IRC log started Thu Jan 6 00:00:01 2000 [msg(TUNES)] permlog 2000.0106 -:- water [water@tnt-10-226.tscnet.net] has left #tunes [] -:- SignOff hcf: #TUNES (Read error to hcf[me-portland-us230.javanet.com]: Connection reset by peer) -:- water [water@tnt-10-226.tscnet.net] has joined #tunes <water> ping 01:50am -:- SignOff water: #TUNES (The Tao went that a-way!) [openprojects]Foreign OperKill: NickServ killed fusion (This nick is reserved by another user) -:- pipiness [bindy@usl7-248-188.netsecure.net] has joined #tunes -:- water [water@tnt-10-226.tscnet.net] has joined #tunes [openprojects]!carter.openprojects.net Net break: sterling.openprojects.net zsoldos.openprojects.net (Ping timeout) [openprojects]!carter.openprojects.net Net junction: irc.linux.com zsoldos.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from zsoldos.openprojects.net. <water> anyone awake? [openprojects]!carter.openprojects.net zsoldos.openprojects.net acknowledged end of net.burst. <Fufie> yes <water> care to discuss prog langs? <Fufie> sure <water> i was looking over beta, gbeta, smalltalk, self, clos, tunes <water> i forget, what's your particular project? <Fufie> I have several projects, loosely tunes related, but usually language related <water> (also i read the great squeak news from oopsla 99) <water> hm <Fufie> gbeta? <water> its a generalization of beta <Fufie> ah, ok <water> it's a student's phd project <Fufie> (though it was GNU beta or something like that) <water> well, it is GPL <Fufie> good.. because there should be a free implementation of beta <water> well, it's not very fast at all 04:40am <water> the nice thing is that it extends beta's dynamic typing quite a bit <Fufie> gbeta? <water> yes <Fufie> have you got an url? <water> oh yeah <water> abi: gbeta is a free implementation of a generalized version of beta at http://www.daimi.aau.dk/~eernst/gbeta/index.html <water> there's also the metaBETA project <Fufie> have you got an url for that one as well? :) <water> yeah <Fufie> and I'll update Languages.phtml with it right away <water> btw, just use cetus-links for proglangs <water> well, gbeta is already on tunes <Fufie> just an ftp-link <water> and metabeta is at the bottom of the page, i think <Fufie> -n <water> oh <Fufie> no metabeta <water> k <water> i'll mail it to tunes-review then <water> and tell abi <Fufie> ok.. I'll commit the minor modification for gbeta then -:- smoke [smoke@15dyn246.delft.casema.net] has joined #tunes <water> abi: metabeta is a meta-level architecture for allowing dynamic typing in beta, or at http://www.daimi.au.dk/~beta/Mli/mli.html <water> abi: beta is an advanced object-oriented language that integrates objects and functions into patterns, or at http://www.daimi.au.dk/~beta/ <water> it's interesting to me because the smalltalk family can't do the same thing, really, without a complicated mop <water> on the other hand, the clos mop is too low-level and insecure, etc <Fufie> the clos mop is supposed to be low-level I guess <water> hm 04:50am <water> well, this kind of thinking is what got me working on arrows in the first place <water> sort of a very-high-level mop <Fufie> have commited Languages.phtml <water> cool, thanks <water> since i'm running 'doze right now, i don't have all of my cool tools to use <Fufie> arrows is in in my reading-list queue :) <water> anyway, it's interesting that gbeta treats inheritance as a type-constraint for dynamic purposes <water> btw, have you read agt? <water> it's a good read, imo, even though it's mostly the meta-theory of logic <Fufie> agt? <abi> i heard agt was Around Goedel's Theorem at http://www.ltn.lv/~podnieks/ <water> thank you, abi :) <abi> sure thing water <Fufie> it is also on my reading-list :-/ <water> the premise has to do with platonic psychology <water> he claims that the proper view for formal reasoning is not to treat symbols as 'tangible' objects, but to treat formal theories of them as such <water> s/such/tangible objects <water> heh. i bet that i can fill up your reading queue rather quickly ;) <water> anyway, most of the paper walks through various meta-mathematical theories and studies their properties in a new light <Fufie> luckily my reading queue is implemented as a linked list and can grow :-) <water> lol <water> just be careful not to swap any pointers around before the next gc generation <water> :) <Fufie> there is a lot of things I want to read and a lot of things I want to hack together.. too often the last list is chosen.. 05:00am <water> that's fine <water> i feel useless very often because i read too much, and have pretty much brought my coding practice to a standstill <Fufie> you named a long list of interesting languages first which you had checked.. how is self? <water> well, self finally got ported to the mac just a month ago, albeit without half of the usual optimizations <water> so, people can actually use it now <Fufie> are you a mac user? <water> but the vm technology in self allowed 50% speed of optimmized c <water> no <Fufie> 50% of optimised c isn't too bad.. it's more than satisfactory in most cases <water> it's terrific when you consider that it's a benchmark of a dynamic population of bytecode objects <water> the authors wrote several papers just on the compiler technology they put into the vm <Fufie> self is prototype-based, no? <water> but the nicest thing about self is that it's oo without classes, and treats variables (internal state) as inheritable behavior <water> right, and it allows multiple dynamic inheritance <water> which "self: the power of simplicity" explains as being one of the major improvements in the efficiency of self code <Fufie> dynamic inheritance isn't too difficult in a prototype-based language <water> both in size and speed [openprojects]!carter.openprojects.net Net break: irc.linux.com zsoldos.openprojects.net (Ping timeout) <water> right, and the self object model is a very simple object-with-slots view <Fufie> I wrote a prototype-based oo-system for a hypermedia-system once.. worked very well <water> yeah i made one based on a toy language that had it's own high-level gui <water> i mean, i wrote the high-level gui in the language <Fufie> in the hypermedia-system I now writes I map objects to the clos system instead <water> even m$'s com is prototype-based <Fufie> -s <water> hm <water> k 05:10am <water> well, my biggest concern right now is what kind of mop system a system would need to get as expressive as tunes <Fufie> difficult to say.. one must experiment <water> i disagree <water> sort of <Fufie> a trade-off between an aop-model and a dynamic clos-mop is probably what is needed <water> yes, but there are no existing dynamic aop systems <water> i mean, creating aspects is difficult <water> s/creating/factoring <water> it's not something you can do on the fly <Fufie> I know.. I err all the time in my thesis.. <water> i'm pretty sure that tunes will eventuall have a group of mop's which are inconsistent overall [openprojects]!carter.openprojects.net Net junction: sterling.openprojects.net zsoldos.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from zsoldos.openprojects.net. <Fufie> apart from the obvious aspects, memoisation, before/after methods, sewing together constructs with special behaviour it is difficult <Fufie> s/,/;/ <water> m [openprojects]!carter.openprojects.net zsoldos.openprojects.net acknowledged end of net.burst. * water/#tunes nods <water> ever studied model theory? <Fufie> too little <water> well, even the basic ideas are enough <water> you have the model's signature, then the realtions and functions defined over the constants and variables in the sig <Fufie> I am probably more a hacker than a theoretician.. <water> k <water> well, my goal for my system is not "hack"-centric <Fufie> one question though, how is model-theory different from category-theory? <water> er <Fufie> or is model-theory a subset of category theory? <water> they're in roughly the same domain <Fufie> I remember that we worked with models in a course on category theory <water> they apply to almost every formal theory <Fufie> ahh <water> but the internal picture is fairly different <water> model theory is almost entirely logic-based <Fufie> ok <water> ct, as you know, just describes types and morphisms <Fufie> yes 05:20am <Fufie> what was it you wanted to discuss initially? (we got carried away slightly) <water> my problem with model theory is that i'm trying to make something where one(?) model's structure can affect another model's logic (inference calculus) <water> hm <water> not sure exactly <water> i mostly just needed to bounce ideas <water> does that statement i just made make sense? <water> about my problem with mt? <Fufie> yes <water> k <Fufie> irc is however a poor substitute for a blackboard :( <water> indeed <water> if only muswiki were up and running :) <Fufie> chalk and blackboards solve most problems :) <water> actually, there is a good blackboard program <water> muswiki essentially *is* a blackboard -:- water_ [water@tnt-10-226.tscnet.net] has joined #tunes -:- SignOff water: #TUNES (Read error to water[tnt-10-226.tscnet.net]: Connection reset by peer) <water_> whoa <Fufie> but it's not working yet? <water_> muswiki? no, not yet <water_> but i know that there are some simple blackboard programs out there <Fufie> I wrote such a system once where one had a talk-channel and a canvas to draw on * water_/#tunes checks <water_> this would be very good for #tunes -:- water_ is now known as water <Fufie> my small system was a crude prototype or "proof of concept/hackability" :) <water> sure <water> so was mine <Fufie> I realised that I had to make a drawing program to make it usable and that wasn't that fun :-) <water> hehe <Fufie> boxes, text and free-hand and some colours I made though <water> that's often good enough <water> muswiki, i believe, will be excellent for such things <water> mostly because of the morphic system 05:30am * Fufie/#tunes must rewrite his component language for his front-end today.. :( <Fufie> (the hypermedia system/thesis) <water> darn it, my explorer windows are getting onery <water> today? <Fufie> I think one day should be enough <water> here's a screenshot of muswiki: <water> http://www.cc.gatech.edu/projects/squeakers/10.html <Fufie> it is lisp after all <water> oh <water> heh. don't overowrk yourself :) <Fufie> or embedded in lisp to be accurate.. I can use the lisp-reader <water> what do you think of that screenshot? <Fufie> smalltalk? <abi> smalltalk is mostly crap, except for Squeak, that is good. <water> heh <water> yes <Fufie> it looks nice enough <water> the idea is that you send the morph data around, vice images <water> so it's high-level and low-bandwidth <Fufie> in my old prototype I sent diffs <water> when both sides have the same class modules, the encoding is extremely dense, but the big idea they're working on is interactive programming and debugging <Fufie> muswiki seems to scale better though as everything is objects <water> yeah, that's the idea <Fufie> and not pixels <water> and we don't have to write new muswiki code for every class <Fufie> squeak seems to have improved considerably since I tested it two years ago <water> yeah, people add quite a lot every few weeks or so <Fufie> ans the squeakers seem to be quite good at community-building <Fufie> -d+s <water> there's now an entire suite of card games, etc for instance <Fufie> -s+d <water> i found out yesterday that disney has a multimedia 3-d game somewhere built on squeak 05:40am <water> at some interactive display or something <Fufie> too little time and too many fun languages/environments to learn <water> and they seem to intend to do a lot more with it <water> heh <water> squeak can only get better... today they released the modularized code image <water> 13 orthogonal modules <water> keep in mind that i follow squeak very closely <water> btw, have you dealt with maude at all? <Fufie> maude? <abi> it has been said that maude is a reflective rewriting logic language at http://maude.csl.sri.com/ <Fufie> no <Fufie> a former norwegian queen was called maud though ;) <water> it's really interesting for it's reflective capabilities <Fufie> (downloading 'squeak) <water> it turns out that re-writing logic covers concurrency schemes and real-time issues, as well as gc schemes <water> i mean, in a straightforward manner, with clean code <Fufie> any of the maude papers you would recommend to a beginner? <water> the intro/manual, i guess <water> it's equational rewrite, so it takes some getting used to <Fufie> hey.. one of the local guys at the institute works on maude.. <water> oh? cool <Fufie> hmm <water> btw, my system for models under arrow winds up looking like a multi-graph that "folds on itself" <Fufie> as said.. too little time.. too many languages to learn <water> the tunesers know it as my ontology system idea <water> hehe <water> yes true <Fufie> maybe I'll drop by Peter and have him explain Maude to me :) <water> sounds like a very good idea <water> i wish i knew people in reall life who understood even half of this stuff 05:50am <Fufie> most of the theoretician at my institute, like Peter, rarely if ever programs anything but three line examples of hello world <Fufie> I am too pragmatic.. I try to learn the theory to be able to use it to make nicer and more elegant systems -:- SignOff water: #TUNES (Ping timeout for water[tnt-10-226.tscnet.net]) -:- water [water@tnt-10-15.tscnet.net] has joined #tunes <water> i hate m$ windows <water> kde never does crap like that <Fufie> (send water 'sympathy) <Fufie> s/kde/gnome/ :-) <water> whatever <water> i use both <water> i find kde to be more coherent and useful is all <Fufie> there doesn't seem to be much mailing-list traffic on the maude-lists <water> the funny thing is that i wouldn't have to boot into windows unless my modem stopped working when i upgraded my linux distro 06:00am <water> which it obviously did :) <Fufie> ouch <Fufie> I had troubles with pppd on rh6.1 when trying to use rh5.2 setup on it <water> that's funny, i just went from rh5.1 to mandrake 6.5 <water> during which the modem stopped playing <Fufie> does the modem work with e.g minicom? <water> nope <Fufie> my problems were pppd related <water> mine doesn't respond at all, even if i enable pnp in bios <water> and it's us robotics <Fufie> *shrug* I have wireless connection to the net.. T1 capability roughly <water> have you seen the cover of the latest linux journal? <Fufie> no.. <water> wow <water> i'm waiting for my cable company to give me the fateful call in a week or two 06:10am -:- SignOff water: #TUNES (Ping timeout for water[tnt-10-15.tscnet.net]) -:- water [water@tnt-10-15.tscnet.net] has joined #tunes * water/#tunes hums to some rockin' ambient electronica -:- pipiness is now known as squinky <water> heh 06:40am [openprojects]Foreign OperKill: NickServ killed chrome (This nick is reserved by another user) [openprojects]!carter.openprojects.net Net break: sterling.openprojects.net zsoldos.openprojects.net (Ping timeout) [openprojects]!carter.openprojects.net Net junction: irc.linux.com zsoldos.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from zsoldos.openprojects.net. [openprojects]!carter.openprojects.net zsoldos.openprojects.net acknowledged end of net.burst. -:- AlonzoTG [Alonzo@216-164-139-236.s490.tnt5.lnhva.md.dialup.rcn.com] has joined #tunes <water> the only whiteboards i can find are in java ;P 07:30am -:- SignOff water: #TUNES (Ping timeout for water[tnt-10-15.tscnet.net]) -:- thomas [thomas@193.217.63.152] has joined #tunes -:- Fare [fare@quatramaran.ens.fr] has joined #Tunes -:- Kaufmann [Kaufmann@200.224.105.207] has joined #tunes <Kaufmann> Foo! <Kaufmann> Aw damn, Tril's not here <Fare> Gakuk <Kaufmann> right back atcha [openprojects]Foreign OperKill: NickServ killed sigKILL (This nick is reserved by another user) <Kaufmann> Hrrrm <Kaufmann> I don't see why it'll take so much longer for Apple to release OSX 08:20am <Kaufmann> I can't wait to start reverse-engineering Aqua <Fufie> isn't it sad that so much energy and money go into an attempt to make Macs prettier than Windows? <Fufie> is there any new functionality in this new OS X? <Kaufmann> hell yeah <Kaufmann> you obviously haven't been reading /. <Fufie> maybe I should read /. then <Kaufmann> :) <Kaufmann> Well, OSX may Not Quite be Tunes, but it brings together the best of NeXTStep and the best of BSD... *and* it has a gorgeous GUI <Fufie> it looks pretty.. and in a few months other wms would look the same <Kaufmann> ha <Kaufmann> _maybe_ when a "stable" release of Berlin shows up <Fufie> pah 08:30am <Kaufmann> The "pretty GUI" people seem to care only about creating absurdly wild n' crazy eye-candy while keeping the worst of Windows <Kaufmann> "pretty GUI people" =~ E/Gnome/GTK guys <Fufie> mash.. bbl <Kaufmann> heh <Kaufmann> Right now, I'm writing a stack-based command-line shell to run under MacPerl <Kaufmann> Am I nuts or what/ <Kaufmann> ? [openprojects]!carter.openprojects.net services1.openprojects.net adding GLINE for *@*.cas-kit.golden.net, expiring at 947158322: Your domain is temporarily restricted from access. For more information, please email support@openprojects.net. * AlonzoTG/#tunes left clicks on Kaufmann * Kaufmann/#tunes ducks with the speed of... uh, of a speeding locomotive. Wait, that's not right. 08:40am <Kaufmann> BTW, Al, nice rant * zarq/#tunes is back from being away: trug 08:50am [openprojects]Foreign OperKill: NickServ killed KENNY (This nick is reserved by another user) [openprojects]Foreign OperKill: NickServ killed Keeper (GHOST command used by Keeper131) [openprojects]Foreign OperKill: NickServ killed Keeper (This nick is reserved by another user) <Fufie> back -:- hcf [nef@me-portland-us245.javanet.com] has joined #tunes 09:00am <hcf> Kaufmann: devised a new project yet this morning? [openprojects]!carter.openprojects.net Net break: irc.linux.com calvino.openprojects.net (Ping timeout) <AlonzoTG> thanks. <AlonzoTG> =) -:- SignOff Kaufmann: #TUNES (Ping timeout for Kaufmann[200.224.105.207]) [openprojects]!carter.openprojects.net Net junction: sterling.openprojects.net calvino.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from calvino.openprojects.net. [openprojects]!carter.openprojects.net calvino.openprojects.net acknowledged end of net.burst. 09:10am [openprojects]Foreign OperKill: NickServ killed inart (This nick is reserved by another user) [openprojects]!carter.openprojects.net Net junction: irc.linux.com lu.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from lu.openprojects.net. [openprojects]!carter.openprojects.net lu.openprojects.net acknowledged end of net.burst. <hcf> abi: gbeta is a generalization of beta at http://www.daimi.au.dk/~eernst/gbeta/ <abi> ...but gbeta is a free implementation of a generalized version of beta at http://www.daimi.aau.dk/~eernst/gbeta/index.html... [openprojects]Foreign OperKill: NickServ killed Jakes (This nick is reserved by another user) 10:00am -:- SignOff Fufie: #TUNES (Ping timeout for Fufie[tunnel-44-22.vpn.uib.no]) <hcf> abi: tos is Tcl Object System at http://cedric.cnam.fr/personne/pawlak/tos.html -:- Fufie [stig@tunnel-44-22.vpn.uib.no] has joined #tunes <smoke> tos is also the os that drives the atari 1040 st and was written by dec 10:10am <AlonzoTG> om <AlonzoTG> anyone here? [openprojects]Foreign OperKill: is killed woot (go to sleep) <Fare> Kaufie: OSX? [openprojects]!carter.openprojects.net Net break: irc.linux.com zsoldos.openprojects.net (Ping timeout) <AlonzoTG> om <Fare> mu <AlonzoTG> my sister put oil on the brakes of her car. <AlonzoTG> to stop them from squeaking. <AlonzoTG> Fare: what is the official response to my poste late last night? <AlonzoTG> post that is. <Fare> on the list? <AlonzoTG> yes * Fare/#Tunes is seriously lagging <AlonzoTG> okay, no problem. <AlonzoTG> I am eager to read responses. 10:30am [openprojects]!carter.openprojects.net Net break: sterling.openprojects.net clarke.openprojects.net (Ping timeout) [openprojects]!carter.openprojects.net Net junction: sterling.openprojects.net clarke.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from clarke.openprojects.net. [openprojects]!carter.openprojects.net clarke.openprojects.net acknowledged end of net.burst. [openprojects]!carter.openprojects.net Net junction: irc.linux.com zsoldos.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from zsoldos.openprojects.net. [openprojects]Foreign OperKill: is killed woot (sl33p alreayd!) [openprojects]!carter.openprojects.net Net break: irc.linux.com zsoldos.openprojects.net (Ping timeout) [openprojects]!carter.openprojects.net Net junction: irc.linux.com zsoldos.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from zsoldos.openprojects.net. [openprojects]!carter.openprojects.net zsoldos.openprojects.net acknowledged end of net.burst. -:- SignOff squinky: #TUNES (Read error to squinky[usl7-248-188.netsecure.net]: Connection reset by peer) <Fare> ls 11:10am [openprojects]!carter.openprojects.net Net break: sterling.openprojects.net norton.openprojects.net (Read error to norton.openprojects.net[216.172.86.2]: Connection reset by peer) [openprojects]!carter.openprojects.net Net junction: irc.linux.com norton.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from norton.openprojects.net. [openprojects]!carter.openprojects.net norton.openprojects.net acknowledged end of net.burst. [openprojects]Foreign OperKill: NickServ killed gxam (GHOST command used by gxam_) [openprojects]!carter.openprojects.net Net break: irc.linux.com norton.openprojects.net (Ping timeout) [openprojects]!carter.openprojects.net Net junction: sterling.openprojects.net norton.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from norton.openprojects.net. [openprojects]!carter.openprojects.net norton.openprojects.net acknowledged end of net.burst. <hcf> abi: MuSwiki is a Swiki-like system which uses Morphic objects instead of HTML at http://www.cc.gatech.edu/projects/squeakers/10.html <hcf> abi: Swiki is Wiki in Squeak, a CoWeb: web-based collaborative space, based on WikiWiki at http://www.cc.gatech.edu/projects/squeakers/14.html [openprojects]Foreign OperKill: NickServ killed magician (GHOST command used by magician_) -:- SignOff hcf: #TUNES (Leaving) 12:00pm * zarq/#tunes is away: Merlin * AlonzoTG/#tunes hurls a laser guided flaming woodchuck at Fare [openprojects]Foreign OperKill: NickServ killed S (This nick is reserved by another user) -:- washort [washort@d118.narrowgate.net] has joined #tunes [openprojects]!carter.openprojects.net Net break: tolkien.openprojects.net asprin.openprojects.net (Ping timeout) [openprojects]!carter.openprojects.net Net junction: varley.openprojects.net asprin.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from asprin.openprojects.net. [openprojects]!carter.openprojects.net asprin.openprojects.net acknowledged end of net.burst. -:- squinky [gigi@208.246.248.136] has joined #tunes [openprojects]Foreign OperKill: NickServ killed Keeper (This nick is reserved by another user) * AlonzoTG/#tunes left clicks on squinky -:- carlito [sabanmr@137.28.142.82] has joined #tunes <smkl> hmm. i'll apply membership for TUNES [openprojects]Foreign OperKill: NickServ killed bignose (This nick is reserved by another user) 01:20pm [openprojects]!carter.openprojects.net Net break: sterling.openprojects.net norton.openprojects.net (Ping timeout) <Fufie> :) <smkl> i want to get smkl@tunes.org email address [openprojects]!carter.openprojects.net Net junction: irc.linux.com norton.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from norton.openprojects.net. [openprojects]!carter.openprojects.net norton.openprojects.net acknowledged end of net.burst. 01:30pm [openprojects]Foreign OperKill: NickServ killed rios (GHOST command used by rios_) [openprojects]Foreign OperKill: NickServ killed \ (This nick is reserved by another user) [openprojects]Foreign OperKill: NickServ killed BlueDraco (This nick is reserved by another user) <Fare> ATG: you got your answer <Fare> smkl: what, not member yet? 02:00pm [openprojects]Foreign OperKill: NickServ killed bd (GHOST command used by bdraco) -:- eihrul [lee@usr5-ppp177.lvdi.net] has joined #tunes <smoke> does the tunes project need an Eyecandy Division? :) <washort> smoke: i doubt it. :) <smoke> damn :) -:- SignOff smoke: #TUNES (brb) 02:50pm [openprojects]Foreign OperKill: NickServ killed Keeper (This nick is reserved by another user) [openprojects]Foreign OperKill: NickServ killed Burtbot (This nick is reserved by another user) [openprojects]!carter.openprojects.net Net break: devlin.openprojects.net lackey.openprojects.net (Ping timeout) -:- NetSplit: lackey.openprojects.net split from devlin.openprojects.net [02:53pm] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [lackey.openprojects.net] [openprojects]!carter.openprojects.net Net junction: sterling.openprojects.net lackey.openprojects.net -:- Netjoined: lackey.openprojects.net devlin.openprojects.net -:- AlonzoTG [Alonzo@216-164-139-236.s490.tnt5.lnhva.md.dialup.rcn.com] has joined #tunes -:- squinky [gigi@208.246.248.136] has joined #tunes [openprojects]!carter.openprojects.net Completed net.burst from lackey.openprojects.net. [openprojects]!carter.openprojects.net lackey.openprojects.net acknowledged end of net.burst. -:- smoke [smoke@15dyn246.delft.casema.net] has joined #tunes -:- hcf [nef@me-portland-us1013.javanet.com] has joined #tunes -:- SignOff carlito: #TUNES (Leaving) -:- SignOff abi: #TUNES (dying by hcf's request) -:- abi [nef@bespin.dhs.org] has joined #tunes <AlonzoTG> my shell crashed. =\ -:- SignOff AlonzoTG: #TUNES (no shell no party.) -:- _BC [dontcha@ptai01m01-72.bctel.ca] has joined #TUNES 03:30pm <_BC> Hi all -:- nepenthe [DIY@ppp020.wi.centurytel.net] has joined #tunes -:- nepenthe is now known as _ruiner_ <smoke> have we all signed the linux hardware petition at http://www.libranet.com/petition.html ? :) -:- AlonzoTG [Alonzo@208-58-192-32.s32.tnt9.lnhva.md.dialup.rcn.com] has joined #tunes -:- SignOff washort: #TUNES (thank you, may i have another?) <Fufie> smoke: what is it about? -:- SignOff _ruiner_: #TUNES (Ping timeout for _ruiner_[ppp020.wi.centurytel.net]) 03:40pm <smoke> fufie; about more hardware support for linux <smoke> fufie; it's to be sent to hardware companies; they hope to get 2,000,000 electronic signatures, according to freshmeat.net <Fufie> that's nice <Fufie> hereby signed [openprojects]Foreign OperKill: NickServ killed technoir (This nick is reserved by another user) <smoke> what number were you? <Fufie> Your signature is number 00013977 * smoke/#tunes forgot his number. somewhere around 13,000 <smoke> aha <smoke> i think they'll have a hard time getting to 2,000,000 03:50pm <Fufie> I can make a script :) <smoke> :) <Fufie> don't use scanf.. use fgets <Fufie> #wrong <smoke> ? <Fufie> it was for another channel <smoke> :) 04:00pm -:- air [brand@p0wer.qzx.com] has joined #tunes -:- _BC [dontcha@ptai01m01-72.bctel.ca] has left #TUNES [] <air> is gnu du part of some other program? <air> there is no du on gnu.org <Fufie> binutils? <Fare> disk-utils? <Fufie> fileutils <Fare> # dpkg -S =du ==> fileutils: /usr/bin/du <air> there is no diskutils <Fufie> [stig@palomba Review]$ rpm -qf which du <Fufie> fileutils-3.16-10 <air> thanks fare <Fare> yes, dpkg -S sucks as far as speed matters <Fare> the dpkg people seem not quite able to handle the concept of "database". 04:30pm <Fare> happily, the automated upgrade tools are unequalled by anything I saw on the rpm side. <AlonzoTG> =\ -:- timfanss [user4985@Clermont-Fd-1-167.club-internet.fr] has joined #tunes <timfanss> i look for tux pitcs <air> wow, fileutils has the biggest configure i have seen <Fare> meant to be very portable <smoke> air; tried glibc ? :) <Fare> albeit using lots of system-specific interfaces to access disk&files <Fare> (stat, ouch) <air> smoke: i have glibc 2.2 <smoke> air; that configure is shorter? * smoke/#tunes is always scared when compiling glibc <air> smoke: i didnt compile it <Fare> ATG: you have your answer * AlonzoTG/#tunes hurls a laser guided flaming woodchuck at Fare <smoke> air: ah :) <AlonzoTG> that's for suggesting I read more books. <AlonzoTG> =\ * smoke/#tunes is off for some sleep -:- SignOff smoke: #TUNES (Z!) <AlonzoTG> They probably cost $400 and are out of print to b00t. <AlonzoTG> =\ <Fare> ATG: they're probably in their nth reprint, and cost a lot, but they are among books that are well worth their price. <Fare> not the kind of book you use as toilet paper after you've read ten pages of it. 04:40pm <AlonzoTG> yeah. I have a growing collection of those. =(((( <AlonzoTG> Okay I guess I mite as well get both and a stack of others on topology after my current stack is on my shelf or flushed. <AlonzoTG> =\ -:- SignOff timfanss: #TUNES (Leaving) -:- Tril [dem@bespin.dhs.org] has joined #TUNES -:- mode/#tunes [+o Tril] by ChanServ <Tril> smkl <hcf> hmm <Fare> Tril! <hcf> http://lists.tunes.org/cgi-bin/wilma/tunes seemingly mishandles y2k <Tril> oh, you just added yourself. I was hoping someone volunteered to maintain that page :) <hcf> jan 2000 should be at the top, not the bottom [openprojects]Foreign OperKill: NickServ killed BigNate (This nick is reserved by another user) <smkl> Tril: did it update correctly? * Fare/#Tunes metaprograms a silly script to mirror wanted parts of debian <Tril> smkl: it looks ok to me... <Tril> Fare : have you seen the ftpmirror scripts they provide <Fare> Tril: nope. Where are they? <Tril> Fare: basically just one rsync command with various --exclude added <Fare> :( * Fare/#Tunes is building a huge exclude regex for lftp mirror <Tril> Fare: look in ~debian on bespin, there are two slightly different versions "anonftpsync" and "ftpsync*" 04:50pm <Tril> Fare: rght now bespin isnt mirroring debian but I'll start again soon. I actually use "~debian/mydsync" from cron when I mirror it <Fare> Tril: thanks <Fare> s/myd/ftp/ ? <Tril> no * Fare/#Tunes forgot to edit /etc/mail/ for smkl <Fare> done <Fare> smkl: you're now officially smkl@tunes.org [openprojects]!carter.openprojects.net Net break: irc.linux.com lu.openprojects.net (Ping timeout) <Fufie> smkl: congrats :) [openprojects]!carter.openprojects.net Net junction: sterling.openprojects.net lu.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from lu.openprojects.net. <smkl> hehe * hcf/#tunes is away: (afk) <AlonzoTG> om [openprojects]!carter.openprojects.net Net break: sterling.openprojects.net lu.openprojects.net (Ping timeout) 05:00pm [openprojects]!carter.openprojects.net Net junction: irc.linux.com lu.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from lu.openprojects.net. [openprojects]!carter.openprojects.net Net break: irc.linux.com lu.openprojects.net (Ping timeout) [openprojects]!carter.openprojects.net Net junction: sterling.openprojects.net lu.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from lu.openprojects.net. [openprojects]!carter.openprojects.net lu.openprojects.net acknowledged end of net.burst. [openprojects]Foreign OperKill: NickServ killed Mongoose (This nick is reserved by another user) [openprojects]!carter.openprojects.net Net break: sterling.openprojects.net lu.openprojects.net (Ping timeout) [openprojects]!carter.openprojects.net Net junction: irc.linux.com lu.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from lu.openprojects.net. [openprojects]!carter.openprojects.net Net break: irc.linux.com lu.openprojects.net (Ping timeout) -:- iepos [iepos@d31.k1-2.tecinfo.com] has joined #TUNES <iepos> hmmm... anyone here? [openprojects]!carter.openprojects.net Net junction: sterling.openprojects.net lu.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from lu.openprojects.net. <Fufie> yes <iepos> oh <iepos> hello <iepos> hmm, do you code C? [openprojects]!carter.openprojects.net Net break: sterling.openprojects.net lu.openprojects.net (Ping timeout) [openprojects]!carter.openprojects.net Net junction: irc.linux.com lu.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from lu.openprojects.net. <Fufie> when forced <iepos> heh heh <iepos> well... i have a little question then <iepos> that you might be able to answer <iepos> what is a good way (in UNIX) to get the contents of a directory... <iepos> an array of filenames or something like it... ? 06:00pm <Fufie> use the DIRENT struct with opendir() and family * Fare/#Tunes is back for a moment <iepos> ah <iepos> thanks <Fufie> I think I have some convenient code for that here <Fufie> two sec <Fare> iepos: the convenient way is to use perl, or zsh, or scsh, or some other language than C <iepos> heh heh <iepos> yes <iepos> i'm not too familiar with those <Fufie> can't find it right now.. on another system -:- SignOff Tril: #TUNES (Tril has no reason) <iepos> that's okay <iepos> i think i can figure it out from the manpages <Fufie> there is an example in the gnu libc manual I think [openprojects]!carter.openprojects.net Net break: irc.linux.com lu.openprojects.net (Ping timeout) <iepos> fare: i've been thinking about nondeterministic systems a little bit <iepos> i think they could be more interesting than i'd thought ... ( :-) ) <Fare> iepos: think logic programming, too -:- carlito [sabanmr@137.28.139.244] has joined #tunes [openprojects]!carter.openprojects.net Net junction: sterling.openprojects.net lu.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from lu.openprojects.net. <iepos> a bit ago I realized that using an inverter it is possible to derive an equality function ... <iepos> (with no other primitives than combinators) <carlito> *is lost <carlito> oops <iepos> hi <carlito> hi <eihrul> Fare: do you know where i can find any information on lisp garbage collectors written in lisp? <Fare> an inverter is a *very* powerful primitive <iepos> yes <Fare> eihrul: in /dev/null ? <Fare> eihrul: no, I exaggerate <eihrul> don't you always? <iepos> also, one could make a sort of "definite article" function that, given a single-membered set, yields its element <Fare> eihrul: in CMUCL sources and internal docs, if you can figure out <Fare> CMUCL is written mostly in LISP. <eihrul> thanks <Fare> eih: what kind of information are you looking at? <Fufie> eih: harlequin's gc heap? <Fare> s/at/for/ <Fufie> they have lots of docs <Fare> eihrul: Languages.html#GC might be a good starting point -:- SignOff squinky: #TUNES (Ping timeout for squinky[208.246.248.136]) <Fare> also see the gclist 06:10pm -:- SignOff zarq: #TUNES (Ping timeout for zarq[9dyn156.delft.casema.net]) <eihrul> Fare: just playing around with making a small semi-reflective lisp... <iepos> humm... <iepos> lisp... [openprojects]!carter.openprojects.net Net break: sterling.openprojects.net lu.openprojects.net (Ping timeout) <Fare> eih: ok [openprojects]!carter.openprojects.net Net junction: irc.linux.com lu.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from lu.openprojects.net. <iepos> fare: do you know of any interesting inference rules/axioms for systems based on an inverter ? <Fare> I haven't thought about it much. <iepos> me neither <iepos> it makes me sleepy, when i try thinking about it <iepos> :-) <Fare> it really looks like logic programming, tho <Fare> after all, inverting relations is precisely what matching is about in logic programs <iepos> by "logic programming", do you mean programming based on relations? <iepos> oops <iepos> yes <iepos> i guess <Fare> (except for the single-mindedness of the search strategy) [openprojects]!carter.openprojects.net lu.openprojects.net acknowledged end of net.burst. <iepos> hm <Fare> by logic programming, I mean prolog, mercury, oz, goedel, etc <iepos> yeah * Fufie/#tunes goes to bed.. afk <iepos> good nite <carlito> night fufie <iepos> fare: earlier you mentioned the problem of having an inverter when there are functions not one-to-one ... <iepos> i'm thinking of making a system, in that case, that has possibly multiple denotations for expressions <iepos> non-determinism, i guess it might be called then <iepos> a sentence would then be acceptable iff all of its denotations were true <iepos> (in a system based on combinators, "x.y.x" and "x.y.y" make nice models for truth and falsity, i think) 06:20pm <iepos> hmmm... <iepos> i'll have to think about it some more, i guess... <Fare> if Rf(x,y) is the relation that holds wwhen y=f(x), then, when y is bound, the query Rf(x,y) will invert f and find a matching x. <iepos> hmmmm <iepos> i remember in your non-determinism paper (i think it was), there was an interesting problem... that normal evaluation (of lambda expressions) leads to bad problems <iepos> on a similar line, a system with non-determinism can't have "X = X" as an axiom schema or it would be inconsistent... <Fare> it's not a "problem". It's just a restatement of the fundamental turing theorem <Fare> X = X CAN be an axiom schema, anywhere. <iepos> well... -:- zarq [zarq@10dyn174.delft.casema.net] has joined #tunes <iepos> if your non-deterministic system has "1 U 2 = 1 U 2" ... <Fare> Any powerful enough system, however, will be incomplete or equivalently will have non-terminating sentences. <iepos> then you can get "1 = 2" <Fare> iepos: stop smoking that carpet <iepos> (where "U" is that ambiguity operator) <iepos> ? <iepos> i am not smoking carpet <iepos> are you saying that it is fine to have "X = X", then ? <Fare> is your = working on *values* or on *expressions* ? <iepos> expressions <iepos> i meant you can't have "X = X" for all expressions "X" <iepos> but you can if "X" is known to have only one denotation <Fare> ok, then, if == is on values, you have (E = F) = ( \->E == \->F ) <iepos> hmm, what do you mean by "\->" ? <Fare> if you don't get confused between expressions and values, there's no contradiction <Fare> iepos: lambda w/o argument <iepos> ok <iepos> right 06:30pm <iepos> but "X = X" is a bad axiom schema (if "X" is unrestrained)... <iepos> however... <iepos> "(\x. x=x) Y" is okay, for unrestrained "Y" <iepos> i think [openprojects]!carter.openprojects.net Net break: irc.linux.com lu.openprojects.net (Ping timeout) <iepos> the beta-reduction is a good thing to restrain... if a lambda refers to its parameter more than one time <iepos> the parameter must be insured to only have one denotation, in that case... <Fare> no. X == X is bad. X = X is good. <Fare> let's use =e and =v instead [openprojects]!carter.openprojects.net Net junction: sterling.openprojects.net lu.openprojects.net <iepos> ok [openprojects]!carter.openprojects.net Completed net.burst from lu.openprojects.net. <Fare> iepos: if you're still there, it's another proof my article wasn't so well written <iepos> i'm still here <iepos> i feel lost <iepos> my mind is foggy this evening... <iepos> heh <iepos> well... it seems clear that "(1 U 2) = (1 U 2)" is bad, at least, don't you agree? <iepos> meaning that it would make a bad theorem of a system <iepos> since it would probably lead to "1 = 2" [openprojects]!carter.openprojects.net Net break: sterling.openprojects.net lu.openprojects.net (Ping timeout) <Fare> (1 U 2) =v (1 U 2) is bad. (1 U 2) =e (1 U 2) is good. <iepos> yes <Fare> although (1 U 2) =v (1 U 2) *might* be true, sometimes. 06:40pm <Fare> so that using may-testing, it appears as "(may be) true" <iepos> yes, it is an ambiguous sentence, then <iepos> I was thinking of a system that only accepted sentences that were true in all their denotations, but maybe not... <iepos> That way there could be an unrestrained "f (x U y)" |- "f x", "f y" <Fare> in my RlambdaND (and master's thesis), a sentence is "true" iff it _may_ terminate. <iepos> (btw, I think "U" can be constructed from an inverter) <iepos> that is interesting... <iepos> a broad definition <iepos> it would include something like "\x.x", wouldn't it? [openprojects]!carter.openprojects.net Net junction: irc.linux.com lu.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from lu.openprojects.net. <iepos> which is an expression with a normal form (that is what you mean by terminate, right?) <Fare> although the canonical truth value (true) will not only terminate, but will also yield *every* possible result as part of its value set. <Fare> yup, any expression that has a normal form couts as "true". <iepos> wait <Fare> expressions that don't have any normal form count as "false" <iepos> hmm <Fare> so that both #f and #t would count as "true". <iepos> but, how can you ever be sure that an expression doesn't have a normal form <iepos> ? <iepos> (that's a tough problem) <iepos> i guess that is the point <iepos> hmm <Fare> although the first is a true _symbol_ or falsehood, and the latter is a true _symbol_ or truth <Fare> 1) u're only interested in things that are true <Fare> 2) among these things, the sentences that say that another sentence is false <iepos> oh <iepos> ah <iepos> that would work... if you just have a negation <Fare> like, you're not really interested in the fact that phi is false; but the fact that (not phi) is true is interesting. <iepos> but, not sure how to construct a negation in such a system though <Fare> of course, you have sentences such that both phi and (not phi) are false. <iepos> a negation that given an abnormal expression yields a normal one and vice versa <Fare> (goedel sentences) <Fare> iepos: you should read my master's thesis :) [openprojects]!carter.openprojects.net Net break: irc.linux.com lu.openprojects.net (Ping timeout) <iepos> oh, the one in french? <Fare> yup <iepos> heh <iepos> hm <iepos> not a real fast french reader <carlito> iepos: I know that one <iepos> carlito, which one? <carlito> if p, then q. and if p is false, it is generally agreed that q could be anything <carlito> not the same thing that you were talking about .... <carlito> but I thought i'd throw my 2 cents in <carlito> =) <iepos> heh heh <iepos> thanks <carlito> welcome <Fare> "P -> false" is a possible definition for "not P" <iepos> yes, i guess if "p" is false, then you can derive that "if p, then q", so the sentence doesn't give you anymore info on q <iepos> yes <iepos> it is <carlito> k <iepos> if you like "->" as a primitive <carlito> yeah. <carlito> that schooling did pay off for something <Fare> then you ask, "how to define ->" ? <iepos> if you take "\xy.x" and "\xy.y" for truth and falsity... 06:50pm <iepos> then "->" can be derived <Fare> no, you can't, at least, not while taking truth as termination <iepos> that's right <Fare> however, you can take them as _symbols_ of truth and of falsity <iepos> truth as termination is something i haven't really thought of <Fare> only some times, you won't be able to establish falsity <carlito> you two go on, I'll be doing some html for a little while <Fare> iepos: it's something I've not seen anyone publish about (before me). But I'm sure some people have thought about it * carlito/#tunes completely lost now <Fare> it's latent in so many works in logic <iepos> hmm, well truth-as-termination sounds does seem interesting... <iepos> although i still like "\xy.x" and "\xy.y" :-| <iepos> i need to try to come up with a concrete system <iepos> are there any particular interesting things you hope to gain from using truth-as-termination ? <Fare> again, at the *meta* level, truth of the *base* level may be represented as \xy.x and falsehood at the *base* level may be represented as \xy.y <Fare> that's what "symbols" are about! <iepos> yes <iepos> but... <iepos> that <iepos> doesn't <iepos> answer my question [openprojects]!carter.openprojects.net Net junction: sterling.openprojects.net lu.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from lu.openprojects.net. <Fare> truth-as-termination was the natural way I came up with to embed logic in a call-by-value system -:- ult [pope@user-38lc6ar.dialup.mindspring.com] has joined #tunes <iepos> ah <iepos> the natural way <iepos> well <ult> blegh <iepos> that is a very good reason <iepos> very similar to the reason i like "\xy.x" and "\xy.y". <Fare> s/embed $1 in $2/extend $2 with $1/ <Fare> iepos: \xy.x and \xy.y are natural ways to express *symbols* of truth and falsehood <iepos> yes, but they could be thought of as the truth and falsehood values themselves <Fare> only if you confuse the symbols with the very things <Fare> meta-level confusion is EVIL. <Fare> it's the kind of stuff OO crooks do. <iepos> i am not confusing symbols with their meanings... <iepos> (at least i don't think i am ) <iepos> heh <Fare> yup, and don't introduce that confusion in your calculus. 07:00pm <ult> darn OO crooks -:- SignOff hcf: #TUNES (Ping timeout for hcf[me-portland-us1013.javanet.com]) <iepos> well... i suppose i'll just defend myself by saying that "\xy.x" and "\xy.y" (which hereafter will be called K and CK) aren't necessarily truth and falsity (but i'll use them as if they are, anyway) [openprojects]!carter.openprojects.net Net break: sterling.openprojects.net lu.openprojects.net (Ping timeout) <iepos> "truth" and "falsity" in the ordinary senses are rather vague concepts <iepos> i don't really have to formally identity my "K" with "truth" (in the vague sense) to make a useful system <iepos> s/identity/identify <Fare> indeed they only arethe generic way to encode two alternatives <iepos> in fact, it might not be necessary to supply a formal interpretation to expressions at all in a system <Fare> in a reflective system, it might help, tho <iepos> yes, and such elegant derivations of other logical operators arise from them... [openprojects]!carter.openprojects.net Net junction: irc.linux.com lu.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from lu.openprojects.net. <iepos> for instance, "xyK" means "x -> y" <iepos> "xKy" means "x | y" <ult> Fare: Has anything ever rivaled the Lisp machines of olden times? <iepos> "xy(CK)" means "x & y" <Fare> I don't follow your latter derivations... <iepos> hmm <Fare> ult: Lisp machines of newer times? <iepos> the first one... "xyK" ... [openprojects]!carter.openprojects.net lu.openprojects.net acknowledged end of net.burst. <ult> Fare: olden times <Fare> ult: in newen times, they were faster <iepos> If "x" is "K" (true), then the expression reduces to "y", which is what would be expected for implication <ult> Back when my grandpa walked 10 miles both ways uphill in the blaznig heat and freezing cold -at the same time-. <Fare> ult: but the community was already dead <iepos> on the other hand, if "x" is "CK" (false), then expression reduces to "K" (the expression is true, which is what would be expected if an implication from a false proposition) <iepos> KyK = y <iepos> CKyK = K * Fare/#Tunes is lost <iepos> 8( <iepos> ok <iepos> i'll try again <Fare> oh, you're talking about boolean calculus <carlito> lol <iepos> hmmm <Fare> but I don't want just booleans. I want logic. <iepos> well <Fare> higher-order propositions. <iepos> you want logic... <iepos> what do you mean, "higher-order propositions"? <iepos> give <iepos> an example of something <iepos> you want the system to do? <iepos> . <Fare> in intuitionnistic logic, not everything can be reduced to booleans <iepos> indeed <iepos> and "xyK" might not necessary reduce to K or CK <Fare> so your nice boolean expressions are of no help <iepos> if x or y are non-propositions then it may be something entirely <iepos> yes i suppose <Fare> well, at the base level, anyway <iepos> my system couldn't reason upon "propositions" that were neither true nor false <iepos> maybe it could though... i don't even have a system <iepos> though <iepos> so it's kind of silly of me to refer to "my system" 07:10pm * Fare/#Tunes learns the joys of CD ripping <iepos> heh <ult> Go Fare! <iepos> anyway <ult> Join the masses! <iepos> i'm getting sleepy... ... <iepos> sometime i'm going to construct a concrete system, and stop babbling <iepos> heh he <carlito> good luck <iepos> heh <iepos> well <iepos> not really concrete <iepos> i'm not going to implement <iepos> why implement <iepos> no need <iepos> well, nice talking to you Fare and carlito... <carlito> see ya <AlonzoTG> om <iepos> bwye -:- SignOff iepos: #TUNES (Leaving) [openprojects]Foreign OperKill: NickServ killed will (This nick is reserved by another user) <ult> b;ah <ult> Fare will appreciate this: <ult> Learning French is trivial: the word for horse is cheval, and everything else <ult> follows in the same way. <ult> -- Alan J. Perlis <carlito> that's awesome <air> damnit, i was about to beat this level in nightmare mode and win98 crashed <carlito> what game? <carlito> wolf? <air> q3 <carlito> nevermind 07:20pm <carlito> anyone want to have some fun? <air> ya do u have q3? <carlito> no no no, online here on mIRC <carlito> =) <air> i dont have mirc * ult/#tunes notes air should get his ass kicked by^H^H^H^H^H^H^H^H^H^H^H^H^Hplay hyrlik. <air> i kicked hyrlik's ass <ult> Did you? <air> yes <carlito> what program are you using? <air> carlito: bitchX <carlito> nice! <air> carlito: and irc != mirc * Fare/#Tunes deep-sixes ult <Fare> (with lag) <carlito> thanks <carlito> there's these chicks on another channel that I feel like harrassing <carlito> incase anyone wanted to join -:- ult_ [pope@user-38lcmts.dialup.mindspring.com] has joined #tunes -:- FareWell [fare@quatramaran.ens.fr] has joined #Tunes -:- FareWell [fare@quatramaran.ens.fr] has left #Tunes [] <air> damn i gotta go buy a 20gig or something bigger than 4.3 for my game machine. this sucks waiting forever for each level to load from cd <carlito> lol <carlito> FareWell just joined the channel! -:- SignOff ult: #TUNES (Ping timeout for ult[user-38lc6ar.dialup.mindspring.com]) <Fare> air: level? <carlito> anyone got any pics I could send these chicks!! =) <Fare> what chicks? <ult_> carlito: give Ricky Martin a GIMP moustache. <carlito> nice! <carlito> I'm playing these girls on a diff channel <carlito> want in? <Fare> why not <Fare> carlito: send a picture of linus <carlito> lol!!!!!!!! <carlito> #casual -:- ult_ is now known as ult 07:30pm <carlito> ult, are you talking to these chicks? <ult> sort of -:- hcf [nef@me-portland-us804.javanet.com] has joined #tunes <carlito> lol <carlito> I must say I'm dissapointed in the level of harrasment <Fare> who's harassing whom? 07:40pm <carlito> I was goofing with these girls..you know typ...I'm 15f looking for a good time <Fare> I don't know the type, but I'm willing to leanr <carlito> #casual <ult> carlito: You mean guys. <carlito> what? <Fare> carlito: on irc, nobody knows you're a dog <carlito> lol <carlito> tell him straight ult <carlito> ult: was I REALLY harrassing? <carlito> not even <ult> Fare: That or they are over the hill, and ugly. <Fare> "over the hill"? teach me english. <carlito> wow you guys are giving me a bad rep <ult> Fare: Over 40 years old. <carlito> Fare: I was doing no wrong <ult> Fare: (about). over the hill generally refers to the position in ones life where the hard nitty-gritty clawing to get to a decent career position is over, and one is just waiting to retire with a decent job and easy family life. <Fare> ok. Will have to remember that <carlito> Fare: where you from? * ult/#tunes ponders <ult> Fare is from the land of food and wine. * carlito/#tunes ponders too <carlito> they have food and wine over there? <ult> (Well, French food is good...) * carlito/#tunes nods <ult> carlito: That's where food and wine basically came from =P * carlito/#tunes shakes head <carlito> I don't think wine came from france...did it? <ult> does it matter? <Fare> *good* wine comes from France. <Fare> That's a big difference. <ult> *cheap* wine comes from California. =) <Fare> :) <carlito> what about Italy? <ult> That's what matters. <ult> Italy? They got vermouth. <carlito> are we having an intellectual conv here on this chat? * carlito/#tunes feels dizzy <Fare> carlito: too much wine? <carlito> doubt it 07:50pm <carlito> wow now I'm not harrasing this chick at all...but she's is puter dumb <carlito> I'm done tutoring for tonight, I'm out of that chat room 08:00pm -:- Downix [down@d-gnaps-13.ici.net] has joined #tunes [openprojects]Foreign OperKill: NickServ killed bd (This nick is reserved by another user) -:- water [water@tnt-9-192.tscnet.net] has joined #tunes <water> 'lo <ult> hiya <water> i'm going to upload the new papers to tunes.org in a few minutes <carlito> high water! <carlito> =) * eihrul/#tunes claps. * carlito/#tunes bows <Downix> HELP! I can't get X to work! <water> i've got the TeX source with me, so i can edit them while online * Downix/#tunes feels like an idiot, X will not compile/work <water> heh <water> compile x? <water> are you crazy? <hcf> water: icuc, http://www.phil.mq.edu.au/isl/ <Downix> Well, I can't install it and have it workeither <hcf> water: seen my review@ posts? <water> c00l! [openprojects]!carter.openprojects.net Net break: irc.linux.com lu.openprojects.net (Ping timeout) 08:20pm <water> i saw one post from you <hcf> the 2nd was sent a few hours ago <water> holy cow! look at the latest tunes posting! <water> this guy even references "vicious circles"! [openprojects]!carter.openprojects.net Net junction: sterling.openprojects.net lu.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from lu.openprojects.net. * water/#tunes does a little dance. <water> today is a very good day <ult> heh [openprojects]!carter.openprojects.net lu.openprojects.net acknowledged end of net.burst. <water> hcf: nice links. thanks <hcf> np * Downix/#tunes does not mind compiling X, but I wantit to compile, dammit <water> downix: you're looking on the wrong channel for that kind of help, dude <Downix> true * ult/#tunes compiles X in his sleep <Downix> ok, onto #linpeople [openprojects]!carter.openprojects.net Net break: devlin.openprojects.net asimov.openprojects.net (Ping timeout) <ult> Downix: whats wrong? <Downix> Ok... how come Lynx hjas a mind of it's own now? * water/#tunes clicks on "pre-order" for the book, substructural logics [openprojects]!carter.openprojects.net Net junction: varley.openprojects.net asimov.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from asimov.openprojects.net. <Downix> I do a scroll down and it hits a link instead <Downix> Ok, this ISP sucks <water> damn it, no ftp client <water> (for win32) <hcf> water: that link was found cuz the author plugged it on sci.logic <water> ok <hcf> water: so *hint* how about u post some msgs? <water> yeah, i guess i really should post a couple 08:30pm <water> hold on, all, for the papers. i have to get a decent ftp client [openprojects]!carter.openprojects.net Net break: varley.openprojects.net asimov.openprojects.net (Ping timeout) [openprojects]!carter.openprojects.net Net junction: irc.linux.com asimov.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from asimov.openprojects.net. <eihrul> no console ftp? [openprojects]!carter.openprojects.net asimov.openprojects.net acknowledged end of net.burst. <water> dunno <eihrul> start->run->ftp <water> wow <water> ack! my window manager died! <water> sigh... <water> brb... rebooting -:- water [water@tnt-9-192.tscnet.net] has left #tunes [] -:- water [water@tnt-9-192.tscnet.net] has joined #tunes <water> whew. saved from rebooting 08:40pm <water> oh cool. the abstract interpretation paper is on my tunes account <water> i thought i lost it <hcf> water: usenet-wise are there no threads u would post to? <water> i haven't gotten around to it <water> keep in mind i spend most of my time reading papers <ult> #!/usr/local/bin/perl -s-- -export-a-crypto-system-sig -RSA-in-3-lines-PERL <ult> ($k,$n)=@ARGV;$m=unpack(H.$w,$m."\0"x$w),$_=`echo "16do$w 2+4Oi0$d*-^1[d2% <ult> Sa2/d0<X+d*La1=z\U$n%0]SX$k"[$m*]\EszlXx++p|dc`,s/^.|\W//g,print pack('H*' <ult> ,$_)while read(STDIN,$m,($w=2*$d-1+length($n||die"$0 [-d] k n\n")&~1)/2) * eihrul/#tunes thinks ult needs a better modem. [openprojects]!carter.openprojects.net Net break: irc.linux.com asimov.openprojects.net (Ping timeout) <water> heh <hcf> water: just saying, along w/ origating msgs, it'd be good to join threads, since the readership is already captured <water> ok, the outline for the new paper is up and an updated intro is up <water> yeah, but i usually don't have/make time <eihrul> water: on ~water/? <water> yes <hcf> water: read papers faster ;) [openprojects]!carter.openprojects.net Net junction: irc.linux.com asimov.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from asimov.openprojects.net. <water> i need to update this home page <water> oh well <water> i'll edit it on my pda :) 08:50pm <water> anyway, i want comments from you guys about individual sections [openprojects]!carter.openprojects.net Net break: irc.linux.com asimov.openprojects.net (Ping timeout) [openprojects]!carter.openprojects.net Net junction: devlin.openprojects.net asimov.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from asimov.openprojects.net. * carlito/#tunes is tired of looking at html code tonight! <water> no no, the papers not the home page <water> hopefully you guys have ps viewers or printers, otherwise just read the .lyx text for the new paper <water> since i don't have a pdf for it yet <eihrul> ps2pdf <water> i'll do that when i get ofline <eihrul> ArrowPhilosophy.pdf :) <carlito> night all [openprojects]!carter.openprojects.net Net break: devlin.openprojects.net asimov.openprojects.net (Ping timeout) <carlito> hcf: thnx for the BOOT! <hcf> carlito: anytime <carlito> hcf: I was being sarcastic [openprojects]!carter.openprojects.net Net junction: varley.openprojects.net asimov.openprojects.net <water> btw, arrowphilosophy=arrow intro (prime) -:- carlito [sabanmr@137.28.139.244] has left #tunes [] <water> which is mostly going to get trimmed down in favor of the technical paper -:- lostsoul [sabanmr@137.28.139.244] has joined #tunes <eihrul> well, its the only paper there that isn't filled with foo's... <water> wb <eihrul> so its the only one i'm capable of reading right now :) <water> well, the outline for now is what i'm offering <lostsoul> huh? 09:00pm <water> i have a bunch of small text files to substitute for the foos <water> lostsoul=carlito? <lostsoul> yes <water> wb <lostsoul> trying to get in a room <lostsoul> someone keeps giving me the boot <lostsoul> hcf that is [openprojects]!carter.openprojects.net Completed net.burst from asimov.openprojects.net. <hcf> lostsoul: its not a room, lamer <lostsoul> channel lamer <lostsoul> whatever <water> why is hcf giving you the boot? * lostsoul/#tunes no idea <water> hcf: what's he talking about? [openprojects]Foreign OperKill: NickServ killed skibum (This nick is reserved by another user) <lostsoul> I was trying to say goodnight to a buddy, "lar1" and he booted me before I could tell him <lostsoul> hcf: if you got a prob, why don't you talk to me about it? <water> so, just how much of a lamer *are* you? :) <lostsoul> =) <lostsoul> a pretty big one I guess...hcf won't even respond <water> hcf is usually the silent type, though. i'd bet he's at work half the time that he's on irc <AlonzoTG> om <AlonzoTG> =\ <AlonzoTG> Hi, water. <water> hi atg <lostsoul> water: you're prob right <AlonzoTG> Looks like I signed myself up for a lot of reading with that last post I made to the tunes list. <water> atg: and you *should* read some more <water> atg: geb and drugs tend to make for some overconfident assertions from people who don't do the real reading <AlonzoTG> Anyone here have quake1.06 and is willing to spend about a buck on LD? [openprojects]!carter.openprojects.net Net break: varley.openprojects.net asimov.openprojects.net (Ping timeout) <lostsoul> night all -:- lostsoul is now known as carlito <water> i think this arrowTechnical outline will work perfectly for the arrow spec [openprojects]!carter.openprojects.net Net junction: devlin.openprojects.net asimov.openprojects.net -:- carlito [sabanmr@137.28.139.244] has left #tunes [] [openprojects]!carter.openprojects.net Completed net.burst from asimov.openprojects.net. <water> it should be really simple to relate the papers i've read into this format <water> unfortunately, the paper won't be complete until the tech spec is complete, of course * ult/#tunes ponders ATG 09:10pm <water> ult: not worth it <water> brb.. * ult/#tunes needs something good to read. <water> abi: tell ult about agt <ult> water, where did you put your papers? <water> in my home directory <water> www.tunes.org/~water/ <water> the squeak code is there too <water> anyway, i'll bbiaf -:- water [water@tnt-9-192.tscnet.net] has left #tunes [] [openprojects]!carter.openprojects.net Net break: devlin.openprojects.net asimov.openprojects.net (Ping timeout) -:- SignOff thomas: #TUNES (Ping timeout for thomas[193.217.63.152]) <air> ok my sleep() code is written, what are the chances it will work on the first test? <eihrul> nil [openprojects]!carter.openprojects.net Net junction: irc.linux.com asimov.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from asimov.openprojects.net. <AlonzoTG> om -:- thomas [thomas@193.217.63.152] has joined #tunes <air> ugh i hate those damn short jumps that are out of range 09:20pm <eihrul> so use long jumps... <eihrul> aka near jumps [openprojects]!carter.openprojects.net Net break: irc.linux.com asimov.openprojects.net (Ping timeout) <air> ya i know but they take up more space so i try to use short until they go out of range <air> damn code gpf'd when it wakes the thread -:- core [core@core.suntech.fr] has joined #tunes <core> people hi <Downix> hey * eihrul/#tunes thinks air needs to learn about the advent of the compiler. <Downix> Back in FR now? <eihrul> quick, someone call fox mulder! <core> wow, lots of cool people here <core> hey Downix, eihrul, air :) <Downix> thaks <eihrul> we've got some really strange phenomenon going on here... -:- water [water@tnt-9-65.tscnet.net] has joined #tunes <core> eihrul: nope, first core sighting of the month :) <water> wow. core <core> downix: yup, i came back last night, and i'm still jetlagged so i'm up at 6:20 AM :) [openprojects]!carter.openprojects.net Net junction: varley.openprojects.net asimov.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from asimov.openprojects.net. <Downix> core: It's not 6:20am,it's midnight, perfect time to be awake <core> water! :O :O <water> hehe <core> downix: it's 6:29 AM here, very early for me, but i kept waking up every hour, and decided i'd toy with code instead of rolling over in my bed :) <water> i've almost got a wroking tech spec <water> although the paper for it is empty so far <ult> 'lo core. <Downix> core: ok. Right now I'm working on the hardware spec for my system <core> water: hehe.. well, i have a postscript printer at work, waiting for it :) <core> ult :) hi :) <water> k :) <water> so where have you been, then? <core> this printer is quite cool actually, i can ftp documents to it; it's probably common, but i never saw that before <water> is it a xerox? <core> water: working insanely.. suntech is growing, so i have to 1) keep driving development of clementine 2) integrate new people 3) give time to my girlfriend 4) idle (not much time for that) and 5) irc :) <core> water: no, kyocera or somesuch <water> core: cool <Downix> hmm <core> water: and jenny got her engagement ring :-) <water> heh * Downix/#tunes has been dealing with the sound mixing channels now <core> downix: cool.. sorted out those sound system issues? <core> ahh. gmta * ult/#tunes wonders how Clem's coming along? <Downix> core: For the most part, now doing the hard part. Trying to speed up the mixing pipes [openprojects]!carter.openprojects.net Net break: varley.openprojects.net asimov.openprojects.net (Ping timeout) <core> ult: well, i personally haven't touched it since dec 28 (probably the last time we talked ?), i was in holidays <core> took a week off w/ Jen in Phoenix, and gave her the engagement ring too. :) <ult> core: Yes. Just a gentle reminder ;) <core> ult: it still runs an http server, plays heretic, and if the people at suntech worked as expected, it should boot off flash memory now too :) <core> downix: analog or digital mixing? <core> the v2_os guy starts monday, he should fill some holes as well. <Downix> core: Pure digital <Downix> There is no analog setup until the final output to speakers <water> wow. the ecoop99 paper linked on the mlist is really nice <ult> odd 09:30pm <ult> core: How does it run Heretic? Is the heretic binary unix-like or dos-like? <core> downix: cool.. harder to do in hardware than amiga-like analog mixing, but probably cheaper on D/A converters :) <core> ult: i ported it from the linux version <ult> core: Is the video just direct access to the VGA hardware circumventing any memory protection? <Downix> core: That's the idea. <core> ult: i wrote a graphics server for clementine (very quickly), and mapped the necessary OS functions to the right clem interfaces, and it more or less immediately worked (after i fixed a few issues in the OS :) <ult> core: Or does Clem actually have a VGA driver that handles things like that? <core> ult: clementine has all GGI drivers, and a vesa 2.0 driver * ult/#tunes nods <core> downix: cool cool.. hope your hardware will at least do 32-bit additions of all the samples before cropping them down back to 16 bits :) <ult> core: Now you just need to get it out the door ;) <core> ult: so yeah it has actual SVGA drivers.. games can still remap the framebuffer and use it if allowed tho <Downix> core: 64-bit actually <core> ult: yeah, i know :) (better is the enemy of best, is that how you say it in English? :) <core> downix: ah, even better :-) <ult> 64bit samples? Overkill? * water/#tunes is afk <core> ult: not samples, just accumulator <eihrul> core: don't know if there's a direct english equivalent... <Downix> ult: That's to allow up to 4 16-bit samles to be be mixed before bit-reduction <ult> I see. <core> hmm, more like, better is the enemy of good, or something. like you have to stop improving it at some point :) * Downix/#tunes nods to core <ult> Downix: sounds efficient <Downix> ult: I hope it's efficient enough to make it worthwhile for computer buyers <core> downix: actually you only need 18 bits for adding 4 16-bit samples * ult/#tunes nudges eihrul in Downix's direction. <eihrul> ult: eh? <abi> somebody said eh was he a drunkard? <eihrul> abi: is your name ult? <ult> eihrul:RISC Lisp machine =) <core> 4*65536 = 262144 = 2^18 <Downix> core: Well, I'm trying to figureout how many bits I need. I'm unfamiliar with digital mixing. Taking a crash course <core> downix: that's what most laserdisc/dvd etc equipment hs <core> s/hs/has/ <eihrul> ult: aka Leak <Downix> core: I just imposeda maximum limit of 64-bits <core> downix: it's just an addition.. so you only need channels * 65536 :) <core> downix: 64 bits is huge. you can mix a ton of channels. that's cool * Downix/#tunes nods,ok <ult> eihrul: Maybe we can talk Downix into giving us a few boards =) <Downix> lol [openprojects]!carter.openprojects.net Net break: sterling.openprojects.net lu.openprojects.net (Ping timeout) <eihrul> ult: in erm, 2 years :) <core> now you might have some 'virtual' channels for some FX, so it's good to have some room :) * Downix/#tunes nods <ult> eihrul: hmm =) <Downix> The sound system is tiny, so the extra bit depth won't hurt the system too much <core> downix: if it doesn't cost you any transistors, go for 64 bits, else 32 bits will allow for 65536 channels :) <ult> I'd just like to get my hands on a nice programmable MIPS demo board. <ult> That or a cheap SGI. <Downix> core: It costs transistors, but nothing to increase the die size (I'm already about 3 million transistors UNDER the makeit or break it point) <core> ult: indy's must be extremely cheap these days. check news://comp.sgi.marketplace or something <core> downix: neat :) but you still haven't allocated transistors to 3D support, flash memory and cdrom interface, etc though, right? <water> yes! yes! yes! <Downix> core: I have for 3D, flash, and CD-ROM is through the firewire/usb <water> sorry <ult> core: Yes, I've ancient ones for under 100$ 09:40pm <water> got a little excited there over reading this paper <core> ult: good enough <core> water: hehe.. excitement is fine. i wish i had some sometimes :) <core> (i don't get excited even at things i should, that is :) <ult> water is the only person I've ever met who gets off on CS papers ;) <water> it's jsut that i now have an excellent basis for part of my spec <core> downix: ahh, usb, right :) <core> water: i'll open the firewall so you can directly print it to kyocera.suntech.fr :-) <water> well, i absolutely need arrow to exist to start my dream company [openprojects]Foreign OperKill: NickServ killed Mike_W (GHOST command used by Mike_W1) <Downix> core: Hey, saves us mice, HD, CD-ROM, etc. controllers <water> that's when i start to make the *really* interesting stuff <core> downix: true.. no need to reinvent too many wheels at once <water> ;) <Downix> core: Right <core> water: seems like there are a lot of entrepreneurs here :) * water/#tunes goes afk again <Downix> yup <core> i'm excited a little though, next week we will close down version 0.2 of clementine (i downsell things quite, 0.2 should be like 0.8 but oh well) * Downix/#tunes nods * Downix/#tunes is upsellinga bit, he'll admit <Downix> Well, I feel like I'm upselling * Downix/#tunes won't be happy till it's working <core> Downix: well, salespeople at suntech do that (actually, they don't upsell, they just say the truth :) .. i usually downsell myself <core> downix: i know the feeling [openprojects]!carter.openprojects.net Net junction: irc.linux.com lu.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from lu.openprojects.net. <Downix> core: So, 64-bits for mixing channels into 5 output digital lines should not be over-the-top? <core> downix: 64 bits for every of the 5 channels you mean? <Downix> core: Right <core> downix: well, that allows for 2^48 digital channels on every output channel :) <Downix> let's see programmers max those out <core> hehe * Downix/#tunes gryns <Downix> THat is PER cycle I'd note <core> programmers no, but artists (graphists or musicians) will :) <Downix> if you buffered, streamed, and re-ran everything, you could do even better [openprojects]!carter.openprojects.net lu.openprojects.net acknowledged end of net.burst. <core> Downix: I think it is a reasonable assumption that 5 * 2^48 channels are enough for the next 10 years :) 09:50pm <Downix> core: I hope <water> yeah right <core> water? :) <water> just you wait <Downix> hehe <water> :) <Downix> there's no way to make programmers happy <core> water: well, true, that's what they said of cd-rom and now of dvd :) * Downix/#tunes nods <water> besides, there's serio i/o architectures to consider in the future <water> s/serio/serial [openprojects]!carter.openprojects.net Net junction: irc.linux.com asimov.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from asimov.openprojects.net. [openprojects]!carter.openprojects.net asimov.openprojects.net acknowledged end of net.burst. <core> well, i suppose you could map every sound source in a game to a digital channel and let the hardware mix all of that painlessly * Downix/#tunes nods <Downix> water: Ihave considered them through a basic local bus port on the chip <water> k [openprojects]Foreign OperKill: NickServ killed fusion (This nick is reserved by another user) <core> downix: mmm, exciting H/W you're designing. <Downix> core: Scary, in my eyes. 8) <core> Downix: the strange moment will be when other people start using it <Downix> core: Very true <core> Downix: i still haven't competely got used to it for clementine <core> s/compe/comple/ * Downix/#tunes nods <core> it's still strange to me to see other people using my tools and work and adding to my ideas. neat but strange :) <Downix> Unlike hardware, you can just compile Clementine and have people useit * Downix/#tunes smiles <core> downix: true. well, you can run your HW in a simulator for many months before first silicon tho, and distribute that :) <Downix> core: true * Downix/#tunes needs the spec done first tho <core> i started linux in the armulator today, first step before the next step of World Domination that starts jan 25 :) <core> Downix: that'll be helpful :) and i saw you're going to use lisp for the simulator, that's a good choice imho. will cut a lot of time. <AlonzoTG> the extremely powerfull hardware abstraction on my OS will allow you to change EVERYTHING between versions and software will still run. =) <ult> eh <ult> What's happening jan25? <water> atg: yeah whatever <core> ult: clementine core port to ARM :) <ult> core: Gee =) <Downix> core: Thank you (I'm growing to like LISP a lot the more I use it) <ult> core: When m68k? 10:00pm <core> ult: in parallel of the "up" growth (ie, applications, etc) that's happening <ult> ColdFire is nice, powerful, and -cheap-. <ult> Downix: What Lisp? <abi> Lisp is ((simply) (perfect)) <core> ult: i don't know yet, probably Q2 2000. it'll be easy once the first one (arm) is done. <Downix> ult: Common <core> downix: i like it, don't have time to practice it much, but when i see what one of my coworkers does with emacs.. geeze :-) <ult> Downix: Which implementation? <ult> Downix: Your writing your emulator in Lisp? <core> ult: well, the #1 customer for coldfire is in the same city as we are :-) <core> (the OS development teams anyway) <ult> core: Mot loves France =) <core> ult: well, they love palm computing at least :) <Downix> ult: Herliquins and yes <ult> core: I would imagine so =0 <ult> downix: How much is a license for HL? <ult> (LiquidLisp or LispWorks)? <Downix> ult: Not sure,I downloaded the free personal one <ult> Downix: Oh, ok. <core> Downix: will you have to write the MIPS and USB emulations too, or is there some code already available for that? <ult> Downix: I was under the impression that as a company you may have purchased it ;) <Downix> core: Everything I'mafraid <Downix> ult: Not yet. <core> downix: ouch :) that'll be a lot of work. i hope you'll have people helping to do that :) <Downix> core: Afraid not. <Downix> heavy are my shoulders <core> Downix: they will feel so much lighter with VC and coworkers :) <Downix> core: Agreed <core> (well, not really lighter, just weight trading i guess) <core> trading development weight for meeting-objectives and keeping-people-happy-and-working weight :) * Downix/#tunes is feeling less and less like contracting out to an EE firm to make the final mold designs <Downix> I'm learning how to make a chip mask, I realized I don't need to out-source <core> Downix: if there is no point in doing the mask yourself, you should outsource tho..? <Downix> core: I ment the pattern on it <core> Downix: oh.. hmm, well, if you're sure you won't fall into any traps or anything <Downix> core: I'm checking out the optionsright now <Downix> I had an option for an established firm to do most of the work for us, for a 20%royalty <core> Downix: is that the normal percentage? that sounds expensive <Downix> core: It is expensive <Downix> core: I am tryinghard to do the design work myself now, as you know <core> Downix: sounds like it to me :) maybe you're better off doing it, i guess <core> Downix: yeah :) * Downix/#tunes is doing wlel he feels 10:10pm <Downix> Still not perfect, I need to learn an HDL <core> hardware is an unknown territory to me. i (obviously) am familiar with its programming interfaces, but inner works are still magic. especially at today's speeds :) i can understand the workings of a 0.9 mhz 6502, maybe a 8 mhz 68000, but that's it :) * Downix/#tunes nods <Downix> yeah, it's scary <Downix> and 27 months ago I didn't know a transistor from an occilator <Downix> You should have seen my original concepts on how to design a computer. They were laughable <core> well, where was i 27 months ago.. that's what, october 1997? <Downix> yes, oct 1997 <core> hmm, i was working on GGI :) had a very vague idea of what i wanted clementine to be like, but didn't start a line of code until april '98. <water> wow * Downix/#tunes laughs <water> not know a transistor from an oscillator and now you have that product in the works? * water/#tunes is non-plussed <Downix> in Sept'97 I began the trek for Eddas <Downix> water: Well, I've had a lot of help. <Downix> water: Helps when you have a grandfather who helps design power transmission stations <core> my grandfather did masonry :P <water> heh [openprojects]!carter.openprojects.net Net break: irc.linux.com lu.openprojects.net (Ping timeout) <water> i descend from farmers <core> my father runs the administrative services of a large university, but he's not very computer-savvy <Downix> Hehe, my grandfather IS a mason.... freemason <Downix> .em laughs <core> both of them taught me that working hard pays off, tho :) <Downix> My grandfather taught me that telligthe truth gets you beaten up, and you know what, he's right [openprojects]!carter.openprojects.net Net junction: sterling.openprojects.net lu.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from lu.openprojects.net. * Downix/#tunes has had so many people berating himon-line for his ideasit's not funny <core> Downix: i know that, but i can't help. i inherited that from my mother (always speaking up that is :) <Downix> core: hehe <ult> only problem with speaking up is being wrong <core> ult: true.. well, if i only say 60% bullshit during a day at work, i'm happy tho. as long as 40% of good ideas come out of it :) * Downix/#tunes laughs 10:20pm <Downix> Same here <core> ult: i don't mind being wrong and being corrected.. have enough self-confidence not to get insulted by it. a lot of people corrected a lot of my thoughts on clem (fare the first) :) <water> every good idea starts as bs... the difference is what you make out of it :) <ult> core: Sure. The problem with being wrong though is when other people believe you =) <ult> Then you end up with 500 people running in the wrong direction -:- SignOff AlonzoTG: #TUNES (Have Nice Day :)) * Downix/#tunes nods <core> ult: oh.. well, i try not to say things i haven't verified. of course it never works 100%.. but usually someone will take pleasure in correcting me :) * Downix/#tunes phews, no more annoying messages about "Advertise in PC Mag" * Downix/#tunes nods <core> people who start at suntech are usually shy and saying OK to whatever i say, but after a month they get the hang of it, and make fun of what i say when i'm wrong ;) [openprojects]!carter.openprojects.net Net break: sterling.openprojects.net lu.openprojects.net (Ping timeout) <Downix> hehe <core> (just like i do when they are.. fair enough :) * Downix/#tunes nods <core> so now we have 12 major, maddening nitpickers. weee :) <Downix> WHgat do I do when people correct me? <Downix> I bitch, moan, get angry, turn around, fix it, improve something else, and make the whole thing smaller and cheaper <Downix> Which is whY I went from a 5million gate machine to a 1.8 million gate machine <core> Downix: hehe.. well, my mood is always even (the good thing being i don't get angry, the bad thing being i never get excited, either), but basically that goes through the same cycle :) <Downix> core: Ihave a problem, I get angry AND excited <Downix> core: Get's worse when an attractive babe walks in the room <Downix> 8) <Downix> lol [openprojects]!carter.openprojects.net Net junction: irc.linux.com lu.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from lu.openprojects.net. <core> Downix: not sure this is a problem :) it's just how i am.. not excited :) <core> Downix: well, i'm an engaged man now, so i'm not supposed to react to that kind of stimuli :) <water> lol <Downix> core: lol <core> jen sends enough stimuli to keep my brain busy 24/7 though :) <water> whoa. cool <core> she's the most restless person i know. can't spend 5 minutes without doing something :) <Downix> core: I was engaged at one point. She ended up becoming excited by that stimuli, and after coming out and 1 week of hot 3-way action, we went our seperate ways. <Downix> 8) <water> whoa <core> Downix: ahh.. well, that's beyond me now, and we still got engaged ;) <core> s/beyond/behind/ :) * Downix/#tunes laughs <ult> Downix: that's something to brag about ;) * core/#tunes pities her friends for the next 2 months (or whoever doesn't know about her ring yet :) <Downix> ult: nah, she IS my business partner now * Downix/#tunes laughs <ult> Well, it's getting rather late here, and I have to get up soon. <core> Downix: you're in business with an ex-girlfriend? <ult> I'll see ya'll later. <core> see ya ult :) <water> bye ult -:- SignOff ult: #TUNES (C'est la vie...) [openprojects]!carter.openprojects.net Net break: irc.linux.com lu.openprojects.net (Ping timeout) <Downix> core: Yes, we broke up friends and we still make good business associates <core> Downix: make sure you have good contracts tho ;) women usually have oscillating moods, but ex-girlfriends.. man.. :) <water> core: don't forget you're logged [openprojects]!carter.openprojects.net Net junction: sterling.openprojects.net lu.openprojects.net <Downix> core: We do. (I am VERY rock solid in contracts when I make them) <core> water: i know i'm logged :) jenny knows she has oscillating moods :) 10:30pm <water> k [openprojects]!carter.openprojects.net Completed net.burst from lu.openprojects.net. <Downix> core: Every woman does. You should see my ex and her gffight! <core> she can go from being excited and clapping her hands, to pout, to being excited, in less than 5 mins :) <core> i came up with something tho, a 3-day no-pouting rule for getting her ring, i liked it :) * Downix/#tunes laughs <core> Downix: oh, she broke up for another girl? lovely :) <Downix> core: Rather <core> ah well, i'll be stuck with the same one for a long time (or the other way around) :) <Downix> I agree <Downix> but try convincing the girl I love of that <core> hehe.. well, no need to hurry :) [openprojects]!carter.openprojects.net Net break: sterling.openprojects.net lu.openprojects.net (Ping timeout) <core> being 'stuck' with jen is great though. she knows me well, never complains about my insane working hours, gets excited for me when i got something done, etc ;) <eihrul> executes evil corporate lawyers? * Downix/#tunes laughs <Downix> sounds like Amy, but Amy hasa big problem about commitment <water> now *that* would be cool gf <core> lol. indeed *that* would rock :) <core> don't give her ideas though :) <Downix> what would rock? <core> if she executed corporate lawyers :) <water> yeah * Downix/#tunes is REALLY sick of Amy and her stringof ex-lovers <water> are you jealous or just annoyed? <core> Downix: well, i'm lucky i guess. she's been my lucky charm, i met her before i started suntech, and things have been going right on both fronts since :) <Downix> madly jealous <Downix> core: Same with Amy <water> btw, most homosexuals are very bad with commitment <core> water: i know two who have been together for 5 years :) <Downix> water: I know that. <water> i used to think that it was a stereotype, but i've been proven wrong too many times <eihrul> Buffy the Corporate Lawyer Slayer... Coming to a cheesy, network television station near you... <water> rofl :D * Downix/#tunes laughs [openprojects]!carter.openprojects.net Net junction: irc.linux.com lu.openprojects.net <water> i think that i might actually *watch* that [openprojects]!carter.openprojects.net Completed net.burst from lu.openprojects.net. <eihrul> cmucl/doc/* is conspicuously empty 10:40pm <core> lol <core> well, she likes buffy (sigh), so she might get ideas. <Downix> hmm [openprojects]!carter.openprojects.net Net break: irc.linux.com lu.openprojects.net (Ping timeout) [openprojects]!carter.openprojects.net Net junction: sterling.openprojects.net lu.openprojects.net <Downix> Ok, I think my mixer is done [openprojects]!carter.openprojects.net Completed net.burst from lu.openprojects.net. <Downix> Ok, any ideas for 3D? <eihrul> 1D > 3D <water> yeah, use matrix transformations :) * Downix/#tunes gets out a pen and paper <Downix> whazzat? <water> huh? <core> matrix transformations are a given, but don't concern downix <eihrul> 3D infers matrices... * Downix/#tunes is planning on focusing on radiocity over polygon count tho <core> downix: it's a transformation-engine issue <water> i know, i was joking <Downix> ok <core> Downix: a neat mathematical thing that allows for many transformations to be merged in one matrix and applied once to every point you transform <core> water: making sure he doesn't get confused :) <Downix> core: hmm, ok <water> np <core> Downix: ie, you can merge a translation/rotation/etc. in one matrix. 9 multiplications per point whatever you do :) <eihrul> core: you used matrix to define matrix <eihrul> silly <core> eihrul: i'm a recursive person ;) <core> s/one\ matrix/one/ <core> :) [openprojects]!carter.openprojects.net Net break: sterling.openprojects.net lu.openprojects.net (Ping timeout) <eihrul> your definitions aren't tail recursive <eihrul> so you must be very susceptable to stack overflow... <core> eihrul: that wasn't a definition, just a vague explanation :) 10:50pm <Downix> core: I'm mostly trying to see what the most RISC'd solution for 3D is I could do <Downix> keep control of the 3D up to the library/compiler <eihrul> isn't that moving against the tide, though? <water> ack! that's like commiting to one type of gfx engine <Downix> eihrul: It is but it's smarter <Downix> water: Wrong, it's committing to no gfx engine <eihrul> Downix: probably slower... <water> well, what are you putting into the gfx kernel then? <Downix> water: what kernel? I'm making hw <Downix> eihrul: Depends, smaller actions can be executed in less CPU time <water> <Downix> keep control of the 3D up to the library/compiler <core> Downix: well, i suppose you could create a sort of 'instruction set' that can be used to do any kind of 3D operation <water> n/m then [openprojects]!carter.openprojects.net Net junction: irc.linux.com lu.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from lu.openprojects.net. <Downix> water: yes, so whatever compiler/library I use can change the way it works <water> so why are you concerned with 3d at all? <Downix> core: X Y Z,rotz, roty, rotz, I know <core> Downix: that's what you'd not use and use matrices instead. i was more thinking of lower level rendering <Downix> water: I need to support it at least limitedly in the chipset <water> um <core> maybe another MIPS directly tied to the 2D graphics controller? *shrug* [openprojects]!carter.openprojects.net Net break: irc.linux.com lu.openprojects.net (Ping timeout) <core> that would probably blow your transistor count though :-) <core> but i don't know if there is anything inbetween an actual cpu that can do what it wants in graphics memory including rendering 3D, and a specialized 3D engine that renders full primitives 11:00pm -:- SignOff hcf: #TUNES (Leaving) <water> yeah, that's a pretty empty continuum <core> maybe a simpler cpu would do.. you don't need things like a write-back cache, an MMU or an FPU for rendering. <water> sure <core> that's maybe the best solution but probably a little expensive :) <Downix> core: Actually, what I was ponderingwasjust that. MIPS has a very nice 3D acceleration method availible [openprojects]!carter.openprojects.net Net break: irc.linux.com forward.openprojects.net (Ping timeout) <water> unless there are chips available with those units already stripped from the design [openprojects]!carter.openprojects.net Net junction: sterling.openprojects.net forward.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from forward.openprojects.net. [openprojects]!carter.openprojects.net forward.openprojects.net acknowledged end of net.burst. <core> water: there probably are.. you want to pick a cpu with the same instruction set as the main one though, or writing code will be hell <Downix> core: but I'd keep the FPU, for Raytracing 3D work. 8) <water> heh <core> Downix: you only want the graphics-cpu for actual rendering. the data would be prepared by the fpu of the main processor <water> mmm... someday i will have my real-time raytracing engine <water> *) <water> 8) <core> i'd go for a mips core without fpu, mmu and cache <core> don't know how expensive that is :) [openprojects]Foreign OperKill: NickServ killed localhost (This nick is reserved by another user) -:- SignOff Downix: #TUNES (changing servers) -:- Downix [down@d-gnaps-13.ici.net] has joined #tunes <core> don't know how much you missed, downix <Downix> I got the "use a second MIPS" and I said.... <Downix> core: Actually, what I was ponderingwasjust that. MIPS has a very nice 3D acceleration method availible <Downix> core: but I'd keep the FPU, for Raytracing 3D work. 8) [openprojects]!carter.openprojects.net Net break: irc.linux.com devlin.openprojects.net (Ping timeout) -:- NetSplit: devlin.openprojects.net split from irc.linux.com [11:06pm] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [devlin.openprojects.net] <water> and i said: <water> mmm... someday i will have my real-time raytracing engine <water> 8) <core> Downix: you only want the graphics-cpu for actual rendering. the data would be prepared by the fpu of the main processor <core> don't know how expensive that is :) <core> water: there probably are.. you want to pick a cpu with the same instruction set as the main one though, or writing code will be hell <core> and whatever else i said :) <Downix> core: well, a MIPS cpu needs the FPU to work [openprojects]!carter.openprojects.net Net junction: sterling.openprojects.net devlin.openprojects.net -:- Netjoined: devlin.openprojects.net irc.linux.com -:- tmf [thomas@193.217.63.152] has joined #tunes -:- smkl [sami@glubimox.yok.utu.fi] has joined #tunes -:- Fare [fare@quatramaran.ens.fr] has joined #tunes -:- eihrul [lee@usr5-ppp177.lvdi.net] has joined #tunes [openprojects]!carter.openprojects.net Completed net.burst from devlin.openprojects.net. [openprojects]!carter.openprojects.net devlin.openprojects.net acknowledged end of net.burst. <core> Downix: ah, blah :) <water> yeah [openprojects]!carter.openprojects.net Net junction: sterling.openprojects.net lu.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from lu.openprojects.net. <Downix> core: but I could use another, licensable CPU core like SPARC for 3D work <water> but just consider the possibilities of marketing a real-time raytracer that can re-optimize itself for the current scene * Downix/#tunes nods to water <water> tunes-style 8) <Downix> that's something else to consider <core> Downix: or a 68K :-) <core> a 68000 core boosted to 250 mhz or so ;-) <core> ok, i have to dash to work, before i suggest using an alpha or something stupid ;-) <core> see you! -:- SignOff core: #TUNES (vroom) * Downix/#tunes laughs <water> heh <Downix> I think MIPSis the one to use <Downix> one for the main int <Downix> and one for 3D/raytracing 11:10pm <water> it's cool to see my mips 4300 run squeak at full speed instead of the "reduced cpu usage" speed used before <Downix> cool <Downix> MIPS is a cool CPU <abi> i already had it that way, Downix. <water> good, abi <Downix> thanks abi [openprojects]!carter.openprojects.net Net break: sterling.openprojects.net lu.openprojects.net (Ping timeout) [openprojects]!carter.openprojects.net Net junction: irc.linux.com lu.openprojects.net <Downix> If I used a MIPS core for a 3D engine it would increase our die size to 3 million gates [openprojects]!carter.openprojects.net Completed net.burst from lu.openprojects.net. <water> hm <water> trade-offs <Downix> A good one <water> unless you don't need 3d <Downix> the second MIPS CPU could beused in other ways when not rendering 3D scenes <water> oh [openprojects]!carter.openprojects.net lu.openprojects.net acknowledged end of net.burst. <Downix> Don't forget, MIPS is the best SMP CPU out there <water> sure <water> so, it's 3d processing would just be a special smp set-up? <Downix> Right <water> ah <Downix> one of the MIPS would tie to the GFX chipdirectly tho <water> i see <Downix> but it would still have thedata bus to the other one <water> with some sort of i/o switch <Downix> right <water> interesting <Downix> would be not too hard to do <water> i'd never considered that kind of setup <Downix> I did in the past <water> but then, i don't design hw :) <Downix> and my GFX system uses it exclusively for a lot of it's more "fancy" effects <water> lol... there's an article from 1900 on wired <Downix> lol 11:20pm <Downix> anyways <water> this is cool... linux distro companies will be going for government contracts <Downix> that is cool <water> and i'm the only guy in my dept who can handle that stuff <water> hehe <water> it will be fun to axe NT <water> just once, it won't be the one to axe itself <water> 8) <water> i'd love to show the guys at work kde [openprojects]!carter.openprojects.net Net break: devlin.openprojects.net adams.openprojects.net (Ping timeout) [openprojects]!carter.openprojects.net Net junction: sterling.openprojects.net adams.openprojects.net [openprojects]!carter.openprojects.net Completed net.burst from adams.openprojects.net. [openprojects]!carter.openprojects.net adams.openprojects.net acknowledged end of net.burst. <water> their jaws would hit the floor if i showed them an open-source oodb like zope <water> anyway <water> (that was quite the tangent) <Downix> hehe <Downix> someone needs a better system than X <water> yeah, x really sucks <water> wait.. did you mean the x windowing system or X as a variable? <Downix> X windowing systsme <water> k <water> wow. the mars polar lander seems to have landed ina canyon <Downix> Someone needs to figure out a better way to handle things over the network <Downix> Yup <Downix> I heard <air> YES!!!! <water> talk about your bad luck <water> air: why excited? <air> one more section of brix is finished <Downix> YES! <water> oh 11:30pm * Downix/#tunes likes hearing of projects being finished <air> well i just finished the sleep() and semaphores so now my floppy driver will work <air> and i will be able to see if my persistence code works <water> ah. very useful, then <Downix> ok <air> well im off to bed now, cya -:- SignOff air: #TUNES (BRiX [http://www.qzx.com/brix] :: sleep) <Downix> Hmm <Downix> I just realized <Downix> all I have left to design now is the local bus * Downix/#tunes is in shock <water> cool <Downix> OMG <water> and hopefully, i'm close to knowing what code i must precisely add to my squeak alpha code that will make arrow work * Downix/#tunes is damned close to being done <Downix> Alright <water> it's been six years since i thought up the idea <Downix> great <water> i studied axiomatic reasoning and formal proofs in this great geometry class, and gradually came to form this idea of a program that "walked" over data-structures on the hard drive that only had structure (structure formed content) <Downix> hmm <Downix> interesting <water> and i realized i could connect all sorts of ideas together <Downix> oooo <water> little did i realize just how much i'd have to learn to get the thing working <Downix> Same with Eddas <Downix> hold on, need the bandwidth -:- SignOff Downix: #TUNES (BitchX: roqz yer nutz!) 11:40pm -:- Downix [down@d-gnaps-13.ici.net] has joined #tunes <Downix> ok, back * water/#tunes is browsing apple's os x decription <water> description, even <Downix> ok <water> not half bad... they went with the next's preference for dps and replaced it with api support for pdf [openprojects]!carter.openprojects.net Net break: irc.linux.com lu.openprojects.net (Ping timeout) 11:50pm <Downix> cool <water> it claims every app will have pdf output support <Downix> grr <Downix> I NEED X TO WORK! <water> sorry <Downix> I just finished compiling it, and it still can't run <water> hm.. i wonder if kde could support pdf-output api <water> probably not <Downix> this is the new /usr/X11R6/bin directory after Imake World:make install <water> k <Downix> does it look complete to you? <water> hardly <Downix> Agreed <Downix> WTF is it missing? <water> is this just for a build or what? <water> not sure what's missing, offhand <water> i'm not that savvy, and i'm in win32 right now <Downix> this is just for me <Downix> I just built it <water> (modem on linux problems) <water> oh <water> dunno... startx? :) <Downix> startx says that files are missing <water> well, they are. but i couldn't tell you which ones they are <water> check xfree86.org <Downix> ok <water> heh. i have a 1280x1024 resolution set up, and i can never find the desktop :) <Downix> lol [msg(TUNES)] newlog 2000.0107 IRC log ended Fri Jan 7 00:00:01 2000