00:00:00 --- log: started haskell/05.05.06 00:00:26 Gahhh - Robert Dewar apparently learned Algol 68 by reading the report. I wish I had that level of skill. :-) 00:01:16 dbremner: you can have given time. After reading the haskell report I felt like I'd rather read reference docs instead of language tutorials. 00:03:00 this method excludes the C++ standard, of course. 00:03:49 Gahhh: the lisp hyperspec is pretty trashy too 00:03:51 Gahhh - I learn languages by immersion. 00:04:39 lispy - it's intended as a reference. 00:05:09 immersion is good. What I don't like is tutorials (having to) leaving out content along the steps. My brain tries to autocomplete the missing pieces then, and I am confused later on. 00:05:51 Gahhh - exactly. I found SoE fairly frustrating. 00:06:01 SoE = ? 00:06:12 school of expre ? 00:06:14 Gahhh - the Haskell School of Expression. 00:06:32 I have that and I read it. I gained almost nothing from it. It collects dust now. 00:06:46 Gahhh - same. 00:07:09 dbremner: yeah, but have you tried to implement common lisp using it as a reference? 00:07:18 dbremner: i tried...and failed :) 00:07:34 lispy - how far did you get? 00:07:48 i got lost when i hit reader macros 00:08:03 the description of the read algorithm was very confusing as well 00:08:28 when that happened i went back to just implementing what i knew of lisp 00:08:47 and figured i'd fix stuff as it became a problem :) 00:09:07 then i ran out of free time (I became a grad student) and stopped working on it... 00:09:11 lispy - there are two ways to write a language standard. One is to formally define it - SML, Algol 68. Then you've got a consistent artifact frozen in time - additions take forever. The other method is describing it in English and dealing with the resulting errors. 00:10:00 blew english. I have an excerpt from the Python reference nailed to my cube's wall. It's unintelligible. 00:12:14 Gahhh - an approach that makes sense to me is to define behavior of the language with a Haskell interpreter. 00:14:06 --- join: iblechbot (~iblechbot@80.81.23.61) joined #haskell 00:15:34 dbremner: I think it's an inherent complexity problem most of the time. You can represent it worse, but probably not much better. 00:16:46 Gahhh - http://blogs.msdn.com/ericlippert/archive/2003/09/18/53053.aspx 00:18:56 heh 00:19:30 I suppose if you know denotational semantics, you can eat your cake too 00:19:32 Gahhh - and the related lambda discussion. http://lambda-the-ultimate.org/classic/message8778.html 00:20:17 Gahhh - I'm currently reading Denotational Semantics: The Scott-Strachey Approach to Programming Language 00:20:17 Theory 00:20:58 Do you find it useful ? 00:21:07 interesting, my PL prof last term decided that no one uses denotational semantics anymore so we didn't discus it 00:21:14 ond now i've been hearing about it a lot lately :) 00:21:18 lispy - I can easily believe that. 00:21:34 lispy - I often feel that CS is in a dark age of sorts. 00:21:38 lispy: what did he say that they were using then ? 00:21:39 which one is denotional semantics? small-step or big-step? 00:22:09 Gahhh: she, and we studied operational and um...the other one that slips my mind at the moment 00:22:56 is the other one formal semantics? 00:23:04 lispy - axiomatic? 00:23:09 ah yeah 00:23:18 lispy - the big three. 00:23:36 isn't (axiomatic > denotational) semantics ? 00:25:26 Gahhh - AFAICT it's a different approach entirely. http://burks.brighton.ac.uk/burks/foldoc/49/9.htm 00:26:02 I just love how Frank Atanassow bashes things. 00:26:21 well, i finally decided which problem to study. The instructions say to write 1 page describing what you want to work on. Hmm...I did it in two sentences. Some how I think I may be missing something... 00:26:41 --- join: Si\ (~simon@host81-152-108-94.range81-152.btcentralplus.com) joined #haskell 00:27:20 dcoutts: ping. 00:27:35 lispy: and what is it ? 00:28:09 a modified version of Knight's tour 00:28:15 lispy - my favorite smart alec response to "any questions?" was "describe the nature of consciousness." 00:28:25 in this version the boards have missing holes 00:28:32 dbremner: have you read that entire thread you posted for me ? It talks about why the guy did the DS for javascript. 00:28:44 --- join: _JusSx_ (~davide@adsl-187-56.38-151.net24.it) joined #haskell 00:28:53 DS? 00:29:00 denot seman 00:29:03 ah 00:29:12 I abbrev aggress tonite 00:29:16 :) 00:29:26 "You can read this for pleasure or personal edification. It's probably quite hard to get hold of now. It's a classic, and it's completely out of date. A bit like Plutarch." 00:29:26 Gahhh - yep. That's where I got the idea from. 00:29:29 that's okay as long as you don't get annoyed when i ask what it means :) 00:29:51 Gahhh - er the quote is from a review of the Denotational Semantics book I'm reading. 00:30:06 --- quit: sbeyer ("fump") 00:31:24 can I read it for profit ? 00:32:02 Gahhh - presumably. 00:32:12 Is it the Strachey? 00:32:42 <_JusSx_> đŋß 00:32:51 --- join: Itkovian (~Itkovian@d51A5D159.access.telenet.be) joined #haskell 00:32:56 <_JusSx_> ßðđŋħ€ŋŧ€łł«ðđŋð»¢«»«»«»«»«»«»»«»«» 00:33:06 --- join: ramen (ramen@ip68-2-168-7.ph.ph.cox.net) joined #haskell 00:33:45 --- join: magr (~magr@patroclus.uebb.cs.tu-berlin.de) joined #haskell 00:33:54 Psuedonym - yes. 00:34:10 Yeah, I read that 13 odd years ago, and it was out of date then. 00:34:30 Gahhh - I got interested in DS after reading a thesis titled Realistic Compiler Generation. 00:34:34 It's not Dragon Book out of date. 00:34:38 Oooh, good title. 00:34:46 Is that online somewhere? 00:34:50 Psuedonym - that's ok, I've got shelves of out of date boks. 00:35:04 the dargon book is out of date?! 00:35:09 Psuedonym - possibly, I've got it as a book. 00:35:11 I have an ACM Computing Reviews from October 1966. 00:35:15 reffie: Sure is. 00:35:19 It's 20 years old. 00:35:28 --- join: shawn_ (~shawn@c-24-7-48-11.hsd1.ca.comcast.net) joined #haskell 00:35:33 Pseudonym that doesn't necessarily mean it's out of date. 00:35:46 i found a book i like better than the dragon book 00:36:00 It doesn't mention SSA form, graph coloring register allocation, cache optimisation, bottom-up code generation... 00:36:05 Code scheduling... 00:36:19 it's called "Modern Compiler Design" by Grune, Bal, Jacobs and Langendoen 00:36:20 Psuedonym - it's always amusing reading descriptions of stacks before the term was used. 00:36:21 Basically doesn't mention anything that modern compilers do except LR parsing. :-) 00:36:47 Pseudonym you mean optimization techniques 00:36:56 it even describes how to make an haskell complire 00:37:08 as one of the motivations of a whole chapter :) 00:37:22 the amusing thing is that they refer to C as the portable assembler 00:37:35 I'd be disappointed if modern compilers weren't doing any new things that are not in the dragon book. 00:37:41 No, I don't mean optimization techniques. 00:37:45 Not just that. 00:37:49 I mean modern intermediate forms, too. 00:38:47 Psuedonym - Realistic Compiler Generation is Peter Lee's thesis. 00:38:57 My favourties are Muchnick and the second edition of Appel. 00:39:02 @wiki Muchnick 00:39:03 http://www.haskell.org/hawiki/Muchnick 00:39:04 Sorry. 00:39:07 @google Muchnick 00:39:09 http://www.arsenyart.com/ 00:39:11 Hmm. 00:39:39 Psuedonym - I've got Comp- w/Continuations and Modern Compiler Implementation in ML 2ed. 00:39:47 hmmm, building ghc-6-4-branch with ghc-6.4 failed for me (x86_64) 00:39:48 http://www.softpanorama.org/Bookshelf/TOC/muchnick1997.shtml 00:40:08 Yes, the Modern Compiler Implementation (not the "basic techniques" one) is one of the best. 00:40:18 And probably the best introduction these days. 00:40:34 Though it helps, if reading the C and Java versions, if you translate the examples back into ML. 00:41:14 Muchnick's "Advanced Compiler Design and Implementation" is an incredibly good book. 00:41:27 If you're into the backends rather than the frontends. 00:41:47 --- quit: jewel (Connection timed out) 00:42:10 Psuedonym - I purchased most of my compiler books after reading the comp.compilers archives. 00:42:16 * Pseudonym nods 00:42:35 --- join: jewel (~jewel@ppp-202.176.99.74.revip.asianet.co.th) joined #haskell 00:44:12 --- quit: Itkovian () 00:45:23 Psuedonym - Concepts of Programming Languages was wretched, though. 00:45:54 Haven't read it, but thanks for letting me avoid it. 00:46:14 where do you guys find the time to read all those books 00:46:22 I've been at it for a while. 00:46:23 Psuedonym - I spotted enough logical and syntax errors in the first 50 pages to give up. 00:46:38 I commute an hour each way on the train. 00:46:47 I tend to get quite a bit of reading in. 00:46:58 Especially if you know how to skim. 00:47:09 reffie - I'm single, obsessive, and have insomnia. My job involves lots of meetings and pointless emails. Abstract CS is how I escape. 00:47:22 I'm reading "Generative Programming" at the moment, and believe me, this is a book you WANT to skim. 00:47:37 So knowing which bits to skim or skip is essential. 00:47:42 Psuedonym - Scott Meyers made a similar comment to me. 00:47:56 About Generative Programming? 00:48:08 Psuedonym - yes. 00:48:15 You know him IRL, or is this an online forum thing? 00:48:33 I met him IRL once. 00:48:37 Psuedonym - he came to speak at the local user group and I talked with him afterwards. 00:48:42 * Pseudonym has a lot of respect for him, even if he still has the same haircut he had in the 70s 00:48:46 heh 00:48:52 very nice guy 00:48:58 Yeah, I got that impression. 00:49:01 And very practical. 00:49:11 He has a good bullshit detector, unlike Coplien. 00:49:27 Though in Coplien's case, he generates so much of it... 00:50:02 Psuedonym - IIRC, we were discussing metaprogramming or some such and that book came up. 00:50:07 Right. 00:50:16 my brain stops after some time. Even if I had insomnia, I couldn't do anything good with it. 00:50:18 --- quit: lisppaste2 (Ping timeout: 14400 seconds) 00:50:50 The more I think about it, the more I think I should spend my commuting time writing rather than reading. 00:51:02 There's a category theory book that needs to be written, IMO. 00:51:05 writing code or articles ? 00:51:27 One that doesn't assume you're already a theoretician, but also doesn't assume you know nothing. 00:52:07 Psuedonym - if you write that book, I will buy it. 00:52:12 Most category theory books can be summed up: All that stuff you already learned the hard way can be simplified! Isn't that cool! 00:52:12 well, good night all 00:52:13 --- part: lispy left #haskell 00:52:15 Night. 00:52:42 how many CT books have you people read ??? 00:52:51 I've read about six, probably. 00:52:57 holy monad... 00:53:08 But you know something? 00:53:14 I still don't understand them! 00:53:30 you mean all of them. You know some CT now. 00:53:31 Psuedonym -off and on, at least 3 or 4. I skim various parts of various books. 00:53:50 Except for "COnceptual Mathematics", which I understood, but it didn't go far. 00:54:11 Psuedonym - I used category theory at work to represent our new SLA. 00:54:21 Oooh, cool. 00:54:25 Oh, I know some of them. 00:54:49 But for example, I've read six descriptions of the Yoneda lemma, and I now know what it's _for_, but I still don't understand what it _says_. 00:54:53 Psuedonym - My boss liked the result. I'd read some of Conceptual Mathematics at the time. 00:55:22 I already have a title worked out, BTW. 00:55:38 I shall call this book: "Abstract Nonsense: Categories for the Armchair Computer Scientist" 00:56:16 Anyway. 00:56:21 ENough pipe dreaming. I must go home. 00:56:22 Psuedonym - I like the description "the social lives of sets". I know there's a pun there waiting to be discovered. 00:56:22 Nytol! 00:56:24 Pseudonym: since you are a sucker for CT text, maybe you read this too: http://www.cs.man.ac.uk/~david/categories/book/book.pdf (Rydeheard's Computational Category Theory). I havent read it. 00:57:19 I just have pdf's lying around everywhere. 00:57:52 Gahhh - I've got a pdf collection also. Lots of standards for odd languages, theses, etc. 00:57:58 I have read it, once upon a time. 00:58:09 I don't ever recall seeing the term "natural transformation" in it, though. 00:58:21 But I might be wrong. 00:59:20 Such scary jargon: "toposes". 00:59:55 "you won't understand shit $#^&" 01:00:11 --- join: timbyr_ (~timbyr@ppp3-5.lns1.syd3.internode.on.net) joined #haskell 01:00:12 "Toposes" is actually fairly straightforward. 01:00:31 A topos is just something that you can play as fast and loose with as you can with a set. 01:00:57 Anyway, I have to go. 01:00:58 Nytol! 01:00:59 --- quit: Pseudonym ("using sirc version 2.211+ssfe") 01:01:20 Gahhh - I've always liked this quote. "No proper program contains an indication which as an operator-applied 01:01:20 occurrence identifies an operator-defining occurrence which as an 01:01:20 indication-applied occurrence indentifies an indication-defining 01:01:20 occurrence different from the one identified by the given indication 01:01:20 as an indication-applied occurrence." --ALGOL 68 Report 01:01:55 Are you KIDDING ME!!! 01:01:57 lol 01:02:12 oh my god 01:02:36 the python quote is outdone. My cube wall will be updated tomorrow. 01:02:48 Python quote? 01:03:00 Gahhh - it was the 60s, drugs were cheap and plentiful. 01:03:08 I have an obscure paragraph from the python lang reference on my wall. 01:03:19 this is 4x better 01:04:05 dbremner: which section is it in ? 01:04:51 --- join: dblhelix (~stefan@j53155.upc-j.chello.nl) joined #haskell 01:06:41 whoa apparently this is a famous quote. many google hits. 01:07:00 Gahhh - I saw it in an email sig. 01:07:14 --- quit: timbyr (Read error: 113 (No route to host)) 01:10:11 dbremner: it's not in the revised algol 68 01:10:18 Gahhh - 4.4.1 01:10:51 Gahhh - of original. 01:11:10 ah 01:11:42 do you have a link to that ? 01:12:16 Gahhh - google found this. http://groups-beta.google.com/group/net.followup/browse_frm/thread/f644eee57a64ad81/6a3e08d0ac0fe9d7?q=algol+68+report+%22No+proper+program+contains+an+indication+which+as+an+operator+%22+-quote+-lipitor&rnum=83&hl=en#6a3e08d0ac0fe9d7 01:13:09 thanks 01:15:24 going to bed. nite dbremner and all. 01:15:46 Gahhh - night 01:27:46 --- join: glimming (~glimming@p89.kthopen.kth.se) joined #haskell 01:30:12 --- quit: blackdog (Read error: 113 (No route to host)) 01:31:46 --- join: Oejet (proxyuser@cpe.atm2-0-1071147.0x3ef2a7ea.boanxx10.customer.tele.dk) joined #haskell 01:32:48 suppose it would help if I were actually building 6-4-branch... ;) 01:33:27 HEAD seems to be building fine on i386 though fwiw 01:36:18 --- quit: Gahhh (Read error: 60 (Operation timed out)) 01:36:43 --- quit: jewel (Read error: 145 (Connection timed out)) 01:39:25 --- quit: dblhelix ("(Mac OS update requires restart)") 01:41:07 --- quit: shawn_ (Read error: 110 (Connection timed out)) 01:42:13 Good day, #haskell. 01:43:22 Morning, Oejet. 01:44:00 --- join: JaffaCake (~simonmar@tide71.microsoft.com) joined #haskell 01:50:29 --- quit: glimming (Remote closed the connection) 01:52:03 --- join: boegel (~boegel@ip-81-11-219-142.dsl.scarlet.be) joined #haskell 01:54:21 --- quit: _JusSx_ ("leaving") 01:57:17 --- join: jewel (~jewel@ppp-202.176.99.109.revip.asianet.co.th) joined #haskell 01:58:05 --- quit: asmodai ("be right back") 02:00:13 --- join: glimming (~glimming@p89.kthopen.kth.se) joined #haskell 02:00:40 Can someone provide me with the TWO definitions of monad in Haskell : Kleisli triple vs. Monad 02:00:46 Which one is Monad type class? 02:00:51 >>= vs >> 02:00:53 ? 02:02:17 glimming: Fisk. 02:02:28 --- join: asmodai (~asmodai@sakko.demon.nl) joined #haskell 02:02:50 glimming: Could you rephrase that? 02:03:27 There are two definitions of monads around. 02:03:33 Which one is the Monad type class? 02:03:58 Both >>= and >> are members of the Monad type class. 02:04:38 And neither are monads. 02:04:56 Argh 02:05:10 the binary operator in the monad class must satisfy certain laws. 02:05:23 That is not my question 02:05:37 I know that a type constructor instantiating Monad is what you calla monad 02:05:38 So 02:05:59 My question is what are the TWO (equivalent) definitions 02:06:00 of monads 02:06:05 that are around in category theory 02:06:07 in terms of Haskell 02:06:12 Kleisli triples vs. Monads 02:06:14 which one is Monad? 02:06:34 --- join: TheHunter (~TheHunter@wwsx24.win-screen.uni-wuerzburg.de) joined #haskell 02:06:42 the hunter, please check backlog 02:07:06 If two definitions are isomorphic then it is an implementation detail what operators happen to be the primitive ones. 02:07:14 Well 02:07:20 I am going to ask my students to prove them isomorphic! 02:07:58 and note also that the actual proof may or may not be expressible in haskell 02:08:24 ;-) 02:09:04 Hehe. 02:10:16 glimming, they are all equivalent. 02:10:23 Yeah 02:10:28 so which one is Monad class? 02:10:34 However, you have to forget about the fail kludge the Monad type class implements. 02:10:41 --- quit: kaol ("/me shuts down X and tries to get his new USB mouse to work") 02:10:51 Which I do not understand anyway - what is the Fail anyway? 02:11:22 Fail is not an issue 02:11:23 the reason for fail is that pattern matching can be handled gracely in do-notation. 02:11:23 ;-) 02:11:30 aha 02:11:34 Can you give an example, or a link? 02:12:06 Given monads S and T, what does it mean to write ST in terms of Haskell - is this combining/composing monads (is there a reference)? 02:12:13 I am still lost as to which definition Monad uses... 02:12:28 http://haskell.org/ghc/docs/latest/html/libraries/base/Prelude.html#9 02:12:45 these are the laws that the standard libs assume. 02:13:15 for fail, check the report (translation of do-notation) 02:13:35 I've found this line at a tutorial: data Bool = False | True 02:13:57 But using ghci I get SyntaxError: invalid syntax. What is wrong here? 02:15:37 MarcWeber: You have to save it in a file and then load it with GHCi. 02:15:57 (and Bool is already defined in the prelude) 02:17:04 glimming, monad transformers: http://homepages.inf.ed.ac.uk/wadler/topics/monads.html (combining monads). 02:17:42 you might also wanna check the references of the "composing monads using coproducts" paper. 02:18:00 Cool 02:18:02 Thanks 02:18:10 I am working with the author of the 2nd paper, neil Ghani! 02:18:11 ;-) 02:18:13 --- part: glimming left #haskell 02:21:03 --- join: bourbaki (~bla@p5091FBF0.dip.t-dialin.net) joined #haskell 02:21:07 moin 02:21:57 moin 02:23:42 hey hows it going? 02:24:18 it's ok. 02:24:58 --- join: goron (~ron@zonnebloem.demon.nl) joined #haskell 02:25:22 boegel are you there? 02:25:23 I know get an executable using ghc. But when I try to execute it I get the errormessage Permission denied. Permission is -rwxr--r-- ? I've also tried to execute it as root. Same result 02:25:59 on what OS? 02:26:06 --- join: thbot (~thbot@wwsx24.win-screen.uni-wuerzburg.de) joined #haskell 02:26:10 gentoo linux 02:26:11 and is the partition mounted noexec or something? 02:26:41 No. I haden't that problems before (using gcc or fpc or something else..) 02:26:44 thbot: @pl (SII(S(K(S(S(K(SII(S(S(KS)(S(K(S(KS)))(S(K(S(S(KS)(SS(S(S(KS)K))(KK))))) (S(S(KS)(S(KK)(S(KS)(S(S(KS)(S(KK)(S(KS)(S(S(KS)(S(KK)(SII))) (K(SI(KK)))))))(K(S(K(S(S(KS)(S(K(SI))(S(KK)(S(K(S(S(KS)K)(S(S(KS)K)I) 02:27:13 TheHunter whats that? 02:27:16 --- join: TheHunter_ (~TheHunter@wwsx24.win-screen.uni-wuerzburg.de) joined #haskell 02:27:22 --- part: TheHunter_ left #haskell 02:27:30 MarcWeber: How are you invoking GHC? 02:27:45 module "base timed out 02:27:45 module "pl" screwed up: thread killed 02:27:50 bourbaki, a test of may latest lambdabot feature. 02:28:04 ghc -o test --make test.hs 02:28:06 --- join: glimming (~glimming@p89.kthopen.kth.se) joined #haskell 02:28:10 and what are these symbols doing? 02:28:10 ooops 02:28:11 back again 02:28:23 So monads in Haskell are in Kleisli form not monoid form! 02:28:39 --- quit: boegel (Read error: 110 (Connection timed out)) 02:28:50 this being the easier thing to program with, yes (though TheHunter disagrees, IIRC) 02:29:15 Is this line sufficient to create a HelloWorld program? main=print("Hello World") 02:31:21 my point was that it's often more sensible to base monads on fmap and join and that there really should be a class Functor m => Monad m relationship. 02:31:58 and, =<< is often more convenient than >>=. 02:32:37 m >>= f = join (fmap f m) 02:32:56 m >> n = m >>= \_ -> n (You can forget that) 02:33:04 MarcWeber: yes, though I'd suggest putStrLn not print 02:33:11 you can forget fail as well. 02:33:21 if you do it with print you'll get the quotes printed too, for slightly complicated reasons 02:33:24 Heffalump: All I want to do is getting started ;) 02:33:34 return = eta 02:33:37 join = mu 02:35:30 --- quit: thbot ("killed by SIGINT") 02:37:30 Marcweber :: are you the owner of the file?, you probably want chmod 0755 filename 02:38:06 Heffalump: Lemmih Sorry.. I thought I was on the right partition.. You were right.. no rights to execute.. ;( 02:40:13 Now it's all working.. (The problem wasn't neither haskell nor the computer.. it was the persion in front of it (again)... ;) 02:41:42 --- quit: jewel (Connection timed out) 02:43:32 --- join: thbot (~thbot@wwsx24.win-screen.uni-wuerzburg.de) joined #haskell 02:43:44 thbot, @libsrc Control.Monad 02:43:45 http://http://darcs.complete.org/fptools/libraries/base/Control/Monad.hs 02:44:16 aarrggh! my firefox has a weird bug that makes this address point to microsoft.com 02:44:38 hehe mine too 02:44:38 thbot, @libsrc Control.Monad 02:44:39 http://darcs.complete.org/fptools/libraries/base/Control/Monad.hs 02:46:01 thehunter :: it isn't a buf 02:46:03 bug 02:46:16 df_, how can it not be a bug? 02:46:21 firefox puts anything that it doesn't know into a google are you feeling lucky search 02:46:50 it sees that http:// is an invalid url and does a search for http using google, which gives microsoft.com 02:47:03 you can disable it by changing a config option 02:47:16 i see. gotta turn that off. 02:47:37 about:config 02:47:45 keyword.enabled = false 02:48:06 hey thanks. 02:50:45 --- join: masm (~masm@c-217-70-66-127.bragatel.pt) joined #haskell 02:51:39 --- quit: bourbaki (Read error: 131 (Connection reset by peer)) 03:01:09 --- join: aleator (~aleator@e82.kekkola.jyu.fi) joined #haskell 03:01:41 --- quit: glimming (Read error: 101 (Network is unreachable)) 03:02:16 --- quit: goron (Remote closed the connection) 03:06:22 --- quit: thbot ("killed by SIGINT") 03:12:22 moin 03:12:42 moin. 03:13:27 ooh, what's this? ~[6~[6~[6~[6~[6~[6~[6~[6~[6~[6~[6~[6~[6~[6~[6~[6~[6~[6~[6~[6~[6~[6~[6~module "pl" screwed up: thread killed 03:13:31 oops. sorry 03:13:43 wireless playing up. 03:13:53 --- join: karingo (~karingo@209-180-169-151.ptld.qwest.net) joined #haskell 03:13:55 i've taken the liberty to make lambdabot a bit more concurrent. 03:14:05 :) cool! 03:15:33 ah, and for that i've merged the LB and IRC monads, renamed IRC.hs to Lambdabot.hs. 03:15:50 yeah, that's ok. 03:16:19 i think we can start thinking about factoring out an IRC library now. 03:16:35 I'd like the Module/MODULE/state stuff factored out 03:17:08 yep. it should be stripped down a lot. 03:17:28 s/stripped/split into relevant modules/ 03:17:39 --- part: sladegen left #haskell 03:18:19 yep, the idea is that the Module/MODULE/state stuff stays in Lambdabot.hs (that's what the modules import). 03:18:40 and Lambdabot.hs use the IRC library to manage the connection. 03:19:11 yep. sounds good. 03:22:07 --- join: glimming (~glimming@dyna222-227.nada.kth.se) joined #haskell 03:23:13 Is indenting done by tabs or spaces (how many)? 03:24:16 both, but it is best to just use spaces. 03:24:57 2 spaces per indent? tab=8 spaces as usual? 03:25:15 http://haskell.org/hawiki/HaskellStyle 03:25:28 * TheHunter uses 2 spaces. 03:25:40 MarcWeber: there are some examples there 03:25:50 most people seem to prefer 3 or 4. 03:25:53 but probably best to just read some code and see which looks good to you 03:26:04 and then there's the braindead indenting that i think haskell-mode does. 03:27:03 What is the mistake in: "tySynD n [] (appT (varT $ mkName "ForeignPtr" ) (varT m))" ? 03:28:01 --- join: goron (~ron@zonnebloem.demon.nl) joined #haskell 03:29:35 dons, sent you a mail. You will have to register a password for lambdabot and put it into the file State/passwd as "password". The current situation where everybody can just /nick ddarius and @msg register foo is very disturbing. 03:30:14 oh, good point 03:30:58 aww, boring :-) 03:31:11 (but yes, good plan) 03:31:15 --- quit: iblechbot (Connection timed out) 03:34:07 --- join: gabor (~Gabor_Gre@gauss.gwydiondylan.org) joined #haskell 03:34:09 TheHunter: I've tried to copy an examplor of a tutorial but I've been getting an indentation error ( suggestion of compiler). Then I've read that haskell uses a kind of python indenting.. So I wondered what's correct 03:34:46 MarcWeber, the number of spaces you indent doesn't matter as long as everything's aligned properly. 03:39:09 Thanks a lot.. Will have lunch now (afk) 03:39:40 --- join: JohnMeacham (~john@adsl-69-231-247-76.dsl.irvnca.pacbell.net) joined #haskell 03:39:50 --- join: shelarcy (~shelarcy@ntchba140253.chba.nt.ftth.ppp.infoweb.ne.jp) joined #haskell 03:43:09 Does anyone know where copies of thomas johnsson's papers are? I am looking for the one that used to be at http://www.cs.chalmers.se/~johnsson/Papers/analysing-heap-contents.ps.gz 03:44:47 --- join: lisppaste2 (~lisppaste@common-lisp.net) joined #haskell 03:47:25 The full title is 'Analysing Heap Contents in a Graph Reduction Intermediate Language' 03:47:41 --- quit: glimming (Read error: 60 (Operation timed out)) 03:56:25 --- quit: lambdabot ("New code") 03:56:35 --- join: lambdabot (~lambdabot@manzano.cse.unsw.EDU.AU) joined #haskell 03:57:50 @version 03:57:50 lambdabot 2p460, GHC 6.4 (OpenBSD i386) 03:57:50 darcs get http://www.cse.unsw.edu.au/~dons/lambdabot 03:59:19 is 460 just a count of the patches? 03:59:20 --- quit: goron (Remote closed the connection) 03:59:21 dcoutts: Kudos. gtk2hs is really neat. 04:01:36 @seen TFK 04:01:37 TFK is in #haskell. I last heard TFK speak 12 hours, 2 minutes and 22 04:01:37 seconds ago, but I have missed 8 seconds since then. 04:01:48 Good morning. 04:01:56 morning. 04:02:28 You rang? 04:02:38 that was fast. Anyway, lambdabot is ready for implementing a announce-listener. 04:02:53 i've created a stub in Plugins/Paste.hs. 04:03:46 Coo! Thanks :-) but I'm afraid I won't be able to hack at it in the following week (or two) at all. 04:04:06 --- join: boegel (~boegel@ip-81-11-219-142.dsl.scarlet.be) joined #haskell 04:04:18 --- join: sbeyer (~fixme@p54B812AF.dip0.t-ipconnect.de) joined #haskell 04:04:31 --- join: CosmicRay (~jgoerzen@2002:4545:7206:1:20e:a6ff:fe5c:55e1) joined #haskell 04:05:00 Real life is getting on my nerves :-/ 04:05:05 oh, nice. 04:08:11 --- part: shaggy left #haskell 04:10:04 --- join: glimming (~glimming@dyna222-227.nada.kth.se) joined #haskell 04:21:45 Heffalump, yep, it's the darcs patch count 04:26:02 --- join: broquaint (~chatzilla@81.187.213.103) joined #haskell 04:32:29 --- join: kaol (~kaol@bickymi.proput.fi) joined #haskell 04:33:30 --- join: cptchaos (~cptchaos@d131219.adsl.hansenet.de) joined #haskell 04:36:47 --- join: basti_ (~basti@p54AE0150.dip0.t-ipconnect.de) joined #haskell 04:37:02 hi 04:37:24 Guten Tag, Herr basti_. 04:37:42 :P 04:38:25 pfft, no German greeting for CosmicRay.... :-) 04:41:24 Aber doch, willkommen zum Kanal, CosmicRay. 04:42:07 danke :-) 04:42:21 * CosmicRay had to look up Kanal :-) 04:42:45 <-- doesn't know much German yet 04:45:22 hmm there's a nice article about german in the german wikipedia 04:46:25 "The rules of declination are very sublte, which makes learning german difficult in the beginning" 04:46:41 (12 declination classes) 04:50:02 --- join: goron (~ron@zonnebloem.demon.nl) joined #haskell 04:50:13 --- quit: glimming (Read error: 110 (Connection timed out)) 05:00:00 --- quit: CosmicRay ("Leaving") 05:02:27 --- quit: sorje (Remote closed the connection) 05:02:39 --- join: sorje (~martin@localhorst.lddd.org) joined #haskell 05:03:46 --- join: iblechbot (~iblechbot@145.16-dial.augustakom.net) joined #haskell 05:12:05 basti_: What is a declination? 05:12:58 uhhm... german words get different suffixes in different "cases" (Kasus?) 05:13:43 I know about cases, but are there 12 of them? 05:13:45 nominativ, genitiv, dativ, akkusativ... 05:13:54 no there's 12 different ways to decline a word 05:14:01 Oejet: in finnish there is :) 05:14:37 12 different ways of how to do it 05:14:44 and some exceptions. 05:14:49 tuomov: Cool, there are six in Russian. But isn't it just shifting from "declinations" to cases? 05:15:16 err? 05:15:18 a case is a syntactical construct, where a declination is a morphological one to reflect a changing case 05:15:21 cases correspond to propositions 05:15:24 s/pro/pre/ 05:15:46 "des hauses" and "der maus" are both genitive, but "haus" and "maus" belong to different declination classes. 05:16:47 basti_: finnish has that many real cases, and more declinations 05:16:52 ibid: i know. 05:16:52 http://www.cs.tut.fi/~jkorpela/finnish-cases.html 05:16:53 =) 05:17:09 --- quit: boegel (Read error: 110 (Connection timed out)) 05:17:11 I don't even know how many ways there are to inflect a word.. 05:17:16 ah, meant to say that to Oejet , apparently :) 05:17:45 http://de.wikipedia.org/wiki/Deutsche_Sprache#Deklination < "Substantive und Deklinationsklassen" 05:17:47 http://www.ling.helsinki.fi/~fkarlsso/genkau2.html 05:18:06 ibid: lol 05:18:46 those are all possible forms of the word "kauppa" 'shop' 05:18:50 the base words change somewhat with the different inflections, but it's all apparent from the syntax 05:18:57 there aren't diffent classes as such 05:19:03 includes independent combination of cases, possessive suffixes and so on 05:19:13 tuomov: actually, there are 05:19:27 tuomov: we finns just aren't taught about them 05:19:36 tuomov: because we get it from our ear :) 05:19:59 tuomov: Nice schema. I think I can recognise instruktiivi from Russian. 05:20:09