IRC log started Wed Feb 3 00:00:00 1999 [msg(TUNES)] permlog 1999.0203 ωνω NetSplit: devlin.openprojects.net split from sterling.openprojects.net [07:10am] ωνω BitchX: Press Ctrl-F to see who left Ctrl-E to change to [devlin.openprojects.net] ωνω Netjoined: devlin.openprojects.net sterling.openprojects.net ωνω KeLp [tcole@feynman.physics.wwu.EDU] has joined #TUNES ωνω iStormy [stormy@rain.futuresouth.com] has joined #Tunes hi hola, iStormy 08:10am hi abi dc: 08:20am ωνω hcf [nef@escher.sdi.agate.net] has joined #tunes ωνω SignOff iStormy: #TUNES (Ping timeout for iStormy[rain.futuresouth.com]) If a trainstation is where the trains stop, what is then a workstation... how bout a space station? 09:10am :) scary! π Fare/#Tunes is away 09:20am ωνω TMF [s720@lake.ii.uib.no] has joined #tunes ωνω iStormy [stormy@rain.futuresouth.com] has joined #Tunes abi, tunes? tunes is in my head or tunes is a useful nevertheless expedient system. or at http://tunes.org/ abi, has ? tmf: wish i knew abi, abi? i am probably the little girl borg with a big brain or #tunes' personal little whore or an infoslut or gaklosmontic or flurivostuginuous or xyvarestoplik or crapulmaniac or or abi, prism? i heard prism was a specification & compiler which attempts to ease programming by allowing the programmer to express programs in a form that closely matches the programmer's conceptual model of the program. abi, be? be is a stupid name.. their web page looks like a joke :) abi, miscrosoft? tmf: bugger all, i dunno be has a nice web page abi, be? rumour has it be is a stupid name.. their web page looks like a joke :) oh well abi, rebol? rebol is not free abi, your're fun for so long! 10:30am abi: be abi look at me, i'm cute! π iStormy/#Tunes looks at abi. iStormy: aren't i sexy? 10:40am abi: i'm married, i can't answer that iStormy: i know what you're thinking though 10:50am abi: i'm married, i can't think that either. iStormy: your loyalty turns me on *blush* abi: i just can't win, can i? i haven't a clue, istormy hows the page coming? !koontz.openprojects.net!! Remote CONNECT devlin.openprojects.net 8004 from ^lilo 11:00am !koontz.openprojects.net!! Remote CONNECT calvino.openprojects.net 8004 from ^lilo !herbert.openprojects.net!! Remote CONNECT calvino.openprojects.net 8005 from lilo !hogan.openprojects.net!! Remote CONNECT calvino.openprojects.net 8005 from lilo !gibson.openprojects.net!! Remote CONNECT calvino.openprojects.net 8005 from lilo 02/03/99: I'm ruining my hands (RSI) on this web page...taking a break. Check back in a few weeks. 11:10am k 11:20am ωνω SignOff TMF: #TUNES (Leaving) ωνω SignOff Fare: #TUNES (Ping timeout for Fare[esmeralda.enst.fr]) !carter.openprojects.net!! Remote CONNECT norton.openprojects.net 8004 from ^lilo ωνω Tril [dem@bespin.ml.org] has joined #TUNES hoy Tril hi how's that irc page coming? heh is it needed? i dunno. what wer eyou going to have on it? i dont know just thot i should do one since im the "irc coordinator" but i havnt given it much thot besides that oh well, we only need a new page if the paragraph about tunes irc on collaboration.html gets very large. what else can i do? (a list of available jobs would come in handy right now) the irc page should be for summaries of the logs, that's the main thing what do you mean? 11:50am have u seen the list on wiki? that disorganized out of date todo list? yeah. did you notice at the top it says one job is to maintain the todo list. :) well, if not, it should the 'check review pages for bad links' thing is done trying to get bineng here from icq he logged off :) π Tril/#TUNES is away: (brb) [BX-MsgLog Off] 12:00pm π KeLp/#TUNES is back from the dead. Gone 12 hrs 11 min 26 secs yo! π Tril/#TUNES is back from the dead. Gone 0 hrs 14 min 13 secs foo 12:10pm bar baz quux corge ωνω NetSplit: asimov.openprojects.net split from varley.openprojects.net [12:14pm] ωνω BitchX: Press Ctrl-F to see who left Ctrl-E to change to [asimov.openprojects.net] !norton.openprojects.net!! Remote CONNECT asimov.openprojects.net 8005 from lilo !carter.openprojects.net!! Remote CONNECT asimov.openprojects.net 8005 from lilo ωνω Netjoined: asimov.openprojects.net varley.openprojects.net ωνω KeLp [tcole@feynman.physics.wwu.EDU] has joined #TUNES what's corge abi: foldoc for foo foo may be sought in foldoc at http://wombat.doc.ic.ac.uk/foldoc/foldoc.cgi?query=foo see that page 12:20pm ωνω NetSplit: asimov.openprojects.net split from carter.openprojects.net [12:20pm] ωνω BitchX: Press Ctrl-F to see who left Ctrl-E to change to [asimov.openprojects.net] ωνω Netjoined: asimov.openprojects.net carter.openprojects.net ωνω KeLp [tcole@feynman.physics.wwu.EDU] has joined #TUNES if it's non-free, how come I'm downloading it :) is debian's def of free the same as rms'? nope slightly different must be more strict but compatible he is starting to help out with debian I think he is more strict you can read it on their web site I notice you didn't say "our" web site ;) !lilo:*! Freshmeat is back. I'm not a debian developer :) !lilo:*! I'll go fix the DNS for devlin :) I thought everyone was a debian developer :) I should be but I think I would rather be a FreeBSD port maintainer 12:30pm Tril: r u pro debian now? I am attached to it now now that I have tried it. is pretty nice π Tril/#TUNES gets frustrated at not having a running tunes stay up all night and write it :) that's too easy. it must be done without caffiene. 12:40pm drug free devel?! I only speak for myself ωνω ultima [ultima@user-37kb9sg.dialup.mindspring.com] has joined #tunes anyone else is free to use stimulants in their quest to code tunes ultima: whats the status of ur review project work? HEY ULTIMA! Are you waiting on me to install mysql?? hcf; Good, i'm trying to work out a way to do partial updates of the SQL db when tril gets MySQL set up on bespin THATS WHAT I THOUGHT Tril heh I am going to install Slink on bespin Tril: Eek. Ok. π ultima/#tunes is considering mysqldump and diff ;) but to do that I need to hook a linux box up to a internet-connected LAN. And the box can't be Bespin. Fun stuff I guess I'll take my 486 into nas.com where bespin is Tril: what day is this stuff happening? I didnt decide yet I just thought of it what is mysqldump abi: what? mysqldump is a program which just dumps a SQL database what are you considering it for 12:50pm >>> Tril [dem@bespin.ml.org] requested PING 918075137 739460 from #TUNES better idea, I'll burn a cd of slink 01:00pm ωνω binEng [bineng@dialup58-4-25.swipnet.se] has joined #tunes hey binEng! hi bineng, how would you make the entire tunes site use a consistent style? cut n' paste mainly you would be cut-n-pasting the navbar? ωνω core [core@core.suntech.fr] has joined #tunes erm keep the navbar in a frame use a target frame ultima: quiet :) the navbar, page header and colors thers a million ways to include the same navbar what header hi tril, hcf, everyone :) cripes, fare isn't here? yo core! hoy core sorry noticed your icq message way too late :) tril: no, didn't he have to go to Reflection'99? i think he's back? he just wrote me the most stupid mail :) last I heard he was writing an article that was overdue hmm.. sounds like him :) 01:10pm he's apparently pissed that it's taking me time to release the whole thing as opensource :P binEng: OK. Now what we need to do is write a program to do whatever cut -n- pasting you would do. By reading colors, headers, and navbars from separate files, and outputting the final pages. Tril: how many pages are there, all in all? core: What exactly are you waiting for? tril: other people :( binEng: that's not the point. the point is we might change the style, or have multiple styles. So it must be automated. core: me? i heard Tril was used red hat now, and hates it passionately tril: first i haven't heard for 3-4 people i sent it to, and i really wanted their feedback tril: well, you gave some feedback, not as much as i'd like, but more than a few others :) core: did I send you the list of typos I found in xcom.txt? and, as Fare calls them, "grammeros"? tril: that's 1); the 2) is that suntech want to be ready when the opensource storm comes, so some of it can shine back on them... i'd be unnice to deny it to them, they've been funding it for a year now binEng: but, there's about 60 html files tril: ok, enough to justify some automation. tril: no, but i have no pride, so be sure to send them.. probably it'll be smaller to send me the list of nontypos :P Tril: I just got your ICQ msg. A bit late :) Tril: what program/etc did u hav in mind for the automation mechanism(s)? In xcom.txt: Fare's email in is out of date :) line 158: "If any of the limits is overcomed..." bad English: use "If any of the limits are exceeded..." line 412: "underlying" should be "overlying"? that's it :) binEng: their servers must be running slow, core just got a msg from me I sent at the same time. I shouldn't count on them! hcf: shell script should do fine tril: that's it? in 1000something line of text written by me? who are you kidding :) Tril: How's your OS model going? Anything you want to discuss? Tril: ok, what do u want for "features"? hcf: you're the one to write the page-generating script? binEng: im trying to stimulate conversation aha :) core:hehe bineng:if I put off upgrading bespin's OS, I could get more time to work on it. then ultima would have to install mysql himself for now. binEng: im not saying i will or wont do the script til i know whats involved tril: seriously, there are no other mistakes? tril: which means, for me? hcf: content should be in one text file, colors in another, header in another. The navbar should be generated from some text files, too. binEng: try to convince me to put off upgrading :) core: there were a few i missed I'm sure. I got into reading the document for its content, and stopped looking for errors :) navbar.txt => html => included in each page ? 01:20pm tril: does it make any sense? it's mostly a boring listing of what xcom does why not just navbar.html => included in each page ? the navbar is different for each page tril: but then again it's been National Emmanuel Bashing Day today so Fare just added the last stone to it and I'm a bit feeling like giving up :P binEng: oh hmm (part of the navbar, anyway) ωνω SignOff ultima: #TUNES (Ping timeout for ultima[user-37kb9sg.dialup.mindspring.com]) it includes path relative to the root, and the title of the current page we're doing 2 rite? overall and page specific core: Fare is always so negative about everything. You should ignore him. hcf: yes, I think ther should be one more: "On this page", with anchors to all the in-page headings. Tril: oiw, toc s/oiw/iow/ the page specific line should start with "Mirrors : USA : Top : Review : Languages" for example tril: i have that thing about respecting what he says usually ... i'llt ry core: first you have to translate it into the positive Fare says "linux sucks!" translation: "linux could be so much better, couldn't it?" tril: there's no positive in this mail :) basically he asks if he needs a license to write opensource code, and if i won't grant it to him, can i ever be so kind and reverse his patches :) tril: true.. or maybe linux can be no better than it is, which would mean that there's no hope for it I think the OTOP project should start to work on hacking up the linux kernel for reflection :) I'm inclined to believe that, at least. ultima didn't come back. bineng: the politics seem to be "don't change what works, even if it's stupid".. so .. i don't know. core:what is suntech doing? can't yo just add a blurb on the main nokernel.org page? ωνω linuxuser [linuxuser@bjas1-p29.telepac.pt] has joined #tunes core:that's just for the main tree. we could fork it :) tril: they've been giving money fulltime for me to work on clementine and release it as OSS.. they have a whole queue of customers for embedded applications, that'd be pushing the thing forward.. π binEng/#tunes suddenly smells Linux tril: there actually have been talks about that, from all the pissed people - GGI, ALSA, Linux Streams.. tril: i wouldn't have time to work on it just in my spare time - beleive me, i tried. and having people hammering it is nice i think... and good things are coming, hiring more people and such, and still putting it as OSS.. tril: Fare doesn't give me much credit though I'm pissed too, since Eberhard Monkeyberg wouldnt allow supermount in the kernel Someone knows how the opcodes of assembler works ?? Moenkeberg, I mean. I think that's who it was tril ;) linuxuser: simple, you give the processor an opcode and it executes it... :) monkeyberg.. *giggle* 01:30pm actually, it decodes it and goes "ouch" first, but yeah basically yah, I simplified π Tril/#TUNES still doesnt understand why clementine is delayed yeah, better not talk about the triple salto the x86 does when decoding the 66 prefix to new people :P tril: it's not delayed, i'm working on it very hard hcf: are you working on some script or what core: oh, so it's not ready code-wise? tril: no, it is. i'm just a perfectionist tril: i could release it today.. i'm just waiting on other people, and nobody seems to understand that Tril: im checking for pre-existing wheels first tril: in the meantime i checked a lot of new code in.. hcf: good :) I tried to make mov es,ax = 0x8E 0xC0 but it seems not to work! linuxuser: use an assembler, assemble what you want, disassemble it, look at the opcodes? dirty but would work core: did you get my bug report? what is the component that I couldn't list the interface for? tril: i did (me and my email .. :/), and i was a bit confused by what you said ;) core : I tried that really now, and with the bochs emulator log, the es don't content the ax value!! linuxuser: you probably reversed the operands then. core : I thinked that too, but I reversed them and they didn't work! linuxuser: then #2: it's not executing your code (or #3: your processor doesn't work at all ;) bineng: it's bochs, so it might be just true :) bochs panics on a keyboard disable command, so why not on that :P hmm 01:40pm no, my PMODE kernel works fine... core: I was running the "show component interfaces" command for all the components listed by "show components list" I had 7 components listed, and 6 of them were shuttle, vmm, cpumgr, pc keyboard, ism, and kgi. I was hoping you could figure out what the 7th was supposed to be, from a component I didn't list. pehraps, when I save the bytes on file, it simple don't convert the chars to the right codes!!!!! π binEng/#tunes was hit by the screensaver on linux that is a problem, writing ascii codes... core: When I tried to list the 7th component, it got in an endless loop core:the data it printed was all hex "Type"s scrolling by faster than can be read actaully, it was the 2nd component in the order in the list sorry, 3rd tril: hmm, did you list the right address? the ISM isn't really protected from mistakes at the moment.. i'll just throw it away for a forth interpreter RSN tril: that's weird, it sounds like the depmgr itself was missing core: that didn't seem too weird to me, since "platform" and "loader" were also missing tril: it should be the 3rd from the start - 1st would be shuttle and 2nd the vmm (there doesn't have to be an order, but at the moment it's like that) I didn't know what components were supposed to be resident tril: oh, platform and loader are exported by the shuttle, at least their current implementations :) tril: all at the moment - the depmgr should also list itself. if it doesn't i goofed something core: I guessed the first interface was the exported one, and the others were ones that component depended on. tril: the one in export { } are exported, the ones in import { } are imported.. you can export and import multiple interfaces without restriction s/one\ /ones\ // export/import aren't shown in 'compnents show interface', just in the source, right? tril: right.. the ISM is a bit raw, it just shows the node types I didn't look in the source :) 01:50pm tril: you can figure out what it is by the bitmasks in include/component/clemdep.h .. but it's temporary; i can't wait for the forth interpreter and soon a shell.. tril: there will have to be a filesystem of sorts for persistant storage.. it'll be transparent to components though tril: right now you can nest imports and exports but it's a fluke, i'll nail that down soon core: how difficult would it be to get JVM in clementine? (just curious) core: I wonder: you had to go through so many hoops to get the components to be separate, and bootstrapped..what's to say developers won't continue to have to go through such trouble to make their work componentizable? ωνω SignOff linuxuser: #TUNES (linuxuser) core: JVM because I am thinking of prism. But now that I think again, wouldnt embedded systems customers be interested in Java compatibility, too? core: I don't know if you know what prism is >>> Tril [dem@bespin.ml.org] requested PING 918079145 51304 from #TUNES 02:00pm I'm leaving. g'night. later! ωνω SignOff binEng: #TUNES (Leaving) tril: sorry, girlfriend talking :) tril: i guess it wouldn't be too hard to get, say, the Japhar JVM running.. most of what's needed is there; the graphics subsystem will be soon tril: i know what you mean.. that's a recurrent question in my mind - the hoops are mostly because i had to spawn components before their handler (the depmgr) is up. that's pretty complicated. you'll notice that the components launched after it, don't have any redtape - the cpu manager for one tril: there are things that could use thinking tho - there is no way to prevent loops for one; no component system i know (corba, dcom..) even tries to fix them, but.. ωνω Fare [rideaufr@esmeralda.enst.fr] has joined #Tunes Gakuk! 'lut fare π Tril/#TUNES punches Fare in the gut π Fare/#Tunes ouches That's for making Core feel bad about himself heh π Fare/#Tunes ouches again j'voulais pas, m'sieur l'agent! j'vous promets, j'recommenc'rai pas! et pis d'abord, c'est pas ma faute! π core/#tunes watches tril run for babelfish :) c'est la faute a l'aut', qu'est pas la! no, I'm just guessing it's Fare groveling, and I laugh core: found the bug? Tril: oh, apparently, the lispos list will survive w/o us. Tril: it'll just be a matter of changing the list address and things like that... Fare: When they move it, it should continue to be archived on bespin, as long as they move all the subscribers 02:10pm ωνω _vulture [vulture@152.173.54.171] has joined #tunes core: the main problem with Clementine from the tunes perspective is that it's a binary format; when we want system-wide source-level reflection. fare: i concentrated on the latter part of the mail :P _vulture: hi! <_vulture> hey Tril <_vulture> which IRC server is this anyway? _vulture: this is irc.openprojects.net (also irc.debian.org, irc.redhat.com, and irc.linpeople.org) <_vulture> okee they're all the same :) tril: i understand that. although your reflection will go through a machine code of sorts, whatever it is, and it can be (i think) packaged in xcom and managed by clementine.. and the existing binary code will allow you to boot, until it is gradually replaced by reflective code .. that's how i see it at least.. else basil wouldn't have tried for KUT either..? looking for classical texts, I found a site by a slave of technology: http://perso.club-internet.fr/ptaieb/ <_vulture> does tunes have any public *working* material yet? vulture: not really i fail one of the keywords, and it's not "working" (although tonight.. ) http://perso.club-internet.fr/ptaieb/index2.html ? Tril: horrible fare: were you serious in the mail? i can never tell with you.. hehe Tril: xcom may be too low-level in its current state; however, the good thing is, the format is completely hours to hack and modify ours hours is time! fare? well, fare is utter or a sick little pervert core: It was a ultimatum. Should you fail to comply within three seconds, 234 thousands, and 703219 nanoseconds, we shall invade your place. fare: sigh 02:20pm abi: damned! Everyone knows, now! s/ands/andths/ ωνω SignOff _vulture: #TUNES (Read error to _vulture[152.173.54.171]: Connection reset by peer) fare: seriously, i don't know how to understand your mail. it's national bashing emmanuel day today, yours was just the final touch core: if you mean "will I seriously ask for an official permission", the answer is yes, but not until more than one week fare: no, more like, do you seriously want me to reverse the patches? i would, but it makes me feel a little terrible. what is this about? do you have to give up your copyright to contribute to clementine? tril: no core: national? fare: apparently fare: it must be a national fest i'm unaware of, but it sure worked. core: or national, national movement? tril: i'll retain all copyright on what i do, like all contributors. suntech just want to pull some fame out of having funded it for a year almost now.. core: no, don't reverse them. so what does it have to do with fare? tril: their main problem is that they are VERY SLOW to do anything. I'll ask for an official authorization from FT. That may take time, but there is no reason why they would refuse. sigh ωνω _vulture [vulture@152.202.123.106] has joined #tunes That shouldn't prevent you from publishing things After all, I submitted patches to the Tunes site, too! their theory (which is quite stupid) is: the day i release it, people will enquire about suntech so they want suntech to look all shiny and pretty for that day so they asked if pretty please i could bang my head on the opposite wall a few more days, weeks, whatever <_vulture> before my computer crashed, I was going to ask... does tunes need any programmers ? what is suntech? noone really knwos knows vulture: sure <_vulture> Fare: okay... would you like me? _vulture: the LLL project can use programmers. Many people want to go ahead coding (and are) vult: sure, why not? There is *plenty* of things to do. <_vulture> and what do I need to do then? Tril: how about a prototypes dir on the tunes ftp? What are your current specialties, and what would you prefer to do? fare: of course i have no clue when i'll make the whole thing available to the entire world, so you might as well stop trusting me now, if you still do. i don't know. i'm tired of this. <_vulture> I can do nearly anything on many different processors, and I usually prefer x86 assembly, but anything will do we need everything, from OS implementation to Language design to web programming to protocol verification to whatever you want <_vulture> just name it.. what are you severely lacking in, and I can probably help hcf: I'll give access to anyone who wants to put their prototype on the tunes ftp vulture: if what you want is to hit the iron, you may work on Clementine and/or on the OS-kit <_vulture> Fare: how so? yeah, the os-kit is public and not all the weight of it stands on one person :P _vulture: at this point,you can contribute to clementine (talk to core, here), prism (see http://www.teleport.com/~sphere), TML (contact s720@ii.uib.no), or Retro (tcn@tunes.org) vulture: or you may work on a persistent memory management. these are all separate projects and you may help in any way you can 02:30pm although hopefully they will all merge together at some time. <_vulture> what would you rather I work on? i sure hope they merge vulture: an interesting project might be making RScheme hit the bare metal with the OS-kit. _vulture: maybe work on making Prism or TML run on Clementine! vulture: what subjects would interest you most? which language do you like, Java, C, or assembly? <_vulture> assembly the most vulture: I suggest you study memory management the three of them are a bad answer :P and implement a flexible memory system that allows for intercommunication between real-time and persistent processes clementine is probably the most functional iron-hitting prototype at the moment, but then again maybe the entire thing is stupid, for all i know. <_vulture> what is "iron-hitting" <_vulture> ? vulture: low-level vulture: talking to the hardware <_vulture> gotcha or more like hitting his programmer with iron today. core: do you know what prism is? prism is a specification & compiler which attempts to ease programming by allowing the programmer to express programs in a form that closely matches the programmer's conceptual model of the program. thanks, abi. de rien Tril <_vulture> is there a kernel yet, or are all of these projects acting like kernels? _vulture: right. We'll probably use one of these as kernels to start with. <_vulture> okay vult: also, I suggest that you learn either FORTH or LISP or both. <_vulture> alrighty and also Scheme or Prolog or Mercury, to understand kinds of things we'd like to do that completely challenge the C programming model. (Scheme is a lisp dialect with call/cc) what is call/cc continuations? Fare: are you saying he should come back 3 years later when he knows all those languages? call-with-current-continuation Tril: not at all. 1) he can do it in parallel 2) it takes much less time to learn them <_vulture> I'm a quick learner 3) it will give him a perspective on the kind of things we're trying to implement RScheme, CMUCL, and Mercury have nice implementation notes, btw. <_vulture> do you have some sort of web site I can goto to learn any of these? FORTH is very simple to learn. Review/Languages.html π KeLp/#TUNES should learn forth 02:40pm then I can do some rad sun eprom hacking look for FORTH, RScheme, Mercury, CMUCL rad? rad what glossary should I look up for that word? _vulture: www.tunes.org/Review/Languages.html, he means Fare: 1980's slang? a unit of absorbed dose of ionizing radiation equal to an energy of 100 ergs per gram of irradiated material rad short for radical , slang for cool, groovy, awesome, far out oh. Btw, that works on recent Mac G3s, too. or maybe he meant R.A.D. = Rapid Application Development, as in Delphi or C++ Builder I meat rad as in cool :) dope sweet groovy etc troll: C has continuations but not escapes or whatever nevermind <_vulture> bbl bye vulture troll? usenet faq no, I mean, who were you answering to? i was just making a random flame-bait comment you troll: Posting deliberately inflamatory remarks to a newsgroup for the sole purpose of stirring up controversy or trying to see how big a thread will ensue. Obvious trolls, like "you are a bunch of sickos" should be ignored. oh, so that was rather C has continuations ya 02:50pm not an irc nick. that was confusing. well, somehow, all languages have continuations, but C hasn't got full-fledged *first-class* continuations (though setjmp/longjmp provide with some kind of dynamic-extent continuations) c doesnt have 1stclass arrays either. it's for "efficiency" :) wrong answers. fast. what? a slogan for C. ahh "wrong answers. fast." also, if C++ is the answer, what is the question? "How do I allow OOP while keeping backwards compatibility in a procedural framework?" C++ is good at OOPses, that's for sure <_vulture> most C++ implementations are slow, as compared to older C implementations add to tunes glossary? OO= the first two letters of the phrase "oops!" as in "oops! I spilled my coffee!" also with OOP / 3 letters oops i set my crotch on fire then the question is, "if C++ is good at OOP, what's OOP to begin with?" "(and what is it for, besides making fun and money out of stupid people who follow OOP gurus)" _vulture: false, dynamic binding only has a constant overhead with respect to the size of the class hierarchy, as opposed to smalltalk where the overhead grows with the class hierarchy Tril: uh? c++ virtual functions only have a constant time factor, even though they are bound at runtime 03:00pm ask the dude who wrote c++ he'll explain it that's purely an implementation issue, trading space for time. <_vulture> umm, C++ is slow lots of space for a bit of time. it referred to the original implementation, I think <_vulture> newer computers are faster, allowing new C++ compiler programmers to make slow code the problem with C++ is not the implementation, but the semantics: it's brainfucked. <_vulture> and, they DO make slower code >>> Tril [dem@bespin.ml.org] requested PING 918082934 914137 from #TUNES π Tril/#TUNES is lagged, and has to go to school anyway ωνω SignOff _vulture: #TUNES (Read error to _vulture[152.202.123.106]: Connection reset by peer) π Tril/#TUNES is away: (NOT HERE! email me) [BX-MsgLog Off] 03:10pm ωνω SignOff iStormy: #TUNES (Ping timeout for iStormy[rain.futuresouth.com]) ωνω iStormy [stormy@rain.futuresouth.com] has joined #Tunes hum pizza is ordered π iStormy/#Tunes needs to learn some strange languages too...he has only touched c, c++, pascal, python, a grasped the basic concepts of asm iS: afterwards, it seems that C++ and Pascal are the strange languages I admit I haven't even looked at python. my language is sso far structured somewhat like python in syntax, though with changes to make typing eaesier but before i nail down the language i wanna learn some new ways of thinking 03:30pm python is c-style, with a more natural syntax a prototyping language designed for quick jobs iS: the syntax is irrelevant what matters is the semantics (well, as far as syntax goes, python is kind of convenient, but this is a petty detail) syntax is obviously irrelevant to language designers...the people who have to use the language have a different viewpoint multiple syntaxes doesn't change your way of thinking (besides than by teaching you not to give any importance to syntax) iS: given proper program editors, they do not care about syntax either. i have a note about that somewhere...one moment 03:40pm well, i won't look for it, but it's basically about configurable editors allowing some leeway of programming style and automatically converting to the global style before saving/reading back.. hi core hi istormy π iStormy/#Tunes waits impatiently for his pizza. π Fare/#Tunes is reading JS Mill's Autobiography (e-text) instead of writing his article :( π core/#tunes is still up at 1 AM for no apparent reason what else is new. iS: that's exactly what I mean about syntax being utterly irrelevant core: copieur! core: t'es qu'un sale recopieur, et j'vais l'dire a la maitresse! fare: en tout cas la mienne tu auras du mal, elle dort profondement (no pun intended). ca fait 1H qu'elle envoyait des NMI pour que j'aille dormir et que peut-etre demain soit un meilleur jour, elle a abandonne. pizza's here! big event of the day? fare: how serious were you about this mail earlier? i know i'm probably ridiculous being touchy-feely, but what else is new. yep my mind is way more limited than yours, so i don't get all the subtleties. core: how serious? if you don't abide by my prescriptions, I'll invest your place with a shotgun, and put holes all over your body! core: t'en fais pas. fare: c'est impossible chez moi ca :P core: juste que tu me donnes envie de faire pareil a FT, sauf que j'ai pas commence la moindre demarche. fare: d'etre paye a faire de l'opensource? fare: le gros inconvenient, c'est que tu peux decevoir des gens qui attendent apres toi, quand tes gros cons de manager, sont des gros cons de managers pas rapides. s/ger\ /gers/\ / 03:50pm mais quand tu es comme moi avec des capacites mentales relativement dans la moyenne et peu de temps en dehors du boulot, y'a pas trop le choix si tu veux bosser sur ce qui te chante. oh well. hey, i know that word! l'opensource core: d'avoir le droit, au cours de mon travail, de contribuer a des projets sous GPL iS: :) iS: are you salaried? i'm a househusband istormy: sorry, i had a pretty harsh day, so i'm complaining in my native language rather than translating .. don't mean to be rude my wife works istormy: gee, that's an idea. i should get my girlfriend to work on expensive proprietary useless code, so i can stay home and improve clementine all day without anyone pissing me off. Lots of beautiful things on J. McCarthy's homepage! core: is she computer scientist or engineer? yeah, that john mccarthy has quite a page he also invented LISP :) fare: well, she knows her way around the unix command line, and she likes pico better than netscape, so she is above all other women i know.. but no she's [striving to become a] journalist π KeLp/#TUNES is away: (home) [BX-MsgLog On] s/pico/pine/ core: oh! fare: it's that bad? well, she can't be any worse than what we have today as "journalists" well, she's full of illusions paper-peeers? where's my french-english slang dictionary, when I need it!? she knows that a media is just a tool for corporations and that absolutely no article published in a major media is sincere but only accompanying the release of a product, but still, she wants to do that 04:00pm and she uses HP/UX too :P oh, *technical* journalist, at that? well, she *really* can't do worse than we have today, then! yeah, well, she manages to keep me interested all throughout rather long papers, so she is good where does she publish, if anywhere? economic/technical part of midi libre at times, but nothing tremendously famous yet. hence the 'striving to' part :) and she's patient with me too, she knows how to translate 'tomorrow' into 'next year' on her own :P tril said he only found 4 typos in my xcom boring doc, i can hardly believe this. oh oh. If you're going to go to sleep "tomorrow", then it's a long time ahead to wait, then! I found none. But then I admit I only overlooked them. well, more like grammos well, you're french :) logically you would do the same :) 04:10pm me, french? can you prove it? [took me five months to prove it, once] anyway, do you think it's worth it that i go on with clementine, knowing that i'm probably a few weeks (in terrestrial time) away from public release, do you think you'll still grant me your insight, and give me some trust? i don't even know if i deserve that. yeah, you told me about that. this country has been strongly influenced by kafka i don't know, i have a friend who supervises the network of the gendarmerie nationale, maybe she can look you up :P !^lilo:*! removing sex spambot gline for infoave.net....please do a /whois and save the output for us, if you see the spam bot again ouch yeah, national 9600 bps radio network, a beauty :) !doogie:*! /whois spambot == spambot: No such nick/channel || /whowas spambot == -:- spambot There was no such nickname bps: coups de Baton Par Seconde ? 'xactly.. probably faster than wanadoo. suntech have a contract with FT in montpellier, they are plain scary No we aren't. We are just a monopoly ;-> you're scary :) All that you say will be recorded, and used against you. they have an incidents database there.. they managed to make one without any relations just seperated tables so most information is in the base 3 or 4 times.. that defeats the whole point :) it runs on top of NT4/IIS, and it serves win 3.1 clients w/ netscape 3.. i wonder how they add the incident that the server blew up. You mean they're not using MS OS/2 1.2, like everyone else in FT? Hum. They have a pretty modern installation, then. yeah, NT4 on a pentium 75 is a real blast 04:20pm One of my colleagues, two doors away from my office (when I come), is responsible for the new information system of FT, but he's having a hell of a bad time with internal bureaucracy BTW, do you know any security expert who could assess security requirements for FT's information system? bun then again i managed to cram nt4 server, oracle web server and database, macromedia generator, ie4, the flash activex control, and a few websites, on a p90 laptop w/ 32 MB of memory so the marketing people could do their marketing thing, *away from me*. He tries to introduce Linux, btw ouch security.. hmm, i don't know, i could always ask the people i know at the CEA, who did theirs the people at the CEA are more clued than i thought, the guy we saw runs debian on his box at work. I met a guy from CEA in Montpellier, in Brest met lots of interesting people there learnt a lot, too not a bald guy with poor taste for clothing i hope, or it'd be the one i know :P including *real scary things* about intellectual property laws, in France and throughout the world. yeah, good thing bosses aren't aware of IP like in the USA yet i got away with things you wouldn't imagine, at mirus notably core: at FT, they try to be :( not too surprising core: as I said, everything you say will be used against you! FT people send faxes like "we take your offer, even though it's the most expensive" fare: bah, i'll take a cegetel line and just modulate 300 bps on their line with my voice core: uh? what are those faxes about? FT can sleep safely; #2 (siris) and #7 (cegetel) are funny at the moment fare: oh, when suntech was chosed for what they asked #2? never heard about them. Are they any active? we tried siris at work, there's so much echo, you'd think you're on undernet yep, for companies at least.. it's the Siris network - one of the oldest operators in france, IP wise at least ISDnet run mostly on siris POPs what did FT contract suntech for? i don't see how you guys can converse so well in your non-native language...i've taken latin, russian, and spanish, and the best i even managed was "hello, how are you, where is the library?" Point-Of-P? fare: redoing their incidents database; and another stupid thing i forgot "hola, como estas? donde esta la biblioteca?" fare: i don't take care of that, that's the SS2I part, with a whole bunch of idiots donde esta la plaza de catalogna? and that's all i can say. right, point of presence 04:30pm my life at suntech is: i arrive at 9:30 AM, go joke with a few people i like, then lock myself into the office and work on clementine all day. "ave. bonus es? ubi bibliothequa?" π iStormy/#Tunes cackles...a joke website mentioned on familymatters is real (bigbooty.com) i've been told i look like jean-michel jarre because of the 4 screens and 5 keyboards on my desk :P core: on my dell laptop, I can suspend-to-disk, boot clementine, reboot, and resume-from-disk to the place I left Linux! fare: cool! (as long as clem doesn't mess around places of the harddisk linux or the S2D use) fare: on my Dell desktop, i can type "bochs" in my xterm, boot clementine in an X window, see debug output *on the console i started it from* (i patched bochs a little bit, to display what i write to an unused isa port..), then reboot and go back to the xterm fare: i'll have proper debug output soon; as well as a serial driver, and a few other goodies.. if you still want to trust me for a few more weeks.. core: did you contribute your patch back? fare: i sent it, along with better keyboard handling; i hope it gets included, it's extremely useful core: looks like your kbd driver didn't not wait enough for things to happen fare: yeah, probably - timing isn't its strong point fare: actually i know how to fix it properly [i think], but it breaks bochs (well, not with the patch) core: reportedly working drivers whose source I read all go at quite some lenght making sure that the kbd controller is not used too fast yeah.. i have the freesbee and linux driver sources always open in a corner core: if bochs groks linux, why not just reuse the linux way of accessing the keyboard? fare: that's where we enter science fiction. bochs doesn't mind linux sending him a keyboard disable (0xF5) opcode, but it minds when clementine does. that, i don't get. (well, maybe the redhat floppy i booted under it, doesn't do that - not sure.) core: timing? sequencing with other commands? what about the way you reenable? maybe.. somehow bochs doesn't grab the opcode when linux sends it i need to look at it into greater detail of course if you want to have a peek, you're welcome.. the pc is a piece of shit in general.. i really love the fact that COM4 and vga share some i/o addresses 04:40pm thanks IBM "engineers" right.. there had to be several. no single person can do that much idiocy not even me hehe "it takes only one man to ruin a good design, but it takes a whole army to achieve a complete disaster such as the PC hardware design" exactly :) no single piece of hardware in the PC can be trusted running Tunes on top of that isn't evident - even if your code is proved to be bugless, you have to account for the hardware. hence the utility of such things as the OS kit. far from a clean piece of code, but it would be difficult to stay clean while touching the PC hardware. well, the OS kit doesn't help you more than a bare linux kernel does well, it's more modular *shrug* i had a look at it, it's COM-modularity whereas Linux ties things to, say, its memory management and process scheduling code ie. it only works in-process sure well, it's an huge limitation true, the linux kernel has things like __get_free_page() all over the place. *shrug* well, you're welcome to use clementine and to hack it all over, it'll be public in a few weeks, oh well ok, the thing i understand the least about tunes: how can you implement a reflective system, if part of it isn't? (the LLL namely) you *always* have some non-reflective axioms, be it only the bare hardware. (or the FPGA programming protocol) well, you'll always have a thin piece of software on top of that at least hardware is mostly designed with the assumption that software will nail bugs down some even can't boot without it, eg. cyrix mediagx chips and the VSA the question is "can you dynamically define the limits, or is there someone having made a static choice for you?" exactly - if you use the oskit, clementine, whatever, there'll be an hard limit between the LLL and the HLL? those bugs are part of the hardware specs. not so how? we hope that an open compiler design will blur the limits between HLL, LLL, asm, hardware we hope to have a continuum appear, instead of fixed limits 04:50pm well, you'll have to rewrite the LLL itself in the Tunes language? (well, in a tunes language i mean) indeed well, i suppose you can do that gradually indeed okay, then it's clearer that'll be pretty funny to do with the OS kit, as modules are not physically independent. you'll probably need some bridging code down to stupid C conventions, that'll you need to duplicate for every module, so that it can link, until you got rid of all C code, so you can then remove all the glue altogether, maybe not too pretty hopefully, we can achieve as much as possible of the glue through automatic generators from a declarative description language or you can implement this glue once for all in say, the dependency manager the depmgr will be dynamically producing code anyway ωνω SignOff hcf: #TUNES (Leaving) ok, i'm starting to feel sleepy finally.. i'll go to bed, maybe tomorrow (well, today) will be a better day and people will find someone else to bash good night. I'm going, too π Fare/#Tunes is away goodnight, fare 05:00pm ωνω SignOff core: #TUNES (I can understand one word out of ten! - http://www.tunes.org) woohoo, firefox is on tv 05:10pm π KeLp/#TUNES is back from the dead. Gone 2 hrs 12 min 30 secs ωνω SignOff Fare: #TUNES (Ping timeout for Fare[esmeralda.enst.fr]) ωνω hcf [nef@escher.sdi.agate.net] has joined #tunes ωνω SignOff KeLp: #TUNES (changing servers) ωνω KeLp [tcole@feynman.physics.wwu.EDU] has joined #TUNES ωνω Tsung [rkirby@194.125.3.246] has joined #TUNES ωνω Tsung [rkirby@194.125.3.246] has left #TUNES [] ωνω KeLp is now known as KeLp_bak ωνω KeLp [kelp@xws083.xtrn03.wwu.edu] has joined #TUNES ωνω SignOff iStormy: #TUNES (iStormy has no reason) π KeLp/#TUNES is away: (trek) [BX-MsgLog On] π KeLp/#TUNES is back from the dead. Gone 0 hrs 9 min 54 secs ωνω SignOff hcf: #TUNES (Leaving) ωνω SignOff KeLp: #TUNES (Read error to KeLp[xws083.xtrn03.wwu.edu]: No route to host) ωνω KeLp [kelp@xws136.xtrn03.wwu.edu] has joined #TUNES ωνω PoopMan [patrick@207.159.73.153] has joined #TUNES yo! what is this channel? check out www2.tunes.org ok ok i went to that page and i still have no clue :) 10:20pm well, if one of these guys who is actualy involved in it would wake up its an OS project but a very strang one I don't really understand it my self 10:30pm ωνω SignOff KeLp: #TUNES (sleep) ωνω KeLp_bak is now known as KeLp π KeLp/#TUNES is away: (sleep) [BX-MsgLog On] ωνω SignOff PoopMan: #TUNES (Leaving) [msg(TUNES)] newlog 1999.0204 IRC log ended Thu Feb 4 00:00:00 1999