00:00:32 <Koen__> if you live a continuous life you'll probably endure an uncountable infinity of states
00:02:00 <Bike> i for one pride myself on living moment to moment.
00:04:53 <kmc> lol continuum is a lie
00:07:04 <Taneb> kmc, don't Zeno's paradoxes pretty much state "both time and distance are neither discrete nor continuous"?
00:08:45 <Bike> most of what i get out of zeno is "dang, calculus is nice"
00:09:25 <Taneb> Although Achilles and the Tortoise is a bit silly
00:09:39 -!- Taneb has quit (Quit: Leaving).
00:13:39 -!- augur_ has joined.
00:13:43 -!- augur has quit (Read error: Connection reset by peer).
00:14:35 -!- carado has quit (Ping timeout: 245 seconds).
00:23:50 -!- Sgeo has joined.
00:27:51 -!- ^v has quit (Quit: Leaving).
00:28:12 -!- ^v has joined.
00:38:08 -!- yorick has quit (Ping timeout: 240 seconds).
00:49:04 -!- conehead has joined.
00:51:54 -!- ^v has changed nick to v^.
00:52:03 -!- v^ has changed nick to ^v.
00:53:49 -!- ^v has changed nick to dan200.
00:58:38 -!- dan200 has changed nick to ^v.
00:59:56 -!- mnoqy has quit (Quit: hello).
01:01:46 -!- ^v has quit (Read error: Connection reset by peer).
01:04:15 -!- ^v has joined.
01:23:21 -!- nisstyre has quit (Quit: Leaving).
01:48:25 -!- dessos has left.
01:55:24 -!- Koen__ has quit (Quit: Koen__).
02:21:34 <Sgeo> I don't think Racket thinks I'm a consenting adult
02:24:05 <Sgeo> Things like not letting me force code that calls raise to continue after the raise unless the code itself supports it
02:24:57 <kmc> do you and Racket have a safeword
02:25:40 <oerjan> he's just forgotten to use the unsafeWord
02:38:31 -!- zzo38 has joined.
02:39:12 <zzo38> I replaced the power supply and now this computer works.
02:40:51 <zzo38> However, this power supply has three lights labeled "A", "S", "T", and a red button below them. The "A" light is lit blue. Do you know what this means?
02:42:57 <elliott> it's abstract, but not syntax or a tree.
02:58:16 <kmc> what happens if you press the button
03:00:25 <zzo38> I didn't push it yet.
03:06:02 <kmc> maybe you ought to
03:09:13 <oerjan> i dunno, what if "A" is for "atomic"
03:12:56 <kmc> zzo38: did that fix the noise as well
03:13:36 <zzo38> kmc: Installing the new power supply fix the noise and everything else that was going wrong.
03:15:10 <zzo38> Now in this Dungeons&Dragons game I have to find some leader of some resistance against Aberration Hater (which runs over the entire city they are in) in order to rescue someone from being imprisoned, but it is difficult to find and I will need some advisor helping with it and will need a boat and there are other problems too
03:17:10 <kmc> what's the boat for
03:17:43 <zzo38> We can travel by water or by land, but there is another war going on in the land; a boat might be better way.
03:17:56 <zzo38> So it isn't strictly necessary but I expect it to help.
03:19:18 <zzo38> Also some dwarf came in to say we finish our quest and gave us three horses and a silver ring; another wizard said he didn't trust him. Maybe that is correct; maybe he trained the horses wrong on purpose or something; I don't know.
03:22:04 <zzo38> I know about this resistance and imprisonment due to some kind of hallucination when trying to escape out of the demon's castle, it interrupted the escape and then the voice of my character's father's friend said some things, I tried to write it down but found only seaweed to write it on and only white ink to write with; but later on I found that it was actually written on paper, not seaweed, and with black ink, not white ink. If he is who he said
03:22:24 <zzo38> Oops did I send this message wrong? Did it get cut off?
03:22:46 <zzo38> (I got a ":cameron.freenode.net 421 zzo38 in :Unknown command" even though I typed no such command)
03:24:54 <oerjan> zzo38: it cut off after "If he is who he said"
03:25:14 <zzo38> If he is who he said he is, then I suppose my character's father's friend is a speaker in dreams.
03:25:27 <zzo38> Why did I get that error message?
03:25:47 <zzo38> (This is a single-threaded IRC client)
03:26:02 <oerjan> zzo38: perhaps you wrote the rest as a raw message?
03:26:34 <oerjan> i mean, the part after what was cut off
03:26:53 <zzo38> oerjan: I didn't; I wrote it all on one line before pushing the carriage return key, and then it came up this error message.
03:29:18 <oerjan> zzo38: well afaiu "in" is the command that was unknown.
03:29:37 <zzo38> oerjan: Yes, I can see that too
03:30:05 <oerjan> maybe the "in dreams." overflowed a buffer and got sent raw?
03:30:09 <zzo38> ("in" doesn't occur after that except as the second to last word)
03:31:30 <zzo38> oerjan: Do you mean after the CRLF is sent somehow? I don't know how.
03:32:53 <oerjan> the "i" is the 519th character.
03:37:51 <zzo38> I tried the red button it changes which light is lit and seems to affect the fan speed.
03:39:40 <oerjan> "Automatic", "Silent" and "Terrifying", clearly.
03:41:00 <zzo38> I get slightly bad video again but only when the CPU is in use a lot, it seems.
03:41:54 <zzo38> And not as bad as before.
03:44:14 <zzo38> Maybe I will need some spell to write on the water.
03:46:02 <zzo38> How do you clone the serial number and capacity of a hard disk?
03:46:43 <oerjan> didn't fizzie do something similar the other day
03:47:06 <oerjan> to move a windows installation
03:47:39 <oerjan> or someone here did, anyway.
03:48:14 <oerjan> except it wasn't actually the same capacity, he moved it from a hard disk to a smaller SSD iirc
03:49:24 <zzo38> I mean if you have a disk with a larger capacity and a different serial number and you want to move it so that it has the same serial number and capacity as the old one (so that a part of the new one isn't used)
03:50:53 <oerjan> ok. well i don't know anyway.
04:27:47 <zzo38> The video now only looks bad when I am standing far away from the computer.
04:28:00 <zzo38> It looks correct from close up.
04:29:32 <HackEgo> slist sb&hj: Taneb atriq Ngevd Fiora nortti Sgeo ThatOtherPerson alot
04:47:35 -!- mnoqy has joined.
04:50:55 <HackEgo> danddreclist 42: shachaf nooodl boily \ http://zzo38computer.org/dnd/recording/level20.tex
04:52:17 <zzo38> The other thing I ought to do before going in the boat is to summon a familiar, but there are no stats for a leech in any of the books. There also isn't any stats for a big leech that we can downgrade (my brother's suggestion).
04:52:49 -!- conehead has quit (Quit: Computer has gone to sleep.).
04:58:57 <oerjan> just take a banker and downgrade that.
05:00:01 <zzo38> Hm... No, that won't work.
05:00:24 -!- Sgeo has quit (Read error: Connection reset by peer).
05:01:30 -!- zzo38 has quit (Quit: Rebooting.).
05:03:05 -!- Sgeo has joined.
05:07:12 -!- zzo38 has joined.
05:07:45 <Sgeo> Watching talk on Go concurrency
05:07:58 <Sgeo> Talks about how concurrency != parallelism
05:08:12 <Sgeo> Then goes on to say how concurrency is
05:08:25 <zzo38> Is the video card or video cable bad? Is there something wrong with the surge protector?
05:08:37 <Sgeo> "(Much nicer than dealing with the minutiae of parallelism (threads, semaphores, locks, barriers, etc.)"
05:09:14 <Sgeo> Yeah, those things are totally features of parallelism rather than poor models of concurrency
05:49:49 -!- asie_ has joined.
05:50:09 -!- asie_ has changed nick to asie.
06:11:23 -!- ^v has quit (Quit: Leaving).
06:14:19 -!- azaq23 has joined.
06:21:56 -!- carado has joined.
06:24:56 <zzo38> I think it does seem to have something to do with the video card.
06:25:27 <zzo38> Now it seems I am getting bad video only when the video memory is being written.
06:35:17 -!- rodgort has quit (Ping timeout: 248 seconds).
06:37:09 -!- rodgort has joined.
07:02:18 -!- oerjan has quit (Quit: leaving).
07:02:39 -!- carado has quit (Ping timeout: 260 seconds).
07:39:23 <Sgeo> onfail $ run $ head -10 "README"
07:39:24 <Sgeo> has the same meaning as:
07:39:24 <Sgeo> (onfail (run (head -10 "README")))
07:44:09 -!- asie has quit (Remote host closed the connection).
07:44:36 -!- asie has joined.
07:46:38 -!- Taneb has joined.
07:54:20 <Taneb> Sweet Sherlock and Hella Watson update
08:07:06 -!- AnotherTest has joined.
08:10:59 -!- nisstyre has joined.
08:12:42 -!- asie has quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz...).
08:37:01 -!- asie has joined.
08:46:21 -!- asie has quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz...).
09:07:15 -!- Lymia has changed nick to Mayushii.
09:13:01 -!- Phantom_Hoover has joined.
09:14:46 -!- MindlessDrone has joined.
09:54:37 -!- nooodl has joined.
09:57:48 <zzo38> Do you like this latest update for Dungeons&Dragons game recordings?
10:22:39 -!- asie has joined.
10:27:38 -!- Uguubee111116 has quit (Quit: Uguubee111116).
10:33:03 -!- zzo38 has quit (Remote host closed the connection).
11:06:50 -!- AnotherTest has quit (Ping timeout: 264 seconds).
11:21:28 -!- asie has quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz...).
11:38:32 -!- shikhin has joined.
11:43:37 -!- asie has joined.
11:44:32 -!- AnotherTest has joined.
11:45:14 -!- shikhin_ has joined.
11:46:59 -!- Taneb has quit (Quit: Leaving).
11:47:41 -!- shikhin has quit (Ping timeout: 245 seconds).
11:55:49 -!- Koen__ has joined.
12:04:42 -!- yorick has joined.
12:27:41 -!- AnotherT- has joined.
12:27:41 -!- AnotherTest has quit (Read error: Connection reset by peer).
12:27:41 -!- AnotherT- has changed nick to AnotherTest.
12:35:55 -!- azaq23 has quit (Quit: Leaving.).
12:40:36 -!- shikhin_ has quit (Ping timeout: 245 seconds).
13:05:59 -!- Phantom__Hoover has joined.
13:07:48 -!- Phantom_Hoover has quit (Ping timeout: 240 seconds).
13:44:18 <impomatic> I've found another three articles about Core War in old magazines and added them to the list http://www.retroprogramming.com/2011/11/bibliography-of-programming-games.html
13:45:43 -!- nisstyre has quit (Remote host closed the connection).
14:01:29 -!- Taneb has joined.
14:02:04 <Taneb> `quote at the same time
14:02:12 <HackEgo> 1031) <Bike> ok confession time <Bike> because of the whole "generation" thing i thought that kids were born in waves until i was like twelve <Bike> and everybody like reproduced at the same time \ 1062) <Taneb> I would like to learn how to use a sword <Taneb> And also how to ride a unicycle <Taneb> Perhaps not at the same time
14:02:31 <Taneb> You'll never guess what societies I've joined
14:02:56 <olsner> the unicycle sword juggling society?
14:09:21 -!- nooodl has quit (Ping timeout: 245 seconds).
14:15:46 <Taneb> JuggleSoc said they'd teach me to ride a unicycle
14:16:07 <Taneb> MedievalSoc said they'd teach me to use a sword
14:16:49 <Bike> are they seriously named like cyberpunk security organizations
14:17:23 <Taneb> There's even HackSoc
14:17:46 -!- shikhin_ has joined.
14:17:54 <Bike> do you get complementary mirrored glasses for joining
14:22:07 -!- sebbu has quit (Read error: Connection reset by peer).
14:22:18 -!- Phantom__Hoover has quit (Ping timeout: 264 seconds).
14:23:03 -!- sebbu has joined.
14:28:31 -!- shikhin_ has quit (Ping timeout: 245 seconds).
14:28:38 <Taneb> Otherwise what is the point
14:30:27 <Taneb> I need to find a song I can sing to show off my singiness
14:31:10 <Koen__> HANG UP THE CHICK HABIT HANG IT UP DADDY YOYO hth
14:31:57 <Taneb> Koen__: is that going to get me into the cast of Jesus Christ Superstar
14:32:23 -!- nisstyre has joined.
14:32:27 <Koen__> but try having some car chasing going on while you sing
14:32:46 <Koen__> that could make a great movie
14:35:47 <Taneb> I was thinking If I Were A Rich Man
14:39:36 <Taneb> I have texted my actor-friend for advice
14:39:46 <Taneb> Hopefully he will not see me as a rival
14:42:30 -!- Phantom_Hoover has joined.
14:42:33 -!- Phantom_Hoover has quit (Changing host).
14:42:33 -!- Phantom_Hoover has joined.
14:42:35 <Phantom_Hoover> <Bike> are they seriously named like cyberpunk security organizations
14:42:51 <Phantom_Hoover> i asked the sociology society if they were called SocSoc and they said yes
14:53:22 -!- mnoqy has quit (Quit: hello).
14:56:45 -!- JWinslow23 has joined.
15:11:14 <Taneb> Apparently there are at least 6 Homestuck fans at the University of York
15:23:21 <fizzie> Is that group called StuckSoc?
15:27:00 <Taneb> You need 15 to start a society
15:27:33 <Taneb> And as far as I know none of them has met more than three
15:28:19 -!- shikhin_ has joined.
15:28:56 -!- nisstyre has quit (Quit: Leaving).
15:29:20 <fizzie> I suppose that's a security measure.
15:30:10 -!- Taneb has quit (Quit: Page closed).
15:32:56 -!- shikhin_ has changed nick to shikhin_THE_mast.
15:33:15 -!- shikhin_THE_mast has changed nick to shikhin_mogw.
15:34:14 -!- shikhin_mogw has changed nick to shikhin.
15:34:22 -!- shikhin has quit (Changing host).
15:34:22 -!- shikhin has joined.
15:36:14 <Sgeo> my smoke/co detector sems to be hallucinating a fire
15:38:32 <Sgeo> It's saying smoke, not co. If it was saying co, I wouldn't assume it was hallucinating
15:54:02 -!- AnotherT- has joined.
15:54:02 -!- AnotherTest has quit (Read error: Connection reset by peer).
15:54:03 -!- AnotherT- has changed nick to AnotherTest.
15:54:48 -!- pikhq_ has quit (Remote host closed the connection).
15:57:26 <JWinslow23> I finally got my Tic-Tac-Toe idea into a wiki page!
15:59:05 <Sgeo> I hope my detector wasn't misreporting CO as fire
15:59:41 <Sgeo> My smoke detector/CO alarm went off, reporting smoke, no evidence of any fire anywhere in sight
16:00:52 <Bike> j's right, what if the fire got time-displaced?
16:00:59 <Sgeo> A few minutes ago
16:01:17 <Sgeo> I threw it away, worried that it would disturb neighbors
16:01:43 <Sgeo> It stopped chirping shortly after reaching the dumpster
16:03:16 <JWinslow23> So, anyone have any suggestions for the Tic Tac Toe language?
16:03:31 <Bike> you just... threw away your smoke detector?
16:03:44 <ion> Makes sense.
16:03:58 <ion> Much easier than, say, removing the battery.
16:04:04 <ion> And you get to buy a new one.
16:04:15 <shikhin> Yeah, what if the old one was faulty?
16:04:27 <Sgeo> ion: I tried to remove the battery. I failed.
16:04:35 <shikhin> What if it burst into flames... would be ironic, won't it?
16:05:02 <ion> (Is it legal to throw smoke detectors to normal trash? They contain a small amount of radioactive material, and in this case, also a battery.)
16:05:07 <Sgeo> I just realized how suspicious an uncomfortable-looking person holding a beeping device wrapped in a blanket must seem
16:05:22 <Sgeo> Some neighbor saw me on my way to throw it out
16:05:43 <Sgeo> ion: my dad claims yes, so... maybe?
16:06:13 <JWinslow23> Every time I come here, I see a lot of smart people...oh look, you're all here too!
16:06:22 <ion> shikhin: I got rid of a smoke alarm that emitted a smell of smoking electronics when i powered it up once.
16:07:04 <ion> (I left it at a recycling center.)
16:07:09 <Sgeo> I am concerned what if there is actually somethign in my apartment triggering it
16:07:31 <ion> Are the other detectors silent?
16:07:50 <Koen__> JWinslow23: I like the idea of a tic tac toe language. but WHY would you make it a brainfuck derivative? :(
16:08:19 <JWinslow23> I wanted other commands, but I didn't have any ideas!
16:08:21 <Sgeo> I only have/had one detector
16:09:09 <shikhin> Well, you'd sure as hell know if it was right, and something really is wrong :-)
16:09:19 <ion> sgeo: ಠ_ಠ Do you only have one room?
16:09:39 -!- asie has quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz...).
16:09:39 <Sgeo> ion: two rooms. Bathroom, and rest of apartment
16:09:59 <Sgeo> Although there's a very short hallway leading into the bathroom, so there's a divider wall
16:10:05 <Sgeo> Although there's a very short hallway leading into the bathroom, so there's a divider wall
16:10:13 <ion> end chorus
16:10:15 <Sgeo> Although, I guess kitchen is semi-separate
16:10:46 <Sgeo> shikhin: unless alarm misreported CO as fire...
16:10:56 <JWinslow23> Anyone have any other ideas for commands on my tic tac toe board? I do NOT want any more bf derivatives! >:-[
16:11:24 * shikhin 's house lacks any sorts of alarm detectors, let alone CO detectors..
16:11:54 <Bike> maybe you should get some.
16:12:20 <shikhin> Bike: I live in a third world country, heh.
16:13:46 <Bike> are they expensive?
16:14:00 <Bike> i think every buildin i've ever lived in has had them preinstalled, now that i think about it
16:14:09 -!- AnotherTest has quit (Ping timeout: 252 seconds).
16:14:44 <JWinslow23> I'll change it to a stack based language. I will have to have help, though.
16:16:06 <Bike> already sounds like underload
16:17:03 <JWinslow23> Well, if no one will help me, we'll be stuck with this BF derivative forever.
16:18:41 <myname> why is there no flood-kicking bot?
16:18:54 <Bike> because most of us aren't stupid enough to spam the fucking channel.
16:18:56 -!- shikhin has quit (Ping timeout: 245 seconds).
16:19:32 <myname> JWinslow23: C is just a BF derivate :p
16:20:30 -!- ChanServ has set channel mode: +o elliott.
16:20:33 -!- elliott has kicked JWinslow23 don't flood.
16:20:34 -!- elliott has set channel mode: -o elliott.
16:21:10 <ion> myname: You just witnessed our fleshy flood-kicking bot in action.
16:21:30 * Fiora god was kicked from #earth by Noah [don't flood]
16:22:00 <Bike> does that mean elliott is satan
16:22:18 <ion> I think he’s just Santana and/or Santa.
16:22:19 -!- ^v has joined.
16:23:46 <Bike> elliott: how did you obtain the power to kick God? what deal have you wrought out?
16:26:41 -!- Phantom_Hoover has quit (Ping timeout: 245 seconds).
16:28:54 -!- Phantom_Hoover has joined.
16:31:18 -!- JWinslow23 has joined.
16:31:39 <JWinslow23> Sorry about the flood, you guys. I promise I won't ever again.
16:32:27 <JWinslow23> I'm still looking for ideas on how to make TicTacToe stack-based, though.
16:34:31 -!- ^v has quit (Ping timeout: 246 seconds).
16:34:45 <myname> i had some ideas on how a program may look like, but it isnn't closely based on the game
16:35:37 <JWinslow23> So you know, I want it to be like the original at least a little bit.
16:35:46 <myname> like, each command is made out of 2 numbers that represent a move, moving again on the same spot will remove that mark, the programm terminates if one player has won
16:36:31 <JWinslow23> Still isn't stack based, though. i also want it to follow the rules of the game.
16:36:53 <JWinslow23> Like, you don't throw the checkers board at someone's face, do you?
16:42:30 <Sgeo> How else would you play?
16:43:23 -!- JWinslow23 has quit (Ping timeout: 250 seconds).
16:43:44 <Koen__> Sgeo: throw the pawns at someone's eyes
16:46:26 -!- ^v has joined.
16:46:58 <Koen__> so imagine you've got a checker's game going on
16:47:15 <Koen__> every pawn carries a value
16:47:38 <Koen__> pawns that get captured go to the prisoners queue of the appropriate colour
16:48:28 <Koen__> whenever a pawn reaches the end row, it gets promoted: a pawn is taken from the appropriate prisoner queue and pushed on it (so it's a stack of pawns)
16:49:04 <Koen__> note that promotion happen not only to single pawns, but to any stack which reaches the end row
16:50:06 <Koen__> popping an empty queue presumably returns a fresh pawn with value 0, so that memory isn't limited
16:50:22 -!- yorick has quit (Remote host closed the connection).
16:50:49 <Koen__> now the question is: how to program this game?
16:51:30 <Koen__> would be fun if a program was a list of directions, strategies, etc
16:51:40 <Koen__> and the game is played at runtime by a bot following those directions
16:53:21 <Koen__> sad thing is, currently the number of stacks is limited to the number of pawns initially present on the board, and even though the total number of pawns is unbounded, the number of stacks can never grow, and it will in fact decrease every time a stack is captured
16:54:53 <Koen__> directions can be of the kind "whenever you can capture, capture" or "if you can promote, promote" or whatever
16:55:06 -!- conehead has joined.
16:55:20 -!- conehead has quit (Remote host closed the connection).
16:55:38 -!- conehead has joined.
16:55:59 <Koen__> currently, interaction between the two sides can only be done through play; black values are completely independant of white values
16:56:06 <Koen__> unless you can give directions that depend on the values
17:03:00 -!- Koen__ has quit (Read error: Connection reset by peer).
17:03:20 -!- Koen__ has joined.
17:07:11 -!- nisstyre has joined.
17:18:56 <Bike> Gregor: thank you!
17:19:22 -!- Bike has set topic: fnord, offset fnord (with no evidence) | PDF while supplies last:u/2023808/wisdom.pdf | http://tunes.org/~nef/logs/esoteric/ | this space intentionally left right.
17:20:12 -!- Bike has set topic: fnord, offset fnord (with no evidence) | PDF while supplies last: https://dl.dropboxusercontent.com/u/2023808/wisdom.pdf | logs: http://codu.org/logs/_esoteric/ or http://tunes.org/~nef/log/sesoteric/ | this space intentionally left right.
17:23:20 -!- ion has set topic: fnord, offset fnord (with no evidence) | PDF while supplies last: https://dl.dropboxusercontent.com/u/2023808/wisdom.pdf | logs: http://codu.org/logs/_esoteric/ or http://tunes.org/~nef/log/sesoteric/ | this spacetime intentionally left right.
17:31:44 -!- nisstyre has quit (Quit: Leaving).
17:40:31 -!- erterrr has joined.
17:40:47 -!- erterrr has left.
17:50:16 <kmc> oh in the url
17:50:19 -!- zzo38 has joined.
17:50:33 <Sgeo> I miss mutation
17:51:31 -!- Bike has set topic: fnord, offset fnord (with no evidence) | PDF while supplies last: https://dl.dropboxusercontent.com/u/2023808/wisdom.pdf | logs: http://codu.org/logs/_esoteric/ or http://tunes.org/~nef/logs/esoteric/ | this spacetime intentionally left right.
17:51:48 <zzo38> What kind of mutation?
17:52:14 <Bike> hardy-weinberg violating mutations
17:53:08 <kmc> wow so you can plug a Firefox OS phone into your computer, open up the Firefox developer tools on the computer, and inspect / debug / mess with the running app on the phone, in realtime
17:53:39 <kmc> so if you don't like the background color or whatever you can just fix that
17:54:44 <kmc> also we have a pure JavaScript implementation of Flash which is good enough to play some real games etc, and it's in FF nightly
17:56:32 <kmc> good for a lot of reasons, one of which is that new versions of Adobe Flash on Linux are Chrome-only :/
17:57:33 <Bike> uh, wow, that's gotta suck.
17:57:36 <Gregor> But can it do the one thing that people actually use Flash for: Play videos
17:58:02 <kmc> Flash games are popular too
17:58:13 <kmc> but I don't know what the status of video is
17:58:39 <kmc> it seems like it shouldn't be that hard, because you only need to run enough Flash to get a video URL to hand off to the browser's already existing video support
17:58:43 <kmc> but I don't know shit really
17:58:54 <Sgeo> kmc: that debugging thing I think other things can do too for web sites
17:58:57 <kmc> of course we'd rather people use HTML5 video
17:59:12 <Sgeo> iOS's browser can be hooked up to Safari on Mac
17:59:33 <kmc> Sgeo: but you can do it for any FxOS app, because they're all using Web tech
17:59:41 <kmc> that's the cool part imo
18:01:58 <kmc> myname: yep
18:16:25 -!- Zuu_ has joined.
18:17:02 -!- Zuu has quit (Read error: Connection reset by peer).
18:25:11 -!- Zuu_ has quit (Ping timeout: 245 seconds).
18:27:00 -!- Zuu has joined.
18:34:45 -!- kmc_ has joined.
18:35:37 -!- kmc_ has quit (Client Quit).
18:38:48 -!- Taneb has joined.
18:44:57 -!- asie has joined.
18:59:12 -!- AnotherTest has joined.
19:00:50 -!- yiyus has quit (Ping timeout: 245 seconds).
19:01:01 -!- Zuu has quit (Ping timeout: 245 seconds).
19:01:20 -!- Zuu has joined.
19:04:30 -!- yorick has joined.
19:25:09 -!- yiyus has joined.
19:32:06 -!- ^v has quit (Ping timeout: 245 seconds).
19:34:18 -!- Zuu has quit (Remote host closed the connection).
19:35:36 -!- yiyus has quit (Ping timeout: 245 seconds).
19:38:45 -!- yiyus has joined.
19:40:37 -!- Taneb has quit (Quit: Page closed).
19:45:53 -!- Zuu has joined.
19:49:07 -!- ^v has joined.
19:49:24 -!- ^v has quit (Client Quit).
19:49:42 -!- ^v has joined.
19:58:26 <kmc> shachaf: do you share my intuition that a hash trie where each node is refcounted would have terrible performance?
19:58:58 <kmc> (in a concurrent setting anyway)
19:59:06 <kmc> due to making every node 1 word bigger + atomic ops overhead + loss of cacheline sharing
19:59:11 -!- AnotherTest has quit (Read error: Connection reset by peer).
19:59:43 -!- AnotherTest has joined.
20:01:13 * impomatic is searching for a copy of CoreMaster by Think Software. Zero hits from Google, so I might struggle...
20:01:30 <Fiora> kmc: what sort of structure would it be replacing?
20:04:30 <kmc> I don't have a specific application in mind, but the advantage over a flat hash table is that you get pure-functional update / cheap snapshotting, and the advantage over a simple search tree is that it's a lot faster
20:05:18 <Fiora> I guess if it's mostly read, and rarely written to, it'd be okay, right?
20:05:33 <Fiora> I guess that'd be a hash table where you can easily remove things, right...?
20:05:41 <kmc> but in that case you could also consider copying a flat hash table
20:06:16 <kmc> well the unique thing is more that your add/modify/remove operations give you a "new" datastructure (which mostly shares nodes with the old) rather than modifying in place
20:06:20 -!- Zuu has quit (Remote host closed the connection).
20:06:33 <Fiora> how does that work O_O
20:06:44 <Fiora> wouldn't each node have like, pointers to its child nodes and stuff?
20:06:50 <kmc> do you know how it works for a plain binary search tree?
20:06:58 -!- Zuu has joined.
20:07:03 <Fiora> I... I guess you replace the node you're changing, and all of its parents?
20:07:10 <kmc> if you implement a BST in Haskell in the most naive way, you get this for free ^_^
20:07:24 <Fiora> so you end up modifying... log(N) of the tree nodes if it's balenced?
20:07:25 <kmc> so if the tree is balanced, you copy only O(log n) data
20:07:46 <kmc> and a hash trie is a tree where instead of going left/right based on a pairwise comparison, you go left/right based on the bits of a hash output
20:08:00 <kmc> and furthermore you probably want a 16- or 32-way fanout instead of 2-way
20:08:25 <kmc> and furthermore most of those 16 or 32 pointers will be NULL most of the time, so you can use clever bit tricks to store nodes sparsely
20:08:39 <kmc> and fit each node into a cache line
20:08:54 <Fiora> 64 bytes is pretty big for a node, isn't it? or am I missing something
20:09:19 <olsner> 64 bytes is a cache line, isn't it?
20:09:31 <kmc> or fit multiple nodes in a cache line, I guess
20:09:41 <kmc> but you take care that they're aligned, anyway
20:09:56 <Fiora> there'd only becontention if you had multiple nodes in a cacheline, right?
20:11:29 <kmc> you mean with refcounting?
20:11:59 <Fiora> oh you mean contention with like, multiple threads updating a single node at the same time? not cacheline contention with like, multiple nodes being modified in the same cacheline
20:12:14 <kmc> the problem is that when you do a functional update, you need to bump the refcount on the nodes you're reusing (even though you aren't modifying them at a semantic level), and this will kill sharing of that cacheline between cores
20:12:29 <Fiora> ohhh. but doesn't that mean modifying the trie is O(N)?
20:12:38 <kmc> so nodes will bounce between caches in exclusive mode, instead of being shared, which is what you would want from ostensible immutable data
20:12:39 <Fiora> because you have to modify O(N-log(N)) nodes' refcounts
20:12:49 <Fiora> or am I totally wrong here
20:12:56 <kmc> I think no, because you only count immediate referants not transitive referants
20:13:12 <kmc> the child of a node you're reusing already has a refcount from its parent
20:13:39 <kmc> but you're creating O(log n) new nodes that refer to some old ones
20:13:58 <Fiora> okay, so you update the refcounts on all the children of your new nodes?
20:13:59 <kmc> and you need to bump the refcount on all the old nodes
20:14:01 <Fiora> erm, immediate children
20:14:06 <kmc> yeah, I think so
20:14:12 <Fiora> not all the old nodes... okay that sorta makes sense
20:14:24 <Fiora> so it's log(N) * some constant refcount updates and log(N) new nodes ish?
20:14:43 <kmc> I think so
20:15:24 <kmc> though a 16-fanout hash trie over a 64-bit hash will have a maximum depth of 4, anyway
20:15:47 <fizzie> If you have 16-fanout, isn't that just 4 bits?
20:15:48 <kmc> so I shouldn't say O(log N)
20:16:01 <kmc> ah, yes, I can't count
20:16:05 <kmc> depth of 16 then
20:16:16 <kmc> what's 4 orders of magnitude between friends
20:16:32 <fizzie> Absolutely nothing. Here, have a flower: https://dl.dropboxusercontent.com/u/113389132/Misc/20131005-flower.jpg
20:16:44 <fizzie> (I just wanted an excuse to share the photo.)
20:16:57 <kmc> what kind is it?
20:17:13 <olsner> maybe instead of addrefing the new children you could just addref the old node you copied
20:17:19 <olsner> that overcounts one reference to something you've removed from the new tree, but you can fix that later
20:17:26 <fizzie> No idea. I think I was told, but I've forgotten already.
20:17:48 <kmc> olsner: do you then need to keep a pointer to the old node so you can collect it when the new one goes away?
20:18:08 <olsner> (wave hands until reaching a working algorithm)
20:18:41 <olsner> yes, you'd probably need another pointer somewhere
20:18:49 <fizzie> Huh, Oxford University Flower Recognition Demo actually maybe even recognized it, I think: http://zeus.robots.ox.ac.uk:8080/flower_demo/demo_result?id=3152
20:18:55 <fizzie> At least the first hit looks quite similar.
20:19:16 -!- asie has quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz...).
20:19:27 <fizzie> Er, not the first, the fourth.
20:19:43 <kmc> woah that's a thing?
20:20:18 -!- Zuu has quit (Remote host closed the connection).
20:20:34 <fizzie> kmc: Their picture was the first hit in a Google "search by image" query, that's how I found it.
20:20:52 -!- Zuu has joined.
20:20:55 <fizzie> (First time that was useful for anything.)
20:22:56 -!- MindlessDrone has quit (Quit: MindlessDrone).
20:24:03 -!- AnotherTest has quit (Read error: Connection reset by peer).
20:24:14 -!- AnotherT- has joined.
20:24:22 -!- AnotherT- has changed nick to AnotherTest.
20:29:30 -!- Zuu has quit (Ping timeout: 240 seconds).
20:32:05 -!- Zuu has joined.
20:35:46 -!- AnotherTest has quit (Read error: Connection reset by peer).
20:36:23 -!- AnotherTest has joined.
20:40:11 -!- Zuu has quit (Ping timeout: 245 seconds).
20:40:17 -!- Zuu has joined.
20:44:36 -!- KingOfKarlsruhe has joined.
20:44:36 -!- AnotherTest has quit (Read error: Connection reset by peer).
20:44:55 -!- AnotherT- has joined.
20:45:04 -!- AnotherT- has changed nick to AnotherTest.
20:47:38 -!- AnotherTest has quit (Read error: Connection reset by peer).
20:48:16 -!- AnotherTest has joined.
20:56:30 <impomatic> Has anyone here got access to the Tamiment Library at New York Uni?
21:05:54 -!- JWinslow23 has joined.
21:06:18 -!- Zuu has quit (Excess Flood).
21:06:58 -!- Zuu has joined.
21:08:18 <Bike> that's a pretty specific question
21:08:27 <HackEgo> ZuU: 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.)
21:12:01 <nortti> `run echo "WeLcOmE $* | rainwords" > bin/ReLcOmE
21:12:16 <nortti> `chmod 755 bin/ReLcOmE
21:12:17 <HackEgo> chmod: missing operand after `755 bin/ReLcOmE' \ Try `chmod --help' for more information.
21:12:24 <nortti> `run chmod 755 bin/ReLcOmE
21:13:33 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: cls: not found
21:13:55 <JWinslow23> `learn cls is a command to clear the screen.
21:14:11 <HackEgo> Runs arbitrary code in GNU/Linux. Type "`<command>", or "`run <command>" for full shell commands. "`fetch <URL>" downloads files. Files saved to $PWD are persistent, and $PWD/bin is in $PATH. $PWD is a mercurial repository, "`revert <rev>" can be used to revert to a revision. See http://codu.org/projects/hackbot/fshg/
21:14:22 <HackEgo> wisdom is always factually accurate, except for this entry, and uh that other one? it started with like, an ø?
21:14:43 <Bike> like most interactive systems, this thing has like four separate help systems, the distinction of which is obscure
21:14:50 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: cls: not found
21:14:57 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: wisdom: not found
21:15:12 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: How: not found
21:15:21 <HackEgo> wisdom is always factually accurate, except for this entry, and uh that other one? it started with like, an ø?
21:15:21 <HackEgo> cls is a command to clear the screen.
21:15:44 <Bike> wow, that would actually have been helpful in this context.
21:18:06 <HackEgo> Ørjan is oerjan's good twin. He's banned in the IRC RFC for being an invalid character. Sometimes he publishes papers.
21:19:13 <nortti> :/ even though IRCNet allows #øæ’jµ it doesn't allow nick ørjan
21:19:16 <impomatic> Bike: the Uni Library has a one of a kind item in their collection I'm interested in.
21:19:40 <Bike> just wondering. i don't have access.
21:20:19 <HackEgo> cls is a command to clear the screen.
21:20:54 <HackEgo> wisdom is always factually accurate, except for this entry, and uh that other one? it started with like, an ø?
21:21:11 <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.)
21:21:38 <HackEgo> cls is a command to clear the screen.
21:22:02 <JWinslow23> `learn Batch is a language that uses the command cls.
21:22:16 <HackEgo> Batch is a language that uses the command cls.
21:22:53 <Bike> elliott: i just realized that indiana jones should probably have taken the grail solely for its archaelogical value, irrespective of the moral choice thing.
21:23:13 <Bike> elliott: well, i mean, in the forties. archaelogists still basically stole crap back then i think
21:23:40 <JWinslow23> `learn Humphrey Bogart is a cousin to Lady Di.
21:23:55 <HackEgo> Humphrey Bogart? ¯\(°_o)/¯
21:24:11 <HackEgo> Humphrey Bogart is a cousin to Lady Di.
21:24:22 <HackEgo> aah ambiguous acronym here
21:24:54 -!- Zuu has quit (Remote host closed the connection).
21:24:54 <HackEgo> HackEgo, also known as HackBot, is a bot that runs arbitrary commands on Unix. See `help for info on using it. You should totally try to hax0r it! Make sure you imagine it's running as root with no sandboxing.
21:25:02 <HackEgo> NIMBY : Not in my backyard? ¯\(°_o)/¯
21:25:24 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: NIMBY: not found
21:25:29 <HackEgo> NIMBY : Not in my backyard.
21:25:31 <HackEgo> NIMBY : Not in my backyard.
21:25:49 <kmc> CAVE People : Citizens Against Virtually Everything
21:26:14 <JWinslow23> BANANA: Build absolutely nothing anywhere near anybody
21:26:35 -!- nooodl has joined.
21:27:23 <impomatic> Bike, elliot: They have some negatives from a Core War tournament
21:27:38 <Bike> oh, like, photos?
21:27:52 <JWinslow23> `? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯?
21:27:54 <HackEgo> ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯? ¯\(°_o)/¯?? ¯\(°_o)/¯
21:28:20 <HackEgo> /hackenv/bin/learn: line 4: wisdom/: Is a directory \ I knew that.
21:28:33 <HackEgo> ¯\(°_o)/¯? ¯\(°_o)/¯?? ¯\(°_o)/¯
21:28:40 <impomatic> Yes. I have an archive or Core War stuff. Am trying to find as much as possible before it disappears forever.
21:28:48 -!- ChanServ has set channel mode: +o elliott.
21:28:52 -!- elliott has set channel mode: +qq HackEgo!*@* myndzi!*@*.
21:28:54 <JWinslow23> I'm trying to make a quine in the shell thing, Phantom Hoover!
21:29:16 <Phantom_Hoover> if you want to do spammy things with the bots, do them in /query
21:29:24 <Bike> I'm good at this.
21:29:34 <impomatic> Also, I'm doing the same for other programming games, e.g. Darwin and RobotWar.
21:29:50 -!- elliott has set channel mode: -qqo HackEgo!*@* myndzi!*@* elliott.
21:29:51 -!- AnotherTest has quit (Ping timeout: 252 seconds).
21:30:21 -!- Zuu has joined.
21:30:37 <Bike> impomatic: have you contacted the library? maybe they can just like, scan it for you.
21:30:44 <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.)
21:31:01 -!- ais523 has joined.
21:32:22 <HackEgo> cat: quines: Is a directory
21:32:29 <HackEgo> cat \ perl \ python \ q \ ruby
21:32:34 <Bike> `ls quines/cat
21:32:39 <Bike> `cat quines/cat
21:34:20 <elliott> `run echo q; cat quines/cat
21:34:28 <elliott> I was hoping it actually said "No output."
21:36:21 <ais523> elliott: it'd still be a quine if it did, ofc
21:36:30 <impomatic> Bike: will be sending an email. Just need to write a decent description of the project.
21:36:50 <Zuu> nortti: thanks
21:36:50 <Zuu> my poopish connection is being poopish again :(
21:37:38 <kmc> popeish connection
21:42:55 <myname> JWinslow23: you are bored, aren't you?
21:44:02 <JWinslow23> Anyone have any suggestions for making Tic Tac Toe (my language) stack based?
21:44:12 <zzo38> I don't have a suggestion
21:44:32 <ion> 0) Add a stack
21:44:52 <zzo38> Is it the cable, the monitor, or the motherboard that is causing the bad video?
21:44:54 <JWinslow23> Well yes, but what 8 commands can I use for stack manipulation?
21:45:11 <myname> JWinslow23: +-><[],. :p
21:45:31 <Koen__> push, pop, swap, dup...
21:45:57 <JWinslow23> Koen, you may be hitting on something!
21:46:00 <Koen__> just pop any page from http://esolangs.org/wiki/Category:Stack-based
21:46:15 <Koen__> and rip off what's there and make a copy and post it
21:46:20 <zzo38> 1 - * DUP there are others too
21:46:38 <JWinslow23> I am! Just need something with 8 commands. I haven't found that yet, though.
21:46:49 <nooodl> i like how +-><[],. is a valid "head -c 1" program
21:47:26 <nooodl> i wonder how many different programs you can get from anagrams of the 8 brainfuck commands
21:47:28 <zzo38> The +-><[] is irrelevant
21:47:47 <zzo38> They cancel each other out and [] is nothing when the value is already zero
21:48:00 <Bike> impomatic: apparently i know somebody who has access.
21:48:39 <Koen__> ,.-[>+<] everythign relevant
21:50:08 <Koen__> JWinslow23: if > means "push 0" and < means "pop" then +-><[],. are the commands for a stack-based language
21:50:53 <myname> Koen__: that shouldn't be turing complete
21:51:15 <myname> where would you pop to?
21:51:25 <ion> Don’t pop where you eat.
21:52:10 <nooodl> <>[,+.-] enterprise brainfuck encryption
21:52:18 <Koen__> or maybe there are two stacks and > and < just means "pop from a to push onto b" and vice versa
21:52:22 <Koen__> which is basically brainfuck
21:52:54 <Koen__> assuming popping an empty stack gives a zero
21:53:03 <Koen__> (well, gives something, anyway)
21:53:43 <ais523> nooodl: I interpreted your [,+.-] loop as just being a caesar cipher that adds 1 to every codepoint
21:53:56 <ais523> that seems pretty much like enterprise brainfuck encryption to me
21:54:12 <zzo38> Can it be made some esolang that is Turing complete if transcendental numbers are allowed and not Turing complete if it isn't?
21:54:15 -!- Zuu__ has joined.
21:54:23 -!- Zuu has quit (Read error: Connection reset by peer).
21:54:49 <JWinslow23> Post on the talk page. I will play Monopoly.
21:54:51 -!- JWinslow23 has quit (Quit: Page closed).
21:54:55 <Koen__> making such an esolang is your task now, zzo38
21:54:58 <nooodl> i was going to call it that, but i predicted someone would point out how it doesn't properly wrap Z to A
21:55:12 <impomatic> Bike: well if they're feeling generous, they're in box 26b, shoot 300030. ;-)
21:55:13 -!- Zuu__ has changed nick to Zuu.
21:55:51 <myname> i thought about making a language that only has half-bits as data types
21:56:44 <Bike> i should probably have a better understanding of how fractional bits work.
21:56:55 <impomatic> Just the way the reference negatives I think!
21:56:56 <elliott> @google turkey bomb cat's eye
21:56:57 <lambdabot> http://catseye.tc/node/TURKEY%20BOMB.html
21:56:57 <lambdabot> Title: TURKEY BOMB | Cat's Eye Technologies
21:57:30 <myname> Bike: my basic idea was to say "half-bit b is the oposite of half-bit a" and so making a decidable full bit
21:57:37 <ais523> is catseye working again, then?
21:57:39 <Bike> well hooking you up with some random undergrad probably won't help you out
21:59:01 <Fiora> myname: is that like quantum entanglement?
21:59:07 <Fiora> like, if you have two particles A and B, and you know they have opposite spins
21:59:14 <Fiora> you have one bit of information about A and B, so half a bit about each, right
21:59:26 <Bike> knowing what a number in 0-9 is exactly is ln 10 bits of information, right?
21:59:51 <myname> Fiora: yeah, but if you don't declare b the opposite of a and you ask of the value of a you either get 1 or it doesn't terminate
22:00:18 <elliott> well, the problem is you're asking for a bit out of a half-bit there, no?
22:00:24 <elliott> so of course it doesn't necessarily terminate, it isn't a total operation
22:00:41 <Bike> wow why did i think ln meant log 2, yes, log 2
22:01:09 <fizzie> You could argue that the natural log should be the one that's most natural given the context.
22:01:18 <Bike> i don't want to, though.
22:01:28 <myname> elliott: well, if the opposite of a semi-decidable language is semi-decisable, then the language itself is decidable and so should be the bit
22:01:47 <Fiora> isn't the natural log always the most natural
22:02:32 <kmc> natural log is best log
22:03:10 <Bike> so knowing the value of a nonnegative integer less than the square root of two should be half a bit of information. clearly. i've got this down.
22:03:45 <myname> Bike: 0 to 9 is more than 1 bit
22:04:18 <Bike> it's log2 10 bits, i said.
22:04:22 <Fiora> fractional bits has me thinking of arithmetic coding -_-
22:07:34 <Bike> that kind of makes sense intuitively, since knowing the value of the third bit of a nonnegative integer less than ten is useless if the other three bits are 111, or 110, or 101, or 100, or 011, or 010, but not if they're 001 or 000.
22:09:05 <zzo38> I have also thought of fraction bits of information, but I have also tried to think of programming with negative number of bits of information, too.
22:10:52 <elliott> Bike: now figure out negative bits.
22:11:18 <Bike> elliott: knowing the naked appearance of a united states supreme court justice.
22:12:20 <Bike> or: the value of a nonnegative integer less than 1/x has -x bits.
22:14:41 <ais523> the negative bits thing in TURKEY BOMB is easy enough to figure out
22:14:58 <ais523> if you have a negative-bits value, then it requires more storage to represent in anything that contains everything except that value
22:15:09 <Bike> oh. right, you just don't use a uniform distribution.
22:18:17 <Bike> if x is discretely uniformly distributed between 0 and 9, the third bit is zero with 80% probability, and the second is zero with 40%, bla bla bla.
22:18:27 <Bike> thank you for permitting me to go over this stupid basic stuff
22:19:42 <ais523> Bike: that works for fractional amount of bits between 0 and 1
22:21:21 <ais523> Bike: using weighted probabilities to get values with less than a bit
22:21:40 <ais523> I guess you'd have a negative number of bits in storage in something that was 120% likely to be a 0, and -20% likely to be a 1
22:22:17 <Bike> Fiora: negative probabilities. quick say something quantum
22:23:37 <elliott> ais523: so we just need more people to give it 120% if we want to store negative bits?
22:23:38 <ais523> well negative bits don't really make sense
22:23:54 <ais523> if you had enough negative-sized files, you could just store a load of them to free up disk space
22:24:31 <Bike> oh, good, and i forgot that the bits aren't independently distributed. i am terrible today.
22:26:21 <Bike> i'm reasonably sure negative information is a used concept. maybe only conditional information, though.
22:27:02 <Bike> only in quantum. great
22:27:51 -!- Sprocklem has joined.
22:28:07 <Bike> "The negative conditional entropy is also known as the coherent information, and gives the additional number of bits above the classical limit that can be transmitted in a quantum dense coding protocol" and here we run into the famous "bike doesn't know know quantum" paradox of einstein's
22:29:36 -!- sebbu has quit (Ping timeout: 245 seconds).
22:30:00 -!- sebbu has joined.
22:32:27 <Koen__> dan simmons has this nice "use the word 'quantum' whenever you don't feel like explaining how it works" paradigm
22:34:34 <Bike> i thinki have shannon's paper lying around somewhere. i could just read it.
22:40:11 -!- mnoqy has joined.
22:44:11 -!- oerjan has joined.
22:46:21 <oerjan> <Sgeo> onfail $ run $ head -10 "README"
22:46:39 <oerjan> <Sgeo> (onfail (run (head -10 "README")))
22:47:04 <oerjan> what language is this? technically haskell does that but -10 doesn't work sensibly for it.
22:47:39 <oerjan> (because the - is binary subtraction)
22:48:27 * oerjan thinks there should be a flag for people who have been idle for hours.
22:48:28 <olsner> is it? I thought that was one of those horrible bits of odd syntax
22:48:31 <olsner> but of course I don't remember which direction it was odd
22:49:01 <zzo38> oerjan: There is in ifMUD (although you can disable this feature). In IRC you can check using WHOIS.
22:49:04 <oerjan> olsner: the odd part is that (- 10) is unary minus, and not a section using the binary one.
22:50:07 <oerjan> zzo38: by flag i mean something that shows up in the initial channel listing.
22:50:38 <lambdabot> Maybe you meant: wiki time list elite dice cide
22:50:39 <zzo38> In the NAMES list?
22:51:25 <oerjan> zzo38: hm well i guess irssi probably sends a NAMES command to get it.
22:51:46 <oerjan> or something like that.
22:52:24 <zzo38> (It is also received when you join the channel, automatically)
22:52:49 <zzo38> (You can push NAMES to make it receive even afterward, though.)
22:54:44 <Vorpal> This is like the stupidest thing youtube has done so far... It decided to start all videos muted as of late afternoon today.
22:54:45 <zzo38> I have the idea of a new kind of trainer card for use in Pokemon card game: Pick one card from your trash, hand, or an energy attached to one of your cards in play, and put it with your side cards. You cannot use this if you already have six side cards.
22:55:03 <Sgeo> oerjan: Any Lispy language that allows sweet-exprs
22:56:37 <Bike> sweet exprs. ok.
22:56:39 <oerjan> Vorpal: does that include the preceding ad? if so that will be a feature for me :D
22:56:49 <zzo38> What is "sweet-exprs"?
22:57:26 <Sgeo> zzo38: http://readable.sourceforge.net/
22:57:29 <Vorpal> oerjan, Hm, no idea, I have been watching ad-free channels most of the afternoon
22:57:50 <zzo38> Can you explain to me so I don't have to open it right now? I will open to read later.
22:57:56 <fizzie> Vorpal: Hey, I got that too.
22:58:16 <Vorpal> oerjan, you could try adblock btw, though of course if you do, that would prevent the independent creators from getting those 0.01 cents from your view or whatever
22:58:19 <Bike> woo, more syntax arguments
22:58:22 <Vorpal> fizzie, really? How to fix
22:58:47 <fizzie> Vorpal: No idea. (Took me a moment staring at alsamixer settings wondering what's wrong before noticing the youtube-level mute.)
22:58:49 <Sgeo> zzo38: a reader-level pass that takes something like {a + b + c} and turns it to (+ a b c), for example
22:58:55 <oerjan> Vorpal: is there adblock for IE?
22:59:02 <zzo38> Sgeo: O, that's what you mean.
22:59:23 <Sgeo> It only takes a "friendlier" (by someone's standards) syntax and converts it in a straightforward way to s-exprs
22:59:33 <fizzie> Vorpal: "Thanks for your patience, all. We have been aware of this issue. It should be resolved soon.
22:59:35 <Sgeo> No semantic interpretation is done
22:59:36 <fizzie> While this issue is live, please try clearing your browser's cache and cookies, --"
22:59:43 <Bike> isn't that just a parser?
22:59:45 <fizzie> (Writes someone on the Google product forums.)
22:59:52 <Vorpal> fizzie, ... seriously? Heh
23:00:05 <Sgeo> Bike: I guess?
23:00:11 <Vorpal> I would assume you would need to clear the flash cookie thingies
23:00:12 <Bike> into sexps instead of asts, i guee.
23:01:48 <Sgeo> At least the {} seems like a really good idea for math stuff. The indentation stuff freaks me out a bit
23:02:37 <Bike> i don't really get why "readable lisp" is such a persistent idea. i mean, have you ever read a math paper, the notation is worse than any programming language.
23:02:43 <Bike> but it's not important anyway.
23:03:04 <zzo38> I do not think all of mathematical notation is necessarily so bad.
23:03:15 <Bike> hopefully i'll never have to hear somebody say SVO is natural for huans ever again.
23:03:58 <zzo38> Also I don't think Lisp and Forth and so on have really bad syntax even though some people says it is.
23:04:32 <oerjan> Bike: is math worse than even apl?
23:05:14 <Bike> The way I figure it, math needs its own typesetting programming language, and no programming language i know of does.
23:05:27 <Bike> quite possible this is comparing apples and oranges, but...
23:05:43 <Sgeo> There's an SFRI in 'final' status for the infix stuff
23:10:08 <zzo38> The kind of card I described above, can have a lot of possible uses. (However, it is usually pretty worthless by themself.)
23:10:28 <Sgeo> I see a potential confusion
23:10:46 <Sgeo> {a get-new-op() b get-new-op() c}
23:10:59 <Sgeo> (get-new-op) is only present once in what that expands to
23:11:22 <Sgeo> ((get-new-op) a b c)
23:11:32 <Vorpal> fizzie, did clearing the cache work?
23:12:23 <oerjan> @tell Taneb i suggest staying away from IngSoc
23:15:00 <Bike> Sgeo: should be ((begin (get-new-op) (get-new-op)) a b c), clearly.
23:15:48 <Bike> that's actually kind of interesting though, i suppose it means no associativity.
23:16:38 <Sgeo> There isn't. There's also no operator mixing like {a + b - c} unless you define a macro... it expands to ($nfx$ a + b - c)
23:17:55 <zzo38> What is the best way to define an ordering in SQL so that you can insert a record in between two specified records (even if some records are deleted before that happens)?
23:18:09 <Bike> i mean associativity in scheme in general. like, (op) can always return an associative operator, and yet you can have ((op) ((op) a b) c) != ((op) a ((op) b c)).
23:18:42 <myname> zzo38: well, if you don't specify any "ORDER BY" clause, sql results are "unordered"
23:19:03 <myname> zzo38: if you do have ORDER BY, just make a value at the position you want
23:22:59 <zzo38> myname: I know that; I mean to define the collation of the field and the function to make it create these ordering values, and then you can use ORDER BY that field.
23:23:49 -!- JWinslow23 has joined.
23:25:29 <Bike> print("hakkkashdfhanuipioquqqqquiusakdfjii10")
23:26:33 <HackEgo> JWinslow23 is a Wisconsinite who doesn't give a BF.
23:27:48 -!- Sgeo has quit (Read error: Connection reset by peer).
23:29:48 <zzo38> So do you know how to do this, then?
23:30:28 -!- Sprocklem has quit (Ping timeout: 240 seconds).
23:32:23 -!- nisstyre has joined.
23:32:41 -!- Sgeo has joined.
23:33:36 <Bike> ^ul (:aSS):aSS
23:34:47 <JWinslow23> ^ul (:aS(:^S^:)Sa:):^S^:(:aS(:^S^:)Sa:)
23:34:47 <fungot> (:aS(:^S^:)Sa:):^S^:(:aS(:^S^:)Sa:)
23:35:15 <fungot> ^<lang> <code>; ^def <command> <lang> <code>; ^show [command]; lang=bf/ul, code=text/str:N; ^str 0-9 get/set/add [text]; ^style [style]; ^bool
23:35:22 <fungot> ^<lang> <code>; ^def <command> <lang> <code>; ^show [command]; lang=bf/ul, code=text/str:N; ^str 0-9 get/set/add [text]; ^style [style]; ^bool
23:35:32 <Bike> the available languages are bf and ul, it says.
23:35:59 <Bike> haha, you skimmed one line huh
23:36:52 <fungot> <CTCP>.. !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~ ...
23:38:18 <oerjan> wait how does that even work
23:39:20 <JWinslow23> ^bf ,[.,]~Oh, yeah. Anything over 1 requires loop.
23:39:35 <JWinslow23> ^bf ,[.,]!Oh, yeah. Anything over 1 requires loop.
23:39:35 <fungot> Oh, yeah. Anything over 1 requires loop.
23:39:42 <fungot> Oh, yeah. Anything over 1 requires loop.
23:41:09 <Bike> inks, as in the end of my message ^Ablinks.^A
23:41:10 <olsner> fungot, the fixpoint out of time
23:41:11 <fungot> olsner: gmane even has a chance of comming back to it and then complain that it `does not work'.
23:42:08 <oerjan> fizzie: i think we found a new fungot bug
23:42:09 <fungot> oerjan: but i'd say haskell introduces extra semantics, then. i'm surprised he doesn't have to be a bit less silly. i can feel it.
23:42:35 <JWinslow23> ^bf ->+>+++>>+>++>+>+++>>+>++>>>+>+>+>++>+>>>>+++>+>>++>+>+++>>++>++>>+>>+>++>++>+>>>>+++>+>>>>++>++>>>>+>>++>+>+++>>>++>>++++++>>+>>++>+>>>>+++>>+++++>>+>+++>>>++>>++>>+>>++>+>+++>>>++>>+++++++++++++>>+>>++>+>+++>+>+++>>>++>>++++>>+>>++>+>>>>+++>>+++++>>>>++>>>>+>+>++>>+++>+>>>>+++>+>>>>+++>+>>>>+++>>++>++>+>+++>+>++>++>>>>>>++>+>+++>>>>>+++>>>++>+>+++>+>+>++>>>>>>++>>>+>>>++>+>>>>+++>+>>>+>>++>+>++++++++++++++++++>>>>+>+>>>+>>+
23:42:54 <olsner> is it a bug? if fungot does it, surely it must be a feature
23:42:55 <fungot> olsner: 19:05 is a funny gmt time to get to the code that implements the foo interface exports the bindings foo1, foo2 and foo3". when i reload it goes to things that exist outside the actor with unique names with the limited input/ output
23:43:30 <oerjan> JWinslow23: irc lines have a < 510 byte limit (which includes some stuff other than the message.)
23:43:50 <Bike> do some compression!
23:44:15 <oerjan> JWinslow23: fungot's ^str command is to get around that, although it's awkward to use.
23:44:15 <fungot> oerjan: though it would be something wrong otherwise. that's how i roll.
23:46:05 <Bike> ^bf [,>][<]>[.>][<]>[.>]!test
23:46:55 <Bike> fungot, buddy, speak to me.
23:46:55 <fungot> Bike: you can do computation with syntax-rules using cps or using other macros ( or non-hygienic low-level macros) wouldn't you have to on solaris
23:47:24 <Bike> i fail to see the connection between Solaris and computation at macroexpansion time, fungot.
23:47:25 <fungot> Bike: i was just busy planning lea's demise of laidback because of its cgi infrastructure. i'm more than halfway through this, i'll be willing to change, and so
23:47:50 <Bike> oh, i see, no output for me.
23:47:52 <fungot> output for me.<CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP><CTCP> ...
23:48:02 <Bike> ^bf ,[,>][<]>[.>][<]>[.>]!test
23:48:12 <Bike> i'm so bad at this.
23:48:44 <kmc> fungot: fungot
23:48:45 <fungot> kmc: one more thing cl stores in symbols, right? but computing the hash value in the middle of interviewing for a job right now,
23:48:49 <oerjan> seems like it doesn't read past the NUL
23:49:03 <Bike> fungot's definitely getting that job
23:49:03 <fungot> Bike: makes it too easy to spot
23:49:09 <oerjan> even if forced to use ,
23:49:32 <kmc> fungot: are you an ascomycete or a basidiomycete
23:49:32 <fungot> kmc: but i already wrote to marc about it.
23:49:36 <Bike> i just want something that reads to newline and then prints it twice.
23:49:39 <fungot> u an ascomycete or a basidiomycete ...out of time!
23:49:52 <kmc> wait so... input is from the last message?
23:50:12 <oerjan> ^bf ++++++++++++++++++++++++++[,[.,]+]
23:50:14 <Bike> somebody exploit this to find fizzie's porn stash
23:50:53 <oerjan> kmc: or rather, from the last message to go beyond the space taken by the ^bf message
23:51:11 <kmc> !bf_txtgen ACTION smacks kmc
23:51:16 <EgoBot> 181 ++++++++++++++[>++++++>+++++>++>++++++++<<<<-]>>-----.++.<.>++++++.++++++.-.>++++.>+++.------.<<+++++++++++++++++++.++.++++++++.++++++++.>.>--.++.----------.<----------------------. [296]
23:51:30 <kmc> ^bf +.>++++++++++++++[>++++++>+++++>++>++++++++<<<<-]>>-----.++.<.>++++++.++++++.-.>++++.>+++.------.<<+++++++++++++++++++.++.++++++++.++++++++.>.>--.++.----------.<----------------------.>[-]+.
23:51:38 <kmc> life is beautiful
23:51:52 <Bike> beautiful and smacktacular
23:53:56 <kmc> why did it output a trailing period
23:53:57 <Bike> ^bf ,[>,]<[<]>[.>]!test
23:54:09 <Bike> kmc: !bf_txtgen apparently always appends a period.
23:54:31 <kmc> icy hot stuntaz
23:54:37 <Bike> ^bf ,[>,]<[<]>[.>]<[<]>[.>]!test
23:54:52 <Bike> ^bf ,[>,]<[<]>[.>]<[<]>[.>]!,[>,]<[<]>[.>]<[<]>[.>]!
23:54:53 <fungot> ,[>,]<[<]>[.>]<[<]>[.>]!,[>,]<[<]>[.>]<[<]>[.>]!
23:55:02 <Bike> JWinslow23: (lol)
23:55:12 <oerjan> Bike: no, it appends a newline, which fungot turns into a period.
23:55:12 <fungot> oerjan: or we could do it rici's way but then i have an idea for a programming language of .net". f includes extensions for working across languages, and accepts everything from c++ to pythong to perl to whatever else can tie into it.
23:55:20 <Bike> oerjan: whichever!
23:55:58 <Bike> who runs EgoBot anyway, i could yell at them slash ask about txtgen's algorithm.
23:56:20 <kmc> !bf_txtgen QUIT seeya suckers
23:56:22 <EgoBot> 192 +++++++++++++++[>+++++>+++++>++++++++>++<<<<-]>++++++.++++.>--.+++++++++++.>>++.<-----.<<++++++++++++++++..>>++++++.<+++++++++++++.>>.<------.++.<++.++++++++.<.>>---.+.>----------------------. [687]
23:56:43 <kmc> ^bf +++++++++++++.---.>+++++++++++++++[>+++++>+++++>++++++++>++<<<<-]>++++++.++++.>--.+++++++++++.>>++.<-----.<<++++++++++++++++..>>++++++.<+++++++++++++.>>.<------.++.<++.++++++++.<.>>---.+.>----------------------.
23:57:26 <oerjan> Bike: it's not txtgen's fault, EgoBot just doesn't strip the final newline from input to commands, while fungot does.
23:57:26 <fungot> oerjan: hello world!"? if so, sorry, haha. i was talking about
23:57:26 <Bike> JWinslow23: i meant, i wrote a "quine".
23:57:29 -!- Phantom_Hoover has quit (Ping timeout: 256 seconds).
23:57:51 <oerjan> Bike: also you can use `interp bf_txtgen on HackEgo
23:57:53 <Bike> oerjan: i'm willing to shift my blame to egobot.
23:58:00 <oerjan> `interp bf_txtgen test
23:58:12 <HackEgo> 54 +++++++++++++[>+++++++++>++++++++>><<<<-]>-.>---.<-.+. [454]
23:58:22 <Bike> `run which interp
23:58:30 <oerjan> i think hat didn't append the newline
23:58:34 <Bike> `run file $(which interp)
23:58:36 <HackEgo> /hackenv/bin/interp: POSIX shell script text executable
23:58:45 <Bike> ^bf +++++++++++++[>+++++++++>++++++++>><<<<-]>-.>---.<-.+.
23:58:53 <Bike> oerjan to the rescue.
23:59:15 <Bike> `run cat $(which interp)
23:59:17 <HackEgo> #!/bin/sh \ CMD=`echo "$1" | cut -d' ' -f1` \ ARG=`echo "$1" | cut -d' ' -f2-` \ exec ibin/$CMD "$ARG"
23:59:22 <oerjan> Bike: i've used `interp bf_txtgen to generate the long bf program for fungot's ^prefixes
23:59:23 <fungot> oerjan: phew! i did, too) has " fnord format fnord".
23:59:32 <Bike> indeed, fungot
23:59:33 <fungot> Bike: what is the egobot?! sounds great though, must be one of the awful limitations of emacs is that it hinder the absorption of some b fnord and can lead to severe health problems.
23:59:42 <oerjan> i think i had to fix a bug in it to make it work
23:59:44 <Bike> `file ibin/bf_txtgen
23:59:46 <HackEgo> ibin/bf_txtgen: Bourne-Again shell script text executable
23:59:57 <Bike> `cat ibin/bf_txtgen