IRC log started Mon Dec 20 00:00:01 1999 [msg(TUNES)] permlog 1999.1220 -:- SignOff zarq: #TUNES (Ping timeout for zarq[10dyn91.delft.casema.net]) -:- zarq [zarq@9dyn6.delft.casema.net] has joined #tunes -:- water [water@tnt-9-204.tscnet.net] has joined #tunes -:- SignOff zarq: #TUNES (Ping timeout for zarq[9dyn6.delft.casema.net]) -:- zarq [zarq@9dyn31.delft.casema.net] has joined #tunes -:- smokie [tw026024@zaalf15.twi.tudelft.nl] has joined #tunes -:- SignOff eihrul: #TUNES ([x]chat) -:- SignOff smokie: #TUNES (Read error to smokie[zaalf15.twi.tudelft.nl]: Connection reset by peer) -:- Kaufmann [kaufmann@amadeus.fluid.impa.br] has joined #tunes FOO! 04:00am bar! ah, I knew /someone/ had to be paying attention :) not always true but anyway what's new? GOD, it's good to be working at a nice place for a change heh lord knows, i don't where do you work? * Kaufmann/#tunes has a Scientific Initiation scholarship at the Institute for Pure and Applied Mathematics, Rio de Janeiro. oh cool `finger kaufmann@impa.br` sorry, not in the habit of touching people ;) but I'm quite user-friendly! :) * water/#tunes is attempting to reset his pda which locked ua game of DoomCE s/ua/during DoomCE?!? Wow, that's fresh yeah, and it runs nice and smoothly amazing... not to mention being free you've got a Cassiopeia, right? yeah e-100 electronics prices in Brazil are extorsively high the only reason i put up with wince tariffs or something? cell phones are an exception, thanks to the current 'everybody's got one craze' urgh cell phones are an exception, thanks to the current 'everybody's got one' craze try buying stuff on the net? that's more like it :) water, import taxes nonetheless ah... the 'catch-all' tariff 04:10am water, tried installng one of the micro-Linuces? micro-linuxes? as in? there are plenty www.linuxce.org whoa hmm squeak'd have to be ported to it for me to cross over, i think but i'll try anything once heck, port it yourself (starting from the regular Linux version) hmm oh man does gcc target mips r4300? "Netscape Communicator: it sucks, but it sucks exactly the same in three dozen different platforms!" heh lemme see someone told me the other day (erroneously) that you had to replace the rom to switch oses on a wince device depends on the specific device, of course. i suppose hmm... no touch-interfacing implemented yet no wait wrong on that count.. i read an out-of-date page found this: http://www.gnu.org/software/gcc/onlinedocs/gcc_3.html#SEC51 04:20am well, it obviously has the code-generation for mips built-in yeah eek... roughly 12mb storage+6mb ram for any useful work that's a tight fit even on mine what? RAM-wise? yeah maybe i can offload the os and stuff into the back-up card until someone offers a rom mo' RAM, babee :) gadzooks! how much is an 8-meg going for these days? they re-compiled gcc on the casio itself! $400 i think * Kaufmann/#tunes promptly adds "gadzook!" to his dictionary of interjections maybe $300 since it's close to xmas heh ahh, Apple Day apple day? Old Isaac's birthday ah So I watched open TV last night, after a two-year cable-only sabbatical how did it feel? (not so strange since most of the crap that Americans get on open channels only plays here on cable) well, Brian, it felt... dirty. heh 04:30am and the ads... good god, the countless crappy Christmas 30-second ads abi: BYOK is Build Your Own Kernel, as opposed to Bring Your Own Kernel LOL that's a very useful piece of information * water/#tunes imagines a byok party * Kaufmann/#tunes feels nausea at the thought... half a million Linux geeks compiling 2.3 series kernels heh that reminds me, i'm still running 2.0.34 heh that's nothing, I'm still running SunOS 5.6 yikes yeah pain and suffering, my friend... pain and suffering i think we have a few of those units at work with CDE 1.2, to boot heh they have cde, i haven't checked the version, though really, my little SPARCstation LX isn't so bad :) hmm.. booting linux on my handheld will be tricky I've been thinking of taking one of these infobots and hacking it to support context "context"? how? 04:40am instead of just serving as a glorified database client, the bot would actually look at the data fed to it, and would use it to build a semantic network as in looking for infrequently-used words in the last 20 lines? hm so when I ask for "assembly", it not only gets me the "assembly" record, but also tells me about machine code, different processors, etc., gives me "Related Topics", etc. have fun! (said the self-proclaimed author of the arrow prototype) LOL btw, are you fammiliar with model theory at all? in a formal sense? yeah, kinda well, no i'm trying to build the same idea in arrow in order to complete the vm spec what idea? idea is that there's a mutually available spec for the data structures passed around that maintains semantic content across nodes * water/#tunes adds "LinuxCE.org" to his bookmarks ROTFL uhh abi forget idea water: I forgot idea -:- SignOff smkl: #TUNES (smkl has no reason) the idea of being able to construct an interpretive structure and then interpret info according to that structure context like logical truth according to a model vs. absolute truth yeah, in a sense I see what you mean brb context seems more like a more specific differentiation of model k back any thoughts? any literature on the subject? online, I mean model theory itself? * water/#tunes web-searches anything to clarify i don't think formal math pages usually "clarify" LOL hrrrm http://www-mgi.informatik.rwth-aachen.de/FMT/ formal model theory homepage 04:50am s/formal/finite how would you do this in an Arrow system? still working that out. functions and operators would be graphs, of course constants, i'm not sure variables are functions, but at another level ignore that FMT page oh, NOW you tell me -:- smkl [sami@glubimox.yok.utu.fi] has joined #tunes sorry, there's a tutorial page i'm looking for j/k how are variables functions? functions from the set of context symbols to values ah i.e. a set of bindings so what differentiates a constant from a variable in that sense? hmm. a meta-contextual constriant? constriant? heh. constraint i really hate the ACM they advertise great papers, but require a membership fee that's more than i can reasonably afford do you have membership? in other words, there's nothing intrinsic (sp?) about a constant that makes it fundamentally different from a variable or any function? water, in the ACM? acm: yeah constant: not really ontological relativism? i guess ontological relativism is the idea that "what is" is relative whee http://www.acm.org/pubs/toc/Abstracts/jacm/147524.html thanks, abi no problem Kaufmann :) that paper i just gave a link for would be a great resource I now understand the Tao cool :) seriously, I finally kind of get it not the Tao Arrow oh ok 05:00am * water/#tunes tries in vain to find good online MT resources well, it (the abstract) looks good... but no online version cheapskates http://www.qucis.queensu.ca/home/dalamb/courses/422/tmintro/tmintro.html that one looks pretty good and programming-applicable, too so, what would determine that, as far as another function is concerned, this given graph represents a constant, and not a variable? Where does the constraint go? good question :) the meta-model, i suppose, but i hate that concept so the meta-model would somehow hold a representation of the graph that constitutes the constant and associate it with another graph that represents the constraint? or something of the sort? yeah, but don't hold me to that * Kaufmann/#tunes grabs Brian by the ankle the central point hinges on how to specify the graph i.e. i need a framework allowing me to express my graphs to a fine enough detail good god, it took half an hour and five reloads for this damn page to load correctly without making the framework too complex to model itself which page was that? when you talk about expressing a graph, what do you mean? water, some sub-page of opengl.org spcifying the contents of a graph precisely oh brb 05:10am back * water/#tunes refines the search to "abstract model theory" a rareer subject, but more useful to the cs person I'm learning OpenGL (we do a lot of modeling here), but it's awfully low-level for everyday hacking s/rareer/rarer yes, it is but then, most gl's are far too low-level so I'm looking for a nice high-level library on top of OpenGL maybe a Perl binding or something of the sort PerlGL? nice I just might hack one up myself mmm... SelfGL * water/#tunes slaps himself heh damn it! i need to stop that fantasizing! Morphic? yeah, morphic is ok i work with the smalltalk version, which is a half-hack I don't quite remember what it /is/, though... seems to have more to do with UI in a general sense, no? yeah, it's a direct-manipulation interface to programming objects the object on the screen IS the object in the store right you essentially get a dialog box which equals the object's programming protocol pretty nifty unfortunately, the smalltalk version needs work (particularly with halos) they're apparently going to merge ThingLab with it so that you can program physical behaviors into morphs and vice-versa 05:20am ThingLab? yeah, developed in the early 80's to give a simple programming interface for making objects act according to physics constraints as I thought I found a set of Perl bindings it's a pretty cool idea, too but they're just one-to-one bindings water, got an URL? for what? oh, amt not yet no, for ThingLab hmm i have an overview of it will do http://www.2share.com/thinglab/ThingLab%20-%20index.html BTW, I'm almost completing the Code Hell project Code Hell? 05:30am It's a compendium on the subject of code obfuscation techniques. ahh, now i rmember i really need to get in contact with UWash CS department why's that? because they're nearby, and they have a lot of resources, and they might be interested in arrow I recall you having a problem with "mainstream" CS researchers yeah, i do nu? i don't know how it will work out, but i definitely need some kind of help MAN, I'm hungry... I'm gonna go grab a byte brb obviously with raw coding, but more so with getting my research done effectively k -:- Kaufmann is now known as KaufmannBRB -:- SignOff smkl: #TUNES (brb) 05:40am -:- smkl [sami@glubimox.yok.utu.fi] has joined #tunes wb 05:50am hello what's new? nothing (except that latest bX is unbelievably buggy) heh -:- KaufmannBRB is now known as Kaufmann back k smkl, bX? bitchx thought so what release? 1.0c13 hrrm mine says 75p3 that's some older version of course the "1.0" version always introduces bugs :P ah well 06:00am cool i found a workaround heh. i just realized that "error 404" counts for jargon abi jargon? it has been said that jargon is at http://www.tuxedo.org/~esr/jargon/html/The-Jargon-Lexicon-framed.html yep. it's there * Kaufmann/#tunes thinks about MIND 06:10am abi, MIND Is Not Delphi, Kaufmann's project for a modular IDE/project management system in Perl. blasted hell I've figured out a rather nice capability system for controlling access -:- yoo [ultima@user-38lcn78.dialup.mindspring.com] has joined #Tunes la de da (just idling how nice hey yoo :) 06:20am ... out there in the cold, getting lonely, getting old, can you feel me? cool song yep darn it, i still have that jargon window open and am still looking up entries in it so? heh so, it's addictive ah i even found "scram switch" in it Damn water, me too =| i've been doing that since 7-ish this morning what time is it there now? 8:20 =) geez Well, I'm coding at the same time. * water/#tunes closes the window and continues on his quest to find an online _grokkable_ intro to model theory model theory? yeah Elaborate? hmm... the theory of theories :) e.g. finite model theory, abstract model theory, ... Gee. yeah, not exactly light reading fun, tho uh sure hmm... abi has a useful link, iirc abi: mice? mice is model-integrated computing environment at http://www.isis.vanderbilt.edu/research/edcs/edcs.html it's a fairly lame implementation, but still interesting oh yeah, it also uses multi-graph's to model it's domain network :) 06:30am (personal favoritte feature, of course) of course it's almost as useful as my ontology graph idea, except that my graphs will allow reflective use of inter-domain links heh oh boy hm? nuthin' 06:40am hum. Gakuk. water: hiya. kaufie: still polluting /. ? Yablutsk! Fare, yeah push @interjections, "Yablutsk!"; heh fare: know any good online resources for abstract model theory? or abstract modelling in programming? * yoo/#Tunes pities poor slashdot editors Mean old kaufie trying to force a clue into their tiny little brains =( hmm. abstract interp counts where at? LOL yoo 06:50am UNIX-heads, the lot o' them not really not really what? too many of the /. screaming kids is not really UNIX-heads.. they're screaming kids.. just like the amiga-kids a few years back.. screaming kids I would trust aUNIX-head to be able to do something, but I can't extend that to the /. screaming kids makes sense with their eye-candy window managers and all :) a /. screaming kid thinks that perl is cool, that ESR is really cool and now even rich, and that Linux rocks because it is not Microsoft ...but Perl /is/ cool! to put it simple; perl sucks, ESR is a rich loser and Linux is cool because it's free I wish it were /dead/ cool. 07:00am heh I wouldn't be so harsh on Raymond, tho kaufie: he has done some good but he lost contact with the planet a long time ago oh well.. need a shower afk lost contact with the planet? brb too damn thee xemacs betas are buggy 07:10am Fare still here? sami: somehow Fare: could you op the bot on ircnet? 07:30am except that ircnet splitted again :) * yoo/#Tunes mumbles about silly mathematics what's silly about mathematics? :) well silly trig stuff yeah, trig is pretty silly 07:40am Vertices of eye: 0.000000,0.000000 10.000000,-10.000000 10.000000,10.000000 yay finally my mathz came out with nice even numbers like they wuz spo'zed to =) * yoo/#Tunes does a little dance abi seen beholder? beholder was last seen on IRC 1 days, 15 hours, 18 minutes and 1 seconds ago, saying: I'll be right back [Sat Dec 18 16:30:02 1999] 07:50am -:- SignOff yoo: #TUNES (Read error to yoo[user-38lcn78.dialup.mindspring.com]: Connection reset by peer) back aaaaaaaah, FINALLY some Real Man's work! 08:00am what's that? math hacking! no more of that silly CORBA nonsense! really? doing what? oh yeah, your new job -:- yoo [ultima@user-38lcn6b.dialup.mindspring.com] has joined #Tunes okie my maths aren't werkin so well again yoo: #math on efnet or udnernet s/udnernet/undernet heh nah its not hard stuff just with lots of numbers and stuff 08:10am oh so why bother us? :) I'm just rambling =) oh boy, it feels good to be hacking again Kaufie: CORBA? i guess CORBA is nice when you don't have time to hack proprietary protocols and when you want to interface with Java heh uh? for Java, here's RMI. (well, ok, that's for Java<->Java). la de da yeah, CORBA -:- smokie [tw026024@zaalf19.twi.tudelft.nl] has joined #tunes 08:20am corBAH Kaufie: what did you do with it? CORBA's a buzzword Oh hey look at me I used CORBA! bzzzz bzzzzz CORBA -- client-server -- three-tier architecture -- Object-Oriented -- Industry-Standard -- Windows NT -- Java -- GUI Fare, I work at the Fluid Dynamics lab; we do simulation software for research. But there's only X people in the world who do research on fluid dynamics, and they decided it'd be a good idea to add remote collaboration features to the software good Kaufie: know anything about multilingual meta-HTML tools? 08:30am such as? damn mulitilingual people....+) =) too * Kaufmann/#tunes opens his mouth, from which sprout countless snake-like tongues lol hmm to compress or not to compress... so Fare... heh. Fare never pays much attention after he asks a question how user-friendly :) hmm and *he's* the one with all the bright ideas about user-interfaces :) 08:40am water: I'm multitasking between several conversations, you know plus work i know just poking fun Fare: such as? see? hopeless :) such as? well, htmlpp, or the stuff used at ffii.org hehe or whatever else I don't know (maybe cl-http or pliant) cl-http is neato well, I don't :P * Fufie/#tunes yawns.. too much stress with all the presents for christmas 08:50am anyone know where to find emulator roms? my li'l brother is a game freak I can give you his email :) heh. no thanks i'm just looking for games to pass the time, mostly for my cassiopeia okie -:- yoo_ [ultima@user-38lc652.dialup.mindspring.com] has joined #Tunes feh! foo! feh? yoo: know of any good rom sites? ROM? Fare, as in game emulators Tril has a lot of roms ftp.apple2.asimov.net ? s/2// yep. i know that one -:- SignOff yoo: #TUNES (Read error to yoo[user-38lcn6b.dialup.mindspring.com]: Connection reset by peer) that's the only one I know. Apple][ rules. well, sure it does hell, i taught myself three programming languages on it you know it, baybee! hmm. i seem to have found a good site Oh? good sites are so hard to find... :) hmm. but i *still* haven't found any good stuff on model theory although jmc's paper turned up, which is good model theory? I know a good book about it for classical logic. but it's in french. Else, the book "Set Theory" by Jech is a great classic. ouch. no translation? not afaik. Maybe in classical aramean, by the author. the book's title is "Théorie des Modèles" what happened to PICT? actually, there's a re-print out called "model theory" this year, but the results look too theoretical to be of use to me or rather, "Cours de Théorie des Modèles", by B. Poizat. classical model theory is sure very theoretic. Lots of stuff about large cardinals and stuff. 09:00am maybe you'll prefer books on intuitionnistic and constructive logic. yeah, i'm more concerned with relating the formal structure with the common notion i've looked at those before, and was not impressed, though -:- SignOff smokie: #TUNES (BitchX-75+Deb1an -- just do it.) Fare: what about *abstract* model theory? -:- hcf [nef@me-portland-us121.javanet.com] has joined #tunes how that "abstract" ? dunno, i've just heard about it in various books sort of an abstract interpretation-style theory 09:10am abstract interpretation is essentially equivalent to typing. it's also equivalent to having adjunct functors in cat. theory essentially, yes. i'm interested in the relating of the various formalisms, though abi: minse? minse is Medium-Independent Notation for Structured Expressions at http://www.lfw.org/math/ * water/#tunes clicks -:- Downix [down@d-gnaps-171.ici.net] has joined #tunes morning all morning Down whats up, Down? YAY! I just hit 20 Karma! nada, designing a plug-in for my Amiga 600 that'll give it 2 ethernet and a modem posting too much on /., eh? 09:20am I suppose too much /. obviously * Downix/#tunes is getting paid for it too, it's cool some people still develop hardware for Amiga 600 ???? Fare: They do if offered $350 to come up with the design only $350 ??? Fare: $350 for a half-hours work is pretty good half-hours? Fare: beauty of the Amiga, it's bus system is very simple even the A600 heh /. is evil Kauf and does hte $350 also include a xorking prototype? Don Don't make me recite from the Necronomicon -:- yoo_ is now known as yoo =) Fare: No, working prototype costs an additional $400 it is not! You people are just jealous of my Karma :) Fare: Win-win deal I feel. This guy has a few hundred Amiga 600's he wants to turn into DHCP servers geez... a few hundred? Downix, figure he'd sell me one dirt-cheap? kaufie: oh.. remember who helped you to the top :p Kaufmann: No idea, likely. He's planning on selling them mid-2000 LOL Fufie BTW, what will we call the next decade? I've already decided on "The Two-Kay's" The Aeon of Death >:) I will call it "the next decade" yoo: Naming it after my girlfriend? Cool * water/#tunes would just be happy with a NeXT(tm) decade water: NeXT(tm) would be cool or maybe the BeBox decade Fufie, it won't be The Next Decade anymore in 12 days with Power series chips 09:30am the <> decade hundred Amiga600 ??? heh kauf: it would using the mc68000 port of Linux? hcf: got me thinking :) kauf: besides.. judgement day is at the same day fufie: darn it, it'll screw my schedule up :) expect long waiting as the court is sure to be full LOL Fufie judgement day? Hmm. Who is the judge? God Jehovah/YahWeh/Tetragrammaton? The God of Lies? just God That one? Jehovah died in a car acciden two years ago * water/#tunes whacks yoo up the side of the head with some Revelations. :) Bah. lots of witnesses lol Oh Fufie =) That was bad =) good :) -:- AlonzoTG [Alonzo@216-164-133-195.s195.tnt3.lnhva.md.dialup.rcn.com] has joined #tunes agnostics will go through a rehab program before court and atheists are sent directly to the cozy place downstairs with the fireplace atheists will deny this of course -:- smoke [smoke@15dyn67.delft.casema.net] has joined #tunes * Kaufmann/#tunes the atheist laughs his ass off * water/#tunes envisions actually seeing Jesus and repeating the content of "Simulation and Simulacra", pp.127-128 all over again GEB? GEB is _Goedel, Escher, Bach_ by Douglas Hofstadter or insanely great tsc water, was that a GEB reference? 09:40am nope nihilism => "God does not exist, no matter how real it is." "Simulation and Simulacra" ? it's a book by whom? hmm haven't read it in a while but surely you remember its author! -:- eihrul [lee@usr5-ppp41.lvdi.net] has joined #tunes sorry, not off the top of my head no one in real life ever asks me about philosophy not a big surprise, mind you :) heh i'm looking it up on the web heh. this is funny. the only nihilist author i've come up with so far is Nietzstche. and i *know* he didn't write SAS 09:50am I somehow fail to see how that's funny Kaufmann: some of the urls i gave water in the 1999.1213 irclog may interest u (model theory, etc) eihrul1 thankee -:- SignOff water: #TUNES (Ping timeout for water[tnt-9-204.tscnet.net]) -:- SignOff yoo: #TUNES (Read error to yoo[user-38lc652.dialup.mindspring.com]: Connection reset by peer) -:- yoo [ultima@user-38lc61r.dialup.mindspring.com] has joined #Tunes no fun no fun no fun 10:00am water: SAS? Save Are Santa? =) Fare: peer got to water * yoo/#Tunes needs a nail clipper yoo: only problem is you mean "our" eihrul: Read that page in the topic some time. It was a joke =) 10:10am how do you say that some pages are clearly superior to the other ones: is "coming out of the batch" valid? or rather "stand out"? "stand out" or "stand above" but the phrase is really overused to the point of becoming ineffectual morning * AlonzoTG/#tunes greets "Salutations Downix!" :-) what's up? well, guys, I'm going. Fare: "clearly superior" sounds fine Night Kaufmann -:- SignOff Kaufmann: #TUNES (Kaufmann has no reason) la de da later -:- SignOff yoo: #TUNES (Leaving) * Downix/#tunes sighs Damned if I do and damned if I don't If I take an abstracted layer, I loose performance but gain safety if I do direct hardware access, the opposite problem -:- yoo [ultima@user-38lc6d5.dialup.mindspring.com] has joined #Tunes Linux is better at trashing Windows NT than even NT is. Fuck. hehe This is the second goddamn time this month. 10:20am oh no beer commercial on the radio 'Mr Ice Smoother Outter' 10:30am la de da so what's up? -:- bdb [jp@bdb.rezel.enst.fr] has joined #tunes hi y'a qqun? any&? any1? -:- NetSplit: merril.openprojects.net split from sterling.openprojects.net [10:39am] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [merril.openprojects.net] 10:40am hmm * yoo/#Tunes waits for the end of the world only a couple of weeks... lol 10:50am zzzzzZZZzzzz I hate these goddanm technicalities. oh? what technicalities feh just a nasty usenet debate Involving, of all things, some woman who claims to be the second coming of Jesus. cool why do people care so much about jesus? it's a 5 letter word... 5 letter words are 1337 om heh Methinks this woman needs an introduction to some egyption gods. Set, Horus, and Nyarlathotep would make a great 'Christmas Carol' trio =) Did you know that federal law requires brewers to age their beer/wine in oak barrels when using an activated carbon filter would be more efficient? DAMN THE GOVT!!! lol Alonzo: Are you serious? would be absolutly serrious. Uh.... Do you have a reference? 11:00am lemme go search for one... I'll probably find it in about 30 minutes... alright thats perfect I'm gonna go shower and wash away the stinkies yoo: Anubis doing a duet with Ma'at * yoo/#Tunes smiles The Egyption Gods. Now *they* had balls. Yup What I hate are people putting Set as the Egyptial "devil" Well Set's wasn't a devil That was just a Xian thing because they wanted to make him look bad. Right But Set was actually a 'good guy' iirc Set responded to his brothers tretchery protected Egypt and all agreed yoo Osiris stole his chance for an heir Set did not want him dead, he trapped him to teach him a lesson anyways * Downix/#tunes always roots for the underdog, go Set! Your into Egyptian Mythos, no? Yes which is why my company is called Anubis 8) I see =) Makes sence, we're bringing a dead concept back to life Your dead concept being? Amigas? de-centrilized processing Downix: Anubis? Why not Thoth? performance through acceleration, not CPU 11:10am Fare: Thoth is too well known Downix: Your joking. Downix: Anubis was part of a TV series =) yoo: Gargoyles, I know Anubis was in Gargoyles? Wow. yup s/a/multiple then! * Downix/#tunes nods My partners liked it Popular fellow. Anubis that is I was originally for Set, but you can understand why we decided against it Yah. But I'd say Anubis is more popular than Thoth, and definately doesn't sound as 'threatening' * Downix/#tunes nods Anubis is "cuer" cuter even We thought the son of Set would fit as well a misunderstood god Anyways, that's all image * yoo/#Tunes nods We all know Isis was the best =) Yup Even if she was a slut 8) That's precisely why =) 8) isn't it Seth? Seth is the proper spelling actually it's Sseth Set is the common one Set is just vulgaris =P Right The Egyptians definately had to have something behind em though. * Downix/#tunes nods * yoo/#Tunes ponders about that. Wonder if their gods are real, or if they were perhaps aliens? =) I believe they are real -:- SignOff AlonzoTG: #TUNES (GPF) Of course I'm Pagan * yoo/#Tunes nods Shamanistic? I see Set Theory in a new light. yoo: Yes -:- AlonzoTG [Alonzo@207-172-184-23.s23.tnt6.lnhva.md.dialup.rcn.com] has joined #tunes fuck 11:20am Downix: Ah. I'm a Chaote. Technically I'm a Wiccan * yoo/#Tunes acks as his eyes bulge hrm? All the more bambi power too ya ; ;) hehe I said "Technically" What do you mean by that? I say "Wiccan" since few people know other Paganistic religions I don't even have a real religion, only a set of beliefs that I adhere to Well, I tend to define a Wiccan as someone who believes in the God/Goddess cycle, three fold law, being good and watching disney movies... hi hehe, not me at all I do believe in duality there is Ptah and Ap Good/evil? No am i in the middle of #theology? good and evil is all in the eye of the beholder * yoo/#Tunes is not familiar with ptah/ap Ptah begot the egyptian lineage of gods Ap was his brother in Mesopotamean legend Ap was called Enlil and Ptah was Enki I see. bdb: Well, yes =) mmmkay Religion can be reflective too yanno =P Yup tunes is really not going to ship soon then :) Duality makes sence but which uis good and evil? Sseth killed his brother was that good or evil? It was an evil act but done for a good reason? Perhaps it is a very grey area I see your point though. Thoth is 2 beings if you'd notice there is the ape-bodied Astenu and the ibis headed Tyroth but they are both Thoth he is one being yet two at the same time * Downix/#tunes likes duality, it makes sence i like the number 1 better myself 11:30am 1 is dual one side has a point, one does not are you talking about its graphical representation? :) Right irrelevent i'm talking about the number behind the glyph I'm being silly of course, all duality is based on a singular point Well, there is '1' and then there is the absense of 1 yup Therefore...even 1 embodies duality ok Downix: What's a good offering for Enki? hrm? usually dust actually Why dust? He was the god of the earth and sky Enki Bilal? Believe so * Downix/#tunes has not studied Ptah/Enki much in many years What about Inanna? dunno wow.... * Downix/#tunes feels woefully unprepared 11:40am * eihrul/#tunes just realized that monadic computations explain task migration really well. lol pretty nifty * yoo/#Tunes lights eihrul on fire eihrul: do they? Fare: yes... imperative languages like C++ which require returns are not optimal in network situations.... Tunes, the only channel where we discuss the interrelation between the mythos of mesipotamea and egypt Fare: if you pass the continuation as an argument to be executed on another node of the network... then you don't need to return to the previous server oh, sure. which is just really a way of explaining task migration eihrul: did you read D.A.Halls' phd thesis? checkpointing the state and passing it to a continuation on another node Fare: nope eihrul but thats just implentation bah implementation yoo: no.... its a utility of monadic computation :P monadic computation is just function nesting without explicitly nesting the functions =P syntax of a language =P =P sure but its a useful concept in itself for viewing task migration And the point remains that Bob the Pot-Smoking Discordian is walking around on ground and your trying to migrate tasks =) of an imperative program :P lol * yoo/#Tunes smiles eihrul: dah does everything you say using metaprogramming over Scheme. 'dah'? abi dah? yoo: no idea yoo: seen Haskerl? doh! Fare: yep abi: Haskerl? Haskerl is at http://www.people.Virginia.EDU/~sdm7g/LangCrit/Haskerl Fare: that's how i learned about monads in the first place Haskerl? Haskerl is at http://www.people.Virginia.EDU/~sdm7g/LangCrit/Haskerl wtf is haskerl Fare: but i kind of underestimated their power until this :) NO! EVIL! hehe haskell and perl mixed! There is no greater evil =) oh man Fare: oh, haskerl Fare: i thought you were saying haskell and just made a really bad typo packaged by Microsoft? Visual Haskerl++ lol exactly With the recurser joystick I can't wait until the day when joystick support gets built into VC++ hehe Ok, now to use MFC6.0...just fly the little microsoft plane to the toolbarland, and shoot the MFC building with an instantiator missile... actually lol Eihrul would like Haskerl i like haskell i don't like perl i have no idea what i think about the combination eihrul serious though check it out 11:50am lots of monads i'm reading the page not -too- braindead. Image perl only based on Haskell instead of C and shell script...thats kinda haskerl anyway i need to sh0w3r -:- yoo is now known as yoo-shower -:- yoo-shower is now known as yoo-need-a-shower ol thats good =) oh man yoo, that stinks * Fare/#Tunes goto 61453 12:00pm -:- yoo-need-a-shower is now known as yoo does anybody have the same problem as me with netscape 4.7 on linux? (mine crashes quite often) 12:20pm -:- SignOff Downix: #TUNES (Ping timeout for Downix[d-gnaps-171.ici.net]) -:- Downix [down@d-gnaps-171.ici.net] has joined #tunes stupid 12:30pm use w3m * Fufie/#tunes yawns.. fell asleep on the bed bdb: min as well.. on avg twice a day min(mine cékoidonc w3m? abi: w3m? somebody said w3m was a text-based WWW browser, that can render tables and frames at http://ei5nazha.yz.yamagata-u.ac.jp/~aito/w3m/eng/ mmmkay 12:40pm i've got cl-httpd to run, but it takes 32mb of memory (!) -:- Kaufmann [Kaufmann@dial759.infolink.com.br] has joined #tunes MUAHAHAHAHAHAHAHAHAHA I just mercilessly crushed some pointer-hugging C++-lover. (apache takes 3mb) DAMN, it feels good to be right kaufmann: i've got cl-httpd to run, but it takes 32mb of memory (!) impressive -:- SignOff hcf: #TUNES (Read error to hcf[me-portland-us121.javanet.com]: Connection reset by peer) -:- hcf [nef@me-portland-us121.javanet.com] has joined #tunes indeed, an impressive amount of memory lost I dare you to run CL-HTTPD /and/ EMACS :) _AND_ KDE, with a big beautiful theme kaufmann; i'm always running emacs, which takes about 5mb I guess CL-HTTPD has comparatively little per-request (script) overhead that's to be hoped :) 01:10pm Can you believe this guy tried to push implementational relativism ("there are no good languages, only different tasks") on me? HUAHAHAHAHAHAHA Indeed, it feels good to be right hm it now takes but 16mb after writing it into a lisp image so i guess i'm back to using php3+apache :( what a pity CL is fun * Fare/#Tunes just posted a message on os-ideas@ I don't subscribe to that is someone porting cl-http to EMACS LISP??? In fact, I didn't even know one existed LOL FARE! Where'd you come up with that? Urgh typo 01:20pm what takes the 32MB is not cl-http itself, but the huge CMUCL or ACL compiler and runtime support. Too bad there's no standard way to strip or isolate programs in a CL system. hmm D: yes? * Fare/#Tunes translated his Bastiat page into english abi: Bastiat is a french political economist (1801-1850) abi: Bastiat is also at http://www.tunes.org/~fare/books/Bastiat/ okay, Fare. (maybe soon at bastiat.org) reading the /. on how Microsoft can control the police force at will Down, what? -:- NetSplit: adams.openprojects.net split from devlin.openprojects.net [01:28pm] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [adams.openprojects.net] fare; hm 01:30pm Kaufmann: The NYPD is now M$'s footsoldiers it seems. ah /that/ story I was just about to read it I read it scary and funny at the same time "PD" in french means homosexual man (short for pédéraste) what's the title of the story? what database can easily be used with common lisp? ROTFL New Yorker accidentally gets $1M M% prototype Kaufmann: I've been thinking, what the hell is in the box that costs a million dollars? smoke well you could always use a programmable database. just write lisp code out read it back in Down, a box that costs $999,999.99, plus $0.01 packaging Kaufmann: hehehe "and what the hell is in the box that costs $999,999.99?" I like the comment "The scary part is that someone actually bought a WebTV" smoke: in which lisp does it take 32M? here it takes _much_ less Bill Gate's Cocain + Heroin shipment... lol I think WebTV's suck the next-gen model just as much yah 01:40pm Indeed -:- SignOff AlonzoTG: #TUNES (Ping timeout for AlonzoTG[207-172-184-23.s23.tnt6.lnhva.md.dialup.rcn.com]) Source code is like manure, if you spread it around things grow. If you hoard it it just smells bad. (/. quote) hehehe yup seen that one in some cases thats ok but GNU software tends to suck badl Not all of it Funny thing is, many of these guys have never written a line of code in their lives * Downix/#tunes nods manure alone is no saviour, you still need something that can grow Fufie: Yup kaufie: your /. friends you mean? ;) Fufie, yeah fufie; i used cmucl * Kaufmann/#tunes is a /.-hugger duh i told you slashdot is full of morons! smoke: I use acl and it takes 3-4mb here, which is comparable to apache+php yoo; i meant some sql like database on which i can perform queries fufie; damn this acl :) well there are a few SQL's for acl i've seen just STFW'ing smoke: well.. I use it for my thesis and am quite happy with it.. ;) STFW? yoo; note that i have very little experience with common lisp smoke: do you need db-connection from lisp? CMUCL seems quite bloated. cmucl is somewhat bloated The thing with cmucl is not as much as its bloated, as it seems to be poorly coded and rather hackish. yoo: I think that is a bit unfair fufie; sort of. u all seem to have plenty of opinions on lisps, why not write up some reviews? yoo: cmucl is written by some of the best coders you get smoke: try the sql/odbc library.. ftp://amirani.hit.uib.no/ Fufie: However...its huge, old, and is supported in free time. 01:50pm fufie; i volunteerd to do a forum on linux.scene.org, the main website for the linux demoscene.. I would've done it with php3+postgresql, but since i'm learning common lisp now, perhaps it's better to learn lisp some more yoo: yes, the problem is that it is not maintained anymore as the project lost funding and the team started to go Dylan what the hell is a demoscene? dylan! What a gross mistake, that thing. downix; you know, people who make and watch demos (amiga,c64,pc) dylan was an attempt to reach losers who don't get the () smoke: The only demos I've ever used were for my Amiga * Downix/#tunes needs to figure out a demo for his hardware Awwww, come on, I don't think Dylan is /that/ bad smoke: to be honest here. the easiest is probably to use an existing system in php/perl/python (why do all start with p?) fufie; because they're practical :) Fufie: they like the letter P smoke: if you're new to lisp and start with cmucl+cl-http and make a heavy used site is probably not a good idea (or painful) fufie; oh it won't be painful :) it's all just for fun, no one will mind hmm u all seem to have plenty of opinions on lisps, why not write up some reviews? hcf: I don't know LISP all that well Downix: what do u know well? hcf: I revamped the Common Lisp entry in Languages. hcf: Hardware, specificially the Amiga one very bad thing about dylan is the idea that you could reconcile static modules with a global object system why can't the Tunes HLL be common lisp ? the whole idea just sucks. smoke: global object system Downix: start a hw review hcf: the problem is that evaluating implementations is a minefield like no other hcf: Got one partly done. Reviewing ECS vs AGA fufie: heheheh :) ECS ? AGA ? fufie: i still have one leg left smoke: yeah, but you're new and easily forgiven :) Fare: Two of the Amiga chipsets. There's also the original Lorraine chipset and OCS fare; a global object system is bad? smoke: and you probably never saw Fare vs the comp.lang.lisp dinos a few months back fufie: hm no -:- AlonzoTG [Alonzo@207-172-184-23.s23.tnt6.lnhva.md.dialup.rcn.com] has joined #tunes join #f-cpu abi: haskerl? haskerl is at http://www.people.Virginia.EDU/~sdm7g/LangCrit/Haskerl Fare: In my review, the original Lorraine surpassed the later chipsets in power per mhz R? Lorraine??? Fare: Lorraine == Amiga, it's the prototype's name smoke: Lispers and comp.lang.lisp in particular are a religious bunch. they usually are right about most things though, but it can be stressful Fare: Lorraines chipset was used only in the Amiga 1000 smoke: global object system doesn't scale to a WAN, or to a multiple-people merge-and-split development fare: have you had a look at EuLisp? I think that one could be a good starting point as a more uncommon lisp EuLisp still has this global object system thingie. doesn't it? 02:00pm Fare: The Amiga 500 had a modified version of the Lorraine system, with a new cap to the maximum chipset memory and less expandibility * Fare/#Tunes may be confusing with ISLisp, that is essentially a CL subset not really sure it's much more modular but EuLisp needs to be revised, and it lacks an implementation so there is no incompatibilities to speak of Fare: The Amiga 2000 used the same chipset as the A500 w/ an added chip for the new Zorro 2 slots Fare: And so on down the line it's a Lisp-1 so there is no separation between variable space and function space though, so you get the elegance of Scheme http://www.aiai.ed.ac.uk/~jeff/lisp/eulisp.html Jeff Dalton was part of the working group and he seems like a knowledgable chap Dammit, how much Karma do you need before your default score is 2?!? hmmm, anyone know if Hugs is still being actively maintained? or atleast maintained at all? 02:10pm eulisp? abi: eulisp is http://www.bath.ac.uk/~masjap/EuLisp/eulisp.html and http://www.aiai.ed.ac.uk/~jeff/lisp/eulisp.html and ftp://ftp.maths.bath.ac.uk/pub/eulisp/ hugs is certainly maintained actively ... and it will be made compatible with ghc cool argh how do i destroy shm segments? one sec shmdt then use shmctl with IPC_RMID -:- SignOff Kaufmann: #TUNES (Ping timeout for Kaufmann[dial759.infolink.com.br]) erm Fufie: where'd youtoo go? other way around actually yeah, but i want to have a program that kills them all well just find em all hcf: there is a version on the ftp-site mentioned shmctl em with IPC_RMID and then shmdt hcf: as for eulisp, I think it grinded to a halt because of little funding yoo: ummm... no dude umm... yes dude why would you destroy the id? then unmap the segment? that doesn't flow in my mind dude RMID doesn't destroy the id it just tells the OS to destroy it when all references are gone shmdt physically wastes it (If, of course, there are no references) time to look at the kernel source.... if it doesn't have IPC_RMID when the process exits it'll hang around even if it has no references if (shp->u.shm_nattch <= 0) killseg (id); break; * eihrul/#tunes coughs. from the source code 02:20pm case IPC_RMID: go argue with Linus :P hmm maybe it was a different shmctl? there's only about 3 ah lets see shp->u.shm_perm.mode |= SHM_DEST; you forgot that little thingy =) IPC_RMID will set that *and* detach it if it has no references yes yes so its actually irrelevent in which order you call the two both will work IPC_RMID just tries to remove it automatically just in case its being called in eihrul's case. but generally its a good idea to set IPC_RMID when you create the shm if you don't want it to hang around after the process exits * eihrul/#tunes ponders why sysv shm doesn't have provisions for doing that on creation. because you have shmctl =) * eihrul/#tunes likes mmap () better. yah but you can't share mmap() 02:30pm MAP_SHARED should work for i in $(ipcs -m | cut --bytes=12-21 | tail +4); do ipcrm shm $i; done was what i needed sort of MAP_SHARED doesn't explicitly say the memory will be shared though it just says it will be written back to the file when it is written to memory so that its possible to share it with multiple processes crap with my compiled map renderer i need a new way of finding visible tiles =| i guess a frustrum clip will be adequate for it though yeah that'll be fine since its all aligned nicely later -:- SignOff yoo: #TUNES (Leaving) 02:40pm hello? abi: hello is a multi-threaded operating system written in Standard ML and can be found at http://www.ics.hawaii.edu/~esb/prof/proj/hello/ what, no 'bye?' 03:00pm that is left as an exercise to the reader it is apparently a girl who've written 'hello', but the MS thesis seemed really short (on written pages) ok, so do I need in a VM that would have "domains of allocation"? 03:10pm -:- mibin [mibin@62.11.104.58] has joined #tunes hmm.. yesterday I was really productive and wrote much code.. today I haven't written a single line of code in 12 hours :( I know how that goes today i have tried to make a lispish interpreter I know how that goes i made the parser ... less than 600 bytes of C code 03:20pm that's not a lot 600 bytes is approx 20 lines heh lemme count them exactly 20 lines!! 17, if you dont take in account the #includes -:- zarq [zarq@9dyn31.delft.casema.net] has joined #tunes ej! 03:30pm mibin: I just used my telepathical skills to have a look :p wanna see the code? or .. you have hacked my system? -:- yoo [ultima@user-38lcn46.dialup.mindspring.com] has joined #Tunes -:- SignOff AlonzoTG: #TUNES (Have Nice Day :)) i even have some passwordless accounts in this box -:- AlonzoTG [Alonzo@216-164-129-68.s322.tnt1.lnhva.md.dialup.rcn.com] has joined #tunes dcc it over :) or mail it mibin: mailto: nef@tunes.org hcf: only the parser? 03:40pm -:- SignOff Downix: #TUNES (Ping timeout for Downix[d-gnaps-171.ici.net]) send it all -:- SignOff smoke: #TUNES (Ping timeout for smoke[15dyn67.delft.casema.net]) -:- smoke [smoke@15dyn67.delft.casema.net] has joined #tunes 03:50pm -rw-rw-r-- 1 mibin 1005 1023 Dec 21 00:50 emesch.tar.gz 1023 is the size .. haha heh I once wrote the worlds smallest yet incredibly fast password cracker the first version was 1337 bytes heh, i guess that didnt use a dictionary approach =) om 1337 bytes... why so large? probably due to the copyright notice ahh, i forgot it in my archives archive* well feel free to threat it public domain =) 04:00pm we're good microserfs and will give it to our maker Bill.. muhahah * yoo/#Tunes sighs i hate this trig cra[ crap trig trig trig sin(PI)=? gah -:- SignOff eihrul: #TUNES (Ping timeout for eihrul[usr5-ppp41.lvdi.net]) hard trig crap =) heh 04:10pm !irq:*! any opers awake? what am i missing there msut be an easy way to do triangle bounding? * Fufie/#tunes goes to sleep this gross hack is gross =) om Alonzo? drom søtt Fufie -:- eihrul [lee@usr5-ppp161.lvdi.net] has joined #tunes WUT?????? =\=\=\=\=\=\=\=\=\=\=\=\= huh? 04:20pm hmm, it seems some mp3s i downloaded were downsampled to 22050 and than to 44100 back 04:30pm gross -:- SignOff hcf: #TUNES (Leaving) when you downsaple and upsample again you create a light 'mirror' effect you can observe with spectrum analyzers i am doing a program to attempt to recreate visually the perceptions your ear receives 04:40pm i am using fft for spectral analysis, but the ear behaves differenty from fft with fft you can distinguish a frequency of 10100Hz from 100000Hz as well as 1100Hz from 10000Hz well 100000Hz is above ears capabilities forget this example feels ignored* 04:50pm abi, fft? i think fft is Fast Fourier Transformation abi, spectre? i don't know, mibin -:- smoke is now known as avoooozl abi, fourier? mibin: wish i knew abi: fftw is at http://www.fftw.org/ -- proves that monads are faster than ASM! monads are faster than ASM? abi, monads? monads are a general foo-passing style to express various bar in a smaller (e.g. pure deterministic) language. See CPS, SPS, NPS, EPS or papers by hbaker, Wadler, Rees (W7), etc. heh general foo-passing style <- very explicatory -:- SignOff avoooozl: #TUNES (z) it is smkl: rmm, where's the proof? it is a joke ... but they use monads in their compiler, and it generates as fast code as platform optimized libraries 05:00pm heh. i'm palying with fft now.. the fft part went fine, but ifft produces noise that'd fit Atari Teenage Riot ifft? abi, ifft? inverse mibin: i haven't a clue ahh it seemed to work fine to me abi: ifft is inverse Fast Fourier Transform i used it to make voice pitch changer s/voice/a voice/ but fft has serious problems due to the very low resolution for low frequency heh. 05:10pm the human ear is still better than fft in that i was trying to reproduce recorded sound.. but this.. this is worse than noise it can distinguish 200Hz from 220Hz zarq: are you using fft to make sound compressable? mibin: uhm i'm using fft to try to recognize it ahh but how do i know if i have the correct fft if i can't reproduce the sound? s/fft/coefficients/ can you send me the relevant code lines? you are just fft-ing and ifft-ing back? uh. what relevant code lines? :) i use a fft function from http://www.intersrv.com/~dcross/fft.html ahh, i used fftw i'm told i should use it. yes fftw that is but this is just a temporary hack until it's proven to work zarq: what operating system are you on? fftw was trivial to install in this debian box linux debian potato same here i know, i already have fftw installed (apt-get install fftw-dev) 05:20pm ow duh. i found why it made noise. forgot to add 128 (DUH) :) ooookaay... we're getting somewhere =) -:- SignOff AlonzoTG: #TUNES (Have Nice Day :)) 05:30pm anybody using guile here? anybody using lisp as their default shell? i'd love to but debian's guile seems to lack the completion-stuff 05:40pm -:- SignOff yoo: #TUNES (Read error to yoo[user-38lcn46.dialup.mindspring.com]: Connection reset by peer) ha i have something zarq? mibin: i can now reproduce the input of the fft program ah, kool but it plays way too fast hmm, it plays faster with the same pitch? or with higher pitch? no i think it spits out everything it has, and gets tons of buffer overruns dropping 80% of it ahh in fftw you have to divide the result by the number of samples or something the sample is about 20 seconds, the ifft program finishes in 3 secs hmm, how is it possible? arent the input & output buffers of the same size? i don't know.. i just write to /dev/dsp, maybe that's not trttd trtd? the right thing to do.. i thought everybody knew that abbreviation :) heh, not me 06:00pm i just know imho, afaik, lol and few others anyways /dev/dsp is 8 bit unsigned 8000Hz by default i know, but i ioctl'd it to 12.8 kHz 11.050 you mean? no, 12.8 12800 hmm, strange anyways the playing time depends only on the number of samples yes, but i'm sure you have to wait for the sound card to be ready to accept more data yea, and the sound card is supposed to accept 12800 samples/sec either in recording or playback hm maybe it matters that i'm writing 256 bytes of data at a time? even O_SYNC doesn't help much hmm, i think no O_SYNC is evil =) why are you using asynchronous transfers? ahh, no i mistake O_SYNC for O_NDELAY 06:10pm synchronous hm. i have some success with ioctl(SNDCTL_DSP_SYNC) heh, i dont mess with ioctl's i just pipe the ouput or the input to /from esdrec and esdcat ioctl(SNDCTL_DSP_SYNC) waits for the sound to be actually sent to the soundcard it seems that may lead to 'jumps' in playback it does =P ah well. more of this tomorrow. thanks, mibin. well, time to disconnect is approaching, for me np * zarq/#tunes is away: zzz never sleep with something unfinished !!! * mibin/#tunes will b soon ehm if i don't sleep now, i'll keep working on this 'till 8 o'clock tomorrow morning that's exactly what i don't want :) eheh ok it is said that night brings suggestion, anyways that is absolutely true. though a night's sleep over an issue may solve many things too. 06:20pm a much longer can solve any problem =) a much longer sleep that is heh yes, but the problem with such long sleeps is, they take so long :) by the time you woke up, you've forgotten what it was you were supposed to have slept on zarq: hmm reincarnation? i hope not :) i meant the final sleep it really solves any problems to you, but brings them to others =) ehe. 9 time to disconnect is coming for me 5 nite all but, alas, the time has come for me to look for something they call `bed' 0 later. or deb -ian ? hm no ok nighty nite i'm using that now, and i don't feel much rested :) haha -:- SignOff mibin: #TUNES ([x]chat) 06:30pm -:- yoo [ultima@user-38lcn6g.dialup.mindspring.com] has joined #tunes abi: seen core core was last seen on IRC 1 days, 10 hours, 6 minutes and 56 seconds ago, saying: sort of :) [Sun Dec 19 08:56:51 1999] * yoo/#tunes pokes eihrul in the arm * eihrul/#tunes pokes yoo in the nads. Owie! You bastard! Whatcha looking for core re? nada mucho How's your mud coming along? * yoo/#tunes is trying to decide whether to refine his mapping system or start coding other things its umm, coming hmm 07:10pm So it's not? 07:20pm -:- water [water@tnt-10-36.tscnet.net] has joined #tunes foo FOO?!?!?!?! HOW DARE YOU! heh i hoped that i'd get a rise out of someone :) 08:00pm hmm. there's a new os written in standard ml hello? rumour has it hello is a multi-threaded operating system written in Standard ML and can be found at http://www.ics.hawaii.edu/~esb/prof/proj/hello/ heh yep abi is slow today hmm. that lisp guy on the ship wants to talk though you said there were no people capable of being lisp men on the ship? that changed a few weeks ago there's a new guy on board -:- fire [no@209-68-229-32.dialup.cust.tfb.com] has joined #tunes 08:10pm and get this: he's also into programming language development :D hi water what luck for him to run into a tunes member wow since he hadn't even heard of tunes yet water: what do weakly typed languages like squeck do for type checking or what takes it's place? i told him about our review pages, and our project, and he thought it sounded pretty cool, though of course difficult fire: run-time fire: or you insert type-checks into method code explicitly (run-time = meta-object protocol) so like what if you tried to add two strings in squeck? you'd type 'aString add: anotherString', and the add method for string would check the argument object to see if it was a subclass of string _as the method is called_. (there's also the a+b syntax, which is handled the same way) oh i thought squeak didn't have classes no, that's self ok then and self can type-check too what would self do? in the same scenario? you just replace "isSubclassOf: string" with "isCloneOf: string" er.. stringTraits that's the idea, anyway i see -:- hcf [nef@me-portland-us700.javanet.com] has joined #tunes but languages like that require the basic classes' code to handle such things properly water: though... do you really need that sort of information hanging around internally on the object itself? can't it just be a programmer defined slot? but then, that's what oo has been about for the last few decades eih: i thought it was the same thing well, from what i read of self it sounded like they still had internal state, even though accessable well, the object itself doesn't have the info... it forwards the check to it's parent er rather, i should say implementation artifact state :) yes... but it has to know who its parents are sure that's where vm support of the language helps but, IIRC... that was managed internally even though accessably sure whereas it could just be a removeable slot itself 08:20pm why would you want to manually remove it, though? hmm what about for representing certain small objects? eih: have you read the self papers ? not thoroughly yet i.e. "power of simplicity"? i have it though you should probably chew on that one. it's pretty good well, i did start chewing but sleep intervened heh yeah, it takes time to learn this stuff heh i need to learn how to write scientific papers fire: you and me, both :) um you know more than i water: the paper itself wasn't complicated, its just i've been doing lots of things this week eih: i understand i need to write a lengua paper fire: just describe it in terms of lisp :) :P bah add a little EBNF can you overload objects in self? you mean overload operators? yes definitely overloading seems a little dirty water: though the paper has gotten me to start thinking about certain things of what i've read it... i want to be able to put it all into one definition so? the point of smalltalk and self is simplicity of user-understanding eih: like what? water: like perhaps how to optimize a dynamic type system into a static one if enough information is present.... or how to gather that information yeah, that's an interesting problem water: and more importantly about how to let this dynamic type system be not inherently part of the primitives its built upon... but just something implemented with primitives my goal is the same fire: try coding in squeak a bit to unify as many concepts as possible to achieve near perfect simplicity hmm ok fire: especially look at the code browser -:- SignOff yoo: #TUNES (Leaving) water: though what i've thought of so far is it boils down in some respects to reducing expressions known be invariable water: so you don't have to waste space or execution time calculating them, yielding static information know to be invariable? constant :) oh, persistently storing expression results 08:30pm like caching that or having the compiler determine at compile time either way could work... hm it's more like two views of the same problem you're either going to modify a static system to make it work with dynamic constraints, or you take a completely dynamic system and try to save run-time overhead by predicting where results will be repeated well, if the compiler can look at all code in a dynamic system.... it can determine whether some properties of an object are mutable or not yeah, but that takes time, too :) sure, you wouldn't use it all the time the question is: how much? but for final, production code... perhaps though dynamically loaded code would defeat it completely :) so the facilities would more likely have to be built into the runtime system which i suppose is self..... sure, any system where the difference between 3hours and 72 hours of compile-time is irrelevant can afford high-quality code ;) you would not use such a compiler for debug cycles.... (of course, there are situations where that is relevant) right where can i find info on self? abi self? self is a prototype-based object system, at http://self.sunlabs.com/ abi: self links? self links are at http://www.cetus-links.org/oo_self.html ooh water: probly nothing new to u oh yeah, i forgot cetus-links 08:40pm water: hows ur next arrow post coming? sigh... re-formulating model theory ain't simple and leaving something like that as an open-ended mlist discussion doesn't seem productive 08:50pm darn it, i want to have some good news in tunes for a change, but i'm running into walls adios amigos :D -:- SignOff fire: #TUNES (Leaving) 09:00pm brb folks -:- SignOff water: #TUNES (The Tao went that way!) 09:10pm -:- water [water@tnt-9-253.tscnet.net] has joined #tunes wb thx -:- NetSplit: sterling.openprojects.net split from lewis.openprojects.net [09:19pm] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [sterling.openprojects.net] 09:20pm -:- Netjoined: sterling.openprojects.net lewis.openprojects.net -:- Fare [fare@quatramaran.ens.fr] has joined #tunes -:- Fufie [stig@tunnel-44-14.vpn.uib.no] has joined #tunes -:- abi [nef@bespin.dhs.org] has joined #tunes -:- smkl [sami@glubimox.yok.utu.fi] has joined #tunes -:- zarq [zarq@9dyn31.delft.casema.net] has joined #tunes -:- eihrul [lee@usr5-ppp161.lvdi.net] has joined #tunes -:- hcf [nef@me-portland-us700.javanet.com] has joined #tunes -:- water [water@tnt-9-253.tscnet.net] has joined #tunes -:- air [brand@p0wer.qzx.com] has joined #tunes -:- SignOff Fare: #TUNES (Ping timeout for Fare[quatramaran.ens.fr]) -:- Fare [fare@quatramaran.ens.fr] has joined #Tunes -:- SignOff Fare: #TUNES (Ping timeout for Fare[quatramaran.ens.fr]) -:- Netjoined: merril.openprojects.net sterling.openprojects.net -:- zarq [zarq@9dyn31.delft.casema.net] has joined #tunes -:- NetSplit: merril.openprojects.net split from fontana.openprojects.net [10:54pm] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [merril.openprojects.net] -:- Netjoined: merril.openprojects.net fontana.openprojects.net -:- zarq [zarq@9dyn31.delft.casema.net] has joined #tunes oh yeah *this* is worth keeping a windows open :) see y'all tomorrow -:- water [water@tnt-9-253.tscnet.net] has left #tunes [] -:- SignOff zarq: #TUNES (Ping timeout for zarq[9dyn31.delft.casema.net]) 11:00pm -:- zarq [zarq@9dyn31.delft.casema.net] has joined #tunes -:- NetSplit: merril.openprojects.net split from fontana.openprojects.net [11:13pm] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [merril.openprojects.net] -:- Netjoined: merril.openprojects.net fontana.openprojects.net -:- zarq [zarq@9dyn31.delft.casema.net] has joined #tunes -:- SignOff air: #TUNES (Ping timeout for air[p0wer.qzx.com]) -:- SignOff eihrul: #TUNES (Ping timeout for eihrul[usr5-ppp161.lvdi.net]) -:- NetSplit: merril.openprojects.net split from fontana.openprojects.net [11:43pm] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [merril.openprojects.net] -:- Netjoined: merril.openprojects.net fontana.openprojects.net -:- zarq [zarq@9dyn31.delft.casema.net] has joined #tunes -:- NetSplit: merril.openprojects.net split from sterling.openprojects.net [11:56pm] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [merril.openprojects.net] -:- Netjoined: merril.openprojects.net sterling.openprojects.net -:- zarq [zarq@9dyn31.delft.casema.net] has joined #tunes -:- SignOff zarq: #TUNES (Ping timeout for zarq[9dyn31.delft.casema.net]) [msg(TUNES)] newlog 1999.1221 IRC log ended Tue Dec 21 00:00:01 1999