IRC log started Fri Mar 10 00:00:01 2000 [msg(TUNES)] permlog 2000.0310 -:- SignOff AlonzoTG: #TUNES (Have Nice Day :)) -:- SignOff air: #TUNES (http://www.qzx.com/ :: sleep) -:- Fufie [stig@tunnel-44-33.vpn.uib.no] has joined #tunes hey Fufie 12:30am -:- Closing Link: TUNES[bespin.dhs.org] by turtledove.openprojects.net (Ping timeout for TUNES[bespin.dhs.org]) -:- Connection closed from irc.us.openprojects.net: Success -:- Use /Server to connect to a server -:- Connecting to port 6667 of server irc.us.openprojects.net [refnum 0] -:- BitchX+Deb1an: For more information about BitchX type /about -:- Welcome to the Internet Relay Network TUNES (from king.openprojects.net) -:- Your host is king.openprojects.net, running version u2.10.05.18.(ipcheck4-5) (from king.openprojects.net) -:- This server was cobbled together Wed Apr 28 1999 at 12 02:19 EDT(from king.openprojects.net) -:- king.openprojects.net u2.10.05.18.(ipcheck4-5) dioswkfcg biklmnopstv -:- [local users on irc(84)] 11% -:- [global users on irc(345)] 44% -:- [invisible users on irc(445)] 56% -:- [ircops on irc(13)] 2% -:- [total users on irc(790)] -:- [unknown connections(0)] -:- [total servers on irc(29)] (avg. 27 users per server) -:- [total channels created(272)] (avg. 2 users per channel) !king.openprojects.net Highest connection count: 103 (101 clients) !king.openprojects.net Welcome to Open Projects! You are on 4 ca 1(2) ft 14(14) tr. -:- Mode change [+f] for user TUNES -:- Mode change [+iws] for user TUNES -:- JOIN activated by "TUNES #tunes tunes@bespin.dhs.org " -:- TUNES [tunes@bespin.dhs.org] has joined #tunes -:- Topic for #TUNES: TUNES Free Reflective Computing System http://www.tunes.org/ || Slate Language http://www.tunes.org/~water/slate-home.html -:- topic set by water [Wed Mar 1 03:22:21 2000] -:- [Users(#tunes:8)] [ TUNES ] [ Fufie ] [ Downix ] [ Plundis ] [ water ] [ smkl ] [ Fare ] [ abi ] -:- Channel #tunes was created at Sun Feb 28 08:48:06 1999 -:- BitchX+Deb1an: Join to #tunes was synced in 6.994 secs!! -:- Mode change [-ws] for user TUNES -:- SignOff Downix: #TUNES ([BX] The FDA says 5 servings of BitchX a day increases sexual potency) -:- smoke [smoke@16dyn236.delft.casema.net] has joined #tunes -:- therbosa [therbosa@165.220.24.62] has joined #tunes name -:- therbosa [therbosa@165.220.24.62] has left #tunes [] 01:50am -:- SignOff smoke: #TUNES (Ping timeout for smoke[16dyn236.delft.casema.net]) -:- overfien [overfien@dialupW198.sttl.uswest.net] has joined #tunes yoh 02:10am -:- smoke [smoke@15dyn11.delft.casema.net] has joined #tunes -:- SignOff smkl: #TUNES (Ping timeout for smkl[glubimox.yok.utu.fi]) ' 02:40am -:- NetSplit: clarke.openprojects.net split from sterling.openprojects.net [03:01am] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [clarke.openprojects.net] -:- Netjoined: clarke.openprojects.net sterling.openprojects.net -:- water [water@tnt-9-7.tscnet.net] has joined #tunes -:- SignOff water: #TUNES (Ping timeout for water[tnt-9-7.tscnet.net]) -:- smkl [sami@glubimox.yok.utu.fi] has joined #tunes -:- SignOff smoke: #TUNES (Excess Flood) -:- smoke [smoke@15dyn11.delft.casema.net] has joined #tunes -:- hcf [nef@me-portland-us947.javanet.com] has joined #tunes -:- SignOff hcf: #TUNES (Leaving) -:- AlonzoTG [Alonzo@209-122-212-205.s586.tnt7.lnhva.md.dialup.rcn.com] has joined #tunes -:- dalvarez [ircusr@212.68.72.98] has joined #tunes * AlonzoTG/#tunes greets "Salutations dalvarez!" :-) * dalvarez/#tunes salutes back... -:- killer [user9287@ppp750.tele2.ee] has joined #tunes -:- killer [user9287@ppp750.tele2.ee] has left #tunes [] -:- rares [rares@wtrb-sh5-port76.snet.net] has joined #tunes abi, Downix? rares: i haven't a clue abi, seen Downix? Downix was last seen on IRC 8 hours, 13 minutes and 37 seconds ago, saying: hey Fufie [Fri Mar 10 00:25:36 2000] 08:40am -:- rares [rares@wtrb-sh5-port76.snet.net] has left #tunes [Segflossed] are there any free languages as practical as common lisp, which employ a type system? (those are 3 requirements) haskell? ocaml? mercury? -:- SignOff Plundis: #TUNES (Ping timeout for Plundis[130.238.23.252]) -:- Plundis [plundis@130.238.23.252] has joined #tunes 09:20am -:- SignOff dalvarez: #TUNES (Ping timeout for dalvarez[212.68.72.98]) -:- SignOff overfien: #TUNES (Read error to overfien[dialupW198.sttl.uswest.net]: Connection reset by peer) smoke: probably not CL has free implementations and has a strong and powerful type system though 09:50am smkl/fufie: do your opinions collide ? 10:00am of course the key issue is 'practical' I guess for my question all three issues are equal and `more practical' is allowed too :) practical as in can you use hackerish solutions, or as in is it easy to make applications with it? the only `weakness' i could find with common lisp is the lack of mandatory types smkl: the latter and I would categorise that as a strength :) statically declared types that is fufie: could you give an example of when it's more handy to not have types than to have to all objects in CL has types but they need not be declared in advance yes but it's handy to spot a lot of errors at compiletime instead of at runtime smoke: polymorphism requires dynamic typing, and polymorphism is really useful 10:10am in what sense? static typing might give you warnings or errors for _some_ problems earlier.. but you as a programmer might not have decided the type exactly yet you can't have polymorphism in haskell/ocaml/.. ? fufie: hm interesting point smoke: all those languages have parametric polymorphism smkl: how polymorphism requires dynamic typing or why polymorphism is useful? smkl: do they have dynamic typing too then? "requires dynamic typing" smoke: some have that too, but parametrically polymorphic code doesn't usually care about types at all oop polymorphism needs lookups of course to decide which method to call when having type-subtype relation dynamic typing is also useful in an interactive environment though a friend of mine is making an interactive front-end to sml it can be statically determined if the subtype relation exists but he has a lot of problems smkl: only when the _whole_ system is known at compile-time.. one cannot guarantee that.. in any usual case, the subtyping relations can be checked statically ... additions have to be checked separately 10:20am given B is a subtype of A and one has (java-style) A ptr = new B(); ptr.fn(); can that be checked statically? given that there is a function fn specialised for both A and B? the types can be checked runtime or compile-time? if .fn() would be different types, then B wouldn't be a subtype of A B is an A (ie JPEG-Picture is a Picture) uhh what would be the declaration of fn() ? java-style: class A { void fn(); } and class B inherits A { void fn(); } by removing dynamic typing one also removes a lot of expressiveness and expressiveness is the key-factor imho for practicality can't even java typecheck that? java is a quasi-statically typed language.. it tries to do both ie it keeps the type-info in the runtime system and all method-calls (unless declared specially) is done dynamically ohh .. now i understand ... you are talking about classes, not types is there a difference between types and classes? *dangerous question* ;) which are different things in many languages ocaml for example would have the same structural type for both classes what if I add an extra integer or two to B? (I think ocaml has limited synamic typing, am I wrong?) dynamic even ocaml only has methods in the type, but B would then have different structural type ... it would be subtype of A's type though 10:30am even java has your example statically typed, i don't think you could have a type error in that method call but i don't know java very well java tries to get the best of both worlds they try to figure out as much as they can statically but retain the typing at runtime and use single-polymorphism can you have a type error on that method call or not? it is not a type error in Java, no the dispatch is done on the hidden argument, 'this' at "ptr.fn();" ? ie.. one actually have two functions fn.. fn(A a); and fn(B b); and B is a subtype of A the ptr is the argument.. 'ptr.fn()' is easier syntax than fn(ptr); the argument is only hidden when making a recursive object call the argument is hidden in all normal functions in Java all functions/methods 'belong to' a class/type the argument is only hidden in method declarations (sorta) and when making a recursive object call, not in "ptr.fn()" in java, can there happen another error except null pointer, in a method call? (that is my question) if not, "ptr.fn()" is statically typed i'm not sure how well java checks castings it is statically types _and_ the dispatch is done dynamically through the dynamic type of the object castings might generate cast-errors sure 10:40am but polymorphism on type/subtype relations (inheritance) needs dynamic typing you can call it that in java you only have type/subtype polymorphism on one argument (the 'this' ptr) while CL has dispatch on all arguments i would usually call them "method lookups" or something like that ie. you can have a collision-function where 'method lookups' are done with type/subtype dispatch on both arguments collision (arg1, arg2) multimethods are harder to type, one reason is that the semantics aren't very clear the semantics are very clear and specified even the algorithm is given in the standard i mean, there isn't an agreement what they should be, of course they are specified in the different languages implementing them I think only CL and Cecil implement them but they have different semantics yes there are probably some others ... i think there should be a successor for Cecil i saw some hint about it somewhere 10:50am but multi-methods also adds to expressiveness which isn't useful in 80% of the cases, and one can hack it for 10% of the cases, but for the last 10% multi-methods give elegant and beautiful code (I have some code in the last 10% which imho is easier to read and maintain than a normal hack would've been) hmmph didn't dylan have multimethods too I think Dylan only has single-dispatch dylan has multimethods mm.. true dylan is a CL derivate though and hardly in the statically typed category afk 11:00am back 11:20am -:- dalvarez [ircusr@212.68.72.98] has joined #tunes -:- _ruiner_ [DIY@ppp004.wi.centurytel.net] has joined #tunes -:- SignOff _ruiner_: #TUNES (destroy what destroys you) -:- kc5tja [kc5tja@garnet.armored.net] has joined #tunes -:- SignOff AlonzoTG: #TUNES (Have Nice Day :)) -:- kc5tja [kc5tja@garnet.armored.net] has left #tunes [] -:- rares [rares@wtrb-sh5-port28.snet.net] has joined #tunes anyone know of a `project' system for emacs? 01:00pm -:- AlonzoTG [Alonzo@216-164-139-201.s455.tnt5.lnhva.md.dialup.rcn.com] has joined #tunes -:- rares [rares@wtrb-sh5-port28.snet.net] has left #tunes [Segflossed] om? -:- eihrul [lee@usr5-ppp84.lvdi.net] has joined #tunes 01:20pm what does a 'project' system do? 01:30pm it lists a bunch of files and then compiles them in the order they are input hm sort of yes i've written functions to insert #ifndef _BLA_H and license information in new .h and .c files that i open now i'd like to have different licenses with different projects, and i used to have seperate modes for my major projects, but it's getting to be a bit cumbersome with lots of tiny projects -:- _ruiner_ [DIY@ppp473.wi.centurytel.net] has joined #tunes -:- tcn [r@cci-209150250139.clarityconnect.net] has joined #tunes greetings <_ruiner_> greetings hmm.. not much going on here? <_ruiner_> just got here I gotta free up this line pretty soon 01:40pm <_ruiner_> k....I'm gonna go watch dragonballz soon anyhow <_ruiner_> (c:', smoke: for c/c++ one usually uses make, for CL one uses DEFSYSTEM smoke: and for c/c++ one usually uses autoconf and automake to take care of many of the ugly details smoke: for CL we're discussing how to make a DEFCONFIGURATION at the CLOCC list 01:50pm fufie: well, i was looking for emacs support for the projects, as i tried to explain above.. hooks should be called when new files are created for example, and changelog formatting should be handled in different ways according to the current project &c fufie: what's the CLOCC list btw ? ah open code collection -:- SignOff _ruiner_: #TUNES (destroy what destroys you) smoke: the problem with the emacs-approach is that it becomes very personal as YMMV fufie: how do you mean? smoke: any non-trivial project is different in structure and organisation of just about everything making a few elisp functions available for customisation, e.g to the code snippet thingie at sourceforge is a good start I guess true, but i have quite a few very similar small ones, like the demo/intro material we produce 02:00pm * Fufie/#tunes yawns.. getting tired early today :-/ hmm ... F1 02:50pm -:- SignOff AlonzoTG: #TUNES (Have Nice Day :)) -:- SignOff dalvarez: #TUNES (Ping timeout for dalvarez[212.68.72.98]) fufie: have you seen steel bank common lisp ? 03:10pm -:- _ruiner_ [DIY@ppp062.wi.centurytel.net] has joined #tunes it is on my list for testing but it will take a few weeks before I get around to it I guess -:- SignOff _ruiner_: #TUNES (destroy what destroys you) -:- rares [rares@wtrb-sh5-port28.snet.net] has joined #tunes abi, seen Downix? Downix was last seen on IRC 14 hours, 50 minutes and 17 seconds ago, saying: hey Fufie [Fri Mar 10 00:25:36 2000] 03:20pm -:- SignOff tcn: #TUNES (ircII EPIC4pre2 cLIeNUX. Can you say that?) -:- tcn [r@cci-209150250041.clarityconnect.net] has joined #tunes -:- rares [rares@wtrb-sh5-port28.snet.net] has left #tunes [Segflossed] -:- water [water@tnt-10-79.tscnet.net] has joined #tunes hello all :) who's actually willing to talk? of what? slate/tunes/arrow/retro woohoo! my substructural logics book order will be here soon! 04:00pm eih: so what about shape? what about it? well since slate isn't quite sexp, we can't exactly use cons cells as freely as lisp can, for instance actually, wait are you going to be any more useful than you have been all week? i mean, are you going to sleep in half an hour? good question yeah i should probably find someone else to talk with btw, i've nearly covered everything in "on lisp" and sicp, and i find that they have lots of problems not worth putting up with did you like "on lisp"? yes it was a great book btw, fufie, i recall that someone here wanted to make a #lisp channel i think it was tcn i'd love one 04:10pm there are enough #lisp channels already there is one on EFnet.. it's called #+allegro smkl: i only know one on efnet which has <10 users ok well i thought i'd mention it #lisp on EFnet was _small_ there isn't a #+allegro at efnet water: "on lisp" is a very nice book.. but it falls short on a few points. but as CS books go it's excellent you mean ircnet no doubt ircnet.. ok.. whatever hm how can i make the clisp plural work with foreign language plurals ? smkl: #+allegro is slightly dominated by norwegians though.. and it might have a few people from c.l.l something like 'error~:P -> xxx~:[yy~;zz~]' ? is + a magic letter that makes a channel not be able to get ops? no.. #+allegro is a CL pun.. smoke: attempting to extend FORMAT with plural from foreign languages is probably not worthwhile well, i'll need to harrash you then smoke: it's easy to define such outside of FORMAT though fufie: ehm it's for the clisp translation to dutch, and i think i have to stick with format (at least for now) ok.. are you writing a po-file? fufie: the new clisp snapshot (2000-03-06) does not build on my system (because of the dutch translation i created :-) 04:20pm -:- AlonzoTG [Alonzo@216-164-136-95.s95.tnt4.lnhva.md.dialup.rcn.com] has joined #tunes fufie: yes, it's already finished; but it bugs. ok.. I think I saw build problems mentioned on the mailing list ow? i didn't, but then again perhaps i'm not subscribed to everything it was another problem though clisp-devel@lists.sourceforge.net hm well i've got a .plan for macros in slate, but it's not conventional otoh, i think it's actually better, but then it's not ready yet :) 04:30pm the problem is that macros in slate will be a part of a larger structure transforming library which requires more development but solves more problems in the long run om ARGV!!! gah! my isp's http server is down hard! =\ * water/#tunes contemplates using his mail client for browsing 04:40pm -:- SignOff tcn: #TUNES (ircII EPIC4pre2 cLIeNUX. Can you say that?) -:- SignOff AlonzoTG: #TUNES (Have Nice Day :)) -:- SignOff eihrul: #TUNES ([x]chat) hm i wonder if mentioning arrow to the pcp guys would be helpful? -:- SignOff smoke: #TUNES (sleepy) 05:10pm -:- eihrul [lee@usr5-ppp84.lvdi.net] has joined #tunes wb 05:20pm -:- SignOff eihrul: #TUNES ([x]chat) -:- air [brand@p0wer.qzx.com] has joined #tunes -:- water [water@tnt-10-79.tscnet.net] has left #tunes [] -:- water [water@tnt-9-54.tscnet.net] has joined #tunes -:- eihrul [lee@usr5-ppp84.lvdi.net] has joined #tunes -:- lar1 [larman@dialup-209.244.105.171.SanJose1.Level3.net] has joined #tunes hey lar eihrul's finally admitted to his depression Greetings, water Oh? yeah i think i've got him to talk a bit for the first time this week How did it go? it's trying to go right now Perhaps he would like to take it to a channel with you and me? well, i'll go to #{} and see if he follows Ok Bri is in that channel, but she doens't log it i know Ok, I was just saying incase he cared about that 07:00pm -:- AlonzoTG [Alonzo@216-164-139-94.s348.tnt5.lnhva.md.dialup.rcn.com] has joined #tunes bbl tonight -:- water [water@tnt-9-54.tscnet.net] has left #tunes [] 07:40pm -:- SignOff eihrul: #TUNES ([x]chat) -:- SignOff lar1: #TUNES (Leaving) -:- hcf [nef@me-portland-us1024.javanet.com] has joined #tunes -:- SignOff AlonzoTG: #TUNES (Have Nice Day :)) -:- SignOff hcf: #TUNES (Leaving) -:- SignOff Plundis: #TUNES (Ping timeout for Plundis[130.238.23.252]) -:- Plundis [plundis@130.238.23.252] has joined #tunes -:- smkl_ [sami@130.232.134.107] has joined #tunes -:- NetSplit: carter.openprojects.net split from sterling.openprojects.net [11:14pm] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [carter.openprojects.net] [msg(TUNES)] newlog 2000.0311 IRC log ended Sat Mar 11 00:00:01 2000