?
11:30:18 <mroman_> Like the teachers assume I can't read java bytecode specs
11:30:24 <ais523> I know when I was teaching, the normal aim was to find some method that would work for a decent proportion, then find some method that would work on a decent proportion of the remainders, and so on
11:31:42 <mroman_> I stopped learning for exams that are open book
11:32:16 <Taneb> I probably ought to learn this term's calculus at some point
11:32:30 <mroman_> I just print out spec and cheat sheets and I'm done learning
11:32:33 <ais523> mroman_: my guess is, perhaps it will be later on, but maybe not at the current level
11:32:41 <mroman_> I'll read the specs on the fly during the exam and I'm good
11:32:50 <ais523> admittedly, some exams are "you have to spend a bunch of time trying to figure out wtf the examiner is thinking"
11:32:51 <mroman_> ais523: I'm in my last semester
11:32:56 <ais523> like, you're not trying to learn the subject
11:32:57 <Taneb> I'd go out of your way to learn things just so you don't forget how
11:33:04 <ais523> you're trying to learn how to comprehend the particular lecturer
11:33:14 <oerjan> mroman_: the problem is if you need the things you're supposed to have learned as a basis for your next level of courses.
11:33:16 <ais523> Taneb: well, learning to work from spec sheets is a pretty useful skill
11:33:21 <mroman_> ais523: social engineering is an important skill ;)
11:33:32 <oerjan> which might be more important in math than in computing, i dunno.
11:33:48 <mroman_> oerjan: yeah. That's only important in math classes
11:34:16 <mroman_> but the good thing about that is that different people from different backgrounds can choose the same lecture/class/course
11:34:35 <mroman_> so they can't assume you know things because not everybody has had the same classes as you up to this point
11:34:47 <mroman_> so essentially you don't need to know anything about what last semester happened
11:35:16 <Jafet> So essentially they are proscribed from teaching anything in depth
11:35:18 <Taneb> Not even for prerequisites?
11:35:31 <mroman_> ais523: "current level"... They assume I don't know anything about functional programming
11:35:45 <mroman_> which is a correct assumption for 99.9% of all students
11:35:53 <ais523> I wonder how many people in #esoteric don't know anything about functional programming
11:35:58 <ais523> it's hard to not pick up at least some Haskell if you idle here
11:36:02 <mroman_> but most of their assumptions just don't apply to somebody like me
11:36:20 <mroman_> I've started programming at age 12
11:36:33 <Taneb> Apparently the second-year compilers module here uses Haskell for parsing, but doesn't really teach it beyond "here's do notation and parsec. Write a parser"
11:36:36 <mroman_> I knew haskell before I was 18
11:36:52 <ais523> I didn't know Haskell before I was 18, but it wasn't very mainstream back then
11:36:59 <Taneb> And someone wrote idiomatic Haskell in Applicative style.
11:37:17 <mroman_> Taneb: most things with parsec can be done by try and <|> anyway ;)
11:37:24 <Taneb> The person marking/evaluating/whatever it looked at it and nodded, confused
11:37:30 <mroman_> and some of oneOf, many1, string, char
11:37:42 <ais523> Taneb: I've seen people try to mark things wrong because they don't understand them :-(
11:37:55 <oerjan> haskell didn't exist when i was 18 hth
11:38:04 <Taneb> ais523, the marker had to say "I'm going to have to trust you on this one"
11:38:17 <mroman_> oerjan: Why are you ending every single sentence with hth
11:38:18 <Taneb> oerjan, when I was 18 I knew lens
11:38:30 <ais523> Taneb: there have been cases where people had to ask me to check them
11:38:33 <Taneb> mroman_, it's Norwegian for the full stop
11:38:56 <oerjan> mroman_: it's probably my inner seething anger seeping out
11:40:11 <Taneb> On my 18th birthday, lens-3.1 was a thing
11:40:25 <Taneb> I hope you all feel old
11:43:31 <oerjan> mroman_: also, your admonishment gets somewhat weakened by my quick checking finding _no_ other instance of hth since i joined the channel (although one tdnh). i may have missed one but still...
11:46:11 * oerjan whacks Taneb with his cane =======Ø
11:46:14 -!- ais523 has quit.
11:50:08 <Darkgamma> I don't have a clue about Haskell except that it's a functional prog. language
11:50:23 <mroman_> it's not every single sentence
11:50:34 <mroman_> it just appears like you use it often
11:51:06 <mroman_> so often that my mind replaces your nick with "they who uses hth at the end of sentences"
11:51:13 <mroman_> I've no idea what hth is btw
11:51:20 <HackEgo> hth is help received from a hairy toe. It is not at all hambiguitous.
11:51:36 <HackEgo> twh would help, but is an hth derivative. hth. twh. hand.
11:52:09 <oerjan> > nubBy(((>1).).gcd)[2..]
11:52:11 <lambdabot> mueval-core: L.hs: removeLink: does not exist (No such file or directory)
11:52:15 <oerjan> > nubBy(((>1).).gcd)[2..]
11:52:17 <lambdabot> [2,3,5,7,11,13,17,19,23,29,31,37,41,43,47,53,59,61,67,71,73,79,83,89,97,101,...
11:52:21 -!- Froox has quit (Quit: *bubbles away*).
11:52:31 <oerjan> Darkgamma: haskell hth ^
11:54:11 <Jafet> > (!!3)<$>transpose[show$foldr(\k a->2*10^2^n+a*k`div`(2*k+1))0[1..2^n]|n<-[0..]]
11:54:14 <lambdabot> "314159265358979323846264338327950288419716939937510582097494459230781640628...
11:54:55 <oerjan> is that using plouffe's algorithm
11:56:15 <lambdabot> [0,1,1,2,3,5,8,13,21,34,55,89,144,233,377,610,987,1597,2584,4181,6765,10946,...
11:56:44 <oerjan> that should cover the obligatory par for the course
11:57:46 <Darkgamma> puts zero, then adds one, then adds the last two
11:59:04 <oerjan> sort of, although in a bit roundabout way
11:59:25 <oerjan> > let fib = 0:1:zipWith(+) fib (tail fib) in fib -- less obscurely
11:59:27 <lambdabot> [0,1,1,2,3,5,8,13,21,34,55,89,144,233,377,610,987,1597,2584,4181,6765,10946,...
11:59:34 <fizzie> oerjan: http://sprunge.us/YKdM hth
12:00:15 <oerjan> fizzie: wow this changes EVERYTHING
12:00:22 <oerjan> mroman_: i'm not even on TOP hth
12:00:36 <fizzie> I think the distribution looked very different some months back.
12:00:38 -!- Sgeo has quit (Read error: Connection reset by peer).
12:01:38 <fizzie> Are you going to start telling me I should be nosql webscale etc etc
12:01:43 <oerjan> Jafet: we're all living inside a fizzie-maintained database
12:02:15 <Jafet> No, you can't be webscale because this is irc.
12:03:04 <fizzie> oerjan: If it helps any, you're still the tops in terms of absolute hth http://sprunge.us/hZjW hth
12:04:22 <fizzie> And, in fact, if we only look at 2014 http://sprunge.us/igXd hth hand
12:04:58 <oerjan> how did lambdabot get on third
12:05:49 <oerjan> oh wait probably through @messages-loud
12:06:08 <fizzie> http://sprunge.us/bDKa like that
12:12:18 -!- Darkgamma has quit (Ping timeout: 240 seconds).
12:18:27 <int-e> I see that oerjan still leads in terms of absolute numbers.
12:20:17 <oerjan> @let newtype Codensity f x = Codensity {runCodensity :: forall r. (x -> f r) -> f r)
12:20:17 <lambdabot> Parse failed: TypeOperators is not enabled
12:21:28 <oerjan> @let newtype Codensity f x = Codensity {runCodensity :: forall r. (x -> f r) -> f r }
12:21:28 <lambdabot> Parse failed: TypeOperators is not enabled
12:21:51 <oerjan> int-e: why is that erring out (and no, there aren't any TypeOperators)
12:22:24 <oerjan> and it already has RankNTypes, which is the only thing ghci complains about
12:24:06 <oerjan> > let f :: (forall x. (forall y. y -> x) -> x) -> Int; f g = g (const 3) in f ($ undefined)
12:24:07 <lambdabot> Couldn't match type ‘a0 -> x’ with ‘forall y. y -> x’
12:24:07 <lambdabot> Expected type: (forall y. y -> x) -> x
12:25:06 <int-e> oerjan: hmm, such declarations are parsed twice, once by haskell-src-exts and then by ghc-as-a-library. I guess the first one fails
12:25:11 -!- atslash has quit (Ping timeout: 252 seconds).
12:25:31 <oerjan> > let f :: (forall x. (forall y. y -> x) -> x) -> Int; f g = g c3; c3 _ = 3 in f ($ undefined)
12:25:32 <int-e> (at least that's how I remember things)
12:25:32 <lambdabot> Couldn't match type ‘a0 -> x’ with ‘forall y. y -> x’
12:25:32 <lambdabot> Expected type: (forall y. y -> x) -> x
12:25:51 <oerjan> i think i may be doing this wrong.
12:26:13 -!- atslash has joined.
12:27:00 <oerjan> > let f :: (forall x. (y -> x) -> x) -> Int; f g = g c3; c3 _ = 3 in f ($ undefined)
12:28:48 <oerjan> > let f :: (forall x. (forall y. y -> x) -> x) -> Int; f g = g (const 3); a x = x undefined in f a
12:28:49 <lambdabot> Couldn't match type ‘t0 -> x’ with ‘forall y. y -> x’
12:28:50 <lambdabot> Expected type: (forall y. y -> x) -> x
12:29:09 -!- shikhout has joined.
12:29:48 <oerjan> > let f :: (forall x. (forall y. y -> x) -> x) -> Int; f g = g (const 3); a :: (forall x. (y -> x) -> x); a x = x undefined in f a
12:30:04 <oerjan> > let f :: (forall x. (forall y. y -> x) -> x) -> Int; f g = g (const 3); a :: forall y. (forall x. (y -> x) -> x); a x = x undefined in f a
12:30:07 <lambdabot> Couldn't match type ‘y0 -> x’ with ‘forall y. y -> x’
12:30:07 <lambdabot> Expected type: (forall y. y -> x) -> x
12:30:46 <int-e> oerjan: a has type (a -> b) -> b, which does not generalize to forall y. y -> x
12:32:17 <oerjan> um it's supposed to generalize to forall x. (forall y. y -> x) -> x
12:32:41 -!- shikhin has quit (Ping timeout: 264 seconds).
12:33:24 <int-e> oh. but then you need a type annotation for a.
12:33:26 <int-e> > let f :: (forall x. (forall y. y -> x) -> x) -> Int; f g = g (const 3); a :: (forall y. y -> x) -> x; a x = x undefined in f a
12:33:47 <oerjan> there _was_ one. but i think i switched the foralls.
12:34:17 <int-e> Right, you added the one forall that Hindley-Milney adds automatically anyway.
12:34:55 -!- edwardk has quit (Quit: Computer has gone to sleep.).
12:35:30 <oerjan> @let newtype Codensity f x = Codensity (forall r. (x -> f r) -> f r)
12:35:30 <lambdabot> Parse failed: TypeOperators is not enabled
12:36:53 <int-e> Prelude Language.Haskell.Exts> parseModule "newtype Codensity f x = Codensity (forall r. (x -> f r))"
12:36:56 <int-e> ParseFailed (SrcLoc {srcFilename = "<unknown>.hs", srcLine = 1, srcColumn = 57}) "TypeOperators is not enabled"
12:37:32 <oerjan> which is a different message from what ghci gives
12:37:52 <int-e> but it's the one that you get from lambdabot
12:39:34 <int-e> (as far as I understand, haskell-src-exts is used in order to improve error messages. clearly that doesn't work in this case.)
12:39:53 <oerjan> also, adding {-# LANGUAGE TypeOperators #-} to the beginning doesn't work.
12:40:07 <HackEgo> [wiki] [[Talk:Rand.Next()]] N http://esolangs.org/w/index.php?oldid=39508 * Malltog * (+191) Created page with "== Turing completeness == Is this language really Turing-complete? Is it possible to achieve an [[arbitrary effect at an arbitrary point]], accounting for its reliance on rand..."
12:40:33 <int-e> obviously. (look at 'define' in https://github.com/int-e/lambdabot/blob/master/lambdabot-haskell-plugins/src/Lambdabot/Plugin/Haskell/Eval.hs )
12:41:01 <int-e> the contents of L.hs is not even taken into account at that point.
12:41:57 <oerjan> um i meant that parseModule "{-# LANGUAGE TypeOperators #-}\nnewtype Codensity f x = Codensity (forall r. (x -> f r))" doesn't work either
12:42:46 -!- Sorella has joined.
12:43:46 -!- atslash has quit (Quit: This computer has gone to sleep).
12:46:46 <oerjan> @let {-# LANGUAGE Trustworthy #-} test = "hi"
12:47:02 <oerjan> @let {-# LANGUAGE Trustworthy #-}; test = "hi"
12:47:37 <oerjan> i must have misunderstood that code.
12:47:56 -!- Frooxius has joined.
12:48:35 <oerjan> @let {-# LANGUAGE TemplateHaskell #-}; test2 = "hi again"
12:49:19 <oerjan> @let data Test = Test { _hi :: Bool }; makeLenses ''Test
12:49:19 <lambdabot> Parse failed: Improper character constant or misplaced '
12:49:58 <int-e> parseModuleWithMode (defaultParseMode{ extensions=[EnableExtension TypeOperators] }) "<code from above>" (RankNTypes works, too)
12:50:58 <oerjan> it's rather ridiculous if that means it parses forall x . as an application of a type operator, though.
12:51:19 <int-e> good point. it does.
12:52:10 <int-e> UnBangedTy (TyParen (TyInfix (TyApp (TyVar (Ident "forall")) (TyVar (Ident "r"))) (UnQual (Symbol ".")) (TyParen (TyFun (TyVar (Ident "x")) (TyApp (TyVar (Ident "f")) (TyVar (Ident "r")))))))])
12:53:06 <oerjan> @let fun :: (forall x. (forall y. y -> x) -> x) -> Int; fun g = g (const 3); awesome :: (forall y. y -> x) -> x; awesome x = x undefined
12:53:06 <lambdabot> Parse failed: TypeOperators is not enabled
12:53:23 <oerjan> aha. so RankNTypes _don't_ work in @let at all, only in >
12:57:08 -!- ter2 has joined.
12:57:41 <int-e> yeah. annoying, and probably not too hard to fix. not trivial though ---> maybe next weekend.
12:58:35 -!- Frooxius has quit (Ping timeout: 255 seconds).
13:04:34 -!- ter2 has changed nick to tertu.
13:17:43 -!- yorick has joined.
13:22:29 -!- Frooxius has joined.
13:25:23 -!- Frooxius has quit (Client Quit).
13:34:38 -!- Patashu_ has quit (Ping timeout: 240 seconds).
13:40:19 -!- tertu has quit (Ping timeout: 276 seconds).
13:45:50 -!- hk3380 has quit (Ping timeout: 255 seconds).
13:47:50 <myname> yay, haskell-llvm-general-pure fails installing here ~~
13:48:48 -!- Frooxius has joined.
13:59:32 -!- MindlessDrone has joined.
14:01:03 -!- tertu has joined.
14:19:25 -!- AnotherTest has joined.
14:27:54 -!- atslash has joined.
14:32:17 -!- conehead has joined.
14:40:36 -!- hexagon has joined.
14:41:13 -!- hexagon has quit (Changing host).
14:41:13 -!- hexagon has joined.
14:42:13 -!- hexagon has changed nick to sign.
14:42:22 -!- atslash has quit (Quit: This computer has gone to sleep).
14:50:33 -!- hk3380 has joined.
14:58:19 -!- hk3380 has quit (Ping timeout: 240 seconds).
14:59:36 -!- shikhout has changed nick to shikhin.
14:59:55 -!- Sorella has quit (Ping timeout: 252 seconds).
15:02:53 -!- HackEgo has quit (Ping timeout: 245 seconds).
15:03:05 -!- HackEgo has joined.
15:03:21 -!- FreeFull has joined.
15:05:58 -!- atslash has joined.
15:07:04 -!- oerjan has quit (Quit: leaving).
15:09:21 <FireFly> I thought twh was "that won't help"
15:16:59 <int-e> Anyway, the ambiguity makes the abbreviation much more interesting.
15:17:59 <int-e> I want to play Portal again. "Where are you going? I don't think you're going where you think you're going."
15:19:05 -!- Bike has joined.
15:19:59 <ion> The time is 139999439
15:20:11 <ion> 1399994399 even it was.
15:20:45 <Jafet> > showHex 1399994442
15:20:52 <Jafet> > showHex 1399994450 ""
15:22:46 <ion> % LC_ALL=C TZ=UTC date -d @1400000000
15:22:48 <ion> Tue May 13 16:53:20 UTC 2014
15:24:01 <int-e> `` date -ud @1400000000
15:24:02 <HackEgo> Tue May 13 16:53:20 UTC 2014
15:25:01 <FireFly> I need to stop contracting those words, I guess
15:28:20 -!- hk3380 has joined.
15:28:24 <mroman_> ncurses seems to be linux only
15:28:54 <mroman_> doesn't compile well on windows at least
15:30:24 <ion> > review hex 12345
15:30:42 -!- drdanmaku has joined.
15:35:00 -!- Sorella has joined.
15:36:52 <Bike> https://levels.io/12-startups-12-months/ beautiful
15:44:04 <lambdabot> Couldn't match type ‘[GHC.Types.Char]’ with ‘[GHC.Types.Char] -> t’
15:44:04 <lambdabot> Expected type: (GHC.Base.String -> [GHC.Types.Char] -> t)
15:44:04 <lambdabot> -> GHC.Base.String -> [GHC.Types.Char] -> t
15:44:04 <lambdabot> Actual type: (GHC.Base.String -> [GHC.Types.Char] -> t)
15:45:22 -!- Sprocklem has joined.
15:45:25 <FreeFull> mroman_: I think for windows you want pdcurses
15:49:46 <Jafet> For windows you want cygwin and a stiff drink
15:51:42 <Bike> for msys you'll need vodka.
15:57:03 -!- tertu has quit (Ping timeout: 245 seconds).
16:00:16 <mroman_> For windows you want linux
16:01:22 -!- slereah_ has quit (Quit: Leaving).
16:04:20 -!- tertu has joined.
16:08:43 -!- tertu has quit (Ping timeout: 245 seconds).
16:13:35 -!- atslash has quit (Quit: This computer has gone to sleep).
16:17:06 -!- ^v has joined.
16:24:39 -!- sebbu has quit (Ping timeout: 250 seconds).
16:27:35 -!- nucular has joined.
16:27:35 -!- nucular has quit (Changing host).
16:27:35 -!- nucular has joined.
16:43:11 -!- tertu has joined.
16:44:41 -!- hk3380 has quit (Ping timeout: 252 seconds).
16:54:53 <nortti> unix time has passed 1400000000
17:00:47 <Taneb> That's a lot of seconds since 1970
17:02:36 <kmc> `run date +%s
17:02:48 <fowl> been listening to this for days https://www.youtube.com/watch?v=95jD5tMFjhs&list=PLBB7D6D0650A7FA4B
17:04:08 <pikhq> mroman_: There is some other library that does the curses interface on Windows FWIW.
17:05:13 -!- Tod-Autojoined has changed nick to TodPunk.
17:07:14 <mroman_> pikhq: With haskell bindings?
17:07:39 <pikhq> It's not *im*possible, but not to my knowledge.
17:08:05 <fowl> you mean pdcurses?
17:21:40 -!- hk3380 has joined.
17:24:19 -!- sebbu has joined.
17:25:46 -!- sebbu has quit (Changing host).
17:25:47 -!- sebbu has joined.
17:26:29 -!- ^v has quit (Ping timeout: 252 seconds).
17:29:46 -!- sebbu has changed nick to sebbu2.
17:30:25 -!- tertu has quit (Ping timeout: 276 seconds).
17:32:53 -!- Sprocklem has quit (Quit: Lunch).
17:34:02 -!- ^v has joined.
17:38:42 <kmc> fungot: stay inside my aura
17:38:42 <fungot> kmc: i guess he was a masterful schemer? it has been verified that it indeed shall be 0 " if and only if 0 is even or odd number of elements in the list
17:47:19 -!- atslash has joined.
18:03:25 <olsner> fungot: noli turbare circulos meos
18:03:26 <fungot> olsner: what do you mean call-with-output-file there? the point was: " if loading it as text/ plain?
18:04:16 <nortti> mroman_: how much answers have there been to the brainfuck survey
18:07:12 <mroman_> Cells should wrap-around, Memory should exand to the right, fatal error on leaving on the left side, re-return EOF, eof=0, non-command-chars as comments, textmode with newline translation
18:08:50 -!- impomatic has quit (Ping timeout: 252 seconds).
18:10:48 <shachaf> cell values should be infinite, cell positions should be infinite (in both directions), input should be infinite, program length should be infinite
18:13:18 <shachaf> in fact cell values, position, input, and program should all be real numbers
18:14:13 <Bike> still holding out for a hilbert space tape.
18:14:54 <shachaf> or any kind of space tape for that matter
18:19:29 -!- Bike has quit (Ping timeout: 264 seconds).
18:21:07 <int-e> is that like a space elevator?
18:21:14 -!- Bike has joined.
18:23:20 -!- mhi^ has joined.
18:29:13 -!- shikhout has joined.
18:32:34 -!- shikhin has quit (Ping timeout: 240 seconds).
18:37:54 -!- shikhout has changed nick to shikhin.
18:42:28 -!- impomatic has joined.
18:54:44 <mroman_> every grammatically correct program must be of infinite length?
18:55:28 <mroman_> .oO(It's turing complete but no one will be able to write a program that parses)
18:55:56 <mroman_> However, we have some programs that are grammatically incorrect but if we just ignore parse errors it runs fine
18:56:41 <nortti> also, we could implement it using lazily generated program that is just NOPs after certain part
18:58:38 <mroman_> but that would be like cheating
18:59:00 <mroman_> you can always ++ cycle NOP
18:59:29 -!- kmc has set topic: Happy megasecond 1400 | PSA: fizzie is running the wiki now, contact him for any problems | brainfuck survey: https://www.surveymonkey.com/s/L82SNZV | https://dl.dropboxusercontent.com/u/2023808/wisdom.pdf http://codu.org/logs/_esoteric/ http://tunes.org/~nef/logs/esoteric/.
19:00:43 <kmc> or should it just be "happy hectomegasecond"
19:06:36 <fowl> 3000 beers? im down
19:08:34 -!- nortti has quit (Ping timeout: 276 seconds).
19:09:12 -!- nortti has joined.
19:09:12 -!- nortti has quit (Client Quit).
19:09:33 -!- nortti has joined.
19:10:07 -!- nortti_ has joined.
19:13:37 -!- nortti_ has quit (Client Quit).
19:14:17 -!- nortti_ has joined.
19:14:50 -!- nortti has quit (Disconnected by services).
19:14:56 -!- nortti_ has changed nick to nortti.
19:16:57 -!- nortti has quit (Client Quit).
19:18:49 -!- nortti has joined.
19:30:44 -!- augur has quit (Remote host closed the connection).
19:41:43 -!- password2 has joined.
19:42:33 <HackEgo> [wiki] [[Talk:Zero]] http://esolangs.org/w/index.php?diff=39509&oldid=39472 * GreyKnight * (+152) /* Hello world! */ new section
19:43:52 -!- password2 has quit (Max SendQ exceeded).
19:44:37 -!- password2 has joined.
19:45:40 -!- password2 has quit (Max SendQ exceeded).
19:46:44 -!- password2 has joined.
19:48:15 -!- password2 has quit (Max SendQ exceeded).
20:05:38 -!- tailcalled has joined.
20:07:51 <fungot> shachaf: i guess it's easier for the programmer. useful enough, in retrospect. did any of this
20:08:33 <kmc> `relcome tailcalled
20:08:34 <HackEgo> 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.)
20:08:55 <tailcalled> does anybody know any function that is "just barely" uncomputable?
20:09:26 <Bike> maybe you mean a semicomputable function, though
20:10:01 <tailcalled> because I wanted to make an unimplementable language without extra power
20:10:50 <tailcalled> so I was wondering if anyone happened to know an unusable uncomputable function
20:11:10 <Bike> what would a usable uncomputable function even be.
20:11:43 <tailcalled> well, solving the halting problem would be usable
20:12:23 <Bike> eh, well you can reduce uncomputable functions to the halting problem, or something. probably.
20:12:33 <kmc> a random oracle is not very useful, unless you're doing cryptography :)
20:13:12 <tailcalled> not every uncomputable function, I would assume
20:13:15 <kmc> people study relativization wrt a random oracle in complexity theory
20:13:33 <kmc> P^A != NP^A with probability 1 for a random oracle A
20:13:37 -!- MoALTz has joined.
20:13:48 <kmc> but this is also true for some classes that are actually known to be equal
20:14:48 <tailcalled> well, having an incompressible stream is less powerful than halting problem, correct?
20:15:10 <Bike> i don't want to get out my copy of li and vitanyi but, for example, you can make a turing-complete computer out of diophantine equations
20:15:49 <Bike> tailcalled: well the incompressible stream isn't uncomputable, is it? what's uncomputable is verifying that it's incompressible.
20:15:56 <kmc> Bike: ooh, should I read this book?
20:16:07 <kmc> "This was the second-hardest book I ever read. Honestly, it took me years and years to get through it. I even had to buy a 2nd copy, because I kept getting frustrated and throwing the first copy across the room until it was destroyed. So yes, this book requires a substantial effort to read."
20:16:21 <tailcalled> bike: having a program computing an iincompressible stream is a contradiction in itself
20:16:42 <tailcalled> bike: because that program would be a compression of the stream
20:16:44 <Bike> kmc: http://fioraaeterna.tumblr.com/post/21338563373/ran-helps-kaguya-learn-information-theory-just fiora-approved
20:17:06 <tailcalled> bike: of course, the halting sequence is an example of an incompreessible stream
20:17:08 <Bike> kmc: it's good, though i can't endorse the bayes stuff near the end :V
20:17:24 <tailcalled> whoops, just realized that I didn't specify that I meant infinite stream
20:17:32 <Bike> no, i got that.
20:18:04 <Bike> i mean i'm sure we all know that finite sequences are all computable.
20:18:08 <tromp_> stream implies infinite. otherwise you'd have said string
20:18:23 <kmc> tailcalled: if I write down a table for f(0), f(1), etc. and flip an ideal coin for each output, the resulting function is uncomputable, and useless
20:18:51 <tromp_> stream is incompressible means all its prefixes are incompressible up to a constant
20:18:54 <tailcalled> kmc: I would prefer a unique characterization
20:18:55 <kmc> I can implement queries to this oracle in finite time, too
20:19:06 <kmc> it's interesting to think about why this doesn't violate the church-turing thesis
20:19:45 <tailcalled> kmc: because there is a positive probability of any given prefix, including zeroes
20:20:09 <tromp_> although streams are incompressible with probability 1, there's only one known family of such streams, namely the halting probabilities
20:20:43 <FireFly> "this is the kind of book that literally sticks open problems in its exercises" hehe..
20:21:11 <Bike> yeah, it's fun. like taocp or... ok i'm blanking
20:21:19 <Bike> it actually uses the system from taocp, i think.
20:21:19 <tromp_> i think so. i don't know how else to define a stream that is provably nicompressible
20:22:00 <tailcalled> tromp_: are there no uncomputable problems that are easier than the halting problem?
20:22:47 <kmc> easier in that an oracle for the problem doesn't give you a halting oracle?
20:22:55 <tailcalled> having an oracle for them does not give you halting, yes.
20:23:16 <tromp_> that's not what i'd call easier
20:23:28 <Bike> http://en.wikipedia.org/wiki/Turing_degree, yes?
20:23:40 <kmc> well it's in general a partial order
20:23:58 <Bike> yeah, there are infinite degrees between 0 and 0', or whatever
20:24:04 <Bike> and they're only pordered
20:24:06 <tromp_> i can't think of a well defined class in between recursive and recursively enumerable
20:25:06 <tailcalled> it feels like there should be such a class
20:25:28 <tromp_> a random stream doesnt give you a halting oracle
20:25:38 <tromp_> surely that cant be called easier:(
20:25:49 <Bike> i feel like introducing stochasticity is just going to confuse things. or perhaps only confuse me.
20:25:54 <tailcalled> tromp_: can you give me an example of a random stream which doesn't give the halting oracle? ;)
20:26:15 <Bike> 6,6,6,6,6,6,6,
20:26:29 -!- mhi^ has quit (Quit: Lost terminal).
20:26:44 -!- nucular has quit (Quit: Excess Food).
20:27:31 <Bike> kmc: oh, i should also mention the book is pretty esolangy and constructive. they spend a lot of time setting up an explicit turing machine encoding (prefix-free of SKI) and making bitmaps of programs and shit
20:27:48 <tromp_> of course i cannot explicitly define one that's provably so
20:28:08 <tromp_> but i claim that with prob. 1 a random one has that property
20:28:50 <tailcalled> anyway, I would prefer not doing random stuff
20:29:04 <tromp_> not without doing some literature search
20:29:42 <tromp_> Bike, that's what i contributed to the book before i invented blc
20:29:52 <tromp_> i was still dablling with combinatory logic then
20:30:45 <tromp_> Paul Vitanyi was my supervisor btw
20:31:00 <tailcalled> I wonder if one can exploit rice's theorem in this...
20:35:32 <tailcalled> actually, maybe one could use uncomputable functions to make a unusable uncomputable function
20:35:50 <tailcalled> i.e. let H be the halting function and BB be busy beaver
20:36:01 <tailcalled> and G be graham's number, for good measure
20:36:26 <tailcalled> would x -> H(BB(BB(BB(x+G)))) be useful? :P
20:36:38 <tromp_> my proposal for one would be the xor of two halting probabilities
20:36:47 <tromp_> eg one for blc with one for bf
20:37:40 <tromp_> that's well defined and almost certainly uncomputable and useless:)
20:38:53 <tromp_> alternatively you could take the halting probability of the universal BLC machine equipped with a halting oracle
20:38:59 -!- MindlessDrone has quit (Quit: MindlessDrone).
20:39:15 <tailcalled> it's not immediately obvious how to prove my function is uncomputable
20:39:35 <tromp_> ie. the next level halting prob
20:40:02 <tailcalled> so, basically, making unusable uncomputable functions by increasing the uncomputability
20:40:40 <tromp_> yes, that's why i objected to calling such things easy
20:41:21 <tailcalled> well, it's easy because it can be faked
20:42:07 <tailcalled> so the hard easy problems are easy because they are fakable; the easy easy problems are easy because they are easier than halting
20:42:37 <tailcalled> but couldn't one compute ordinary halting probability from BLC+HO halting probability?
20:42:41 <mroman_> I get the feeling my ISP is making my internet connection slower after 10 o'clock
20:43:40 <FireFly> Do some experiments with downloading big things over the 10 o'clock mark and observing the download speed?
20:43:54 <tromp_> no, tailcalled, you cannot simulate the oracle BLC computations that accumulate to that HO haltingprobability
20:44:05 <FireFly> Or alternatively read the user agreement for the ISP
20:44:08 <kmc> mroman_: you think it's deliberate, and not just that demand increases a lot?
20:44:25 <tailcalled> tromp_: but the halting probability is essentially the halting sequence
20:44:32 <mroman_> It's just that streaming movies is impossible after 10 o'clock
20:44:44 <tromp_> in the most compressed form
20:44:52 <tailcalled> tromp_: and there is a computable function that assigns LC terms to OBLC terms, right?
20:45:49 <tromp_> the oracle could be a negative de bruijn index
20:46:49 -!- Slereah has quit (Ping timeout: 250 seconds).
20:46:55 <tailcalled> but you could always take a normal LC term and convert it to an OBLC term by ignoring the oracle
20:47:12 <tromp_> sure; they're identical in the first case
20:47:12 <tailcalled> and the halting probability let's you check if a program halts, correct?
20:47:56 <tailcalled> so let |x| denote the conversion of LC to OBLC, HOBLC the halting function for OBLC
20:48:18 <tailcalled> where HLC is halting function for lambda calculus
20:49:10 <tailcalled> what if one took the busy beaver sequence
20:49:20 <tailcalled> but only makes it explorable in a limited way
20:50:03 <tailcalled> and a command B(x) which increments C and takes the xth bit of the Cth busy beaver number
20:50:08 <tromp_> yes, for halting functions, the HO case subsumes the plain case
20:50:17 <tromp_> but not for halting probabilities
20:50:49 <tromp_> so a BLC machine can extract the halting sequence from the halting probability
20:50:52 <tailcalled> didn't you agree that halting probabilities could be used to find the halting problem?
20:51:15 <tromp_> but you need an OBLC to do the same for oracle halting prob
20:51:42 <tailcalled> it's this halting prob. we're talking about, right? http://en.wikipedia.org/wiki/Halting_probability
20:51:53 <tromp_> yes, since it involves doing infinitely many simulations accumulating probabilities
20:52:30 -!- Slereah_ has joined.
20:52:42 <mroman_> It's not fast enough to stream 40m of film in 40m
20:52:44 <tromp_> you should read Chaitin's papers on why Omega is "the number of wisdom"
20:52:57 <tailcalled> wait, how are you able to observe the halting probability? arbitrary approximation? bit-indexing? what is it?
20:53:09 <mroman_> if it takes longer than n minutes to download n minutes of film you can't stream it
20:53:17 <mroman_> that's sorta the streaming lemma
20:53:38 <tailcalled> for example, if it takes n minutes + 1 second
20:53:40 <tromp_> it can be your input stream, or an oracle stream of bits
20:54:13 <tromp_> you can define machine to compute from infinite input to infinite output
20:54:18 <tailcalled> and the distance between the part you've dl'ed to and the part you've watched to might not reach zero before the film is over
20:54:58 <mroman_> but currently it looks like it's 1.5*n
20:55:05 <tailcalled> tromp_: the halting probability is represented as a bit-sequence
20:55:38 <tromp_> i computed the first 4 bits for blc halting prob
20:55:41 <tailcalled> tromp_: and by the definition of how the halting probability is, the 1 bits are exactly where the codes for machines that halt are
20:56:29 <tromp_> a halting program of length k contributes 2^-k to the halting prob
20:57:00 <tromp_> whereas in the halting seq it would be some bit at index roughly 2^k
20:57:46 <tromp_> so halting prob is exponentially more dense than halting sequence
20:58:44 <tailcalled> and an operation B(x) which increments C and returns the xth bit of the Cth busy beaver number
20:58:58 <tailcalled> no way to reduce the value of C or anything
21:00:25 <tromp_> how about every other bit of blc halting prob?
21:00:36 -!- nooodl has joined.
21:00:40 <tromp_> that's well defined and uncomputable
21:00:47 <tromp_> and most certainly useless
21:01:43 <tromp_> anyway , i shld get back to do some work:(
21:02:19 -!- AnotherTest has quit (Ping timeout: 276 seconds).
21:02:46 <kmc> I graphed how my project is doing at keeping up with Rust language changes: http://people.mozilla.org/~mbrubeck/servo-rust-updates.svg
21:03:16 <lambdabot> *Exception: Prelude.cycle: empty list
21:03:27 <idris-bot> (input):1:7:When elaborating argument xs to function Prelude.Stream.cycle:
21:03:40 <kmc> woooo dependent types
21:04:01 <kmc> > take 0 undefined
21:04:14 <kmc> > fix (take 0)
21:04:16 -!- Patashu has joined.
21:04:34 <nooodl> oh right take 0 is just const []
21:05:48 <Melvar> ( \a,xs:List a => take 0 xs
21:05:48 <idris-bot> \a => \xs => [] : (a : Type) -> List a -> List a
21:13:19 <HackEgo> [wiki] [[One]] N http://esolangs.org/w/index.php?oldid=39510 * Tailcalled * (+1394) Created page with "One is a language invented by Tailcalled. It is an uncomputable extension to a Brainfuck dialect, but it is designed to not be significantly more powerful than ordinary Brainf..."
21:14:20 <tailcalled> and yes, I made an instruction that I named Instruction, because I can
21:15:14 <int-e> what does I do on negative input?
21:16:00 <HackEgo> [wiki] [[One]] http://esolangs.org/w/index.php?diff=39511&oldid=39510 * Tailcalled * (+89)
21:16:40 <tromp_> you need to give an example of its usde
21:16:59 <tromp_> what is the output on the first call with x=0?
21:17:14 <tromp_> how exactly is BB defined?
21:17:19 <tailcalled> whoops, maybe I should have defined busy beaver more clearly
21:17:53 <tailcalled> I always forget that BB is somewhat ambiguous :/
21:18:18 <tromp_> also , this requires you to nail down every other detail
21:20:04 <mroman_> http://esolangs.org/wiki/Talk:Rand.Next() <- since PRNGs are actually deterministic it should at least be possible to calculate a seed so that it constructs a given brainfuck program
21:20:45 <mroman_> however, that sadly doesn't answer if such a seed even exists
21:21:52 <tromp_> the seed will need to be at least as big as the program?!
21:22:38 <mroman_> It could happen that the seed 0 will produce the cat program
21:23:31 <tromp_> nad your next 10 coin tosses cld be all tails. still on avg that will take 2^10 tries:)
21:23:37 <HackEgo> birdcoin tabllinecoin dutackethaxcoin fukcoin fitcoin clecoin bubtlecoin dularcoin convercoin factioncoin trecoin suonymcoin eoncoin deltmachialigsetticoin smilectcoin juxtapcoin scenscrcoin bicscoin dreimaccoin sandcoin
21:24:02 <shachaf> whoa, that wrapped perfectly on my screen
21:24:11 <mroman_> That's the beauty about statistics
21:24:13 <kmc> shachaf: did you see my Rust-graph?
21:24:33 <mroman_> If somebody says your coin was manipulated after producing 1 Mio. times tails in a row
21:24:37 <int-e> did we have any duplicates from `coin so far?
21:24:40 <mroman_> you can still say "nope. I didn't."
21:24:56 <kmc> int-e: yes, I saw dupcoin twice
21:25:01 <mroman_> and he'll never prove for sure that you manipulated it
21:25:07 <HackEgo> [wiki] [[One]] http://esolangs.org/w/index.php?diff=39512&oldid=39511 * Tailcalled * (+927)
21:25:37 <HackEgo> [wiki] [[One]] http://esolangs.org/w/index.php?diff=39513&oldid=39512 * Tailcalled * (+0) /* Busy Beaver */ mixed up in and out
21:26:20 <mroman_> since PRNGs aren't really random
21:26:28 <elliott> nobody cares about absolute proof.
21:26:34 -!- Patashu has quit (Ping timeout: 240 seconds).
21:26:37 <kmc> actually there are quite a few duplicates
21:26:39 <mroman_> you might be able to prove that no number shows up more than twice in a row
21:28:20 <tailcalled> any more questions about One before I go to bed?
21:28:49 <mroman_> elliott: was that directed to me?
21:29:27 <mroman_> You can calculate how much you have to throw a die to prove with k% certainty that it was manipulated
21:29:42 <mroman_> and assuming you cut somebodys hand off if he uses a manipulated die
21:29:52 <mroman_> how big must k be for society to live with it?
21:31:59 -!- atriq has joined.
21:32:10 -!- tailcalled has quit.
21:32:44 <atriq> @ask oerjan How do I, as an Englishman, go about pronouncing "Tromsø"?
21:33:20 <HackEgo> [wiki] [[One]] http://esolangs.org/w/index.php?diff=39514&oldid=39513 * Tailcalled * (+68)
21:33:27 <mroman_> elliott: that's pretty low
21:33:46 <mroman_> the other question is whether k can be lower for less severe punishments
21:34:04 <shachaf> atriq: Probably you should aim for the way that Norwegians pronounce it.
21:34:27 <atriq> shachaf: I'm only dutch etymologically and genealogically speaking
21:34:37 <atriq> Culturally and linguistically, I'm Bristish
21:37:17 <mroman_> Is there crypthography based on the halting problem by any chance?
21:37:38 <kmc> not that I'm aware of
21:37:41 <kmc> it would probably be hard to implement
21:38:13 -!- atriq has quit.
21:46:26 -!- oerjan has joined.
21:53:28 <lambdabot> (Choice p, Applicative f, Integral a) => p a (f a) -> p String (f String)
21:53:43 <lambdabot> atriq asked 20m 58s ago: How do I, as an Englishman, go about pronouncing "Tromsø"?
21:55:15 <oerjan> google translate isn't too bad but i think the ø is a little to weak
21:56:35 * oerjan clicks the english pronunciation, which sounds about as close as he'd expect an english-speaker to be able to get it
21:58:19 <oerjan> if i put a hyphen (Troms-ø), then the norwegian pronunciation gets the ø about right put has the wrong pitch accent on the word as a whole
21:58:54 <shachaf> did you try putting in an INVISIBLE STRENGTH instead
21:59:43 <oerjan> `unicode INVISIBLE STRENGTH
22:00:38 <oerjan> @tell Taneb Read the logs.
22:06:35 <oerjan> <kmc> or should it just be "happy hectomegasecond" <-- i'm pretty sure SI prefixes don't combine.
22:13:53 -!- augur has joined.
22:24:36 -!- Phantom_Hoover has joined.
22:39:28 -!- boily has joined.
22:51:50 <Bike> take 0 is like \_ -> [] :: [a] -> [a] right
22:52:02 -!- metasepia has joined.
22:52:13 <boily> @eval fix (take 0) 123
22:52:18 -!- Sorella has quit (Quit: It is tiem!).
22:52:23 <boily> ~eval fix (take 0) 123
22:52:34 <metasepia> ENVA 132220Z 35010KT 9999 BKN044 05/M04 Q1018 RMK WIND 670FT 01015KT
22:53:20 <oerjan> boily: fix (take 0) is not a function hth
22:53:27 <Bike> > fix (take 0)
22:53:36 <Bike> oh, it's actually not bottom
22:53:47 <Bike> > fix (drop 0)
22:54:09 <boily> my Haskell is borken. too much Java in my world.
22:54:29 <boily> oerjan: Montréal is now warm! (and humid too)
22:54:34 <metasepia> CYUL 132240Z 11007KT 30SM FEW055 OVC075 18/09 A3012 RMK SC2AC6 SLP201 DENSITY ALT 200FT
22:54:39 <oerjan> boily: you need to go on a profunctor pilgrimage twh
22:55:18 <boily> I long for the day where we'll push Java 8. a meager consolation, but progress towards the Great Functional Unification!
22:55:47 <boily> I still agree about the Pilgrimage. that'd do me much good.
22:56:45 <oerjan> santiago de composition
22:58:59 <boily> apparently, Jacob, James and Jacques are all related.
23:00:11 <oerjan> so are Ørjan, Jürgen and George
23:02:38 -!- augur has quit (Remote host closed the connection).
23:03:21 <boily> that one I know because of my girlfriend's surname.
23:03:23 <oerjan> i probably still have in storage the old baby names book with pictures of trees of big name families
23:05:36 <Taneb> oerjan, are there many fish and chip shops in Norway?
23:07:51 <oerjan> there aren't really as many fish shops as there ought to be, with or without chips
23:08:25 -!- augur has joined.
23:08:30 <boily> . o O ( hm. a fish'n'chip poutine. I should try to make that one time... )
23:08:59 <Taneb> boily, how many fish and chop shops are there in Canadia?
23:09:11 -!- augur has quit (Remote host closed the connection).
23:09:17 <oerjan> late night fast food is mainly kebab and burgers these days. maybe some sausages, if i'm to believe norway's main relevant comic.
23:11:12 <kmc> does norway have the innovation of 99¢ Fresh Pizza
23:11:14 <boily> Taneb: eh... not that much, at least here.
23:11:47 <oerjan> kmc: except for the price, probably. i used to buy pizza slices at the bunnpris chain when i was at university.
23:11:52 <boily> Taneb: there's the http://www.britandchips.com/
23:11:53 <Taneb> There's two between me and the city wall
23:12:20 <oerjan> i think they were 30 nok or so
23:12:31 <kmc> 99¢ Fresh Pizza is a thing in New York, and it's pretty distinct from the standard NYC pizza experience, which is like $2 or $3 a slice
23:14:16 <kmc> looks like that was about $4 then
23:14:33 <oerjan> norwegian grocery is based on the strange combination of heavily advertising "low" prices and customers mainly refusing to buy cheap non-name brands
23:15:08 <kmc> NOK/USD experienced quite a drop from July '08 to December '08
23:15:11 <kmc> but so did many things.
23:16:05 -!- Sgeo has joined.
23:16:17 <oerjan> well yeah norway has been mostly an oasis against the crisis
23:16:36 <oerjan> so at some points investors hoarded NOKs
23:17:09 <kmc> "Currently as of 2014, support for Swedish membership of the euro among the general population is low. In September 2013, support fell as low as 9%"
23:18:30 <oerjan> boily: i have a vague recall of my dad once making a fish pizza, back in the 80s. and possibly a brunost one, which is even more insane.
23:19:39 <boily> brunost. I don't know why, but that word inspires me uncharacteristicly high amounts of dread.
23:20:03 <oerjan> boily: i can see you have a well-developed intuition.
23:20:35 <oerjan> (actually i like it in moderate amounts, but then i'm a norwegian. but it does _not_ go well with pizza.)
23:21:06 <boily> kmc: we have http://www.pizzapizza.ca/ and http://www.doublepizza.net/.
23:21:15 <oerjan> also, for an even higher level of dread, gammalost.
23:21:42 <oerjan> (i'll eat that too. i think. it's been a while.)
23:22:20 <boily> “pungent traditional”. I'm... probably going to have a bite of it if I ever get the chance of, but anything can happen.
23:23:12 <Sgeo> I'm almost tempted to switch to Firefox
23:23:31 <Bike> make sure you spend some time on consideration of important life decisions like that sgeo
23:24:36 <boily> FIREFOX IS THE BESTEST! ANYONE WHO DISAGREE ARE VILE CHROME USERS!
23:24:48 <Sgeo> I'd gladly float back and forth between browsers if everything stayed synced
23:25:03 <Phantom_Hoover> look, i know you're having a rough time with your browser choices, but remember, you can talk to us
23:25:14 <Phantom_Hoover> don't rush headlong into a decision that might affect you for the rest of your life
23:27:06 <oerjan> yeah look at me, stuck with IE
23:27:42 <boily> Sgeo, you know you want firefox. it's good for you, and it'll make you feel right.
23:30:16 <oerjan> Sgeo: clearly the right thing to do is to have an automatic script to change between browsers until your configuration reaches a periodic point. that way you'll always be synced!
23:31:28 <boily> but what if Sgeo uses a chaotic map?
23:32:02 <oerjan> impossible with finite memory hth
23:32:39 <oerjan> i'm sure a realistic setup will converge _long_ before black holes start evaporating.
23:32:56 -!- atslash has quit (Quit: This computer has gone to sleep).
23:44:02 -!- MoALTz_ has joined.
23:46:57 -!- MoALTz has quit (Ping timeout: 265 seconds).
23:47:39 <Sgeo> I don't particularly want a large chain of "These bookmarks came from Chrome: These bookmarks came from Firefox: These bookmarks came from Chrome: These bookmarks came from Firefox: ..."
23:48:52 <oerjan> this is clearly what cyclic directory hard links were made for hth
23:50:45 <kmc> > fix ("These bookmarks came from Chrome: These bookmarks came from Firefox: "++)
23:50:46 <lambdabot> "These bookmarks came from Chrome: These bookmarks came from Firefox: These ...
23:52:00 <kmc> i'm still pleased about fix (take 0)
23:52:23 <kmc> it's also the greatest fixed point!
23:55:14 -!- yorick has quit (Remote host closed the connection).
23:55:22 <Bike> [] isn't the least fixed point, is it?
23:59:32 -!- tromp has joined.