00:13:23 -!- adu has joined.
00:18:05 -!- ProofTechnique has joined.
00:22:44 -!- ProofTechnique has quit (Ping timeout: 264 seconds).
00:27:33 -!- Phantom__Hoover has joined.
00:29:13 -!- fractal has quit (Ping timeout: 265 seconds).
00:30:11 -!- aretecode has quit (Ping timeout: 265 seconds).
00:30:15 -!- Phantom_Hoover has quit (Ping timeout: 250 seconds).
00:32:50 -!- aretecode has joined.
00:34:25 -!- Tritonio has joined.
00:37:02 -!- fractal has joined.
00:46:21 <oren> I usually run long-running jobs on the university's computer
00:47:56 <oren> because as some of you know, my personal computer is total crap
00:54:42 <boily> helloren. what kind of long-running jobs are they?
00:59:49 <oren> graphics rendering, ai training, that sort of thing
01:19:06 -!- ProofTechnique has joined.
01:24:25 <boily> elliott: helliott. I find this acceptable: the +1 long sword of Punishment (weapon) {drain, +Blink +Fly rElec rPois Dex+2}
01:26:11 <elliott> that's a very caster weapon, heh
01:27:16 -!- zemhill_ has joined.
01:27:54 -!- zemhill has quit (Remote host closed the connection).
01:29:17 <oren> #define up2(i,n) for(i=0;i<(n);i++)
01:29:33 <elliott> boily: your term is slightly less huge than usual. I find this commendable.
01:33:32 <oren> #define up3(i,q,n) for(i=(q);i!=(n);i++)
01:34:00 <pikhq> #define malloc(x) ((void*)&(x))
01:34:44 <oren> #define new(T) (T*)malloc(zv(T))
01:34:58 <oren> #define zv(T) sizeof(T)
01:35:41 <pikhq> #define return for(;;)
01:36:54 <oren> #define zzzzz do{printf("FUUUUUU");exit(374872);}while(1)
01:36:57 -!- zzo38 has joined.
01:37:20 <pikhq> On a vastly more serious note, unshare() is neat.
01:38:32 <boily> elliott: my term is slightly huger when I'm on my laptop hth
01:38:43 <elliott> that sounds a bit backwards.
01:38:46 <boily> elliott: also, I'm experimentaling with my skills.
01:39:10 <boily> my laptop's display is 1920×1080, whereas my desktop's is 1680×1050.
01:39:47 -!- Phantom__Hoover has quit (Read error: Connection reset by peer).
01:41:32 <oren> #define randi(l,u) rand()%(u-l+1)+l
01:42:42 <boily> I got splattered to smithereens.
01:43:03 <boily> mismanaged the mob and I was exploded into tiny slimy bloody fragments.
01:43:44 -!- Tritonio has quit (Ping timeout: 252 seconds).
01:47:13 -!- Lymia has quit (Ping timeout: 250 seconds).
01:47:21 -!- oerjan has joined.
01:48:13 -!- Froo has joined.
01:52:01 -!- Froox has quit (Ping timeout: 264 seconds).
01:57:16 -!- FireFly has joined.
02:17:19 -!- hjulle has quit (Ping timeout: 245 seconds).
02:29:33 -!- Lymia has joined.
02:37:46 <boily> quinthellopia! DCSS.
02:38:08 <boily> well, playing is a fancy word. I'm more down to earth, back to dying simply and effectively.
02:39:33 <oerjan> always look at the bright side of death
02:40:18 <boily> hellørjan. the more practice you have dying, the better you get at avoiding it!
02:46:05 * boily is thinking to himself “I am not being called Girly Molly. I am not being called Girly Molly. Brain, stop being dyslexic.”
02:46:43 <oerjan> ye brane, staff beeing disliking
02:46:57 -!- dianne has quit (Ping timeout: 245 seconds).
02:48:04 <boily> I could be Molly Grue for all I know... self identity is such a hazy concept.
02:48:55 -!- dianne has joined.
02:55:24 -!- boily has quit (Quit: UNICORN CHICKEN).
02:59:54 -!- GeekDude has quit (Quit: {{{}}{{{}}{{}}}{{}}} (www.adiirc.com)).
03:19:36 -!- GeekDude has joined.
03:22:08 -!- GeekDude has quit (Client Quit).
03:46:19 <oren> confirmed: a single-layer perceptron utterly fails at doing xor
03:48:24 -!- MDude has changed nick to MDream.
03:49:18 <oren> But ti works with a 2-2-1 network, thus showing that my code is correctly implemented. ieiii!
04:13:29 -!- ProofTechnique has quit (Ping timeout: 246 seconds).
04:52:13 -!- bb010g has joined.
04:54:07 <Sgeo> https://www.reddit.com/r/videos/comments/2ujjl8/he_may_be_british_but_the_point_applies_in_the_us/
04:54:20 <Sgeo> Horrifically anti-intellectual video or important point about missing priorities?
04:54:40 <Sgeo> I tend to think it's both, that the things people aren't taught in school that he mentions absolutely should be
04:54:55 <Sgeo> But don't really see a downside to teaching those 'useless' things as much
04:55:06 <Sgeo> (Except that priotization is a thing)
05:12:01 <oren> Most of the things he mentions were taught to me by my parents
05:12:59 <Sgeo> I don't think it's necessarily a good idea to rely on that in general though
05:13:33 <Sgeo> Especially for important-but-not-yet-commonly-known things, such as recognizing signs of mental illness. Can't teach how to treat or perfect detail maybe, but can teach at least some recognition signs
05:14:51 <oren> True, that one is a good idea
05:15:37 <oren> But how to vote, pay taxes or budget, and the laws, were all taught to me by my parents
05:17:01 <oerjan> hm i don't think my parents taught me any of those
05:17:03 <Sgeo> I still don't know how to budget >.> I just sort of assume that if I set aside 20% of my income as 'allowed to spend frivolously' I'm good. Except that that account has become my -pay stuff online- account, which isn't perfectly correlation with -stuff I want but don't need-
05:17:18 <oerjan> although my mom was darn good at budgeting so i'm sure she would have if i'd asked
05:17:55 <Sgeo> Also going to need to figure out taxes for the first time this year
05:18:22 <oren> One thing that he didn't mention is cooking. Absolutely everyone should be able to cook a meal for themselves.
05:18:23 <Sgeo> (Last year my dad did them for me. Totally educational. I need to stop relying on him)
05:18:26 <oerjan> (i suppose they may have told me _some_ laws. i'm still unsure about many.)
05:20:55 <oerjan> i knew how to do my taxes once, but the government has made it so easy that i don't need to any more
05:21:10 <oren> The best point he makes imo is the one about history. I only know what the Cold War was abut because my father explained it. it was never covered in school.
05:22:25 <oren> They covered c. 1912-1945 when they should have covered 1950-2000
05:23:16 <lifthrasiir> education was never free from the influence of politics
05:25:23 <oerjan> (basically everyone who doesn't have a complex economy (like running your own business or the like) gets a suggested tax form from the government, and if you basically agree with it you don't need to do _anything_)
05:28:52 <oerjan> the insane part is that this makes the tax authorities the most service-minded branch of the norwegian government, with the best publicity.
05:28:57 <oren> "you won't have a calculator every day" -- only true in primary education system. In University either the numbers are so small you don't need them, or you DO get a calculator
05:29:27 <oerjan> ...until you actually _disagree_ with them, that is. then things can get rather ugly.
05:30:19 <oren> I don't know my times tables, and this has not been a detriment since grade 11.
05:30:33 <oerjan> (but i don't expect they're worse than other countries in that respect.)
05:32:55 <Sgeo> There were a few current event projects in my school, where we were tasked with writing about whichever current event we wanted
05:33:12 * oerjan does know his times tables. helps a lot with that calculator crossword in the newspaper. also, prime numbers.
05:33:31 <Sgeo> The thing about calculators I disagree with though. Yes, we carry around calculators, but there's no way any of my teachers could have predicted that.
05:33:50 <Sgeo> Most teachers can't see the future.
05:34:04 <oren> When i was in school hand calculators were already universal
05:34:21 <oren> it was the 90's after all
05:34:26 <Sgeo> Would you have believed you would carry one in your pocket every day?
05:34:32 <Sgeo> For your adult life?
05:35:03 <oren> My friend had one on his keychain, so yes.
05:36:33 <oren> I mean-- I was born in 1993. There were cell phones, laptops, etc throughout my childhood
05:37:04 <oerjan> the year september never ended
05:37:42 <oerjan> that's how i know i'm old on the internet, i was there _before_ that.
05:38:23 <oren> My dad told be email addresses used to look like name!name!name
05:38:34 <Sgeo> I had crappy laptops during my childhood
05:38:59 <Sgeo> I remember bringing a DOS laptop to my babysitter's house when I was a kid
05:39:21 <zzo38> I often make calculation by hand or by mind if it is simple enough, although I also have a TI-92 calculator that I usually have when traveling somewhere
05:39:35 <oren> My mom had a IBM DOS laptop that she made my dad carry because it was heavy as hell.
05:42:57 <oerjan> i still have the Oric-1 down in the storage room (technically it's my dad's), i think if i tried to turn it on the neighbors would call the cops on me, that had some _ugly_ radio interference.
05:45:12 -!- adu has quit (Quit: adu).
06:28:45 -!- Froo has quit (Quit: *bubbles away*).
07:31:00 -!- Patashu has joined.
07:34:32 -!- chaosagent has quit (Read error: Connection reset by peer).
07:59:19 <zzo38> The rules of golf says that ice can be considered as casual water or as loose impediments at your choice. If it is loose impediments then you can remove it, but what happen if both balls are on the putting green, whoever ball is farther away from the hole wants to remove it, is the other player allowed to put it back exactly where it was before in order to treat it as casual water or you have to do without it instead?
08:03:50 -!- FreeFull has quit (Ping timeout: 246 seconds).
08:05:40 -!- tromp__ has joined.
08:05:40 -!- zzo38 has quit (Read error: Connection reset by peer).
08:09:16 -!- tromp has quit (Ping timeout: 255 seconds).
08:09:25 -!- zzo38 has joined.
08:12:05 <zzo38> What is "/w00tw00t.at.blackhats.romanian.anti-sec:)"?
08:18:55 <b_jonas> zzo38: well, if both players can choose to remove or replace the ice at will any number of times, then clearly the non-active player gets to decide the outcome due to the infinite loop rules.
08:22:07 <zzo38> I get other strange HTTP requests too such as for "/lolo/lol/lo.php" and "/tmUnblock.cgi" and "/wqwq/wqw/wq.php" as well as one that doesn't have a valid request method; the entire request is "\xa7\x02" without GET or anything else.
08:22:57 <b_jonas> zzo38: maybe that tried to be some request for some protocol other than HTTP
08:23:42 <zzo38> Well, it isn't a protocol I recognize. Other malformed requests include "\xff\xff\xff\xff@\x01r" and "\xff\xff\xff\xff@\x01s"
08:24:10 <zzo38> (I also don't know what "/muieblackcat" is.)
08:24:16 <zzo38> Do you recognize any of these?
08:25:01 -!- Frooxius has joined.
08:25:10 <oren> I reimplemented this font http://www.arcavia.com/Software/ProgFont/. http://snag.gy/lkMAc.jpg
08:26:52 <oren> All kinds of crazy code is running out there. Could be a malformed student project, a malfunctioning botnet, who knows
08:30:10 <zzo38> Also "/rom-0" and "/back.css" and "/tag_products.php?id_tag=%27" and "/w00tw00t.at.ISC.SANS.DFind:)" and "/1gophserv" (this last one seems to be a misinterpretation of a gopher URL as HTTP). And even "/user/soapCaller.bs" and various requests for stuff in /cgi-bin/ (I have nothing there) and many things that are trying to be proxy requests.
08:31:52 <oren> requests for cgi-bin are possibly hacking attempts?
08:32:24 <zzo38> I would think they try to see what software is installed. Well, there is no such file.
08:33:34 <zzo38> However, they try the same file too often, and they shouldn't do that after you can already see there is no such files.
08:46:40 <b_jonas> zzo38: I've got /muieblackcat in my apache logs, let me see the others
08:47:44 <zzo38> Do you have files with any of those names on your computer?
08:48:01 <b_jonas> well, let me locate, but definitely not in the HTTP
08:48:04 -!- Frooxius has quit (Quit: *bubbles away*).
08:49:48 <b_jonas> I've got requests for /rom-0 (maybe it's some router or other embedded device thing), requests with User-Agent starting with () to try to use that bash vulnerability,
08:50:59 <Jafet> http://blog.malwaremustdie.org/2013/10/a-disclosure-of-whats-behind-w00tw00t.html
08:51:41 <b_jonas> and yes, I've got /w00tw00t.at.blackhats.romanian.anti-sec:) too
08:52:04 <b_jonas> what other files did you say? let me see
08:53:16 <b_jonas> and also /user/soapCaller.bs
08:53:27 <b_jonas> so basically all of them, except /1gopherserv
08:57:08 <zzo38> /1gophserv is specific to my computer because probably someone tried to follow a gopher:// URL but it changed to http:// and then it didn't work.
08:58:54 <b_jonas> I get hits to files that actually exist, very often apparently from the bots of google, yahoo, and baidu
09:01:05 <b_jonas> zzo38: have you got any of the bash bug requests?
09:03:27 <zzo38> I don't know; I cleaned the log now and don't log user-agents
09:03:34 <mroman> If you multiply an MVar Double with a Double
09:03:56 <mroman> in a function like foo someMVar someDouble
09:04:07 <mroman> it will say: Expected type MVar Double -> MVar Double
09:06:36 <mroman> http://codepad.org/dE059AxK
09:06:55 <mroman> or is there a Num instance for MVars?
09:07:08 <mroman> I'd expect it to report No Instance Num (MVar Double)
09:07:30 <b_jonas> wouldn't you need to read the mvar and then >>= multiply it
09:08:37 <b_jonas> like, fmap (2.7*) :: MVar Double -> IO Double
09:48:10 -!- Tritonio has joined.
09:59:22 <oerjan> mroman: pretty sure your last $ var should be $ value hth
10:02:16 <oerjan> also that error message doesn't contain "Expected type MVar Double -> MVar Double" so what are you babbling about
10:06:53 <oerjan> you _could_ actually make a Num instance for MVars, couldn't you. or even an Applicative instance. it would be weird and not very thread-safe...
10:07:41 <oerjan> not without unsafeSomething.
10:07:57 -!- Phantom_Hoover has joined.
10:08:59 <oerjan> otoh if you had newtype CrazyApp a = CrazyApp (IO (MVar a))
10:30:36 <Jafet> instance (Appicative f, Num a) => Num (f a)
10:31:32 <oerjan> that's what i was alluding to
10:33:10 <oren> Between 1788 and 1868, approximately 165,000 convicts were transported to various parts of Australia. All of them were category theorists.
10:41:47 -!- SopaXorzTaker has joined.
10:58:58 -!- Patashu_ has joined.
10:58:58 -!- Patashu has quit (Disconnected by services).
10:59:58 -!- Patashu_ has quit (Client Quit).
11:06:27 -!- Patashu has joined.
11:10:52 -!- oerjan has quit (Quit: leaving).
11:17:50 <Jafet> The nvicts were hanged.
11:24:32 -!- Tritonio has quit (Ping timeout: 252 seconds).
11:26:40 -!- boily has joined.
12:00:31 <J_Arcane> Well, this is some wild stuff. Like Haskell in Lisp-ish syntax. https://github.com/ympbyc/Carrot
12:08:55 -!- oren has quit (Quit: Lost terminal).
12:22:18 -!- boily has quit (Quit: WHITEPOINT CHICKEN).
12:27:59 -!- Koen_ has joined.
12:28:44 -!- Phantom_Hoover has quit (Ping timeout: 264 seconds).
12:31:13 -!- vanila has joined.
12:33:32 <vanila> http://esolangs.org/wiki/--C-%3DC-C--
12:33:35 <vanila> how is this turing complete?
12:34:54 -!- hjulle has joined.
12:40:34 <Jafet> You can write hello world and 99 beers in it, qed
12:49:29 -!- oren has joined.
12:53:52 -!- Patashu has quit (Ping timeout: 240 seconds).
13:29:04 <vanila> is C++ templates an esolang
13:31:06 -!- copumpkin has quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…).
13:31:45 -!- ProofTechnique has joined.
13:31:46 <oren> c++ templates is some kind of language
13:43:05 <vanila> http://en.wikipedia.org/wiki/Barton%E2%80%93Nackman_trick
13:43:22 <vanila> this is like math or something
13:50:45 <oren> everything is like math or something. usually like something, but sometimes like math.
13:54:11 <Jafet> Nah, this is a pointer.
14:01:47 <oren> But basically this whole trick is just a workaround of the fact that c++'s algolish syntax doesn't define which operand of == is the "self".
14:02:06 <b_jonas> algolish as opposed to smalltalkish?
14:03:03 <oren> as opposed to all the excellent alternatives
14:03:53 <oren> for example, you could make a rule that a==b is always a.equals(b)
14:04:36 <b_jonas> oren: that would be a bad idea, because then you could no longer write (0 == smartpointer) or write (nullptr == smartpointer)
14:05:24 <Jafet> You can, in fact, have two different definitions for a==b and b==a
14:05:52 <oren> good point. maybe b.equals(a) would be better for the constant == variable convention (which is only necessary because they used = for assignement)
14:06:24 -!- FreeFull has joined.
14:16:16 <oren> http://snag.gy/uZKT8.jpg
14:19:02 <oren> C++'s resolution rules are way too complex enyway
14:19:06 <vanila> lol the size difference
14:19:16 <vanila> btw you might be interested in PNG
14:22:20 <oren> vanila: Oh, you mean in the picture. yah, artists cometimes screw up the perspective like that
14:25:06 <myname> quick haskell question: in an Array Int (Map foo bar) will i be able to somehow update the map without changing the array?
14:26:24 <Jafet> Array e a is just a faster [a]
14:27:07 <myname> that doesn't answer the question
14:27:29 <Jafet> You're right, it doesn't
14:27:58 <myname> like: can i update a single element?
14:28:11 <oren> does haskell have mutable state?
14:28:31 <vanila> you could use a mutable array instead of Array
14:29:04 <myname> isn't that horribly slow at writing?
14:29:15 <vanila> I don't think so, I dont really know though
14:30:16 <tromp_> not slow at all. it just mutates in place
14:31:39 <dulla> was there any progress on diff arrays to not suck
14:31:43 -!- FireFly has quit (Changing host).
14:31:43 -!- FireFly has joined.
14:32:35 <tromp_> yes, mutable arrays live in IO or ST monad
14:33:14 <Jafet> If you don't know, use Data.Sequence.
14:33:59 <vanila> ST is really cool because it is mutable internally, but can be used in a pure way
14:36:10 <dulla> set aside the state
14:36:28 <dulla> when you use the state, it's impure
14:36:29 -!- ProofTechnique has quit (Ping timeout: 245 seconds).
14:37:06 <dulla> boxed versus unboxed
14:42:31 <myname> okay, sequence is too slow
14:45:49 -!- FreeFull has quit (Ping timeout: 256 seconds).
14:48:57 <TieSoul> http://www.reddit.com/r/dailyprogrammer/comments/2ww3pl/2015223_challenge_203_easy_the_start_of_something/coyxvoj
14:49:21 <TieSoul> most of the program is just gibberish.
14:52:37 <TieSoul> of course, the program isn't actually square
14:52:48 <TieSoul> it just looks square when printed in plaintext
14:54:35 <vanila> http://www.netfunny.com/rhf/jokes/88q3/0tenny.html
14:54:41 <vanila> saw this on the wiki..
14:57:32 <TieSoul> https://gist.github.com/TieSoul/0e116f1156e2ae1d5ffe here's the "clean" version
14:58:47 <TieSoul> also, Gist has a Befunge language category, but no highlighting.
15:02:41 -!- `^_^v has joined.
15:06:24 <J_Arcane> TieSoul: I think HackerRank supports doing exercises in Befunge.
15:07:30 <J_Arcane> Ahh, no, I'm mistaken, I think I was thinking of somewhere else.
15:07:40 <J_Arcane> It does have Brainfuck and Whitespace though.
15:08:31 <fizzie> I promoted Befunge at work the other day.
15:08:44 <fizzie> Mainly by showing fungot sources to colleagues.
15:08:44 <oren> how complex would the highlighting regexes for befunge be?
15:09:00 <fizzie> They commented that it all seemed very practical.
15:09:33 <fizzie> (I also showed them the graphified version.)
15:09:37 <b_jonas> did someone say fungot? is he back?
15:09:57 <fizzie> A BT engineer has promised to visit on March 3rd.
15:10:06 <TieSoul> sounds like an interesting challenge
15:10:30 <oren> in most editors, regexes are used to decide the syntax colorings.
15:10:53 <fizzie> Of course, I'll be off-country the next two weeks or so, so the fungot shortage shows no signs of abating.
15:11:31 <fizzie> You could do the trivial context-free highlighting trivially.
15:11:40 <TieSoul> I don't think you could accurately tell what's a string and what's not without running it
15:11:47 <fizzie> And the more complicated thing not at all, pretty much.
15:12:02 <TieSoul> and some things could be both a string AND not
15:12:19 <b_jonas> and the contents of cells can change during running the program too
15:12:30 <TieSoul> so context-free highlighting is pretty much the best you're going to get
15:12:38 <fizzie> Being both string and commands is just a matter of translucency.
15:13:11 <fizzie> For a "well-behaved" program, you can do quite a bit with static analysis, but the editor context is worse for that kind of stuff, since the code would spend most of the time being invalid.
15:13:38 <fizzie> Every time you misplace a >, the rest of the program would be re-highlighted completely differently.
15:13:39 <b_jonas> fizzie: translucency if it's achieved using bridges (trampoline instructions), polarized light if it's achieved using different directions of execution
15:14:14 <TieSoul> befunge highlighting can't be properly done with regexes.
15:14:17 <fizzie> b_jonas: Combined with one of those polarized 3D glasses, that'd be nice.
15:17:08 <fizzie> http://users.ics.aalto.fi/htkallas/fungot.png for those who haven't seen it yet.
15:17:52 <oren> it can't be displayed because it has errors
15:18:07 <fizzie> Works for me. It's a biggish file, though.
15:19:04 <TieSoul> automatically generated Befunge flowcharts?
15:19:16 <fizzie> http://users.ics.aalto.fi/htkallas/fungotsmall.png is a 1:5 scale replica, if 7485x15016 sounds scary.
15:20:02 <fizzie> There's a bit of crude heuristics to handle things like j.
15:20:45 <oren> jesus christ... pkill -9 firefox
15:21:30 <fizzie> Also I think there was some heuristic about ignoring reflections from fingerprint instructions that don't "look like" they're being handled, since those are a big source of potential static control flows that don't (or shouldn't) occur in practice.
15:22:36 <TieSoul> is the flowchart generator up for download? :P
15:22:38 <b_jonas> what are fingerprint instructions?
15:23:21 <Deewiant> oren: https://groups.google.com/d/msg/comp.unix.questions/Pg0Gv1Vk9G4/O-k8lym2DXoJ
15:33:32 -!- CADD has quit (Ping timeout: 265 seconds).
15:37:13 -!- tomlukeywood has joined.
15:37:23 -!- tomlukeywood has left.
15:37:37 -!- dulla has quit (Ping timeout: 264 seconds).
15:47:17 -!- dulla has joined.
15:48:24 <fizzie> TieSoul: I don't think so, I think I did it as a ugly one-off Java thing or something.
15:49:28 <fizzie> b_jonas: The A..Z commands can be redefined by loading fingerprints. They have a general contract (of sorts) that they reflect on error, but of course for many fingerprints there are often instructions that never fail.
15:53:49 <oren> I use pkill -9 firefox when firefox has started swapping and is threatening to hang my system.
15:55:27 <Melvar> oren: Do you need swap?
15:57:23 <oren> I have only 2GB memory
15:59:19 <oren> Some programs need morethan that, but unlike Firefox they don't need it all at once
16:01:04 <Melvar> oren: I see. Perhaps you could run firefox with a resource limit on memory if you know its behavior in this regard?
16:02:11 <oren> I tried that, it doesn't seem to work well. If only I could ban firefox from swap while letting other procs use it
16:03:34 -!- paul2520 has quit (Ping timeout: 245 seconds).
16:04:01 -!- GeekDude has joined.
16:06:23 -!- FreeFull has joined.
16:07:16 <oren> hmm... maybe the problem would be less if I made a sawp file on a n SD card...
16:07:37 <Lymia> Why are all your strings backwards.
16:08:43 <fizzie> Lymia: You know, stacks.
16:09:24 <oren> why not by convention put all strings in < direction then
16:10:19 -!- mihow has joined.
16:13:42 <Koen_> `befunge <@."hello"
16:13:59 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: befunge: not found
16:14:59 <Lymia> Clearly strings should all be in ^ direction
16:16:09 <Koen_> `funge <@_.<0"hello"
16:16:09 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: funge: not found
16:16:52 <Koen_> ergo: forward strings don't work in befunge
16:24:49 -!- Lymia has quit (Remote host closed the connection).
16:25:14 -!- Lymia has joined.
16:33:25 -!- paul2520 has joined.
16:33:31 -!- paul2520 has quit (Changing host).
16:33:31 -!- paul2520 has joined.
16:34:58 -!- TieSoul_ has joined.
16:35:52 -!- TieSoul has quit (Disconnected by services).
16:35:58 -!- TieSoul_ has changed nick to TieSoul.
16:43:53 -!- augur has quit (Read error: Connection reset by peer).
16:44:12 -!- ^v has joined.
16:44:38 -!- augur has joined.
16:55:36 -!- ^v has quit (Read error: Connection reset by peer).
17:00:01 -!- Phantom_Hoover has joined.
17:00:02 -!- oren has quit (Ping timeout: 252 seconds).
17:05:02 -!- SopaXorzTaker has quit (Remote host closed the connection).
17:07:53 -!- oerjan has joined.
17:09:40 <oerjan> <J_Arcane> Well, this is some wild stuff. Like Haskell in Lisp-ish syntax. https://github.com/ympbyc/Carrot <-- i am pretty sure there used to exist a "liskell"
17:10:15 <J_Arcane> oerjan: There's been a couple. There's even a "Haskell Lisp" Twitter account. But sadly none seem to have ever gone anywhere.
17:11:33 -!- AnotherTest has joined.
17:11:51 <vanila> MLs don't work well with lisp syntax
17:12:04 <vanila> that's why they usually use ML-like syntax
17:22:17 -!- FreeFull has quit (Ping timeout: 265 seconds).
17:22:30 <Koen_> vanila: is there a proof or something?
17:23:27 <J_Arcane> vanila: Well, what's weird about Carrot is, it does use an ML-like syntax, but with everything still being in prefix notation.
17:23:41 <J_Arcane> So = is still the declaration operator...
17:23:50 <J_Arcane> I also couldn't make heads or tails of the conditional syntax.
17:25:07 <lambdabot> (a -> c) -> (b -> c) -> Either a b -> c
17:25:16 <lambdabot> ArrowChoice a => a b d -> a c d -> a (Either b c) d
17:25:29 <oerjan> i guess either is the specialization
17:25:59 <lambdabot> (Data.Bitraversable.Bitraversable r, Applicative f) => (a -> f b) -> r a a -> f (r b b)
17:26:09 <lambdabot> Arrow a => a b c -> a b c' -> a b (c, c')
17:28:24 <oerjan> maybe because c is not to c' as a is to b (because a is the arrow itself)
17:29:08 <oerjan> except, of course, ||| does it anyway
17:29:35 <oerjan> conclusion: it's to confuse people
17:30:25 <Koen_> I seem to recall a haskell manual stating "giving the type of a simple function is usually enough to understand what it does"
17:31:03 <oerjan> it has to be very simple and very polymorphic, though
17:31:05 <Koen_> I guess if I knew what Arrow meant I'd understand though so that's ok
17:31:22 <oerjan> Arrow is kind of a mess
17:32:08 <oerjan> in afterthought it is like a mix of Applicative and Category, but not exactly the former because it has an extra argument
17:32:10 <Koen_> to be honest I don't even know what "a b c" means
17:32:40 <oerjan> a good approximation is to say a = (->), so a b c = (b -> c)
17:32:58 <oerjan> that's the arrow used for 90% of use cases, anyhow.
17:33:38 <Koen_> so &&& merges two functions or something
17:33:46 <oerjan> (it turns out that several of the Arrow methods happened to be useful, not already defined functions for the (->) case)
17:34:29 <oerjan> > (succ *** pred) ('b', 'K')
17:34:34 <b_jonas> they aren't already defined _because_ Arrow existed early enough in the library that people didn't bother to define specializations
17:35:58 <Koen_> and ||| despecializes
17:36:19 <Koen_> takes two functions and merges the domains
17:36:35 <oerjan> > (succ ||| chr) (Left 'a')
17:36:48 <oerjan> > (succ ||| chr) (Right 48)
17:39:35 <oerjan> i think Arrow was invented before Applicative and had some of the same intended uses, so when the latter got more popular Arrow (which is really complicated in comparison) got much less use except for the (->) specializations
17:39:49 <oerjan> (both can be used for parsers, e.g.)
17:39:50 -!- MoALTz has quit (Quit: Leaving).
17:40:20 <Koen_> do you have an example of when Arrow isn't -> ?
17:41:02 <oerjan> when m is a Monad, Kleisli m is an Arrow
17:42:07 <b_jonas> so these days we just use Applicative instead? that's good, because I know what Applicative does.
17:42:25 <Koen_> well thank you for the lesson :)
17:42:32 <oerjan> > (Kleisli $ \x -> [x]) &&& (Kleisli $ \x -> [x+1]) $ 3
17:42:33 <lambdabot> Couldn't match expected type ‘s0 -> t’
17:42:33 <lambdabot> with actual type ‘Control.Arrow.Kleisli [] c'0 (c'0, c'0)’
17:43:19 <oerjan> b_jonas: there's a sense in which Arrow = Applicative + Category + a law or two
17:44:11 <b_jonas> I don't think I understand that but ok
17:44:26 <oerjan> well first of all every Arrow is a Category
17:44:56 <oerjan> secondly, if a is an Arrow then a b is morally an Applicative
17:45:25 <oerjan> (although haskell's class system doesn't allow this to be expressed as a superclass)
17:45:51 <oerjan> and you can define the functions going back and forth
17:45:58 <Koen_> the limits of hskell
17:46:41 <oerjan> although you need some extra laws to get the full Arrow laws
17:47:37 <oerjan> https://cdsmith.wordpress.com/2011/07/30/arrow-category-applicative-part-i/
17:47:45 -!- MoALTz has joined.
17:48:27 <b_jonas> oerjan: and does it work backwards too? can you take any Applicable m => m and make an Arrow a => a from it such that a b c is similar to b -> m c ?
17:48:47 -!- vanila has quit (Quit: Leaving).
17:48:56 <b_jonas> wait, that's what you said Kleisli does
17:49:06 <oerjan> b_jonas: no Kleisli needs Monad
17:49:33 <TieSoul> I'm trying to figure out how to write a binary counter in Brainfuck with a non-wrapping implementation
17:49:36 <oerjan> you need Category as well, and some compatibility laws
17:50:07 <TieSoul> (it throws an error if you decrement from zero or increment from 255)
17:50:31 <oerjan> TieSoul: well you should be able to do it with just 0,1 values :P
17:50:59 <oerjan> i think you want internal padding
17:51:33 <oerjan> 0 bit0 1 bit1 1 bit2 ... bitn 0
17:52:11 <oerjan> i guess you can do without the padding then, in principle
17:52:27 <oerjan> probably may need a temporary cell somewhere, though
17:52:38 <TieSoul> the thing is, I can't figure out how to check for 1s without decrementing from zero.
17:53:15 <b_jonas> juse use a couple of temp cells
17:53:58 <oerjan> everything's better with temp cells
17:55:22 <TieSoul> ...I don't get how temp cells would help
17:56:07 -!- oren has joined.
17:56:12 <oren> I like dempster-shafer theory
17:56:42 <TieSoul> I want to convert a number under 32 to its binary representation.
17:56:51 <oren> use a lookup table
17:57:31 <TieSoul> it's a code golf thing btw
17:57:45 <oren> ohhhh then don't use a lookup table
17:59:03 <TieSoul> I want to use a binary counter
17:59:18 <TieSoul> but I have trouble figuring out how to do that in a non-wrapping implementation
17:59:31 <TieSoul> (checking for 1s without decrementing from 0)
18:00:39 <oren> check for zero, then decrement, then check for zero again?
18:01:33 <TieSoul> I could... have a 0 cell before/after every 'binary cell'
18:01:39 <TieSoul> and then increment to 2 if it's 0
18:02:14 <oren> yeah, you need a zero cell handy anytime you want to do an if instead of a while
18:04:14 <TieSoul> I think I got confused there
18:04:19 <TieSoul> what I said doesn't make sense
18:04:34 <TieSoul> it's an 'if-nonzero', not 'if-zero'
18:05:35 <oren> that doesn't make much difference except the order of blocks
18:06:12 <Koen_> of, if-zero is ] [ right?
18:06:19 -!- FreeFull has joined.
18:07:00 <oren> Koen: something like >(moving to zerocell) ] [
18:08:12 <oren> sorry no [<]>[ ... ]
18:08:58 <TieSoul> that should be a binary counter?
18:09:40 <TieSoul> and then returning to the original cell... somehow
18:09:44 <oren> if you reset to the first cell before it
18:10:20 <oren> you need some postioning data in a parallel set of cells
18:10:39 -!- MDream has changed nick to MDude.
18:10:45 <TieSoul> I have a group of 'filled' cells to the right
18:10:57 <TieSoul> and a fixed-size binary counter to the left (5 bits)
18:11:11 <TieSoul> so I could just do >>>>>>[<]> to return to the original cell
18:11:36 <TieSoul> assuming there's a 0 cell between them
18:14:06 -!- Frooxius has joined.
18:15:00 -!- Froox has joined.
18:18:23 -!- Frooxius has quit (Ping timeout: 246 seconds).
18:23:04 <oren> What if allocation and garbage collection were hardware features?
18:23:28 <TieSoul> then that'd be some slow hardware
18:24:24 <oren> in particular, what if you had a separate, primitive core that did the gc continuously
18:24:28 <TieSoul> my binary converter seems to work
18:24:51 -!- mitchs has quit (Quit: mitchs).
18:25:42 <oren> did you test it on all 0-32
18:34:58 -!- oren has quit (Ping timeout: 272 seconds).
18:38:02 -!- FreeFull has quit (Ping timeout: 246 seconds).
18:39:27 <oerjan> `! befunge <@_.<0"hello"
18:39:46 <HackEgo> Unsupported instruction '' (0xffffffff) (maybe not Befunge-93?) \ 0
18:40:24 <oerjan> `! funge98 <@_.<0"hello"
18:40:28 <HackEgo> /hackenv/bin/!: 4: exec: ibin/funge98: not found
18:40:37 <oerjan> `! befunge98 <@_.<0"hello"
18:53:16 -!- mitchs has joined.
18:56:44 -!- oren has joined.
19:04:45 -!- Lymia has quit (Ping timeout: 265 seconds).
19:08:12 -!- Lymia has joined.
19:20:42 -!- adu has joined.
19:20:42 -!- adu has quit (Client Quit).
19:21:07 -!- oren has quit (Ping timeout: 245 seconds).
19:21:19 -!- vanila has joined.
19:36:02 -!- mihow has quit (Quit: mihow).
19:48:47 -!- mihow has joined.
20:10:05 <TieSoul> https://www.hackerrank.com/challenges/baconian-cipher
20:10:16 <TieSoul> I just made a 209-character solution
20:10:29 <TieSoul> I'm pretty proud of myself :P
20:10:36 -!- oren has joined.
20:10:45 <TieSoul> especially since I've never seriously done anything in brainfuck before
20:12:13 <oren> I have tried to write programs that /write/ BF prgorams, but i have not tried to write bf prgrams.
20:12:50 <TieSoul> https://www.hackerrank.com/challenges/baconian-cipher I just did this challenge
20:13:00 <TieSoul> my solution is 209 characters
20:13:05 <TieSoul> could probably shorten it a bit
20:13:46 <J_Arcane> Huh. If I'm reading that correctly, isn't that just 5-bit text encoding?
20:14:19 -!- mihow has quit (Quit: mihow).
20:14:21 <TieSoul> that's what I needed the binary conversion for
20:14:33 <vanila> i couldnt write this in BF at all, let alone 200 chars
20:14:41 <TieSoul> hackerrank's implementation has no wrapping
20:14:46 <TieSoul> so that makes the challenge harder
20:14:52 <J_Arcane> I couldn't write jack squat in BF.
20:15:05 <TieSoul> https://gist.githubusercontent.com/TieSoul/17f9670ed239ad94f2be/raw/1d3ea0c59b409b76a70f3ba8140f9e26619c0298/bacon.b
20:15:20 <J_Arcane> But I might have a stab at it in F# just for fun. Give me an excuse to try out fsharp-mode in emacs now that I've got the intellisense working.
20:15:45 <TieSoul> I could probably do a Ruby/Python one liner
20:15:59 <oren> you use an advanced editor, J_arcane.
20:16:33 <oren> I only have synatax highlighting
20:16:45 <oren> because I use mcedit
20:17:26 <oren> so emacs is very advanced compared to what i use, no saracsm
20:17:28 <J_Arcane> What I use really depends on the language.
20:17:51 <TieSoul> > people with scores of 98.46
20:17:53 <lambdabot> <hint>:1:20: parse error on input ‘of’
20:29:18 -!- TieSoul has changed nick to TieSleep.
20:33:44 -!- CADD has joined.
20:38:38 <oren> a language in which it is impossible to write an interpreter for a language unless that language is TC
20:39:22 <coppro> oren: that's impossible
20:39:55 <coppro> actually... hmm, no, it's not *obviously* impossible.
20:40:03 <coppro> if you maintain a source/input distinction
20:40:18 <coppro> but I would be very surprised if you could do that
20:40:34 -!- oerjan has quit (Quit: ZZZZ).
20:42:40 -!- arjanb has joined.
20:43:03 <oren> wait. trivially, a restricted compiler for any language, which only accepts the source code of a BF interpreter
20:43:17 <Gregor> It depends on whether that is your only restriction.
20:43:29 <Gregor> If it is only impossible to write ONE interpreter for ONE language, that's easy :)
20:43:56 <Gregor> But a language which will accept ANY interpreter for ANY TC language, but will reject any other program is a halting problem solver.
20:45:29 <pikhq> Yep, you either managed to make it trivial and uninteresting *or* impossible.
20:45:32 <oren> it might be even worse than that actually. it basically has to tell whether the halting problem is unsolvable on the langugae defined by its input program.
20:46:43 <oren> wait. DOES the haltingproblem thing go both ways?
20:46:53 -!- Patashu has joined.
20:47:15 <oren> are there languages where the halting problem is unsolvable, but the language is not TC?
20:47:49 <vanila> there are undecidable problems in context-free grammars
20:51:45 <oren> http://www.pentacom.jp/pentacom/bitfontmaker2/
20:51:56 <oren> this site is awesome
21:07:07 <FireFly> Hm, I just use a generic bitmap editor
21:08:35 <oren> I was doeing it with mtpaint
21:08:54 <oren> but this thing is way easier if a little limited
21:09:41 <J_Arcane> Wee! I have my char lookup table now.
21:14:00 <oren> So i made an utterly insane font, which I posted a screenshot of earlier
21:15:30 <b_jonas> oren: where's the screenshot?
21:15:42 <oren> http://snag.gy/GZVB3.jpg
21:15:47 <oren> heres a new screen
21:16:19 <FireFly> That's really hard to read
21:16:31 <b_jonas> no, it's not very hard to read actually
21:16:38 <oren> it uses underdot instead of capitals
21:16:48 <b_jonas> but the v looks ugly to me
21:16:49 <FireFly> Well, apart from t and h I guess it's fairly readable
21:17:09 <oren> what bout the f
21:17:16 <FireFly> My brain really doesn't like the way "that" looks
21:17:25 <b_jonas> oren: nice, thank you for sharing
21:18:07 <FireFly> the f isn't too bad for me from a readability perspective
21:19:20 <FireFly> I made this bitmap font a while ago: http://xen.firefly.nu/up/pixfont/index.html
21:19:31 <b_jonas> oren: what does the "z" look like? "z" is ugly in many fonts but you don't find that out from screenshots like this
21:19:34 <FireFly> Though I didn't try it out properly until yesterday
21:20:40 <oren> `1234567890-=qwertyuiop[]\asdfghjkl;'zxcvbnm,./~!@#$%^&*()_+QWERTYUIOP{}|ASDFGHJKL:"ZXCVBNM<>?
21:20:41 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: /hackenv/1234567890-=qwertyuiop[]\asdfghjkl;'zxcvbnm,./~!@#$%^&*()_+QWERTYUIOP{}|ASDFGHJKL:"ZXCVBNM<>?: No such file or directory
21:21:10 <FireFly> ...that's one way to do it
21:21:53 <oren> http://snag.gy/0dBxW.jpg
21:22:29 <b_jonas> wow, that dollar sign is ugly
21:23:11 -!- mihow has joined.
21:24:54 -!- ais523 has joined.
21:24:54 <oren> here is the font file for pasting into the editor https://gist.github.com/orenwatson/55b014da50347d032b1f
21:26:24 <FireFly> Hmm, I should try to convert my bitmap font to a proper font file
21:26:48 <oren> once again the numbers are the way I write them...
21:26:49 -!- augur has quit (Ping timeout: 252 seconds).
21:27:16 <oren> well, except I usually put a dot in my zeros
21:29:23 <oren> firefly: thta font is nice and small
21:30:06 <oren> I like small fonts when I'm programming
21:31:03 <FireFly> I actually prefer larger fonts personally, but I wanted to make a small but legible bitmap font for platforms where screen space is more limited
21:31:14 <b_jonas> oren: wait, I have an old crazy font I'd like to show to you
21:32:36 <oren> In particular, I like fonts that conserve vertical space
21:32:58 <FireFly> As for smallness, I have this thing I made a long time ago: http://xen.firefly.nu/up/pixel-font.png
21:33:11 <FireFly> Though it uses manual anti-aliasing so it kinda doesn't count
21:33:43 <FireFly> It's surprisingly legible for being 3x5px glyphs, to me anyway
21:34:02 <oren> that's pretty damn good
21:35:49 -!- hassa-aravit has joined.
21:37:20 <FireFly> The S looks pretty ugly, hmm
21:38:12 <fizzie> I've made a 3x5 (well, 4x6 cell) font too, for rfk86.
21:39:35 <fizzie> https://zem.fi/rfk86/ should render using it, if things work out right.
21:42:27 <fizzie> I like the lowercase s/z trick it uses, which I think I adapted (or even just copied) from mooz's TI-86 befunge interpreter.
21:45:31 -!- idris-bot has quit (Quit: Terminated).
21:45:52 -!- Melvar has quit (Quit: rebooting).
21:47:39 -!- augur has joined.
21:48:46 <FireFly> Hmm, it shouldn't be too hard to tweak my JS thing to generate a BDF font file
21:50:11 -!- Melvar has joined.
21:51:48 -!- Lymia has quit (Ping timeout: 256 seconds).
21:54:13 <b_jonas> oren: http://www.math.bme.hu/~ambrus/pu/stickfont-screenshot0.png
21:54:36 <b_jonas> it's a 16x9 font because I originally used it for VGA console
21:54:49 -!- idris-bot has joined.
21:54:53 <b_jonas> not for serious purposes though, it was always intended as a silly font
21:55:08 -!- ais523 has quit.
21:55:23 -!- ais523 has joined.
21:55:35 <J_Arcane> Part 1: http://rextester.com/UUWO55700
21:56:18 <b_jonas> I didn't draw more characters than what you can see there though
21:57:11 -!- AnotherTest has quit (Remote host closed the connection).
21:57:32 <b_jonas> FireFly: that small font looks nice
21:57:48 <b_jonas> though I'm not sure I like the "S"
21:58:05 <oren> it is a little too light isnt it?
21:58:08 <b_jonas> nor the way the N and H look very similar
21:58:32 <b_jonas> the "S" is too closed, that is, the top and bottom curve too much down
21:58:55 <b_jonas> the N and H might be a bigger problem though
21:59:47 -!- callforjudgement has joined.
21:59:51 -!- ais523 has quit (Read error: Connection reset by peer).
22:00:05 -!- callforjudgement has changed nick to ais523.
22:01:25 <FireFly> b_jonas: the 3×5 one, or the bigger one?
22:01:51 <b_jonas> By the way, I like the “Munka és béke” poem (displayed in that screenshot) partly because it has almost every letter, and has most letters in both the first four lines and the last four lines, which is quite remarkable for such a short poem.
22:01:59 -!- tromp_ has quit (Ping timeout: 244 seconds).
22:02:07 <b_jonas> So it's not really a pangram, but close.
22:02:11 <b_jonas> Useful for demonstrations like this.
22:02:57 <FireFly> Poems for demoing charsets reminds me of Iroha
22:03:16 <b_jonas> FireFly: is that the Japanese one that helps demo kana?
22:03:32 <FireFly> It uses every syllable exactly once
22:04:00 <oren> it now contains ones that don't exist in modern japanese like wi
22:04:21 <oren> (well, wi does occur in names, like arawi keiichi)
22:05:00 <ais523> hmm, I wonder if using every syllable exactly once is possible in English
22:05:01 -!- Lymia has joined.
22:05:04 <ais523> it seems like it might be
22:05:37 <b_jonas> ais523: there are too many syllables so you'd need a longer text, which is hard without reusing “the” more than twice
22:06:05 <ais523> yep, you'd have to avoid common words
22:06:14 <ais523> but long English prose that avoids common words has been done before
22:06:14 <FireFly> Well, if you can write a book without using the letter 'e'...
22:06:21 <ais523> FireFly: the great thing about that book
22:06:28 <ais523> is that it's a translation of a French book that also doesn't use the letter 'e'
22:06:41 <ais523> ('e' is the most common letter in French too, IIRC)
22:06:43 -!- FreeFull has joined.
22:06:50 <oren> http://snag.gy/tRIpP.jpg
22:06:56 <FireFly> that's what I usually say when that book is bought up--the translator's job is more impressive than originally writing it
22:07:04 <oren> my font works well at small sizes too
22:07:18 <FireFly> (not to say that authoring it isn't also impressive)
22:08:12 <oren> without 'e' in FRENCH?? very impressive,
22:08:47 <b_jonas> The fun part of Munka és béke is that it has two variants. The other variant has an “ű” but doesn't have “ő” in exchange
22:08:58 <b_jonas> no wait, that can't be true
22:09:07 <b_jonas> there are two “ő” so one must remain
22:09:20 <b_jonas> it doesn't have some other letter
22:11:01 <Taneb> I have written a quine in Haskell
22:11:33 <b_jonas> hmm, maybe it has “ű” but doesn't lose anything in exchange?
22:11:45 <b_jonas> but I seem to remember it wasn't more close to a pangram than this version
22:12:03 <b_jonas> whatever, I'm too tired to figure it out now
22:12:17 <oren> maybe there are 3 versions?
22:12:29 <zzo38> Does not 'e' in French also mean doesn't have any 'e' with accent mark too, or not?
22:13:05 <b_jonas> oren: could be, but I only know of two
22:16:18 <zzo38> Do you know of any Z-machine implementation that actually supports small-endian other than ZORKMID?
22:16:29 <zzo38> (ZORKMID does support big-endian too)
22:16:51 <b_jonas> isn't it called "little-endian"?
22:17:54 <zzo38> I don't know, I called it "small-endian"; Infocom called them "byte-swapped" story files (and as far as I know, never created any such story files nor ever implemented an interpreter that would support them).
22:19:41 -!- gamemanj has joined.
22:24:52 -!- augur has quit (Remote host closed the connection).
22:31:29 -!- aloril has quit (Remote host closed the connection).
22:32:22 -!- gamemanj has quit (Ping timeout: 246 seconds).
22:38:46 <Koen_> ais523: e is far more common in french than in english, i reckon
22:39:08 <ais523> Koen_: does that include é and ê and è or not?
22:40:10 <Koen_> not sure they make a relevant difference though
22:40:59 <Koen_> but that's also just based on my impressions as a native french speaker
22:41:04 -!- aloril has joined.
22:41:28 <Koen_> i think it's easier to write in alexandrines than without the letter e
22:41:46 -!- `^_^v has quit (Ping timeout: 244 seconds).
22:43:03 -!- gde33|2 has quit (Read error: Connection reset by peer).
22:45:11 -!- tromp has joined.
22:57:14 <FireFly> Hmm, I don't think it's possible to do an 8×5 monochrome '½' that doesn't look like crap...
22:57:38 <Koen_> the bar is too long imho
22:59:04 <FireFly> http://xen.firefly.nu/up/2015-02-27_235830.png here's my best attempt at ¼ ½ ¾
22:59:55 <FireFly> Oh, I meant 8-high and 5-wide
23:01:42 <Koen_> "just tilt your head right"
23:02:07 <Koen_> are you designing a game boy?
23:02:32 <ais523> the trick's to leave out the slash
23:03:04 <ais523> I'm juts being creative
23:03:15 <elliott> here's my glyph: [0.5 just mashed into the available space]
23:04:07 <int-e> 1/2 works, but the rest is hard :)
23:04:18 <Koen_> ais523's try was friendlier to proportionate fonts irc clients
23:04:35 <ais523> Koen_: I use a proportional font in my client
23:04:35 <FireFly> I found the 2 to be hardest actually
23:04:57 <ais523> actually, this reminds me of a table I saw on a proportional-font-using forum
23:05:07 <ais523> that used ' and ` as spacers in order to make all the columns line up
23:05:08 <fizzie> The "leave something out" trick is a good one, it's the basis for the rfk86 s/z too.
23:05:11 <ais523> despite the proportional font
23:05:31 <ais523> fizzie: that reminds me of the trick I used when trying to fit the whole alphabet on a 7-segment display
23:05:44 <ais523> (M was the top segment, middle segment, and the two lower sides)
23:05:52 <FireFly> Koen_: the font? I had the TI-84 in mind at first, now I'm just adding some more glyphs because I was happy with how it turned out
23:06:52 <vanila> can we generate a new font
23:06:54 * Koen_ had a ti-84 but it was stolen during mardi gras in prepschool
23:07:01 <ais523> I remember that Windows 3.1 shipped with a set of raster fonts of very small sizes
23:07:35 -!- Fleur has joined.
23:09:38 <fizzie> I had a script somewhere for IRCing with the rfk86 font, based on the 2x2 Unicode quarter-block things.
23:10:31 <fizzie> There's the full set of them, so 3 lines is enough for a font 6 pixels high.
23:11:01 <int-e> FireFly: http://int-e.eu/~bf3/tmp/6x13frac.png .. the fixed font here also leaves out the / (or bar)
23:11:03 <FireFly> You can do better if you abuse the Braille Unicode block
23:11:15 <fizzie> I'm typing with a speech recognizer again
23:11:25 <int-e> FireFly: but of course those extra vertical pixels help a lot
23:11:26 <FireFly> That gives you 8×2 "pixels" per character
23:11:41 <HackEgo> Fleur: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: <http://esolangs.org/>. (For the other kind of esoterica, try #esoteric on irc.dal.net.)
23:11:43 <nortti> int-e: that link gives not found error for me
23:11:46 <elliott> happy birthday 30 pussy Busy Busy fizzy
23:11:55 <fizzie> Elliot how did you get the backtick
23:12:05 <int-e> nortti: right, forgot to press enter for the scp command. should be ok now.
23:12:06 <vanila> Hey i have an idea lets make an esoteric font
23:12:32 <fizzie> I think they have improved the voice typing
23:12:41 * FireFly ponders whether to remove the fractional slash or keep his current glyphs
23:12:44 <zzo38> vanila: OK, what are you going to put on it, and is it METAFONT or PC 8x8 bitmap font or whatever?
23:12:49 <fizzie> Although this is far less hilarious than last time
23:12:52 <FireFly> I could do alternate characters
23:13:10 <elliott> Fleur: came here from the wiki?
23:13:19 <int-e> FireFly: I believe you'll find that it's 4x2 pixels
23:13:20 <vanila> zzo38, I thought it could be generated randomly, by trying to find a set of images which look very different (by some metric)
23:13:21 <fizzie> Maybe this phone just has a better microphone
23:13:27 <FireFly> int-e: oh. yes, I meant that
23:13:38 -!- boily has joined.
23:14:07 <Fleur> elliott: Sort of - I've always loved browsing the wiki for pleasure, and then nortti told me there was an IRC channel so I thought I'd join
23:14:48 <elliott> well, there are significantly fwer brainfuck derivatives here
23:14:57 <elliott> also significantly fewer esolangs in general though really...
23:15:05 <vanila> Fleur, did you invent any esolangs or have some favorites
23:15:58 <Fleur> vanila: I don't really have any particular favourites, I just keep clicking the random page button until something catches my eye :P
23:16:25 <Fleur> But I have attempted in the past to create my own (but nothing has particularly taken off further than ideas in my mind)
23:19:45 <boily> Flellour! first time on the channel?
23:19:54 <Koen_> elliott: am I allowed to delete the pages of brainfuck derivatives I made or would that be vandalism?
23:20:15 <ais523> Koen_: if they suck, I personally wouldn't mind
23:20:22 <ais523> however, you don't have the permissions to delete pages
23:20:37 <ais523> put {{delete}} on them, if there's been no significant activity from others on them
23:20:40 <ais523> I'll delete them for yo
23:21:45 <vanila> Koen_, how many are we talking
23:27:04 -!- Fleur has quit (Remote host closed the connection).
23:31:27 <boily> she left. we had a talkative newcomer...
23:36:39 -!- mihow has quit (Quit: mihow).