12:20am hmmmm......oh man....I'm really out of it now...I'm trying to visualize memory as geographic shapes to visualize other ways to manage the memory hahahhaah i think i may have hit the best way... :) I was thinking trees.... the I thought....well....if you tree it....it'd help to visualize it sideways... so it turned into a triangle... then I was thinking...maybe a tree where it starts at the lower addresses, and one starting at the highest going lower..... so it turned into a ring 12:30am wow.. I dunno I have a lot more studying up to do before I really have to worry about it hhhahaha.... * billyboof/#tunes remembers that time... ;-P what does nop stand for? no operation ah i forget which one exactly, but it's the same as either: or al, al or maybe: or ax,ax since that does nothing see you guys later it just or's itself.... thanks for the help billy later nice talking to you folks! goodnight -:- carlito [sabanmr@137.28.142.238] has left #tunes [] night ruiner: yeah.... if ax=5, what's 5 or 5 ?? :P 5 yep... whats 5 or 5 not? or ax, ax doesn't modify any registers or flags... i have to figure out my new memory map.... right actually.....wouldn't 5 or 5 not be 5? bad example...heh heh no wait it wouldn't it'd be....7 how do you figure?? 5 = 1001b.... do the binary math.... 1001b or 1001b..... =1001b no. 5 is 101 5 not is 010 oh.... hehehe or them and you get 111 or 7 who said anything about not I did...scroll up oh... i missed that.... heh heh start end contents 0 05FF bios (1 page, 400h free) 600 0607 memory size (free after init) 1000 17FF idt (1 page, 200h free) 2000 9FFF 8 free pages 10000 1FFFF GDT (16 pages) 20000 22FFF startup code32 23000 2FFFF 12 free pages 30000 4FFFF PAT (32 pages) 50000 9FFFF 80 free pages A0000 12:40am A0000 0FFFFF ROM (80 pages) ----------------------[1 MB boundry]----------------------- 100000 100FFF page directory 101000 108FFF page table 0-7 109000 1FFFFFF free space --------------------------[PAT]---------------------------- shit did that come up right? heck...I wouldn't know right from wrong I hate working in hex hahaha i see one thing wrong with my memory map already 2000 9FFF 8 free pages should be from 2000 to FFFF 12:50am hmmmmmmm 01:10am I'm gonna go night ok... talk to you later -:- SignOff billyboof: #TUNES (hatefull@antisocial.com) -:- SignOff ruiner: #TUNES (Leaving) 01:50am -:- smoke [smoke@195.96.122.107] has joined #tunes hello 02:00am hum 02:20am -:- SignOff smoke: #TUNES (Read error to smoke[195.96.122.107]: Connection reset by peer) -:- smoke [smoke@195.96.122.240] has joined #tunes -:- BORT [JAC@infovia249-54.dcec.cica.es] has joined #Tunes -:- Yo [MO@infovia249-54.dcec.cica.es] has joined #Tunes DO YOU KNOW J?CL -:- Yo [MO@infovia249-54.dcec.cica.es] has left #Tunes [] -:- BORT [JAC@infovia249-54.dcec.cica.es] has left #Tunes [] 05:10am -:- BORT [JAC@infovia249-54.dcec.cica.es] has joined #Tunes -:- BORT [JAC@infovia249-54.dcec.cica.es] has left #Tunes [] -:- Tomy [JAC@infovia249-54.dcec.cica.es] has joined #Tunes hola abi niihau, Tomy guau abi -:- SignOff Tomy: #TUNES (Leaving) 05:30am -:- SignOff Fare: #TUNES (Connection reset by pear) -:- AlonzoTG [Alonzo@client-151-200-120-8.bellatlantic.net] has joined #tunes -:- Fare [fare@balance.wiw.org] has joined #Tunes y0 06:20am -:- SignOff AlonzoTG: #TUNES (Have Nice Day :)) -:- SignOff smoke: #TUNES (One day sheep will rule the world) -:- iStormy [stormy@rain.futuresouth.com] has joined #Tunes hi lo 08:00am -:- smkl [sami@MCCXXXIV.rdyn.saunalahti.fi] has joined #tunes -:- mith_gone [mithris@t2o22p54.telia.com] has joined #tunes hello -:- mith_gone is now known as mithris 09:30am hi heh..:D i didn't think any1 was alive.. could you tell me a little about tunes? how far work has come? i dunno...i'm not associated with tunes, i just hang out here sometimes www.tunes.org has lotsa info hmm okay more info than you could read in a week yeah..:D thats the problem.. i wanna know more about one specific piece right now.. 09:40am noon1 else alive? 10:00am * Fare/#Tunes is dead, but hush, don't tell anyone fare: could you tell me a little of the state of TUNES? 10:10am mithris: what do you want to know? smkl: well... is there an official TUNES version? or is there only prototypes? 10:20am only prototypes. there will never be "official" versions how far has is the stdlib project? any system that fullfills TUNES requirements is TUNES mithris: nowhere humm.. okay... has any1 started working on it? i don't think so. there is not much point because HLL doesn't exist is any1 worling on the HLL project then? woeking gah..:D working Fare, Tril and BRice how far has they come? there is nothing usable, too much design problems :/ 10:30am cya -:- SignOff mithris: #TUNES (Leaving) 10:40am -:- AlonzoTG [Alonzo@client-151-200-127-217.bellatlantic.net] has joined #tunes -:- SignOff AlonzoTG: #TUNES (Ping timeout for AlonzoTG[client-151-200-127-217.bellatlantic.net]) -:- smoke [smoke@195.96.122.246] has joined #tunes hi. hello smoke why are you on #TUNES ? 11:40am smkl: i'm interested in the project smkl: trying to read up what it is about 11:50am -:- ruiner [nate@ppp314.wi.centuryinter.net] has joined #tunes hello ruiner hi hey, ruiner hi smkl? abi has nothing on you yet 12:20pm ruiner? i think ruiner is Nater1464 -:- SignOff ruiner: #TUNES (Leaving) 12:30pm -:- Tril [dem@sloth.wcug.wwu.edu] has joined #TUNES -:- mode/#Tunes [+o Tril] by ChanServ -:- AlonzoTG [Alonzo@client-151-200-127-13.bellatlantic.net] has joined #tunes -:- SignOff AlonzoTG: #TUNES (Have Nice Day :)) -:- ruiner [nate@ppp398.wi.centuryinter.net] has joined #tunes greetings 01:20pm -:- SignOff ruiner: #TUNES (Read error to ruiner[ppp398.wi.centuryinter.net]: Connection reset by peer) -:- ruiner [nate@ppp114.wi.centuryinter.net] has joined #tunes grrrrrr 01:50pm -:- SignOff ruiner: #TUNES (Leaving) hum 02:10pm -:- SignOff smoke: #TUNES (Ping timeout for smoke[195.96.122.246]) -:- somiador [vet234900@i178.infovia.xtec.es] has joined #Tunes hello somiador hello!!!!!!!! -:- SignOff somiador: #TUNES (Leaving) 02:30pm -:- SignOff smkl: #TUNES (sleeping ...) * Tril/#TUNES is back from the dead. Gone 15 hrs 59 min 23 secs -:- NetSplit: clarke.openprojects.net split from koontz.openprojects.net [03:20pm] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [clarke.openprojects.net] -:- Netjoined: clarke.openprojects.net koontz.openprojects.net -:- abi [nef@bespin.dhs.org] has joined #tunes -:- Tril [dem@sloth.wcug.wwu.edu] has joined #TUNES -:- mode/#Tunes [+o Tril] by ChanServ -:- tcn [tcn@cci-209150250104.clarityconnect.net] has joined #tunes hey core, fare, tril.. blah what's up? hmmm... up is above down haha so is anyone here? * Tril/#TUNES is present 03:50pm lemme guess, you're trying to fix the IRC logging no what's wrong with it? too many files in one dir? no, it just forgot to start a new file today -:- ON listings: * JOIN " TUNES #TUNES * " umode -s NORMAL 0 * MSG " TUNES NEWLOG * " set logfile $2 SILENT 0 * MSG " TUNES PERMLOG * " exec chmod +r $2 SILENT 0 * PUBLIC " * #TUNES * " @activity=1 QUIET 0 * TIMER " %:?0 " baz SILENT 0 * TIMER " *00:00* " exec -msg TUNES date +"newlog %Y.%m%d" SILENT 0 * TIMER " *00:01* " exec -msg TUNES date +"permlog %Y.%m%d" SILENT 0 .......................LOGFILE 1999.0513 * iStormy/#Tunes stretches. * iStormy/#Tunes dislikes functional languages. what DO you like? would you rather have a broken language? Tril: i thought that was an amusing choice of names too (c: declarative is ok :) gakuk! tunes will be the first fully functional OS. Fare did you fall asleep or what? you asked me a question and then ... iStormy: the choice of name is not entirely coincidental, functional gurus DO think other languages are broken. (with good reason!!!) no, went for dinner at 4:20 am? oh, you mean yesterday? I guess I fell asleep you said "what is the purpose of this infrastructure?" but was more like 5 am and i was wondering if you read my response i figured. istormy: Maybe you could answer the question, what do you like? istormy: or, what would you like? Tril: there's Fox already as first functional OS I wonder if their code is online... -:- ruiner [nate@ppp184.wi.centuryinter.net] has joined #tunes also the proprietary LISP OSes of old IRC log ended Fri May 14 15:57:51 1999 -:- Starting logfile 1999.0514 IRC log started Fri May 14 15:58:38 1999 -:- Value of LOG set to ON Tril: who knows...i find python to be the most usable, but it's still not my dream language. part oof my problem is i don't know anything other than imperative languages. trying to broaden my horizons at the moment by fiddling with other stuff like smalltalk and lisp. is the most important part of os creation the hardest or easiest? as for the infrastructure, I believe it should be codeveloped with either practical code or theoretical concerns (or both) ruiner: I can't answer, for I'm not sure I understand the question, and even less that I have enough information to answer if I understood it well, design is the most important part, but is that harder or easier than actually coding? ruiner: they both tie together.. about the same, I think. tcn: OK I fixed the dumb irc log ah 04:00pm (what good is a design, if you can't code it?) what good is being able to code, if you haven't defined the problem and how you're going to go about solving it? ruiner: because the idea of the problem either doesn't fit in your head, or isn't expressible in words what good is being able to code if you spend all your time writing documents. (c; heh heh -:- _QZ [brand@p0wer.qzx.com] has joined #tunes ruiner: One thing I've learned: Don't try to design everything perfectly before you code it well written code in a badly designed system is worse than poor code in a well designed system -:- AlonzoTG [Alonzo@client-151-200-126-46.bellatlantic.net] has joined #tunes tcn: i haven't learned that yet...i'm a perfectionist. )c: i've never finished anything in my life. not even mention of the code on the foxnet server <_QZ> geez, linux has finished 2 work units and windows is only 30% done with its first unit work unit? rc5? is this that benchmark? <_QZ> seti@home QZ: same hardware? _QZ the windows one only runs when the screensaver pops up can anybody explain to me how exokernels are faster than other kernel types? ruiner: are they? <_QZ> the radio telescope collects 35gig of data per day and breaks it up in 300k work units Hey, my quad AXP21264@700 under Linux is faster than my 386sx16 under WindowsNT! according to the tunes page <_QZ> Tril: it also has the option to run full time ruiner: aegis is faster because it merges application and system code into one ah ruiner: AND it is able to co-optimize them <_QZ> Fare: linux is on a celeron463 and win98 is also on a celeron463 cripes ruiner: well, a microkernel is slow because it runs separate tasks for each kernel driver right qz: are the drives the same speed? Fare: how does that compiler work? is it incremental? did you read the glossary entry on microkernels? ;-> how about monolithic kernels? <_QZ> they are both udma nope Tril: no, they have domain-specific languages for block device access I don't really see what slows them down monolithic kernels are just harder to manage <_QZ> tcn: but it would not be affected by drive speed in which you can code your access patterns as well as your filesystem consistency policies <_QZ> tcn: the program can keep teh 300k in ram and its all FFT processing qz: oh what's the principle of seti@home? to seek out & destroy new life <_QZ> win98 has 128meg ram and no apps running, linux has 192meg ram and 30+ apps running QZ: maybe the Win system is doing lotsa MMX<->FPU switches? Fare: Analyze radio signals from space to detect extraterrestrial life * ruiner/#tunes yawns the question is: what kind of "analysis" could possibly detect extraterrestrial life??? well, I'm out of here....see you later folks -:- SignOff ruiner: #TUNES (Leaving) <_QZ> seti@home utilizes the millions of computers at home to help process the data faster doing dummy calculations faster won't lead you anywhere <_QZ> Starting Gaussian curve-fitting <_QZ> Starting work for Doppler shift rate -7.571536, FFT length 2048 it does FFT's to make the data understandable FFT won't make *any* data understandable I hope they use FFFTW, btw (the Fastest FFT in the West) <_QZ> http://setiathome.ssl.berkeley.edu/faq.html what about the East? Does Russia have a better FFT? ;) (developed by people at MIT, I think, using ad-hoc metaprogramming techniques, IIRC) <_QZ> Fare: go read that page Agreed fare rtfm I could donate some time on my 486 to SETI :) <_QZ> hahaha 04:10pm <_QZ> tcn: windows or linux? all the computers in this lab are already running rc5.. hehe <_QZ> it took 9 hours to process one work unit on a linux celeron463 what does rc5 prove? <_QZ> i calculated it to take 23 hours on my win98 celeron463 that RSA ain't that bad after all? om all computer at home are already running random text generation, and analyzing the results to see if it's better than Shakespeare... If I had any internet hours and it worked on win3.11 I'd do it. :( but for now I have to go. <_QZ> u dont need internet time but first: any comments on the specs I published? <_QZ> u download a 300k work unit and then get off the net try coding it fare: that'll never work. :((( <_QZ> when the unit is done it gets another QZ maybe alonzo needs new units sent on disk via snail mail <_QZ> :) i wonder if I can subscribe to electronic mailing lists by floppy disk.. -:- SignOff iStormy: #TUNES (iStormy has no reason) -:- iStormy [stormy@shell.futuresouth.com] has joined #tunes afk <_QZ> this super soaker pen isnt a very good pen to write with -:- SignOff AlonzoTG: #TUNES (Have Nice Day :)) 04:20pm does redhat call their os "Redhat GNU/Linux"? how that "by floppy disk" ? we could snail mail floppies with the list archives. I remember using a floppy to transmit mail back in the old times before ppp was available at school since I wasn't root to push things in the sendmail queue, I just had elm -s here-scripts -:- water [big.dog@206.81.201.20] has joined #tunes I tried closure. Its rendering engine is impressive. But there's a bug in picture handling, and the user-interface is mostly inexistant, apart from the bare minimal sirvival commands hi all hey i was wondering when you'd show up.. :) brian? it took long enough. yes, it's brian. brian! where are ya? I admit I was confused by your arrow paper ina computer lab on the navy base fare: really? what about in particular? too much motivational things, and too little technical content I understood a few things you said earlier, but I still don't have the whole picture ok. what sort of technical points would you need? * Fare/#Tunes has bought MacLane's "Categories for the Working MAthematician" ah. i see. i'm sorry that i didn't spend enough time explaining undeerlying theory. water: No you're not :) man, I have a ways to go before I'm up to your level in math (fare and water) do you agree that KL-Reflection paper is prerequisite for reading the Arrow paper? then I should write that on the web page Tril: good idea.. also mention Cybernetics tril: yes, although i did cover the basic ideas. KL is actually a pretty common idea in AI research. knowledge representation, yes but reflection? 04:30pm I didn't understand everything in the KLR paper either tril: well, the reflection part of it is just as new as my own model-level idea. ok, just checkin * Fare/#Tunes knows a top AI researcher in Reflection, but he's retiring next year, and focusing on using his already reflective system as a tool for higher-level problems, rather than improving it * water/#tunes is now downloading a new version of BETA and has just update his Squeak. fare: really? can Squeak save individual programs? (as opposed to the whole image) tcn: sure, you just export a class (or a set of classes) to a file. tcn: not afaik you may export a class, but then, it won't save the dynamic data, only the stati source, isn't it? fare: yes will that work for your Arrow thing? tcn: well, i'm making an set of classes for arrow environments that will be self-contained, including editors and examples the biggest hurdle (very small, actually) is implementing graph specifications for lazy evaluation. 04:40pm I just installed cmucl, now trying to get Garnet to run on bespin so I can try it (graphical stuff for lisp) cool anyway, there should be some pleasing code by tomorrow late. WTF is #+string ? what relationship is there between your arrow system and category theory, if any? what are arrows meant to represent, to begin with? is there any _exact_ relationship you can draw with something that I might know in theoretical (or practical) CS? Tril: conditional compilation under common lisp Tril: #+toto (foo) will read (foo) iff the compile-time feature toto is defined ok then what is the compile-time feature indicating a linux system? #+linux ? not any that I know it's trying to detect a version fare: graphs are categories that can reflect, and do not assume relationships among arrows. brice: I'm not sure I understand at all your kind of arrows I understand the basics of category theory fare: arrows are just objects whose only attributes are two references. fare: graphs are just 'sets' of arrows. 04:50pm how do you store information (besides relations)? tcn: my approach is to cast all information in terms of relations (sort of). I've been recommended the book "Category Theory for Computing Science, 2/e " but can't find it at either Prentice Hall or Amazon water: like peano's axioms, right? tcn: sure. water: are arrows composable, as in cat th? why talk about "arrows" and not "relations", then? fare: yes, but composition is a predicate, not necessarily a function or operator. fare: relations (first-order) are non-degenerate graphs. you mean, g o f may have several values? non-degenerate? fare: or be undefined. be undefined, it can in cat th hehe.. look at this: You can download a binary archive of William Walter Patterson's port of INTERCAL 0.15 for DOS, which finally has the programming language it truly deserves. rofl but cat th mostly assumes usual (deterministic total) functions fare: yes, but cat th doesn't treat it as a predicate (formula, etc). although its arrows can describe more than that water: cat th has many predicates; it's just that its basic constructs are total deterministic functions is the INTERCAL compiler itself written in INTERCAL? fare: in the arrow system, the composition object is just that: an object, not a primitive. Fare: I don't want to know suddenly, scp and cvs-over-ssh don't work for me! Fare: works for me (update tunes) * Fare/#Tunes doesn't understand what's wrong ssh -v shows me it connects; but afterwards, data doesn't pass, or extremely slowly mere ssh works Fare: is this with bespin? yes Fare: The only thing that should affect that is the reverse DNS changed from bespin.cx to bespin.dhs.org. although it looks like other hosts can connect w/ bespin w/o problem 05:00pm but connection *does* complete; it's afterwards that it doesn't work ssh is working fine for me.. check your setup, fare. ssh1, right? Fare: RhostsRsa authentication not available f or connections from unprivileged port. hum. I removed the pcmcia NIC, inserted it back, and now it works do NOT ask me why it works. It just does. Fare: stop using port 64357 and use 1023 err. never mind that's probably for different services perhaps the NIC is unhappy with changing from an IP network to another (home, office, school, other school) Fare: I dunno if this is related, but sshd generated a new RSA key 15 minutes ago I think it does that all the time, though maybe it fucked up that shouldn't matter, afaik (not the first time a reset of the pcmcia nic enhances network response) water: what are primitives? are they turing equivalent? can they express the whole logic? fare: primitives are objects in a language that cannot be effectively identified by language structures. (btw, please stop pinging me!) afaik, when you write cat th in an algebraic way, you actually have something essentially isomorphic to higher-order typed lambda calculus just why couldn't they be identified by language structures? you can *always* write things in an algebraic fashion even if awkward * Fare/#Tunes doesn't ping water * Fare/#Tunes is sleepy just look at the VM's for our favorite languages. what about them? 05:10pm they all contain instructions which are "protected" or "necessary". dunno what you mean but they sure contain primitives there is no logic without axioms their semantics aren't encapsulable by the language. Tril: can you check whtehr you can find the book anywhere? ISBN 0-13-323809-1 Michael Barr & Charles Wells Category Theory for Computing Science, 2/e * water/#tunes is now downloading his favorite old apple-II games. water: no logic w/o axiom that's an unescapable reality. See Goedel's theorem. water: where'd you find them? The failure of Hilbert's dream water: can you get Prisoner II to run? fare: goedel's theorem is about first-order languages! fare: i don't have that one. I have a copy, but either it is buggy, or my a2 emulator is fare, looks out of print.. water: ftp://ftp.apple.asimov.net/pub/apple_II/ maybe a used bookstore http://www.asimov.net/apple_II/site/ fare: i'm there used bookstore? 1996! ok I found a used copy want it? $45 out of print at amazon.com hum. $45. How many FRF is that? ... well, looks an ok price (expensive, but if the book is good...) maybe I should double-check the recommendation about the book? I've seen *lots* of other titles. -lilo(lilo@varley.openprojects.net)- [GlobalNotice] Please welcome verne.opirc.nu (leiden.nl.eu.opirc.nu) back into the fold, courtesy of Wichert Akkerman and the Leiden University Free Desk Collective :) !wichert:*! You're very welcome lilo :) !wichert:*! And online just before I'm leaving on vacation as well... checking currency exchange, does 276.75 FRF sound right? Ive got 6.15 FRF to a dollar sound right ok, let's go for it 05:20pm water: there's a a2 emulator for linux, but I suspect it be buggy, for many games break also, the X mode is broken. hmm one of the first apps I want for Tunes is an apple2 emulator hehe at least, the first emulator I want, before PIG gotta go to bed. later by fare any question/comment/explanation? ok. Bye, then! * Fare/#Tunes is away hehe