00:02:01 <tswett> Unless, of course, its preimage is an even function.
00:04:20 -!- adu has quit (Quit: adu).
00:15:55 -!- kcm1700_ has joined.
00:19:59 -!- kcm1700 has quit (Ping timeout: 246 seconds).
00:27:36 -!- adu has joined.
00:30:36 -!- Phantom_Hoover has quit (Read error: Connection reset by peer).
00:35:49 -!- adu has quit (Quit: adu).
00:42:48 -!- adu has joined.
00:44:21 -!- Sprocklem has joined.
01:11:01 -!- boily has joined.
01:11:17 -!- dorei has quit.
01:11:46 <lambdabot> CYUL 140100Z 25007KT 15SM SKC M22/M30 A3012 RMK SLP204
01:12:21 <oerjan> for once, the outside weather is immaterial.
01:12:22 <boily> the sickness disprovement worsened again?
01:12:26 <lambdabot> ENVA 140050Z 13002KT 1000 R09/P2000N R27/P2000N SN VV003 M01/M01 Q1012 RMK WIND 670FT 17001KT
01:12:41 <boily> the outside weather is the kind where your eyelids try to freeze together.
01:12:48 <boily> (seriously. no joke.)
01:13:26 <boily> oerjan: are you still mainly oerjan, or has the flu made a coup and installed a new government in you body?
01:13:33 <oerjan> well the _flu_ isn't getting that much worse now, but my back is really longing for an outside walk. and then there had to pop up a canker sore just for completeness.
01:15:46 <quintopia> cold as in temperature or illness?
01:15:59 <boily> temperature. re the previous @metar.
01:16:34 <boily> > (-21) * 9/5 + 32
01:17:08 <boily> quintopia: it's -21 °C (-6 °F) outside. much fun.
01:17:25 <oerjan> the last one sort of interfers with plans to gorge on comfort pizza and chocolate
01:17:45 <oerjan> (that's generally the right thing to do with a flu, right?)
01:18:05 <quintopia> feed a flu, starve an eating disorder?
01:18:10 <boily> for me it's large doses of chicken noodle soup and OJ.
01:18:16 * oerjan fetches some chocolate anyway
01:18:28 <quintopia> thank goodness they invented kanka eh
01:18:41 <quintopia> i always buy some and can never find it when i need it next
01:18:55 <boily> speaking of pizza, today we ordered a special gigantic oversized poutine, with sausage, ham, mushrooms and bell peppers in it.
01:19:03 <boily> quintopia: what's a kanka?
01:19:05 <oerjan> well i _have_ had OJ, although i only drink half a cup per day as my stomach doesn't hold well with too much sour
01:19:08 <quintopia> that sounds like a good approach to cold weather
01:19:16 <quintopia> boily: medication for canker sores
01:19:54 <quintopia> i probably wouldn't need it if i brushed more regularly eh
01:20:06 <quintopia> so what you gotta do tonight boilyface
01:20:08 <boily> ah! something like anbesol!
01:20:17 <oerjan> my understanding is that chicken soup for healing is an american thing, possibly jewish in origin.
01:20:26 <boily> don't care; has chicken.
01:21:08 <boily> hm... I'll have to investigate that mysterious product.
01:21:28 <boily> meanwhile, tonight is the night where nothing happens much.
01:23:22 <boily> uhm... <_<'... mainly reddit browsing hth
01:23:59 <quintopia> so hard to find people who want to game with me
01:24:25 <boily> sorry, completely exhausted. I don't know how many hours I worked this week.
01:27:28 -!- Patashu has quit (Remote host closed the connection).
01:28:43 -!- Patashu has joined.
01:29:27 -!- Sprocklem has quit (Ping timeout: 245 seconds).
01:30:11 <tswett> Depends on the game and how you play it.
01:30:34 <tswett> When I play Starcraft, if I'm trying hard to win, it's pretty stressful.
01:31:21 <quintopia> just fucking around and messing with stuff
01:34:20 <zzo38> If you don't like Starcraft then play game "I don't know"craft
01:35:39 <quintopia> i don't know if i like starcraft, but i'll play "I don't know"craft when you finish writing it.
01:37:03 <zzo38> Or, try to play Washizu mahjong by internet if you have any software to do so.
01:37:31 <quintopia> I already beat and 100%ed that game
01:53:39 -!- boily has quit (Quit: SPLIT CHICKEN).
02:02:14 -!- skj3gg has joined.
02:11:48 -!- Tritonio has quit (Ping timeout: 252 seconds).
02:12:37 -!- skj3gg has quit (Quit: ZZZzzz…).
02:24:07 -!- ais523 has quit.
03:21:07 -!- GeekDude has quit (Quit: ZNC - http://znc.in).
03:44:48 -!- doesthiswork has joined.
03:47:18 -!- bb010g has joined.
03:58:02 <zzo38> Why is it permitted for names of elements and attributes in XML to contain non-ASCII characters? My opinion is this is stupid and it shoudln't be. Non-ASCII characters should only be allowed in text (both normal and CDATA), values of attributes, and comments.
03:58:51 <pikhq> XML is overly complex as a rule.
04:00:48 <zzo38> Yes it is too complex, although it presumably is doing what it needs to do. Still a few things are a bit stupid.
04:01:17 <pikhq> Take a quick look at the processing instructions. :)
04:05:14 <zzo38> While it does have a lot of extensions, most of them are optional (and does not affect the syntax of the document) so it is not a problem. If you can ignore unknown processing instructions then it might work in some cases.
04:07:06 <zzo38> It does have a few good ideas such as namespaces, although the method of doing this might be a bit more complicated than it should be.
04:09:07 <zzo38> It does have far more complexity than it should have, although some files/services are using XML so a XML parser would be needed.
04:12:48 -!- Sprocklem has joined.
04:19:25 -!- dulla_ has joined.
04:19:29 -!- dulla has quit (Ping timeout: 264 seconds).
04:19:36 -!- dulla_ has changed nick to dulla.
04:20:48 -!- shikhin_ has quit (Ping timeout: 245 seconds).
04:21:00 <coppro> pikhq: wasn't "overly complex" a design goal?
04:22:47 <coppro> also related: oh my god I cannot wait for HHTP/2
04:23:54 <oerjan> hyperhyped transfer protocol
04:25:11 <pikhq> coppro: Not really.
04:25:21 <pikhq> It was supposed to be a simplified subset of SGML.
04:26:02 <oerjan> then they forgot to shoot the enterprise people before they started designing.
04:43:16 -!- nys has quit (Quit: quit).
04:48:49 <pikhq> oerjan: Common problem.
05:01:18 <pikhq> Clearly the proper answer is to murder enterprises.
05:25:56 <zzo38> I want to read XML files in SQLite. I don't need DTD and encryption and fancy stuff like that, although I do want namespaces, and it would also help to support queries.
05:30:51 <zzo38> Is there suitable C library to do it with? With simple one file and don't need a configure script and so on
05:31:50 -!- ProofTechnique has quit (Ping timeout: 246 seconds).
05:48:08 -!- bb010g has quit (Quit: Connection closed for inactivity).
05:50:19 -!- MDude has joined.
06:09:22 -!- chaosagent has quit (Remote host closed the connection).
06:18:57 -!- copumpkin has joined.
06:32:24 -!- vanila has joined.
06:37:41 <MDude> I keep getting distracted from writing things, and need to go to bed.
06:38:30 <MDude> Also, I got a microSD card but immediatley forgot the adapter to plug it into normal sized SD slots.
06:39:23 <MDude> Maybe I left that downstars actually.
06:41:13 <MDude> But yeah, I should write about ideas on chatbots and AI and stuff already instead of waiting until I implement stuff to write about it.
06:41:42 <MDude> Since getting my idea down somewhere probably helps get to work actually following through with it.
06:42:53 -!- adu has quit (Quit: adu).
06:47:32 -!- ProofTechnique has joined.
06:57:55 <zzo38> What is your idea about it?
07:06:01 -!- doesthiswork has quit (Quit: Leaving.).
07:07:17 -!- doesthiswork has joined.
07:08:25 -!- doesthiswork has quit (Client Quit).
07:08:40 -!- bb010g has joined.
07:14:13 -!- Sprocklem has quit (Ping timeout: 264 seconds).
07:17:22 <MDude> I don't want to distract myself from sleeping currently.
07:18:03 <MDude> I'll try to get some stuff written down tomorrow, I'll try to do that first and then link the posts to talk about them.
07:18:19 -!- MDude has changed nick to MDream.
07:19:36 <MDream> It's all very simple stuff.
07:19:57 <dulla> I'm still interested
07:20:20 -!- Tod-Autojoined has joined.
07:22:19 <MDream> I think I mentioned some of it before, about making variants of chatbots.
07:22:41 <MDream> But yeah, I need to make sure I'm not up too late.
07:23:37 -!- NihilistDandy has joined.
07:23:49 <MDream> Of the general idea of how they work.
07:24:19 <dulla> or they have more general functionality
07:24:41 <dulla> but that'd require run-time reloading/factoring, or a dynamic structure which is a pain nin the cock
07:24:48 <MDream> The first idea I had on it was when thinking about how chatbots left in public tend to pick up rude behavior.
07:25:44 <MDream> If you string two chatbots together, the first one can do the learning, but isntead of speaking directly to the user it can pass it message through a more hardcoded chatbot that paraphrases it.
07:26:15 <MDream> And from there I came up with other ideas in which a chatbot is considered a node in a larger network.
07:26:59 <dulla> so more or less a multi-layer ANN for information representation
07:27:05 <dulla> So that shit stays pg
07:27:16 <dulla> and that they can't teach your dog to respond to a retarded name
07:27:40 <dulla> With multiple inputs, it can gather more data in the long run
07:27:58 <dulla> Though you'd need to enforce some kind of channel-locality to some of the endpoints
07:28:06 -!- ProofTechnique has quit (*.net *.split).
07:28:07 -!- TodPunk has quit (*.net *.split).
07:28:07 -!- paul2520 has quit (*.net *.split).
07:34:04 <MDream> This seems like a good place to leave it for now, rather than stringing things along further.
07:35:56 -!- paul2520 has joined.
07:46:42 -!- vanila has quit (Quit: Leaving).
07:55:48 -!- Lymia has quit (Ping timeout: 245 seconds).
07:59:59 -!- Lymia has joined.
09:20:04 -!- Taneb has quit (Ping timeout: 250 seconds).
09:22:19 -!- Taneb has joined.
09:30:10 -!- Patashu has quit (Disconnected by services).
09:30:10 -!- Patashu_ has joined.
09:33:06 -!- Patashu_ has quit (Remote host closed the connection).
09:33:18 -!- Patashu has joined.
09:36:44 -!- Patashu has quit (Disconnected by services).
09:36:44 -!- Patashu_ has joined.
09:47:29 -!- mihow has joined.
10:01:19 -!- Phantom_Hoover has joined.
10:23:23 -!- augur has quit (Ping timeout: 246 seconds).
10:35:12 -!- heroux has joined.
10:38:06 -!- augur has joined.
10:51:17 -!- mihow has quit (Quit: mihow).
10:59:23 -!- augur has quit (Ping timeout: 246 seconds).
11:14:02 -!- SopaXorzTaker has joined.
11:14:31 -!- augur has joined.
11:16:33 -!- Phantom_Hoover has quit (Ping timeout: 246 seconds).
11:22:33 -!- augur has quit (Ping timeout: 256 seconds).
11:29:13 -!- Patashu_ has quit (Ping timeout: 264 seconds).
11:29:13 -!- augur has joined.
11:55:17 -!- ^v has quit (Read error: Connection reset by peer).
11:55:44 -!- ^v has joined.
12:03:14 -!- AndoDaan has joined.
12:17:13 -!- SopaXorzTaker has quit (Ping timeout: 255 seconds).
12:55:15 -!- Tritonio has joined.
13:13:35 -!- boily has joined.
14:12:12 -!- SopaXorzTaker has joined.
14:21:08 -!- Phantom_Hoover has joined.
14:32:27 -!- GeekDude has joined.
15:17:21 -!- shikhin has joined.
15:35:42 -!- Tritonio has quit (Remote host closed the connection).
15:36:16 -!- Tritonio has joined.
15:41:18 -!- skj3gg has joined.
15:41:26 -!- AndoDaan has quit (Read error: Connection reset by peer).
15:41:54 -!- AndoDaan has joined.
15:45:38 -!- Phantom_Hoover has quit (Ping timeout: 265 seconds).
16:02:33 -!- SopaXorzTaker has quit (Ping timeout: 265 seconds).
16:04:08 -!- Tritonio has quit (Ping timeout: 252 seconds).
16:14:44 -!- SopaXorzTaker has joined.
16:31:33 -!- hjulle has quit (Ping timeout: 246 seconds).
16:47:51 -!- boily has quit (Ping timeout: 250 seconds).
16:50:43 -!- boily has joined.
16:51:21 -!- ais523 has joined.
16:56:24 -!- Tritonio has joined.
17:01:40 -!- oren has joined.
17:03:11 -!- MDream has quit (Read error: Connection reset by peer).
17:06:31 -!- conehead has quit (Quit: ZNC - http://znc.in).
17:15:56 -!- oren has quit (Quit: Lost terminal).
17:28:54 -!- adu has joined.
17:33:29 -!- adu has quit (Read error: Connection reset by peer).
17:34:15 -!- adu has joined.
17:41:54 -!- skj3gg has quit (Quit: ZZZzzz…).
17:55:46 -!- conehead has joined.
17:57:54 -!- skj3gg has joined.
18:00:40 -!- mrohman has joined.
18:01:06 <mrohman> This irc client eats 20% cpu
18:02:21 <mrohman> Most cpu consuming client ever
18:04:21 <mrohman> Is there yet an or-typesystem?
18:05:48 <mrohman> (List (if p 5 5.0) "hi") has the type 《int or double or string》
18:08:37 <oerjan> mrohman: i've seen such a system for lambda calculus but damn if i can manage to google it
18:09:27 <ais523> mrohman: I was working on this in my thesis
18:09:29 <elliott> there are type systems like that. i forget the name. gradual typing systems often have it. typed racket is like that i think
18:09:32 <ais523> the normal search term is "intersection types"
18:09:47 <oerjan> i tried "disjunctive" or "conjunctive" :P
18:10:13 <ais523> my favourite prior work on the subject was by Kfoury
18:10:21 <ais523> let me see if there's a publicly available link
18:10:46 <oerjan> it seems rather obscure to find regardless
18:11:30 <oerjan> actually i'm not sure this is the same thing
18:11:40 <oerjan> what i remember was _untagged_ types
18:12:05 <ais523> here we go: http://hdl.handle.net/2144/1597
18:12:18 <ais523> it's a technical report, so should be available even if you don't have subscriptions to every journal in existence
18:12:23 <ais523> (I like technical reports for that reason)
18:12:35 <ais523> that's very close to the stuff I was doing in my thesis, btw
18:12:38 <oerjan> e.g. \f x -> f (f x) had a type like ((x -> y) /\ (y -> z)) -> x -> z
18:13:16 <ais523> oerjan: that's exactly the type it gets in my bounded intersection types
18:13:23 <ais523> and regular intersection types also give it the same type
18:13:41 <oerjan> also, if you added an omega type that everything belonged to, you could characterize terminating terms in untyped LC as those that could be typed with the omega in only negative positions
18:14:48 <oerjan> (where negative position corresponds roughly to "you can take this as an argument but not return it")
18:15:08 <oerjan> and also to contravariance in functor terms
18:15:13 -!- skj3gg has quit (Quit: ZZZzzz…).
18:15:27 <oerjan> note also that there was no quantification in this type system
18:15:28 -!- skj3gg has joined.
18:16:22 <oerjan> oh and you didn't need omega for strongly normalizing terms
18:16:48 -!- ^v has quit (Read error: Connection reset by peer).
18:17:14 -!- ^v has joined.
18:17:28 <oerjan> come to think of it, this was precisely the subject which got me to make my first SO post, way back
18:21:58 <oerjan> i guess "top" type mentioned in that report abstract is the same as omega
18:38:15 <mrohman> I thought intersection types had something to do with value ranges
18:38:56 <mrohman> Certain expressions in my type system would bind a type
18:39:01 <oerjan> that's probably interval types
18:40:04 <mrohman> (If (isint q) (add 1 q) (strreverse q))
18:40:40 <mrohman> This would imply that q is (int ot str)
18:40:42 <oerjan> pattern matching is so much more elegant
18:40:58 <mrohman> So the add wouldnt typecheck
18:41:50 <mrohman> However you can add a rule that the if check binds q in subexpressions to a specific type
18:42:31 <mrohman> Which allows that add there typechecks statically as well
18:54:27 <mrohman> So in an essence the only reason for dynamic typed languages is being too lazy to write a typechecker
19:02:45 -!- Sprocklem has joined.
19:02:45 -!- Sprocklem has quit (Changing host).
19:02:45 -!- Sprocklem has joined.
19:02:50 -!- Sprocklem has quit (Read error: Connection reset by peer).
19:04:08 <boily> mrhello.. mrohello? mrohmhellon??? hmm... hellohman, perhaps?
19:05:38 <boily> mrhelloman. I guess dynamic typed languages are "it works until it doesn't and you hit some runtime error", and static are "it doesn't compile until it does".
19:06:26 <ais523> I think you can write programs in dynamic languages that can't be decidably type-checked (without rewording them)
19:08:30 -!- shikhin_ has joined.
19:08:36 <oerjan> f(n) = if n == 1 then "hi" / 0 else if odd n then f(3*n+1) else f(n `div` 2)
19:09:46 <boily> I'm not sure ghc will accept a blatant use of ‘"hi" / 0’.
19:11:14 <ais523> > f(n) = if n == 1 then "hi" / 0 else if odd n then f(3*n+1) else f(n `div` 2)
19:11:15 <lambdabot> <hint>:1:6: parse error on input ‘=’
19:11:27 <ais523> > let f(n) = if n == 1 then "hi" / 0 else if odd n then f(3*n+1) else f(n `div` 2)
19:11:28 <lambdabot> not an expression: ‘let f(n) = if n == 1 then "hi" / 0 else if odd n then f(...
19:11:31 <oerjan> boily: this is supposed to be a dynamical language hth
19:11:32 -!- shikhin has quit (Ping timeout: 246 seconds).
19:11:38 <ais523> > let f(n) = if n == 1 then "hi" / 0 else if odd n then f(3*n+1) else f(n `div` 2) in 0
19:11:39 <lambdabot> Could not deduce (GHC.Real.Fractional [GHC.Types.Char])
19:11:39 <lambdabot> arising from a use of ‘GHC.Real./’
19:11:43 -!- shikhin_ has changed nick to shikhin.
19:12:04 <ais523> I think that would work in a hypothetical dynamically typed Haskell
19:12:18 <oerjan> yes, but it cannot be typechecked then
19:12:18 <ais523> although it has to end in either type mismatch error, or an infinite loop
19:12:54 <oerjan> i assume the point is to have a typechecking that is equivalent to "only err if the program would dynamically do so"
19:13:20 <oerjan> because otherwise, what's the question really.
19:13:59 <ais523> in which case it's obviously undecidable due to TCness
19:14:14 <ais523> err, due to halting problem
19:17:28 -!- GeekDude has quit (Quit: ZNC - http://znc.in).
19:24:25 <dulla> > let f n = if n == 1 then "hi" 0 else if odd n then f (3*n+1) else f (n `div` 2) in 0
19:24:26 <lambdabot> Couldn't match expected type ‘a0 -> t’
19:24:26 <lambdabot> with actual type ‘[GHC.Types.Char]’
19:24:26 <lambdabot> Relevant bindings include f :: a1 -> t (bound at <interactive>:1:5)
19:24:44 <dulla> > let f n = if n == 1 then 1337 else if odd n then f (3*n+1) else f (n `div` 2) in 0
19:24:54 <dulla> > let f n = if n == 1 then 1337 else if odd n then f (3*n+1) else f (n `div` 2) in f 0
19:25:04 <dulla> > let f n = if n == 1 then 1337 else if odd n then f (3*n+1) else f (n `div` 2) in f 1
19:25:10 -!- Tod-Autojoined has changed nick to TodPunk.
19:25:33 <dulla> > let f n = if n == 1 then 1337 else if odd n then f (3*n+1) else f (n `div` 2) in f 3
19:25:42 <dulla> > let f n = if n == 1 then 1337 else if odd n then f (3*n+1) else f (n `div` 2) in f 9
19:30:25 -!- boily has quit (Quit: ACORN CHICKEN).
19:37:58 -!- skj3gg has quit (Quit: ZZZzzz…).
19:52:27 -!- mitchs has quit (Quit: mitchs).
20:04:48 -!- mitchs has joined.
20:16:53 -!- mrohman has quit (Read error: Connection reset by peer).
20:18:30 -!- skj3gg has joined.
20:19:12 -!- SopaXorzTaker has quit (Remote host closed the connection).
20:28:44 -!- skj3gg has quit (Quit: ZZZzzz…).
20:29:55 -!- skj3gg has joined.
20:39:55 -!- Tritonio has quit (Remote host closed the connection).
20:44:25 -!- Tritonio has joined.
20:45:39 -!- adu has quit (Quit: adu).
20:47:41 -!- GeekDude has joined.
20:47:43 -!- GeekDude has changed nick to Guest87001.
20:48:44 -!- Guest87001 has quit (Changing host).
20:48:44 -!- Guest87001 has joined.
20:48:52 -!- Guest87001 has changed nick to GeekDude.
21:02:04 -!- skj3gg has quit (Quit: ZZZzzz…).
21:09:26 -!- oren has joined.
21:14:41 -!- shikhin has quit (Quit: leaving).
21:20:46 -!- Patashu has joined.
21:23:15 -!- mrohman has joined.
21:24:01 <mrohman> Of course i can typecheck it even if its tc
21:25:10 <oerjan> not in such a way that exactly the non-erring programs fail to typecheck
21:25:43 <oerjan> PLEASE ADD/REMOVE A NEGATION AS APPROPRIATE
21:26:39 <oerjan> mrohman: you cannot make a typechecker that accepts all programs that wouldn't give a dynamic error and none that would, and does so before running the program
21:27:46 <mrohman> You mean (if p a b ) where b produces a runtime type error
21:28:31 <mrohman> But thats not what i'm aiming for
21:29:30 <mrohman> It's just that i want the style of such languages with static typechecking
21:29:58 <oerjan> i think the word "hybrid" belongs in there somewhere then hth
21:30:56 <mrohman> Which means that you have to account for ifs not always returning the same type
21:31:25 <oerjan> did you know that "hybrid types" gives you a lot of information on cars tdnh unless that's what you're looking for i guess
21:31:57 <mrohman> You wont need runtime typechecks then
21:33:41 -!- mrohman has quit (Remote host closed the connection).
21:39:32 -!- skj3gg has joined.
21:47:26 <zzo38> You could make the compiler to defer the error until runtime, although this doesn't seems like very good in most cases (although maybe in some cases it might help, if it is possible at all in such case).
21:49:52 -!- skj3gg has quit (Quit: ZZZzzz…).
21:53:05 -!- adu has joined.
21:58:22 -!- skj3gg has joined.
22:09:01 -!- oren has quit (Quit: Lost terminal).
22:22:05 -!- skj3gg has quit (Quit: ZZZzzz…).
22:39:34 -!- skj3gg has joined.
22:40:08 -!- ais523 has quit (Ping timeout: 246 seconds).
22:45:44 -!- Phantom_Hoover has joined.
23:02:33 <Sgeo> I can't tell if this soundfont has piano keys that last too long or if this music is just bad
23:19:48 -!- skj3gg has quit (Quit: ZZZzzz…).
23:26:10 -!- copumpkin has quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…).
23:27:22 -!- ais523 has joined.
23:27:46 -!- copumpkin has joined.
23:36:48 <Vorpal> Sgeo, you could say that is a key question
23:39:34 * Sgeo goes to buy all Ray Lynch music on Amazon
23:46:03 -!- augur has quit (Quit: Leaving...).
23:52:14 -!- yorick has quit (Remote host closed the connection).