00:00:59 <oerjan> so it boils down to a type restricted version of typeRep that explicitly _avoids_ using the evidence for the whole type.
00:01:25 <shachaf> Right, that's the same sort of thing.
00:01:50 <shachaf> Except for the non-obvious fact that (k => t) is of the form (a b), I guess.
00:02:20 <oerjan> they should sort of change that, if they really want it to count as impredicative
00:02:32 <shachaf> Calling it impredicative is kind of strange too.
00:02:45 <shachaf> But so is instantiating a type variable to (k => t).
00:02:47 <oerjan> well spj seems to like doing that
00:05:12 <shachaf> variable: Who are you, anyway?
00:05:25 <shachaf> You're in all these channels and you have all these nicks.
00:05:52 <oerjan> clearly variable is secretly Lilax
00:06:29 <Sgeo_> "ACME is a protocol for automating the management of domain-validation certificates, based on a simple JSON-over-HTTPS interface. This repository contains the specification for ACME.
00:06:37 <Sgeo_> Words more informative than 'API'? What is this madness?
00:47:23 <orin> the new way to spell "and" is amp
00:47:58 -!- variable has quit (Ping timeout: 272 seconds).
00:52:37 -!- oerjan has quit (Quit: Nite).
01:14:44 -!- GeekDude has quit (Quit: {{{}}{{{}}{{}}}{{}}} (www.adiirc.com)).
01:20:14 <zzo38> While watching a show recorded on a VHS tape I noticed that the tape counter's hours and minutes showed the same time as the time I was watching it.
01:34:19 <quintopia> that's bound to happen if you start watching at noon. (and if it's the minutes and seconds that match...well it's pretty much guaranteed to happen once per movie)
01:36:20 <zzo38> Yes, but I didn't start watching at noon (and the tape wasn't started from the beginning either; it was from where it left off)
01:41:53 -!- doesthiswork has joined.
02:03:29 <zzo38> Now I invented RDF-survey specification; it is pretty simple and here is the full specification:
02:03:31 <zzo38> A single URI is defined for this purpose: <gopher://zzo38computer.org/1ns/foaf:survey> This will be the predicate of a triple where the subject is the person/agent answering the survey and the object is the node identifying the answer set; the rdf:type of that node is the URI identifying the survey itself.
02:06:14 -!- Decim has joined.
02:06:46 <Decim> Ok so like what programs other than minghc from the haskell site can run haskell or is that all
02:09:10 <zzo38> There are other Haskell implementations but most programs are meant for GHC
02:09:49 <orin> iirc GHC is considered the defqcto standard
02:11:12 <zzo38> If no extensions are used then you can use a different implementation but the GHC extensions are very useful to have.
02:14:02 <orin> You know what would be nice? A version of CPP that expands macros but not includes.
02:15:31 <zzo38> Possibly with AWK you can make it to comment out all of the includes and then you can make it like that
02:24:49 <zzo38> /^#include/{$0="//"$0}1
02:27:00 <zzo38> AWK is very good for that kind of stuff.
02:27:29 <zzo38> As it turns out, AWK is also good for implementing Deadfish.
02:35:53 <Decim> Did someone say deadfish!
02:36:15 <zzo38> Yes....?? Isn't it?
02:38:00 <zzo38> We can implement Deadfish with everything including C and AWK and dc and assembly language and TeX and so on but not Deadfish (but it is possible to implement Deadfish while sitting in the Deadfish cafe)
02:40:01 <zzo38> Even on Nintendo Family Computer, with more than 16-bits.
02:40:31 <zzo38> Do you know how to program a Famicom?
02:50:45 -!- variable has joined.
03:02:45 <zzo38> I also worked a lot making the Z-machine implementation in Famicom. It is faster than Commodore 64 (almost twice as much)
03:10:47 -!- variable has changed nick to function.
03:11:41 -!- function has changed nick to executes.
03:11:47 -!- executes has changed nick to function.
03:17:25 <zzo38> I have not quite decided how to implement vocabulary lookup though
03:18:23 <Decim> I'm looking into famicom
03:34:18 -!- bb010g has joined.
03:43:39 -!- sebbu has quit (Ping timeout: 244 seconds).
04:05:58 -!- Decim has quit (Remote host closed the connection).
04:08:29 -!- doesthiswork has quit (Quit: Leaving.).
04:18:07 <zzo38> This is I made up most of the Famicom Z-machine implementation http://sprunge.us/PjUc http://sprunge.us/YLZE Do you think is OK, do you have other complain about it please?
04:29:56 -!- jsilver has quit (Ping timeout: 252 seconds).
04:30:42 -!- Sprocklem has quit (Quit: 'Tis time to say goodbye, as I bid y'all farewell, my time for sleep is nigh, and bed seems oh so swell).
04:31:03 -!- function has quit (Quit: 1 found in /dev/zero).
04:31:19 -!- variable has joined.
04:41:08 -!- doesthiswork has joined.
05:30:23 -!- sebbu has joined.
05:31:00 -!- sebbu has quit (Changing host).
05:31:00 -!- sebbu has joined.
05:32:08 -!- doesthiswork has quit (Quit: Leaving.).
05:35:44 -!- AndoDaan has joined.
05:44:59 -!- PinealGlandOptic has quit (Ping timeout: 250 seconds).
06:00:12 -!- pikhq has quit (Ping timeout: 272 seconds).
06:01:07 -!- rdococ has joined.
06:05:47 -!- hjulle has quit (Ping timeout: 250 seconds).
06:19:11 -!- zadock has quit (Quit: Leaving).
06:56:15 -!- zadock has joined.
07:11:30 -!- Insipid has joined.
07:15:12 -!- iamevn_ has joined.
07:18:30 -!- variable has quit (Read error: Connection reset by peer).
07:18:45 -!- wieldablesquid has quit (Ping timeout: 265 seconds).
07:24:40 -!- Insipid has quit (Quit: Leaving.).
07:24:53 -!- Insipid has joined.
07:29:16 -!- Insipid has quit (Ping timeout: 256 seconds).
07:30:18 -!- b_jonas has joined.
07:35:39 -!- f|`-`|f has quit (Ping timeout: 256 seconds).
07:36:15 -!- f|`-`|f has joined.
07:44:26 -!- AndoDaan has quit (Read error: Connection reset by peer).
07:45:19 -!- AndoDaan has joined.
07:57:09 <lambdabot> ‘id’ (imported from Data.Function),
08:02:19 -!- Insipid has joined.
08:13:07 <mroman> ghci has it as "last result"
08:30:10 -!- Insipid has left.
08:43:55 -!- Patashu has joined.
08:48:32 -!- Patashu has quit (Remote host closed the connection).
08:48:49 -!- Patashu has joined.
08:57:48 <HackEgo> [wiki] [[420]] http://esolangs.org/w/index.php?diff=42673&oldid=42671 * Vriskanon * (+4962) /* Sample programs */ Added Hello, world! program (copied from Chicken's page)
09:17:08 -!- orin has quit (Ping timeout: 250 seconds).
09:19:10 -!- orin has joined.
09:34:15 -!- orin has quit (Ping timeout: 264 seconds).
09:35:36 -!- orin has joined.
09:43:31 <b_jonas> fungot, is it true that to remain binary compatible with user-space context/coroutine switching, ABIs cannot designate new registers in future CPUs as callee-saved, but must keep them as either scratch registers or registers you must restore at all boundaries?
09:43:31 <fungot> b_jonas: that's why i learned to get hugged without feeling strange. but the same idea
09:44:16 <fungot> AndoDaan: i don't plan to enter it, the full range of 256 values per channel is lessened understand why cons should have two loop variables, then... wait... uhm...
09:45:15 <b_jonas> fungot: no, we actually support 16 bit depth here, and in fact some of our channels have over 9000 nicks joined
09:45:15 <fungot> b_jonas: laml is two things in te second statement of a program written in scheme
09:45:39 -!- rdococ has quit (Ping timeout: 246 seconds).
09:46:55 -!- Insipid has joined.
09:47:09 <mroman> It's somehow weird that movies claim the world has gone bad
09:47:19 <mroman> like it was better 500 years ago
09:47:46 <Taneb> 500 years ago serfs could be serfs, lords can be lords, and everyone knew what there place was
09:48:08 <Taneb> Nowadays it's all mixed up
09:48:26 -!- Insipid has left.
09:54:01 <FireFly> fungot: what do yout think about the mix-up?å
09:54:01 <mroman> and Tanebs could be Tanebs
09:54:01 <fungot> FireFly: kuribas, because syntax-rules doesn't integrate scheme the pattern language can include, e.g. ( blah :some 3 ( something) is instance local ( afaik)
09:54:56 <FireFly> (blah :some 3 (something) is instance local (afaik)), got it
09:59:37 <mroman> fungot: Would you like the domain "fungot.sucks"?
09:59:38 <fungot> mroman: the swingle singers... which may be what you're looking for? :) that just makes you look like that.
10:10:29 -!- orin has quit (Ping timeout: 245 seconds).
10:12:21 -!- orin has joined.
10:12:24 <Taneb> Hmm, I seem to have ended up with a tendency to write emoticons with multiple mouths
10:17:36 <HackEgo> [wiki] [[420]] M http://esolangs.org/w/index.php?diff=42674&oldid=42673 * Vriskanon * (+1) /* Commands */ Changed 'Chicken' name name
10:18:24 <mroman> Taneb: That's just your double-chin acting out.
10:18:57 <mroman> fungot: Are you a fitness trainer?
10:18:58 <fungot> mroman: of course. thanks. it'll need a few temp cells need to be specified in a certain time of year" is the name
10:20:08 <mroman> fungot: Would you train me?
10:20:09 <fungot> mroman: nobel watchers say neither mr bush nor mr blair has much chance of that before...
10:21:09 <mroman> I'm trying to beat 15 pull-ups
10:21:13 <mroman> but I'm stuck at around 12 :(
10:35:19 -!- boily has joined.
10:52:34 <Taneb> mroman, that's about 11 more than I can do!
10:54:38 <boily> taking it quieter and slower after a week of social life.
10:54:46 <boily> how's it on your end?
10:55:08 <Taneb> Well, last night I tried out a board game I bought recently
10:55:11 <Taneb> I found it pretty fun
10:55:16 <Taneb> (Avalon is the game)
10:55:27 <Taneb> (It's essentially The Resistance in Camelot)
10:56:00 <boily> still boardgaming :D
10:56:03 <mroman> Taneb: then you should start training ;)
10:58:14 <boily> Taneb: recently I played a few new games. I think I'm a deckbuilding addict.
10:59:02 <Taneb> I've challenged myself to see how good I can get at Magic without paying any money
10:59:44 <boily> we play Magic every lunch at work. I have Dominion, Tanto Cuore, Mage Wars, Space Realms, Quarriors...
11:00:04 <boily> going to buy Race for the Galaxy soon, but then there's Roll for the Galaxy too.
11:03:29 <shachaf> never mind, i need to go to sleep
11:05:39 <mroman> Taneb: the real secret is actually that it's not really hard nor does it take up a lot of time.
11:05:57 <mroman> You just need to devote like 15min of your day to fitness and that's really all there is.
11:06:36 <mroman> People seem to think you need to train for hours to get 30 push-ups, 15 pull-ups and the like which is utterly wrong.
11:07:04 <mroman> I sit in front of my computer all day except like 15min where I do some body weight exercises.
11:09:20 <Taneb> mroman, my problem is more space than time
11:14:09 <boily> let's try not falling asleep in the shower.
11:14:24 -!- boily has quit (Quit: OTOLITHIC CHICKEN).
11:15:08 -!- oerjan has joined.
11:15:31 -!- AndoDaan has quit (Ping timeout: 256 seconds).
11:16:29 -!- orin has quit (Ping timeout: 246 seconds).
11:18:34 -!- orin has joined.
11:18:56 -!- Lymia has quit (Changing host).
11:18:56 -!- Lymia has joined.
11:19:00 <Taneb> Also that doesn't really help with my pull-ups
11:19:26 <mroman> no trees around? No bars?
11:21:04 <Taneb> Most of the trees around lack low branches
11:22:40 * oerjan finds himself literally laughing maniackally
11:22:52 <oerjan> that means it must be a very good idea.
11:23:13 <Taneb> mroman, and I can't install a bar because this is rented accomodation and I'll only be here for another 2 months or so
11:23:43 <oerjan> just make it a minibar hth
11:24:45 <Taneb> I'm too tall for a minibar
11:26:29 <oerjan> but too short for the branches
11:26:48 <oerjan> solution, put the minibar under the branches and stand on it.
11:32:09 <Taneb> Unfortunately I don't have a minibar
11:34:53 <mroman> If you live in a city there's gotta be something you can do pull ups on
11:35:00 <mroman> even if it's a street sign :D
11:37:12 <Taneb> mroman, I'm damn near the edge of the city
11:37:35 <oerjan> i think they usually have a sign at the edge of a city hth
11:37:58 <oerjan> but i guess it might be crowded.
11:54:53 -!- zadock has quit (Quit: Leaving).
12:04:17 <mroman> https://www.youtube.com/watch?v=upUUWuJhvXk hth
12:08:27 -!- Patashu has quit (Ping timeout: 255 seconds).
12:38:46 -!- AndoDaan has joined.
12:45:05 <orin> go to a playground
12:45:12 <orin> use the monkey bars
12:50:12 <elliott> nobody has ever done one, ever
12:50:29 <Taneb> orin, I do not know any playground with monkey bars around here
12:59:05 <HackEgo> [wiki] [[Λιμπ]] http://esolangs.org/w/index.php?diff=42675&oldid=11870 * Esowiki201529A * (+14)
13:03:56 -!- hjulle has joined.
13:05:10 <HackEgo> [wiki] [[AFTL]] http://esolangs.org/w/index.php?diff=42676&oldid=42672 * Esowiki201529A * (+14) /* Hello, world! program */
13:19:31 <mroman> 18 Reasons Why You Don’t Get Laid"
13:19:54 <mroman> Can't I even read news know without these fucking pop-ups
13:20:04 <oerjan> thanks, now i'm wondering if sumo wrestler get laid (probably)
13:20:05 <mroman> that aren't really the good old pop-ups from the old days
13:20:31 <mroman> pop-ups used to be windows.
13:20:50 <mroman> now they are "static divs" injected into pages with javascript
13:21:21 <mroman> although I wonder what those 18 reasons could be...
13:21:41 <mroman> I can only think of one. And that one is enough so I don't even need the other 17 reasons.
13:21:44 <oerjan> 1. you spend all your time browsing web
13:22:31 <mroman> but I guess that could be one of those 17 other reasons.
13:23:56 <oerjan> Taneb: did that fetish society have any people wearing paper bags on their heads
13:24:07 <Taneb> oerjan, I do not believe so
13:24:11 <Taneb> But I did not stay very long
13:24:17 <oerjan> i guess there's no hope then
13:24:36 <oerjan> oh wait, blind people exist
13:24:38 <mroman> What would a paper bag over your head do?
13:24:42 <mroman> You can still have an ugly body.
13:24:43 <Taneb> I know exactly why I don't get laid
13:24:50 <mroman> Besides having an ugly face/head.
13:25:01 <mroman> fungot: Are you ugly too?
13:25:01 <fungot> mroman: how are you doing there? :p)
13:25:11 <mroman> Is it mocking me??!?!?!?
13:25:32 <b_jonas> fungot: give me test data because it's hard to write a program without it
13:25:32 <fungot> b_jonas: may i add that? since only " the result(s) of the objections about plt-specificness we need to know ( my " behaviors" you mentioned
13:26:05 <mroman> Taneb: you're butt ugly too?
13:26:10 <Taneb> mroman, nah, I'm asexual
13:26:21 <mroman> That's my reason number 2.
13:26:46 <Taneb> I do not believe so
13:27:53 <oerjan> yay an asexual pairing
13:33:07 <mroman> I actually have no idea if I'm butt ugly. I just assume so.
13:33:19 <mroman> It's something that you can't really judge by yourself I think.
13:34:29 -!- Sprocklem has joined.
13:34:52 <mroman> Judging by the reflection in the mirror and other people I'm at least ugly but not necessarily butt ugly.
13:40:25 <mroman> Hm. Theres 402 - Payment required.
13:47:05 <fizzie> I don't think it sees much use.
13:47:39 <b_jonas> mroman: that's just the temporary name assigned to the reserved code in earlier versions of the standard. the more correct localized message these days is 402 Virgin sacrifice required
13:48:46 <mroman> fungot: Apple uses it apparentely
13:48:47 <fungot> mroman: anywhere but sweden, apparently. i won't actually pay 50 cents for that :p
13:48:56 <mroman> fizzie: Apple uses it apparentely.
13:49:07 <mroman> fizzie: BUT PAYMENT IS REQUIRED! I WANT THOSE 50 CENTS!
13:49:29 <mroman> that should've been directed to fungot
13:49:30 <fungot> mroman: http://www.99-bottles-of-beer.net/ language-malbolge-995.html the last comment in http://esolangs.org/ wiki/ quadtree work, would just be better to have names like that
13:50:01 <fizzie> There's also the "418 I'm a teapot" response.
13:50:26 <b_jonas> mroman: the HTTP/1.1 rfc specifically says "reserved for future use" "http://www.w3.org/Protocols/rfc2616/rfc2616-sec10.html#sec10.4.3"
13:50:51 <b_jonas> I think HTTP/2 defines the meaning correctly and calls it its proper name
13:51:40 <b_jonas> only that's one of those w3c standards that are always under development and never released, like HTTP 5 or the new CSS standard
13:52:00 <mroman> hasn't that been released?
13:52:31 <fizzie> It's also a "living standard".
13:52:47 <fizzie> Or the version by WHATWG, anyway.
13:52:50 <b_jonas> apparently I'm not relaly following this junk
13:53:25 <b_jonas> now the question is, which response header gives the bounds for the age of the virgin?
13:53:38 <fizzie> I got the impression that W3C released some particular fixed version, while the WHATWG standard is just... living.
13:54:32 <fizzie> (And the W3C standard doesn't exactly match WHATWG.)
13:55:25 <b_jonas> probably WWW-Authenticate with some approperiate challenge typ
13:58:37 -!- b_jonas has set topic: Stenosteganography | https://dl.dropboxusercontent.com/u/2023808/wisdom.pdf http://codu.org/logs/_esoteric/ http://tunes.org/~nef/logs/esoteric/ | http://esolangs.org/.
14:04:48 <elliott> whatwg's standard is just "HTML" with no version
14:06:36 -!- Sprocklem has quit (Ping timeout: 272 seconds).
14:21:26 -!- `^_^v has joined.
14:28:02 <tswett> Duh, "claptrap" is what the hostname should be.
14:30:06 <b_jonas> fungot, why does this function have so many arguments?\
14:30:06 <fungot> b_jonas: i came across srfi-83 last night, couldn't get to sleep now. goodnight everybody!! i've flown in leslie gore and two dozen
14:30:17 <fungot> b_jonas: sarahbot is very much in the recursive call must be optimized. just a non-american living in american indeed.
14:32:31 <tswett> fungot: wait, what's *your* opinion of what the hostname should be?
14:32:31 <fungot> tswett: no clue really.) the name with a
14:32:55 <tswett> I guess "claptrap" is, in fact, a "name with 'a'".
14:42:27 -!- AndoDaan_ has joined.
14:45:34 -!- GeekDude has joined.
14:45:47 -!- AndoDaan has quit (Ping timeout: 246 seconds).
14:50:15 -!- TieSoul has joined.
14:52:41 -!- orin has quit (Quit: Lost terminal).
14:53:50 -!- `^_^v has quit (Ping timeout: 246 seconds).
14:53:58 -!- Sprocklem has joined.
14:56:59 -!- `^_^v has joined.
15:00:27 -!- Vorpal has quit (Ping timeout: 272 seconds).
15:00:57 -!- rdococ has joined.
15:12:06 -!- Sprocklem has quit (Quit: Leaving).
15:25:52 -!- pikhq has joined.
15:46:19 -!- fractal has quit (Ping timeout: 245 seconds).
15:46:21 -!- Vorpal has joined.
15:46:21 -!- Vorpal has quit (Changing host).
15:46:21 -!- Vorpal has joined.
15:50:46 -!- variable has joined.
15:58:12 -!- GeekDude has quit (Remote host closed the connection).
15:58:40 -!- GeekDude has joined.
15:59:55 -!- oerjan has quit (Quit: leaving).
16:02:28 -!- fractal has joined.
16:08:48 -!- bb010g has quit (Quit: Connection closed for inactivity).
16:11:11 -!- vifino has quit (Remote host closed the connection).
16:12:14 -!- vifino has joined.
16:21:50 -!- zzo38 has quit (Remote host closed the connection).
16:27:33 <HackEgo> [wiki] [[SMETANA]] http://esolangs.org/w/index.php?diff=42677&oldid=35747 * 62.216.211.203 * (+39)
16:31:39 -!- zadock has joined.
16:39:34 -!- Insipid has joined.
16:39:48 -!- Insipid has left.
16:45:32 -!- Sprocklem has joined.
16:51:04 -!- AndoDaan_ has quit (Read error: Connection reset by peer).
16:51:35 -!- AndoDaan has joined.
17:07:56 -!- b_jonas has quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client).
17:37:36 -!- bb010g has joined.
17:41:27 -!- AndoDaan has quit (Read error: Connection reset by peer).
17:41:30 -!- AndoDaan_ has joined.
17:50:24 -!- TieSoul has changed nick to TieSoul_.
17:50:27 -!- TieSoul_ has changed nick to TieSoul.
18:09:49 -!- supay has quit (Ping timeout: 272 seconds).
18:14:36 -!- supay has joined.
18:18:16 -!- Sprocklem has quit (Ping timeout: 245 seconds).
18:25:33 -!- Sprocklem has joined.
18:25:39 -!- b_jonas has joined.
18:31:52 -!- AndoDaan_ has quit (Ping timeout: 256 seconds).
18:49:36 -!- PinealGlandOptic has joined.
18:50:30 -!- Phantom_Hoover has joined.
18:53:28 -!- AndoDaan has joined.
18:54:01 -!- nortti has changed nick to nortiecat.
18:54:09 -!- nortiecat has changed nick to nortti.
18:54:40 -!- nortti has changed nick to hvidiecat.
18:54:45 -!- hvidiecat has changed nick to hvidie.
18:54:54 -!- hvidie has changed nick to nortti.
19:06:09 -!- shikhin has joined.
19:30:24 -!- rdococ has quit (Quit: Page closed).
19:38:22 -!- nycs has joined.
19:41:25 -!- `^_^v has quit (Ping timeout: 256 seconds).
20:10:59 -!- roasted42 has joined.
20:14:58 -!- roasted42 has quit (Changing host).
20:14:58 -!- roasted42 has joined.
20:14:58 -!- roasted42 has changed nick to TheM4ch1n3.
20:36:02 -!- daria has joined.
20:43:19 -!- perrier has quit (Read error: Connection reset by peer).
20:43:33 -!- nycs has quit (Quit: This computer has gone to sleep).
20:44:33 -!- perrier has joined.
20:48:33 -!- `^_^v has joined.
20:52:16 -!- AnotherTest has joined.
20:52:35 -!- AndoDaan has quit (Ping timeout: 272 seconds).
20:53:30 -!- nortti has changed nick to lawspeaker.
20:53:49 -!- lawspeaker has changed nick to nortti.
21:02:05 -!- Patashu has joined.
21:03:05 -!- b_jonas has quit (Remote host closed the connection).
21:14:23 -!- `^_^v has quit (Quit: This computer has gone to sleep).
21:16:17 -!- `^_^v has joined.
21:23:46 -!- Sprocklem has quit (Ping timeout: 255 seconds).
21:24:40 -!- Patashu has quit (Ping timeout: 255 seconds).
21:28:50 -!- TheM4ch1n3 has quit (Ping timeout: 245 seconds).
21:35:20 -!- AndoDaan has joined.
21:38:20 -!- daria has quit (Ping timeout: 244 seconds).
21:44:44 -!- AnotherTest has quit (Remote host closed the connection).
21:45:04 -!- `^_^v has quit (Quit: This computer has gone to sleep).
21:45:28 -!- shikhin has quit (Quit: leaving).
21:46:22 -!- shikhin has joined.
21:47:33 -!- Phantom_Hoover has quit (Ping timeout: 246 seconds).
21:48:01 -!- Phantom_Hoover has joined.
22:12:08 -!- MoALTz has quit (Ping timeout: 250 seconds).
22:12:54 -!- MoALTz has joined.
22:19:13 <shachaf> oerjan: your fame grows every day hth
22:32:16 -!- staffehn has quit (Ping timeout: 265 seconds).
22:32:19 -!- staffehn_ has joined.
22:41:55 -!- oerjan has joined.
22:42:34 <tswett> I think I like Object Oriented Thue.
22:42:48 <tswett> I definitely like the idea. I'm not sure about the implementation.
22:43:53 <tswett> It's not clear from the specification (excluding the Libraries section) that it's actually possible to create objects.
22:44:25 <tswett> Whoops, there's the part where it says you can do that.
22:46:39 <tswett> Yes, I think I do like it.
22:50:24 <oerjan> shachaf: the fix was one of the things i suggested in comment 92, too.
22:52:18 <oerjan> edwardk: now i'm vaguely wondering if any of this will break your constraints package
22:52:42 <shachaf> Have you been following #9858?
22:52:43 <edwardk> breaking the constraints package would kinda suck
22:53:04 <oerjan> edwardk: a => b will no longer unify with t u
22:53:27 <edwardk> i don't use a => b explicitly as a type anywhere
22:54:56 <shachaf> A previous change was to make Constraints un-Typeable.
22:55:35 <oerjan> edwardk: i suspect you'd be happy if they implement what spj mentions in https://ghc.haskell.org/trac/ghc/ticket/9858#comment:95
22:56:55 -!- staffehn_ has quit (Ping timeout: 256 seconds).
22:56:55 -!- fractal has quit (Ping timeout: 256 seconds).
22:57:10 <oerjan> the previous comment is something i know you've complained about before
22:57:11 <edwardk> been too busy playing with fractions to notice =)
22:57:43 <edwardk> working on exact real arithmetic using nested linear fractional transformations and continued fractions
22:57:46 <shachaf> oerjan: see #haskell-lens hth
22:58:11 <edwardk> but really its all just really fancy lazyish fractions
22:59:00 -!- staffehn has joined.
22:59:02 <edwardk> pi = 3, no 3 + 1/7 no 3 + 1/(7 + 1/15), no...
22:59:46 <oerjan> i vaguely thought exact arithmetic with continued fractions was something that had already been done in haskell
23:00:01 <edwardk> i'm not using continued fractions per se
23:00:08 <edwardk> but rather "nested linear fractional transformations"
23:00:14 <edwardk> which has _also_ been done in haskell
23:00:27 <edwardk> in fact, SPJ has a paper on continued fractions.
23:00:43 <edwardk> what i'm working on is extending the scope of what we can compute with exactly
23:00:54 <edwardk> by extending the work here to so called "Hurwitz numbers"
23:01:00 <edwardk> and then generalizing the notion of hurwitz numbers
23:01:03 <oerjan> linear fractional transformations, is that like möbius transformations except with integers/rationals
23:01:30 <edwardk> matrix representation used instead multiply matrices to compose.
23:01:53 <edwardk> now we represent numbers as potentially infinite sequences of linear fractional transformations
23:02:29 <edwardk> an LFT with all positive (or all negative, they are the same up to scalar multiples) coefficients takes the interval from [0,infinity) to a small interval.
23:03:25 <edwardk> you can view a continued fraction [a;b,c,d,e...] as the composition of matrices [a 1; 1 0]*[b 1; 1 0]* ....
23:03:56 <edwardk> quadratic irrationals (e.g. sqrt 2) have repeating representations as continued fractions
23:04:26 <edwardk> "hurwitz numbers" have near-repeating representations. the representations repeat with a polynomial in the number of times you've repeated
23:04:40 <edwardk> 2;1,2,1,1,4,1,1,6,1,1,8,1...
23:05:33 -!- fractal has joined.
23:05:46 <edwardk> so i can write that out the repeating part as a matrix with polynomial coefficients: the product [1 1; 1 0] * [2k 1; 1 0] * [1 1; 1 0]
23:05:56 <edwardk> and we can convert the prefix to another ft
23:06:33 <edwardk> so it is [ 2 1; 1 0] * Pi_k^{1..infinity} [1 1; 1 0] * [2k 1; 1 0] * [1 1; 1 0]
23:06:52 <edwardk> which if you truncate it at any point takes the range from [0..infinity) to a window very very close to e.
23:07:15 <oerjan> this zooming with matrices reminds me of some of my PhD work
23:07:17 <edwardk> then we can use algorithms for adding and multiplying/dividing such numbers, etc.
23:07:52 <edwardk> the algorithms for adding/multiplying uses a 3d tensor to merge results from two different sources into one output
23:08:30 <edwardk> and then finally i can represent a meromorphic function like 'exp x' or 'log x' as a funny infinite tensor chain, where i've got the coefficients as polynomials like the hurwitz case.
23:08:42 <edwardk> and put all that into one number type and i can compute fast nice reals.
23:09:06 <edwardk> and hopefully still get things like e/e = 1 to cancel exactly because of a small closed form.
23:09:49 <edwardk> i can get computations that never step outside of quadratic surds to cancel, but now i want computations that never step outside of hurwitz numbers to cancel.
23:09:59 -!- shikhin_ has joined.
23:10:03 <tswett> data SmallReal where { toSmallReal :: Stream Bit -> SmallReal; mid :: cons Zero (repeat One) = cons One (repeat Zero); down :: forall (b :: Bit, x :: Stream Bit, y :: Stream Bit). (toSmallReal x = toSmallReal y) -> (toSmallReal (cons b x) = toSmallReal (cons b y)) }
23:10:10 <tswett> Real numbers in the interval [0, 1], in Homotopy Haskell.
23:10:11 <edwardk> basically looking for 'not-symbolic' but close enough representations where you can compare them nicely.
23:10:17 <tswett> (NB: there's no such thing as Homotopy Haskell.)
23:10:29 <oerjan> oh so you are keeping finite representations as far as possible
23:10:44 <edwardk> rather than unrolling them into a lazy mess like everyone else
23:11:28 -!- shikhin has quit (Ping timeout: 264 seconds).
23:11:47 <edwardk> LF = Rat Integer Integer | Hom Integer :* LF | Hurwitz (Hom Poly) | Meromorphic (Bihom Poly) LF
23:11:52 <edwardk> where Hom is a homographic transformation
23:12:21 -!- Decim has joined.
23:12:38 <edwardk> and Bihom is a bihomographic transformation which is a 3d analogue. or equivalently z(x,y) = (axy+bx+cy+d)/(exy+fx+gy+z) which can be viewed as z(x) in Z[y] or z(y) in Z[x]
23:13:37 <edwardk> LF = Rat Integer Integer | Hom Integer :* LF | Hurwitz (Hom Poly) | Meromorphic (Bihom Poly) LF | Tensor (Bihom Integer) LF LF
23:13:46 <edwardk> because i can't apparently cancel that
23:14:17 <edwardk> and the finite form simplifies whenever it can
23:15:04 <edwardk> phi = Hurwitz (Hom 1 1 1 0) gives the golden ratio, sqrt2 = Hom 1 1 1 0 :* Hurwitz (Hom 2 1 1 0)
23:15:48 <Decim> Sorry Im just dead
23:16:24 <edwardk> anyways i have algorithms now for 'spigoting' off partial matrices from each of those forms
23:16:31 <edwardk> and modifying the remainder
23:17:05 <edwardk> and ways to construct them from continued fractions, redundant binary, decimal, etc.
23:18:00 <edwardk> basically the only requirement is that every homographic transformation after the first 'contracts' some interval.
23:18:03 <oerjan> the z(x,y) makes me think vaguely of quaternions
23:18:30 <Decim> What's Brainfuck used for?
23:19:16 <Decim> I know but what are its general purposes or is it just a fun language
23:19:49 <tswett> As far as I know, the sole purpose of Brainfuck is to be interesting.
23:20:04 <oerjan> edwardk: can you differentiate with this too?
23:20:23 <edwardk> i can treat these as raw numbers for AD
23:20:35 <edwardk> and you can build pade approximants using the same techniques
23:20:46 <edwardk> and those are used as good forms for differentiation/integration
23:21:12 <Decim> Well in general all of you are pretty neat
23:21:36 <Decim> Except that one guy >_>
23:21:47 <edwardk> i got started with pade approximants when jacques carette earwormed me with the idea that they would be the best form for dealing with more or less "automatic integration"
23:21:59 <edwardk> as that is what all the big boys use in things like Mathematica / Maple, etc.
23:22:27 <edwardk> but then put it away for a couple years
23:22:36 <edwardk> until now, when i now have more of the groundwork in place
23:22:51 <Decim> Please teach at our school, Ill pay you like a dollar
23:27:01 <tswett> Why the hell did I decide I wanted to go into industry instead of academia...
23:30:14 <oerjan> probably involved money
23:37:57 <Decim> Did you want to teach tswett
23:38:17 <tswett> No, I'm not really interested in teaching professionally.
23:38:57 -!- boily has joined.
23:39:21 -!- Decim has quit (Remote host closed the connection).