IRC log started Mon Mar 6 00:00:02 2000 [msg(TUNES)] permlog 2000.0306 -:- SignOff smkl: #TUNES (Ping timeout for smkl[glubimox.yok.utu.fi]) -:- SignOff air: #TUNES (http://www.qzx.com/ :: sleep) -:- Closing Link: TUNES[bespin.dhs.org] by lewis.openprojects.net (Ping timeout for TUNES[bespin.dhs.org]) -:- Connection closed from irc.us.openprojects.net: Success -:- Use /Server to connect to a server -:- Connecting to port 6667 of server irc.us.openprojects.net [refnum 0] -:- BitchX+Deb1an: For more information about BitchX type /about -:- Welcome to the Internet Relay Network TUNES (from varley.openprojects.net) -:- Your host is varley.openprojects.net, running version u2.10.05.18.(ipcheck4-5) (from varley.openprojects.net) -:- This server was cobbled together Wed Apr 28 1999 at 12 02:19 EDT(from varley.openprojects.net) -:- varley.openprojects.net u2.10.05.18.(ipcheck4-5) dioswkfcg biklmnopstv -:- [local users on irc(36)] 5% -:- [global users on irc(330)] 45% -:- [invisible users on irc(400)] 55% -:- [ircops on irc(16)] 2% -:- [total users on irc(730)] -:- [unknown connections(0)] -:- [total servers on irc(30)] (avg. 24 users per server) -:- [total channels created(261)] (avg. 2 users per channel) !varley.openprojects.net Highest connection count: 58 (53 clients) !varley.openprojects.net Welcome to Open Projects! You are on 3 ca 1(2) ft 14(14) tr. -:- Mode change [+f] for user TUNES -:- Mode change [+iws] for user TUNES -:- JOIN activated by "TUNES #tunes tunes@bespin.dhs.org " -:- TUNES [tunes@bespin.dhs.org] has joined #tunes -:- Topic for #TUNES: TUNES Free Reflective Computing System http://www.tunes.org/ || Slate Language http://www.tunes.org/~water/slate-home.html -:- topic set by water [Tue Feb 29 21:13:22 2000] -:- [Users(#tunes:4)] [ TUNES ] [ smoke ] [ abi ] [ Plundis ] -:- Channel #tunes was created at Sun Feb 28 08:48:06 1999 -:- BitchX+Deb1an: Join to #tunes was synced in 7.108 secs!! -:- Mode change [-ws] for user TUNES -:- smokie [tw026024@zaalf13.twi.tudelft.nl] has joined #tunes -:- SignOff smokie: #TUNES (BitchX-75+Deb1an -- just do it.) -:- Fare [rideaufr@esmeralda.enst.fr] has joined #Tunes -:- smkl [sami@glubimox.yok.utu.fi] has joined #tunes -:- AlonzoTG [Alonzo@216.164.131.57] has joined #tunes -:- SignOff AlonzoTG: #TUNES (Have Nice Day :)) -:- SignOff smoke: #TUNES (Ping timeout for smoke[16dyn188.delft.casema.net]) -:- smoke [smoke@16dyn188.delft.casema.net] has joined #tunes -:- thomas [thomas@193.217.63.152] has joined #tunes -:- AlonzoTG [Alonzo@216-164-138-152.s152.tnt5.lnhva.md.dialup.rcn.com] has joined #tunes -:- rares [rares@wtrb-sh4-port142.snet.net] has joined #tunes -:- SignOff AlonzoTG: #TUNES (Ping timeout for AlonzoTG[216-164-138-152.s152.tnt5.lnhva.md.dialup.rcn.com]) -:- SignOff smkl: #TUNES (My damn controlling terminal disappeared!) -:- smkl [sami@glubimox.yok.utu.fi] has joined #tunes -:- AlonzoTG [Alonzo@216-164-138-152.s152.tnt5.lnhva.md.dialup.rcn.com] has joined #tunes -:- rares [rares@wtrb-sh4-port142.snet.net] has left #tunes [Segmentation] -:- SignOff AlonzoTG: #TUNES (Have Nice Day :)) -:- SignOff smoke: #TUNES (Ping timeout for smoke[16dyn188.delft.casema.net]) -:- smoke [smoke@16dyn188.delft.casema.net] has joined #tunes -:- SignOff thomas: #TUNES (Read error to thomas[193.217.63.152]: EOF from client) -:- rares [rares@wtrb-sh8-port166.snet.net] has joined #tunes >>> rares [rares@wtrb-sh8-port166.snet.net] requested PING 952382186 from #tunes -:- SignOff rares: #TUNES (KVirc 1.0.0 Millennium BETA 3) -:- eihrul [lee@usr5-ppp38.lvdi.net] has joined #tunes -:- tcn [r@cci-209150250172.clarityconnect.net] has joined #tunes hey hey 03:10pm what up homey? 03:20pm just pondering how to handle memory mapping -:- SignOff smkl: #TUNES (Ping timeout for smkl[glubimox.yok.utu.fi]) 03:30pm hmm how about "don't"? not my decree :) what are you working on? dolphin ah i've been pondering whether memory mapping is really necessary the answer is no not really :) not even for Tunes especially not for tunes not with proof security I was worried about persistent storage.. then I realized that's what Forth blocks are doing, without memory mapping hardware 03:50pm it's crude but not slow hehe wish I'd had Forth on the apple ][ -:- rares [rares@wtrb-sh1-port41.snet.net] has joined #tunes those floppies were slow, with files hey rares hey eihrul, what's a good lostsouls race deryni, invae, nyloc, sekh, danpiru everybody hates everyone else in thios place 04:00pm well, it's an individualist mud that's nuts not individualism but makes quake look easy -:- SignOff rares: #TUNES (Ping timeout for rares[wtrb-sh1-port41.snet.net]) 04:10pm -:- rares [rares@wtrb-sh6-port155.snet.net] has joined #tunes are these still text muds? yup text muds rule show me a graphical mud that let's you rip a guy's leg and attack him with it 04:20pm hehe that's good to hear I used to play a little so is that what all these people in here do all the time? hehe probly obviously you and eihrul :) actually i'm just getting back into this crazyiness heh i forget how addictive it is i mean, i could die without accomplishing a damn thing in the real world :) umm that's good the world needs less fixers like me i should be locked up 04:30pm trying to save the world are ya? -:- water [water@tnt-9-94.tscnet.net] has joined #tunes 'lo anyone want to discuss any ideas? hey (not just slate stuff) persistent object stores? sure what about them? are you working out some ideas with them? gakuk trying to simplifiy things fare! t c n ! hey fare water! how ya been? * water/#tunes picked up 'On Lisp' finally brb (taking a shower) k will you be there in 10 minutes? definitely cu then he'll be quick, they only get 4 minutes of water in the US navy :) um tcn I'M in the navy, not fare oh yeah ;) 4 minutes a day, is that right? for the most part, yes anyway, persistent stores? hehe.. takes discipline 04:40pm yeah, um.. maybe I could do a simple POS in forth yeah forth is always good for a simple implementation of an idea no use of MMU hardware.. just forth's block i/o * water/#tunes reads tcn's post to the mlist store the objects in blocks.. have a table of objects, with block#, offset, length, and 'dirty' flag good, mmu reliance i've heard can cause a performance penalty, to say nothing of the loss of portability right.. guys like Chuck Moore can make chips, but MMUs? I dunno, probably too much work for him :) have you been able to look at 4os's code, btw? 4os, no.. where? guh i don't think so abi 4os 4os is, like, a FORTH OS at http://www.itvc.com/Technology/4os.htm oh that :) yeah *that* :) looking... i thought you'd be all over the 4os stuff ok was it more or less than 10' ? i wasn't keeping track... check the logs btw, on current hardware there _is_ already the MMU, so we pay a large part of the penalty, even if we don't use it, so that using it is marginally low-cost, for good usage ratio 04:50pm fare: it can be turned off. for memory-intensive tasks, doesn't the overhead go way up? now, if I had to design new hardware, I would investigate a lot beforeto implement a full MMU in hardware. Most likely, the end result would be basic functions (lookup, bitops, rollops) to access programmable cache hierarchies tcn: the overhead of software validation isn't trivial either. I have got no figures either way, so I can't tell. We have to experiment so as to know. * water/#tunes thinks he should just stick to hll stuff :) My uneducated guess is that since we already pay most of the price, using MMU might be desirable on current hardware. water: he who does not watch his feet is likely to trip :) or something like that anyway, my opinion is that this is all preposterous optimization before we get to optimize such behavior, let's implement it at all eihrul :) there's no point in overoptimizing a tiny part when the rest is not here yet hm squeak ported to dreamcast and playstation 2 8) now, this is a very interesting part indeed to investigate later fare: i'd like to do it without the MMU, then see what sort of hardware would be useful water: does that mean 3d morphs will run at acceptable speeds now? :) eih: they're working on that right now tcn: that's most sensible tcn: be portable first, be optimized on special hardware later yeah eih: i gave a whole lot of thought to slate over the last 2 days eih: i'm not going to be very forthcoming about the specs until i have them all ironed out well, what have you decided and on what? btw, the apple ARM palmtops did their object marking in software. and i'm most likely going to change some of the stuff slightly that we'd already decided on -:- SignOff rares: #TUNES (Read error to rares[wtrb-sh6-port155.snet.net]: Connection reset by peer) just make sure they're changes for the better (although that shouldn't be hard) eih: be nice (:P) fare: a persistent *object* store would definitely be easier without an MMU * eihrul/#tunes ponders the Pinky Processor. hmm 05:00pm and reading 'On Lisp' helped me sort out just exactly what lisp is good for and why i don't want slate to be clos Retro has gotten close to Tunes lately tcn: s/easier/more precise/ We don't know how much precision matters fare: more efficient if you're not low in memory, less precision can be a good efficiency tradeoff water: well, what are you siginifigantly changing with slate now, though? :) anyway, I agree that object-wise marking is a good strategy in absence of MMU. eih: mostly moving the primitives around a bit eih: and cleaning up the syntax ;) and I agree we should avoid the MMU to begin with water: not touch the sexp though? eih: of course not, i've got a rewrite system to lay on top of it ;) btw, if you're going the persistent way, it is foremost to differentiate objects from projects (in more traditional terms, values from objects; or in more elaborate terms, extensions from intensions) that's where C++ goes _way_ awry fare: hm fare: if the programs use non-contiguous memory, they won't need an MMU, and there'll be no MMU to get bogged down by 'random' memory accesses. water: shape and data? :) fare: is there a good paper on extensions vs intensions in oo or fp? (and maybe Java, too, although it looks like you can do the right thing in a contorted way, in Java) eih: kind of, yes ok, I wanna see the Lisp discussion :) (Java can do a lot of things right, in very contorted ways) non-contiguous memory? as in, several memory banks? tcn: well, what about cache? one reason to differentiate extensions from intensions: ones can be duplicated/thrown away, whereas the others must be committed. sure this is a particularly large hit in parallel/distributed processing eih: cache is finer-grained water: what design for the pattern-matching language? fare: huh? tcn: so rmm, how many more days till retro3? :) rewrite systems are easier with pattern matching eih: heh.. 3 days of work :) fare: well sure fare: most likely an open design I guess an ML-like (or mercury-like) pattern-matcher is a good start eih: FYI, I spent the last 2 days setting up a decent Linux install having the matching infrastructure backtracking-friendly helps, too fare: i'd prefer mercury-like by default tcn: ouch fare: nothing about extensions vs intensions in papers, then? "decent linux install". Ain't that an oxymoron? eih: it's the forth of linuxes :) tcn: cLIeNUX ? yeah, a customized clienux had to do some scripting and stuff is there a shorter word for "retrospective summary" ? fare: review? 05:10pm tcn: agree to put cvs and/or rsync in a cron job? fare: never! :) tcn: nope, like, in my architecture1 paper, I summarize my general system factoring at end of part 1 bash: rsync: command not found tcn: rsync is a must. Unless you install unisson (written in OCAML) instead rsync is kinda handy an ftp script would do just as well tcn: know a Russell Fish? for my small files tcn: rsync is much more resource-friendly than ftp fare: nope url for rsync? rsync.samba.org apt-get install rsync anyway Slate, Tunes, and Retro? Slate--> Tunes <-- Retro sure If Slate's based on Lisp, and we add Scheme to Retro, umm, that's great water: what about starting with a pattern compiler as the basis for the implementation? * water/#tunes shrugs sure that would work right now i'm concentrating on getting the semantics worked out properly maybe starting from the mercury implementation notes (did they document pattern matching specifically?) water: keep us tuned k abi mercury? somebody said mercury was purely declarative logic programming language at http://www.cs.mu.oz.au/research/mercury/ or to Prolog what Haskell is to Lisp or to Prolog what Haskell is to Lisp -:- SignOff smoke: #TUNES (z) hm my bookshelf is starting to look like a cs phd's haha actually #advanced cs books = #advanced math books well, my floor looks like an undergraduate degree :) interesting the mercury fellows are working on a self-hosted compiler cool hm but it's implemented in prolog I played with prolog a bit.. looks like fun it's always seemed like a toy to me 05:20pm i mean, have you ever heard of someone using *prolog* as an extension language? water: uh? wasn't that long ago before they bootstrapped it? mercury is now fully self-hosted (well, above gcc) i'm looking to find clearer info one problem w/ mercury is it went the SML/NJ memory hog way (all the more as it calls GCC as a backend) hm yeah gcc back-end ok prolog was the absolute first implementation (they should really update some of their site) hm here's an appropriate statement: "using GNU C as a portable assembler" :) well, that's common knowledge eih: btw, i figure out that very often people will want to move around object's meta-behavior along with their behavior, to varying degrees 05:30pm eih: except that it makes GC a HELL (check the gclist archive about it) (it's archived on lists.tunes.org well, C-- is a better portable assembler :) is meant to be I don't think C-- really exists at the time being plus, I greatly dislike many of C-- options oberon? screw C--, we have Forth an imperative infrastructure for the implementation of declarative languages, this sounds fundamentally wrong to me heh oberon is imperative. Bad. Plus they have split-encouraging proprietary licenses yeah, FORTH is a reflective language; good for implementation of reflective systems i'm just now realizing that Forth is almost Tunes except that ANS FORTH really sucks a lot yeah, but that's not REAL forth :) uh sure tcn: did I tell you that my first enlightenment as to what a system could and should be was due to the HP28 RPL ? hehe * water/#tunes tries to contemplate the dictionary structure of a large-scale distributed forth system fare: you have said they're cool I learnt FORTH afterwards, because I had been told RPL was derived from FORTH; but I always preferred RPL to most of FORTH. water: the dictionary would be used for everything! tcn: exactly :P fare: RPL has lists, right? yup one of the meanings of the L in RPL was "LISP", I had been told 05:40pm hm i remember learning to use cadforth and autolisp although again, I wouldn't learn LISP until (much) later water: cadforth? was that in autocad? no, it was in cadkey, iirc At that time, I didn't see books about non-ancient LISPs, anyway. i had access to a lot of cad packages when i was younger water: any preference? -:- lar1 [larman@dialup-209.245.129.111.SanJose1.Level3.net] has joined #tunes tcn: not anymore they all suck now hehe i thought lisp would be better for cad it is or at least RPL autocad 10 years ago was awesome it sucked when they went to windows today it's like coca-cola * water/#tunes nods 2000 is a little better than 13/14 they've removed functionality in exchange for multimedia -:- rares [rares@wtrb-sh4-port171.snet.net] has joined #tunes so how do I fix it rares: take it to #f-cpu, please -:- SignOff eihrul: #TUNES (/quit) anyone seen ColdForth? not i allright fare, I got rsync installed :) now I need ssh1, goddamn pain in the fucking ass 05:50pm you don't have ssh1? not in clienux oh rick says it compiles fine under clienux though 13/14 ? under Debian or something I really had to hack ssh1's guts out to get it compiled goto 3053 water: openssh.com, not ssh1 fare: autocad r13/r14 Which ssh do I get??! ssh1 is proprietary crap. yeah i suppose openssh is ssh1-compatible free software (derived from ancient free ssh1) (made by the security-paranoid openbsd folks) most likely that's what i have -:- SignOff Fare: #TUNES (Read error to Fare[esmeralda.enst.fr]: EOF from client) -:- Fare [rideaufr@esmeralda.enst.fr] has joined #Tunes gr damn it, my book on sub-structural logics seems like it will never get released anyway. Good Gnight! s/my book/the book i bought/ Fare did you see f-cpu later fare 06:00pm -:- smkl [sami@glubimox.yok.utu.fi] has joined #tunes -:- air [brand@p0wer.qzx.com] has joined #tunes heh.. Netscrape or Internet Exploiter -:- SignOff Fare: #TUNES (:3053) 06:10pm hm the discussion definitely went south would anyone care to try again? what topic? bad movies? slate, tunes, or retro or something related rares: i decided to use those stupid parenthesis :) okay seriously it's the kind of attitude that will move your game forward 06:20pm heh -:- SignOff rares: #TUNES (KVirc 1.0.0 Millennium BETA 3) air: lisp? 06:30pm -:- water_ [water@tnt-9-94.tscnet.net] has joined #tunes -:- SignOff water: #TUNES (Read error to water[tnt-9-94.tscnet.net]: Connection reset by peer) -:- water_ is now known as water arrgh, need SSL for SSH I told fare this was gonna be a pain in the fucking ass 06:40pm tcn: no not lisp but now it looks like lisp for brix? openspace list-based, then like euphoria 06:50pm christ! now I need perl5! (FUNCTION1 x y (FUNCTION2 x y)) like that its logo with parens hm i'm gonna shoot someone! Heh but kinda not logo cuz i dont use logo words * water/#tunes hands tcn a pump-action shotgun s/shoot/strangle/ (more satisfaction) * air/#tunes hands tcn a rope hehe larry wall here I come! nah, shigawire's much better :) ya but that would cut and kill em faster, rope will make em feel it longer hm true Make him pass out, wake him up, repeat you may as well go the torture route yeah, these bastards need to squirm awhile tie the rope around their neck and to the back of yer vehicle and drive em up and down the street at a few mph. if they try to stand up gun it so they fall back down tie it to their ankles Ouch s/ankles/balls/ hah Double ouch all cause of bad hacking :) if u had some kinda rope tied to yer balls and were being dragged u would beat yer skull into the ground to kill yerself fast i'll be on #modtunes, guys I would just take the rope off... ;) water: Ok, bye 07:00pm lar1: along with your nuts haha heh this really sucks! it doesn't just use perl5 for the configure maybe I can find an rpm binary that isn't libc6 -:- SignOff tcn: #TUNES (Ping timeout for tcn[cci-209150250172.clarityconnect.net]) -:- tcn [r@cci-209150250039.clarityconnect.net] has joined #tunes -:- ult [ult@user-38lc6b2.dialup.mindspring.com] has joined #Tunes Can somone please explain what the heck the DCMA is? hey Hey, ult abi, dcma? bugger all, i dunno, ult doh. I know its the Digital Melenium Copyright A But thats all a/A/Act hmmmm good god there must be an info site on it linked to from /. that would be a major overhaul lar: try www.loc.gov/copyright or just loc.gov Heh...my stepmom did part of the LoC web page... loc? Library of Congress? cool Hmm. 07:10pm Ok This freelance web crap sucks. ult: ? you stepmom does freelance web design and it sucks? i can see where you're coming from. It's bad enough just doing my own web pages.. I can't be bothered with graphical web pages, java, cookies and crap My stepmom does freelance graphic design. I do freelance web design. I'm poor. Well, I do freelance everything. Right now I'm doing web design. It pays much better than coding, ironically. but oh well. 07:20pm what's up with amazon? abi akos akos is, like, A Kind Of Special game or at http://tpp.dhs.org/AKOS.html 07:30pm -:- rares [rares@wtrb-sh2-port134.snet.net] has joined #tunes -:- SignOff ult: #TUNES (Ping timeout for ult[user-38lc6b2.dialup.mindspring.com]) abi until? water: i haven't a clue 08:10pm -:- hcf [nef@me-portland-us920.javanet.com] has joined #tunes hey hcf hey good lll discussion just finished on modtunes k some good stuff in tunes-lll today too -:- rares [rares@wtrb-sh2-port134.snet.net] has left #tunes [Segmentation] see ya all k -:- SignOff tcn: #TUNES (Core dumped) i'm hoping beh will show up tonight water: slate is mentioned here, http://www.ntlug.org/~cbbrowne/osnovel.html seriously? amazingly, yeah "which suggests that some portion of the "analysis paralysis" may be complete..." great quote he has another link to it 08:30pm well that was entertaining he must be subscribed to tunes@