IRC log started Thu Aug 12 00:00:00 1999
[msg(TUNES)] permlog 1999.0812
-:- SignOff liar: #TUNES (BRiX [http://www.qzx.com/brix] :: sleep)
-:- SignOff Plundis: #TUNES (Ping timeout for Plundis[chaosdev.org])
-:- Plundis [plundis@chaosdev.org] has joined #tunes
* Plundis/#tunes is gone. Gone since Tue Aug  3 18:38:00 1999
-:- SignOff eihrul: #TUNES (Ping timeout for eihrul[usr5-ppp73.lvdi.net])
-:- ElGato [dave@bespin.dhs.org] has joined #tunes
* ElGato/#tunes is away: (Auto-Away after 10 mins) [BX-MsgLog On]
<ElGato> fare: you're a math teacher?
06:50am
-:- SignOff ElGato: #TUNES (ElGato has no reason)
-:- hcf [nef@me-portland-us1029.javanet.com] has joined #tunes
-:- Crimson [crimson@chaosdev.org] has joined #tunes
<Crimson> hello...
08:00am
-:- hcf has changed the topic on channel #Tunes to: Reflective Computing System <http://www.tunes.org> | http://www.dilbert.com/comics/dilbert/archive/images/dilbert1993043090804.gif | http://osdevnew.listbot.com | http://www.researchindex.com
-:- hcf has changed the topic on channel #Tunes to: Reflective Computing System <http://www.tunes.org> | http://www.dilbert.com/comics/dilbert/archive/images/dilbert1993043090804.gif | http://osdevnew.listbot.com | http://www.researchindex.com | http://www.tsia.org
<hcf> abi: r5rs is The Revised(5) Report on the Algorithmic Language Scheme at http://www-swiss.ai.mit.edu:80/~jaffer/r5rs_toc.html or http://www.cs.rice.edu/CS/PLT/packages/doc/r5rs/
09:00am
!is:*! Hi all. Any #linpeople admins awake?
-:- smkl [sami@MCCCLXXVII.rdyn.saunalahti.fi] has joined #tunes
-:- SignOff hcf: #TUNES (Ping timeout for hcf[me-portland-us1029.javanet.com])
-:- hcf [nef@me-portland-us724.javanet.com] has joined #tunes
-:- ElGato [dave@bespin.dhs.org] has joined #tunes
* ElGato/#tunes is away: (Auto-Away after 10 mins) [BX-MsgLog On]
* ElGato/#tunes is back from the dead. Gone 0 hrs 1 min 33 secs
<ElGato> is anyone around?
* hcf/#tunes is
<ElGato> hello
<abi> hi, ElGato
<ElGato> hi abi
<ElGato> abi: good girl
<abi> :)
12:30pm
<ElGato> is there anyway to save your code in rsc?
<hcf> rsc?
<ElGato> rscheme
<ElGato> an interactive scheme interpreter
<hcf> i know what rscheme but havnt used it
<hcf> s/but/is but/
-:- water [water@ppp-tnt-5.tscnet.net] has joined #tunes
<ElGato> hi water
<water> hey
<water> good! someone looked at the TSIA site.
<hcf> water: i just put in there so someone else will
<water> oh
<ElGato> hey im full of questions
<water> k
<hcf> water: u get ur water@tunes.org mail right?
<ElGato> what are continuations and at that what are non-linear continuations
<water> hcf: yes, but right now it's set to forward to my work account
<water> elgato: continuations are the "rest of the computation" passed around by coroutines or threads in a lisp
<water> hcf: which reminds me, i will change that now
<ElGato> uhhh...k
<water> the lisp computation actually points to the cell containing the next lisp instruction
12:40pm
<water> like the Instruction Pointer in a processor
<ElGato> like the IP?
<water> hehe
<water> yep
<water> except it works for SEXP's
-:- eihrul [lee@usr5-ppp83.lvdi.net] has joined #tunes
<water> Call/CC is what allows a lisp thread to pass the continuation on
<water> i'm not sure what Fare means by nonlinear continuations, but i'm guessing it means those that can't be guaranteed to return (i.e. those that might hang)
<smkl> he probably means continuations that are used more than once
<ElGato> and when you say passing the continuatrion on you mean "incrementing it's 'IP'"
<water> explain
<water> elgato: actually, it's a high-level thing
<smkl> you can store the continuation and call it as many times as you want .. to get some really weird behaviour
<smkl> he posted about them to comp.lang.scheme and tunes mailing list
<ElGato> he's all worked up about them too
<water> that _would_ give some really weird behavior
<ElGato> so this continuation is like the state of the computation
<water> yep
<water> "nonlinear" behavior, even :) .  now i get it
12:50pm
<ElGato> and this state contains what?
<water> (not sure how to explain that, myself)
<ElGato> that's ok
<ElGato> i'll just take what i got
<water> get a lisp and try to use 'em
<water> they're supposed to be pretty powerful
<ElGato> well i have this online book that talks about "call-with-current-continuation" but doesn't really tell what it does
<water> that's Call/CC
<water> it passes control from the current thread to another one, and gives it the continuation as the second part of the parameter
01:00pm
<ElGato> if fare gets so worked up about trying to find a _good_ lisp to do the HLL in i think he should just do it in the tunes LLL
<water> i think that Scheme was _supposed_ to be the LLL
<ElGato> bah
<ElGato> scheme is way high level
<water> not enough so
<ElGato> not low enough to be called a  LLL though
<smkl> i think HLL was supposed to be made by starting with scheme and modifying it
<water> right
<ElGato> that's what the HLL page said
01:10pm
<water> hcf: nice research index
<hcf> water: yes quite
<water> hcf: is it just an index, or does it have any online material or links to online material?
<water> hcf: n/m, it does
* ElGato/#tunes is away: (Auto-Away after 10 mins) [BX-MsgLog On]
<water> this index rocks!!!!
01:20pm
<hcf> water: ur welcome (x 18)
<water> hehe
<water> "Imperfect Information Flow of Agents Communication in Arrow Logic"
<water> :)
01:30pm
-:- SignOff ElGato: #TUNES (Hey!  Where'd my controlling terminal go?)
<hcf> of parseable, parsable, parsible which is the correct spelling?
<water> parsible, i think. check an on-line dictionary
<hcf> i cant find an online dict that has it
<water> hmm
<eihrul> i vote for the first... but have no concrete evidence to back it up
01:40pm
-:- ElGato [dave@bespin.dhs.org] has joined #tunes
<water> wb
<ElGato> thanks
<ElGato> bonjour
<ElGato> fare aime toucher de petits garcons
<ElGato> oui?
<ElGato> Un de ces petits garcons est hcf.
<ElGato> :D
<hcf> eh?
<ElGato> im messing around with bable fish
<water> oops. i have to go. be back in an hour or two
<ElGato> bye
-:- SignOff water: #TUNES (Leaving)
<hcf> One of these little boys is hcf
<ElGato> try the top line
<hcf> le poing de hcf baise de petites asiatiques filles
02:10pm
<hcf> that didnt work right
<ElGato> hehehe
<ElGato> im gonna learn to read french
<ElGato> and i'll talk to fare
<hcf> urinates de hcf sur de vieilles personnes
<ElGato> hahaha
<ElGato> you urniate on old people?
<hcf> whenever possible
<ElGato> Je ne Pense pas que les personnes agrees apprecient le urinating sur eux.
<ElGato> put hcf between le and urinating
<ElGato> Oh mon penis faisant mal!
02:20pm
* ElGato/#tunes is away: (Auto-Away after 10 mins) [BX-MsgLog On]
-:- pyro [tcn@cci-209150250123.clarityconnect.net] has joined #tunes
<ElGato> hey pyro
<pyro> hey
<ElGato> what's up?
<pyro> nothin much.. see ya
<ElGato> D:
03:00pm
<pyro> checking out some linux software to see if it's better than the proprietary crap we're using at work
<ElGato> oh
<ElGato> im hacking retro :)
<ElGato> how large is the TSS register?  do you know?
<ElGato> i'd think 32 bit
<pyro> umm
<pyro> 16-bit
<ElGato> ok
<pyro> it's a GDT selector.. the descriptor points to the actual TSS address.  To switch tasks, you can just change the descriptor in the gdt.
<pyro> you've seen task.asm, I take it
<ElGato> yeah
<ElGato> ok
<pyro> allright, time to get my ass in gear and add DOS file support
03:10pm
<pyro> hey, I finally saw the Matrix
<pyro> it was cool, until the end
-:- SignOff eihrul: #TUNES (Ping timeout for eihrul[usr5-ppp83.lvdi.net])
03:20pm
<ElGato> is retro gonna be FAT?
<pyro> hell no!
-:- SignOff hcf: #TUNES (Read error to hcf[me-portland-us724.javanet.com]: Connection reset by peer)
03:30pm
<ElGato> uh...Denial of Service fs? (hehehe)
-:- eihrul [lee@usr5-ppp40.lvdi.net] has joined #tunes
<pyro> this is just a temporary expedient
<ElGato> so this temporary expediant will be FAT?
<pyro> yup.. it's just for loading Forth programs, XCOM modules containing asm/C/etc programs, and other data files (text, bitmaps, fonts...)
<pyro> that'll help me cut the umbilical cord to Linux
<ElGato> yeah
<pyro> edit the Forth programs from within Retro
<pyro> once I have an assembler I can compile the whole thing from Retro
<ElGato> :D
<ElGato> retro almost has FPU support
* pyro/#tunes is getting a GIMP
03:40pm
<pyro> feel the lag :)
<pyro> hey.. what lag?
>>> pyro [tcn@cci-209150250123.clarityconnect.net] requested PING 934499059 851716 from #tunes
<eihrul> hrmm, how bad is disabling paging temporary? really, really bad right?
<ElGato> where in the TSS can i store the FPU state?
<pyro> if you disable paging, programs with any translated pages can't run
<pyro> ElGato: before the tss
<eihrul> well, i mean temporarily
<pyro> yeah
<eihrul> it will be reenabled before any programs are run again
<pyro> sure, you can do that
<eihrul> does it flush the TLB?
<pyro> besides, each task has its own CR3.. 
<eihrul> i know it does...
<ElGato> pyro: uh like TSS - 94?
<eihrul> but i need to copy between those different page tables :)
<eihrul> eh?
<pyro> whatever
<eihrul> well, i just want to know what are the effects of disabling paging after paging is reenable?
<pyro> ElGato: Look at the new file, uvars.inc... those are the per-task variables
<pyro> eihrul: probably none
<pyro> try it :)
<eihrul> also, do you map your page tables into the task's address space?
03:50pm
<pyro> ElGato: You can optimize task switching by using 256 bytes for the TSS.. but it's 104 for the tss, 108 for the fpu, and 44 available for the OS
<ElGato> i thought it was 94 for the FPU
<ElGato> mabye 108 is for newer FPU's
-:- hcf [nef@me-portland-us834.javanet.com] has joined #tunes
-:- shamino [sham@phila-dialup138.nni.com] has joined #tunes
<ElGato> er...i gotta read up on the tss again
<ElGato> all i need to do now is figure out what fsave's opcode should be
<ElGato> ohhh...
<ElGato> i see
<ElGato> just save the FPU state at TSS plus 108?
04:00pm
<pyro> no
<pyro> I just changed uvars.inc
<pyro> I put the TSS at the end of the 256 bytes, so I can put an ioperm map in the next 2 pages (which can normally all be mapped to the same 2 physical pages)
<ElGato> so where does the FPU fit in?
<pyro> uvars.FPU
<ElGato> i mean like where in the TSS?
<pyro> right before it
<ElGato> which is TSS minus 108?
<pyro> yeah, which is also UVARS plus 44
<ElGato> mov ax, TSS
<eihrul> dern it... there's no information anywhere on disabling paging
<ElGato> sub ax, 108
<ElGato> fsave ax
<ElGato> ?
<pyro> nononono... for each task there will be a UVARS address.. the first 44 bytes at that location are system variables, followed by 108 bytes of FPU state, then the TSS
* ElGato/#tunes is a little slow
<ElGato> i see
<ElGato> so UVARS plus 44?
<hcf> ElGato: upgrade ur brain
<pyro> uvars.FPU
<pyro> uvars.fpu, I changed my mind. keep it lowercase..
04:10pm
<ElGato> ok
<ElGato> im guessing it's double words aligned?
<pyro> 256-byte aligned
<ElGato> no uvars.fpu
<ElGato> yeah i think thats double word
<pyro> that's because I just added that 2 second ago
<pyro> abi: 44/4
<abi> 11
-:- NetSplit: carter.openprojects.net split from irc.linux.com [04:12pm]
-:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [carter.openprojects.net]
<pyro> abi: 44%4
<abi> 0.616
<pyro> abi: 44 mod 4
<ElGato> abi: 256/4
<abi> 64
-:- Netjoined: carter.openprojects.net irc.linux.com
-:- fare [fare@quatramaran.ens.fr] has joined #Tunes
-:- shamino [sham@phila-dialup138.nni.com] has joined #Tunes
<pyro> abi: 4*11
<abi> 44
<pyro> do a cvs update, elgato
<ElGato> ok
<ElGato> uh...
<ElGato> where in cvs is retro btw
04:20pm
-:- water [water@ppp-tnt-7.tscnet.net] has joined #tunes
<water> back
<ElGato> wb
<pyro> you don't have cvs?
<ElGato> i do but i can't find retro with it
<ElGato> where is it hidden away?
<pyro> under retro
<pyro>  /cvs/retro
<ElGato> oh
<ElGato> ok
<pyro> on bespin
<ElGato> yes yes
<ElGato> nevermind
<ElGato> hey it's 4:20
<ElGato> im high on retro :)
<ElGato> where is print_string defined?
<pyro> i dunno
<pyro> ever heard of grep?
<ElGato> yeah
<ElGato> :(
<water> heh
<pyro> just so you're not tempted to ask me stupid questions, I'm going for groceries
<ElGato> :|
<water> lol
<pyro> btw, the 'sprint' macro is useful.. 
<water> gato: i'll answer whatever of your stupid questions i can. ;)
<pyro> like this:  sprint "blah...",10,"ljsdfja",10
<ElGato> and lemme guess, you don't know where that is either?
<pyro> it's %included somewhere
<pyro> it's in the kernel directory
<pyro> :)
<ElGato> heh
<ElGato> great
<ElGato> what's wrong with print_string anyway?
<pyro> anyway, it puts the string in the data section, and a call to print it in the code section
04:30pm
<pyro> print_string is only used in realmode
<pyro> it's defined in the boot sector
<ElGato> well im not in real mode yet i don't think
<pyro> yet? realmode is where you start
<ElGato> er
<ElGato> pmode rather
<pyro> it ain't THAT retro
<pyro> hey brian, you saw the matrix right?
<water> yep
<pyro> I finally saw it last night
<water> i think i saw it the 7th time just recently
<water> what did you think?
<pyro> oh, it was cool, until the end
<water> the end?
-:- Tril [dem@bespin.dhs.org] has joined #TUNES
-:- mode/#Tunes [+o Tril] by ChanServ
<water> what about it?
<water> tril!
<pyro> that, and the idea that you start bleeding for real when your virtual body gets hurt
<pyro> that's just plain stupid
<water> that's not bs
<pyro> maybe if you bit yer tongue :)
<water> your nervous system has electrical feedback loops built in
<water> yes, they probably do that, too
<water> are you there, tril?
<pyro> and god only knows why they were using people as batteries when they had fusion.. but that was pretty damn funny when morpheus holds up the battery 
<water> it's sci-fi, not sci-fact
<water> they had to make _something_ that you coudln't rationalize
<Tril> hi
<abi> hi, Tril
<pyro> hey tril
<water> what's new, tril?
<Tril> just messing with freebsd at my new jiob
04:40pm
<water> btw, if we had a vote about the tunes member with the most contributions, i'd have to vote for hcf
<pyro> what kinda job?
<Tril> tiredf from game playing
<Tril> unix stuff
<pyro> sounds like fun
<Tril> at the isp bespin is at
-:- SignOff eihrul: #TUNES (Leaving)
<water> any progress on the object-type system?
<Tril> i decided my scope is too small and that is why i'm not making progress
<Tril> i will have to diagram the *entire* tunes system
<Tril> dynamic compiler, user interface, machine model, ...
<pyro> and programs!!
<ElGato> i call hcf the official tunes knowledge distribution engineer
<water> you may be able to automate that diagramming process
<Tril> huh?
<Tril> using arrows, i suppose
<pyro> just get a big sheet of paper
<water> hehe
<water> no, not necessarily arrows
<water> i'm saying that you could describe generic models and then specialize them into the specific things you want to model
<Tril> not really following
<ElGato> pyro: um i can't make retro.  i thought you could now
04:50pm
<water> tril: well, you could work on describing the number system first, and use that and your basic types (function, set, collection, ...) to build up the common things like boolean logic or Pascal
<pyro> ElGato: why not?
<ElGato> should it be the cvs or the snapshot?
<pyro> cvs
<pyro> it makes for me
<water> tril: what makes your problem any more difficult than mine, then?
<ElGato> oh wait neverming
-:- SignOff smkl: #TUNES (sleeping ...)
<Tril> where is fare
-:- eihrul [lee@usr5-ppp14.lvdi.net] has joined #tunes
<Tril> abi: seen fare
<abi> fare was last seen saying something on IRC Tue Aug 10 15:10:19 1999
<ElGato> um
<ElGato> "1 error"
<ElGato> yep it won't make
<Tril> water: what do you mean
<water> tril: i see my problem as the same in nature for arrow, even after my theories have been completely formalized
<pyro> _what_ error?
<Tril> which problem?
<water> tril: having to 'diagram' everything in the system
<ElGato> options.default: no such file or directory
<Tril> is the emphasis on diagram, or on everything?
<water> both, i guess
<water> i'm just trying to figure out why your progress would slow down
<Tril> lack of concentration
<water> :(
<water> focus, David-san! :)
<Tril> brb
<ElGato> i am master miagi
05:00pm
<ElGato> :)
<pyro> ElGato: I see options.default there..
<ElGato> this sucks
<pyro> you're in ~dave/retro ?
<ElGato> cvs/retro
<pyro> on whose computer?
<ElGato> bespin
<Tril> ok
<pyro> dohh
<Tril> someone trying to compile *,v files?
<ElGato> is there something you did not tell me?
<water> hehe
<ElGato> ~dave/retro has just the snapshot
<Tril> elgato do you know how cvs works
<pyro> you're supposed to run cvs to check it out from /cvs/retro!!
<pyro> don't you ever rtfm?
<ElGato> Tril: heh, i guess not
<ElGato> :(
<Tril> the way cvs works is each user has their OWN COPY of all the files. The central repository (/cvs on bespin) is only suposed to be written by the cvs program
<water> lol
<Tril> (and read by the cvs prograsm)
<Tril> to get a copy, you "check out" using the cvs program.
<Tril> there should be instructions for this on both the tunes page and teh retro page
<pyro> i'm not sure I want him checking in.. :)
<ElGato> hey you told me /cvs/retro
<ElGato> mabye you should have been a little clearer
<pyro> yeah, that's where CVS gets it from
<pyro> hehe
* ElGato/#tunes feels quite stupid
<Tril> ElGato: rm -rf ~/retro;cd;cvs -d /cvs co retro
05:10pm
* Tril/#TUNES blames this on pyro for not explaining properly
<water> hehe
<water> if we could continue
<ElGato> nah
<ElGato> im sorta dense sometimes
<pyro> it's easy when you're not doing cvs over the net
<Tril> i'm thning
<water> k
<ElGato> let'sjust forget this
-:- shamino [sham@phila-dialup138.nni.com] has left #tunes []
<ElGato> continue water
<pyro> I just borrowed the Tunes directions for using CVS.. 
<Tril> they should work! :)
<Tril> ok i'm trying to read this tsia thing from the topic
<water> i haven't read their papers yet
<pyro> allright.. time for dinner
<pyro> see ya all
<Tril> bye
<water> bye
-:- SignOff pyro: #TUNES (pyro has no reason)
05:20pm
<Tril> they just want to separate execution details such as parallel scheduling, distribution etc
<Tril> from a program specifiation
-:- liar [brand@p0wer.qzx.com] has joined #tunes
<Tril> it sounds a lot like AOP
<Tril> but they dont cite it
<water> a little strange that they don't
<liar> u guys any good at C programming?
<eihrul> yessuh
<water> haven't in a while, but sure
<liar> whats the best way to get all the command line arguments in a single string?
<Tril> use strlen to count the length of all the arguments, allocate a buffer big enough, then strcpy them one at a time
<eihrul> after they've already been so carefully separated?!?!?
<eihrul> strcpy () is slow
<eihrul> since it has to find the end of the buffer every time
<liar> i dont need speed
<eihrul> then strcpy () :)
<Tril> strncpy, whatever!!!
<liar> does strcpy append or replace?
<Tril> oh, i guess i meant strcat
<eihrul> me too :)
<water> tsia doesn't look like it wants to make coding simpler by simplifying the language
<liar> so strcpy the first and strcat the rest
<eihrul> or shove a zero in the first part of the string
<eihrul> and strcat all of em
<water> it doesn't necessarily separate aspects, it just provides a module (or black box) that handles them
<liar> and it doesnt add the spaces does it?
<eihrul> liar: are there any effects on the TLB for temporarily disabling then reenabling paging?
<eihrul> it puts in everything you have in the string but the delimiting zeros
05:30pm
<liar> eihrul: not that im aware of
<liar> eihrul: its just when u load cr3
<eihrul> ah... because i want to implement message passing that way
<liar> eihrul: it flushes the all tlb entries
<liar> eihrul: except for global pages
<eihrul> yep... that's what the intel docs say...
<liar> i enable/disable paging alot
<eihrul> but the intel docs omit anything about the other registers
<eihrul> ah, i couldn't find even a PEEP
<eihrul> on whether it was okay or not
<eihrul> am relieved to hear that :)
<liar> oh and global pages only work on k6+ and ppro+ so u need to check if the cpu supports them before enabling global pages
* eihrul/#tunes really needs to stop putting words in all-caps for emphasis.
<eihrul> yeah, i noticed that in linux code
<water> who was it asking me how os's do multi-tasking in hw vs sw?
<liar> u can set the global page flag in the page entries tho, i do it and it seems to work ok. intel says not to but...
-:- Tril [dem@bespin.dhs.org] has left #tunes []
<liar> PROGRAM this is a string
<liar> argv[] doesnt have spaces does it?
<liar> i thought the spaces separated the argv's
<ElGato> uhhh...hehehe
<ElGato> this is wierd
<ElGato> i tried to make retro and there are like 300 errors
<ElGato> and it isn't my code
<liar> ElGato: paul has been screwing retro up
<eihrul> i might as well check for large 4MB pages as well
<eihrul> if i must add code to do that
<ElGato> heh...serial.asm has lie 200 errors
<ElGato> like
-:- SignOff eihrul: #TUNES (Leaving)
-:- eihrul [lee@usr5-ppp14.lvdi.net] has joined #tunes
<liar> brb eating
<eihrul> that stunk
-:- SignOff eihrul: #TUNES (Ping timeout for eihrul[usr5-ppp14.lvdi.net])
05:40pm
-:- eihrul [lee@usr5-ppp14.lvdi.net] has joined #tunes
<ElGato> wb
<liar> back
<ElGato> ynwb
05:50pm
<ElGato> ga
<ElGato> ysf
* ElGato/#tunes is playing around with acronyms no one knows :)
<liar> http://borg.qzx.com/bi/compressed_name.html
<liar> anyone know the best way to convert a string, using that page, into a 64bit number?
<liar> 12 5-bit characters
<liar> doesnt need to be fast just simple
<ElGato> uh..hash it
<liar> how
<eihrul> eh?
<ElGato> i don't know
<ElGato> it just makes me sound cool
<liar> but now u sound like an idiot for not knowing :)
<eihrul> what are you trying to do exactly, liar?
<liar> im writing a program so i can convert strings into 64bit numbers for brix
<liar> it will be easier to have a program then to do them all by hand
<eihrul> 8 bit chars -> 5 bit?
<liar> ya
<eihrul> use look up table to do conversion
<eihrul> and after that, the bitwise stuff gets a bit messy
<liar> the bitwise is cake
<eihrul> sure... but it's messy :)
<liar> and i already thought of the lookup table
<eihrul> actually, maybe not
<eihrul> since you can just use shifts
<liar> and i know its only 32 characters but thats still alot of typing
<eihrul> okay...
<eihrul> simple way to map the alphabet then:
<eihrul> if (isalpha (some_char) != 0) offset_from_a = some_char - 'a';
<eihrul> you can use isdigit () and '0' to do numbers
<eihrul> then add 1 or 27 depending
<liar> ah thnaks
<liar> thanks
<eihrul> look up table still faster :)
<liar> i dont need speed
<liar> as long as the program can compute it faster than i can then its good
<eihrul> i know... i know... j/k
06:00pm
<ElGato> k this sucks
<ElGato> after all this work i can't compile retro
<ElGato> cause some other file is fux0red
<liar> rl: send a bitch email to paul
<liar> s/rl/ElGato/
<liar> does c have a function to convert a string to lowercase?
<liar> or do i have to dig out mine?
<eihrul> it has a function to convert a character to lowercase
<ElGato> yes 
<ElGato> paul
<liar> eihrul: that will work
<eihrul> make a function to filter a string through a function pointer :)
<eihrul> it's the pseudo-higher order thing to do
<liar> void strlower(char *string) {
<liar>         while(*string!=0){ *string|=32; string++; }
<liar> }
<liar> there is mine
-:- fare is now known as Fare
<Fare> hum
<Fare> wassup?
<eihrul> tolower () does the same thing
<water> fare!
<liar> well u didnt tell me so i had to get mine
<Fare> water!
<water> i'd give you my spec, but it's only on paper now
<water> for reflective arrow logic, at least
<water> (my laptop is very sick lately)
-:- Tril [dem@bespin.dhs.org] has joined #tunes
-:- mode/#Tunes [+o Tril] by ChanServ
* Fare/#Tunes is still reading "The Mind's I"
<ElGato> hi fare
<eihrul> whatever works
<eihrul> haha, sorry... i forgot
<ElGato> Me Donnerez-vous principal?
06:10pm
<ElGato> A la difference du hcf, je pas urinate sur les personnes agees autant que possible.
<Fare> full hand tonight?
<ElGato> hmmm?
<Fare> s/hand/house/
<liar> is there a 64bit integer in C?
<water> oh. some here are lurkers
<ElGato> double i think
<Tril> long long, gcc specific
<ElGato> Fare: D'ou avez-vos cache mon grand cheval?
<liar> ok what is the %letter to print a long long
<liar> still d?
<ElGato> d
<Tril> got to log off
<ElGato> d es para decimal
<Tril> bye water!
-:- SignOff Tril: #TUNES (Tril has no reason)
<Fare> lo, HS
-:- NetSplit: carter.openprojects.net split from sterling.openprojects.net [06:15pm]
-:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [carter.openprojects.net]
-:- Netjoined: carter.openprojects.net sterling.openprojects.net
-:- Fare [fare@quatramaran.ens.fr] has joined #Tunes
-:- eihrul [lee@usr5-ppp14.lvdi.net] has joined #Tunes
-:- SignOff Fare: #TUNES (changing servers)
<liar> hmm this stupid program isnt giving me the same numbers that i computed
<water> ?
<liar> isnt there a way to tell %x to only print 1 digit?
<liar> a nibble
<ElGato> Je sais des choses, ce dieu egal ne connait pas et je sais que votre pere, il me respecte maintenant. Et c'est de shit de taureau.
<ElGato> mask the other bits
<liar> nm
<water> sorry, i don't mess with that stuff any more
<eihrul> liar: %1x
<liar> ah thats it
06:20pm
<ElGato> Je me trouvais sur l'herbe dimanche matin de la semaine derniere.  Se livrer a mon defaite d'individu.
<ElGato> Ne volez pas mon soleil.
<liar> man i hate c
<water> so do i. that's why i don't use it
<liar> u leave out a single = sign and the program still works, just wrong
<liar> if(c=' ') p = 27;
<eihrul> #define equals ==
<eihrul> there ya go
<water> wow. that's the same kind of bug i remember dealing with two years ago
<eihrul> no chance of misrepresenting == with = then
<ElGato> i don't mind c
<liar> eihrul: heh
<water> eihrul: << instead of <<=
<ElGato> it's just c++ i hate
<eihrul> i've never had that happen
<eihrul> i've only had one incident of == and =
<water> you bit-shift as much as you can in a good graphics engine or file-system
<eihrul> yep... even so
<eihrul> i've never accidently typoed in that way
<ElGato> ditto
<water> it's easier than you might think
<ElGato> i've had like 3 incidents
<eihrul> it's easy in principle yes
<eihrul> but it's not frequent
<liar> is there a %letter for binary?
<eihrul> nope
<liar> damnit
<eihrul> is in C++
<ElGato> what's 82h in decimal?
<eihrul> 8 * 16 + 2
<water> 130
<ElGato> ok
<ElGato> hey cool
<ElGato> now i know how to do that
<water> sigh...
<ElGato> :D
<ElGato> binary to hex is easy
<ElGato> same with binary to octal and vice versa
<liar> ok now my little program works
06:30pm
<liar> are there any easy ways to squeeze more characters into 5 bits?
<eihrul> 5 bits is pretty scarce as is
<ElGato> ...hash it :D
<liar> i would like to have 26 alpha, 10 digits, and space
<liar> ElGato: die :)
<eihrul> use 6 bits
<water> lzw?
<liar> 6 bits gives me a smaller string length
<ElGato> liar: mangez un penis
<eihrul> you could put restrictions on the locations of letters and numbers
<liar> water: im using a 64bit number to store a string into
<eihrul> and after a certain escape bit, accept only numbers
<liar> water: with 5bits i can get 12 characters, will lzw give me more?
<water> liar: oh. hmm
<water> lzw works better with a larger stream
<liar> ya i know
<ElGato> i don't thinnk you'll get anymore in
<water> so it wouldn't be very good for 64 bits at all
<liar> hmm maybe i can use the last 4 bits to hold a full number
<liar> and i dont need a null
<ElGato> what about letters though?
<eihrul> you have 60 bits for em
<water> why only a 64-bit string?
<liar> water: its for keywords and i want fast lookup
<liar> and small storage
<water> ah
<water> like forth does, sort of
<liar> ya
<liar> keywords are the only type of filenames in brix
<liar> well almost
<water> i hate to mention it, but why not a hash key?
<ElGato> rivaux rouges de bande reutilisant des bibles :D
<ElGato> go water!
<ElGato> see liar
<liar> i dont know howto do hash, but someone told me my way would work better
<water> hash isn't terrific
<liar> and that hash wouldnt really work for what i need 
<water> what's your idea? fitting letters into bit-fields?
<liar> 5bits per character
<liar> 12 character in a 64bit number
<water> wait... here's an idea
06:40pm
<ElGato> what happens to the last 4 bits?
<water> you could put all of the keywords into an lzw dictionary
<water> similar patterns in keywords would allow compaction
<liar> how
<water> it wouldn't be straight-forward, as far as code is concerned, though
<water> you just treat your keywords as being arranged in a series in a text-file (nuls instead of whitespace between words)
<ElGato> why don't you turn the keyword into a hash key by addingthe values of each characte in the string together and let it search using that
<water> you can then compress it and access it that way
<ElGato> don't ask me what to do about hash clash though
<water> if you aren't really familiar with lzw, i'm sure that there are good explanations on the net
<water> elgato: hash-clash would result in inconsistencies in look-up times
<ElGato> uhhh...yeah
<water> elgato: on the other hand, lzw wouldn't help much either
<ElGato> :)
<liar> so im better doing it my way :)
<water> liar: your original idea will probably be the best, if only for simplicity
<ElGato> hashing would be faster
<water> elgato: it depends on how many keywords his system will allow
<water> even liar's idea will require him to search the keywords' databse
<water> which could take a while
<ElGato> just make it an arbitray prime number and then when that fills up make another list
<liar> ya but the database is sorted and uses access counts to move often accessed keywords to the top
<ElGato> but with that you still have to search
<liar> ya but the most accessed keywords are at the top
<ElGato> the point of hashing is not to have to
<water> so, it's sorted by most-often/recently accessed
<water> elgato: right
<liar> not recently
06:50pm
<water> elgato's right that sorting could take a lot of time
<ElGato> just give them a fixed place in the dictionary
<water> lair: but over time, often-accessed entries get bumped by once used more recently anyway
<ElGato> plus with qz's keyword idea, there would only be a certain number of possible character combinations for keywords
<liar> no really
<ElGato> you only have 12 charcacter places
<liar> now i have 13
<ElGato> ok 13
<liar> the last 4bits will hold 0-9 and a-f
<ElGato> how many combinations can you have with 13 characters (i don't know the math)?
<liar> oh and i do have aliasing to alow bigger keywords
<water> elgato: plenty, but most aren't readable
<water> ok, so aliases would use unreadable keywords
<ElGato> just make your dictionary a prime number size around there
<liar> ya or they can just be used to map lots of keywords to a single keyword
<water> basically, you're optimizing for short word/number combinations, particularly for look-up, at the cost of adding new words, since you have to re-sort
<liar> ya
<ElGato> i gotta go
<ElGato> laterr guya
<ElGato> blarg
<ElGato> later guys
-:- SignOff ElGato: #TUNES (hash it i tell you)
<water> that's cool, as long as you know where the slow points will be
07:00pm
<water> bbiaf
<water> b
07:10pm
-:- ruiner [nate@ppp432.wi.centuryinter.net] has joined #tunes
<water> hi ruiner
<ruiner> hi
<ruiner> what is os/2's mascot?
<ruiner> is it just that lame os/2 circle?
<ruiner> at least linux has an animal
<water> i didn't think that it had one
<liar> windows has a little guy with a black ski mask
<water> linux is the only os i can think of with a mascot
<water> you could say that the old amiga screen-saver is a mascot, but that's a stretch
<liar> well linus likes penguins, bill likes to run over penguins
<liar> freebsd has a mascot
<liar> and hurd is gnu so its mascot is that gnu
<water> so does gnu
<ruiner> yeah....
<water> mascot's aren't really special, they're just stand-in logos
<eihrul> that's degrading to mascots everywhere
<water> so?
<water> :)
<eihrul> so... you're going to make the BSD Daemon cry!
<water> and... ?
<eihrul> he might hurt himself... he's a daemon
07:20pm
<water> that's fine with me :)
<ruiner> mascots rule over any dopey logo anyday
<ruiner> what should the windows logo be?
<ruiner> a turtle? a hog?
<water> logos are terrible
<ruiner> skunk
<ruiner> ?
<eihrul> i thought it was a 900 pound baboon?
<water> a meaningless symol
<ruiner> launchpad mcquack?
<eihrul> meaningless in itself... but powerful in interpretation :)
<ruiner> nah, baboons are cool
<ruiner> turtles are nice and slow though
<ruiner> and windows is a hog...
<eihrul> a cockroach?
<ruiner> launchpad mcquack crashed all the time...ever watch ducktales?
<eihrul> sure
<eihrul> but cockroaches have so many interesting traits:
* ruiner/#tunes is waiting....
<eihrul> they just don't seem to die... they generally carry diseases... they sneak in everywhere... and when they get stuck on their backs, they can't turn over and die
<water> cockroaches have been around for hundreds of millions of years, surviving many extinctions and disasters
<ruiner> hmmmm......
<ruiner> I think windows' mascot should be a herpes virus...you don't really want it, but once you get it you're stuck with it
<eihrul> that'd work
<ruiner> then when somebody came into a chat room they could go "I need help with my herpes"
<eihrul> but do you go into an irc channel and say: i need help with my (penguin/daemon)?
<eihrul> not likely
<ruiner> true
<water> the penguin means nothing. it's an advocacy gimmick
<ruiner> but I think that would become adopted by people who use linux
<eihrul> we're well aware of that, water
* water/#tunes invites ruiner and eihrul to #linpeople
<water> :)
07:30pm
<eihrul> you need to put something over that "Made for Windows 95" plate on your computer
<eihrul> it might as well be a penguin or some other mascot
<ruiner> I'm saying like...somebody would say that they have windows, and two linux users would go "he has herpes"
<water> "Made for You"
<eihrul> not necessarily
<eihrul> that's a lie in and of itself
<water> which is why it's so ironic that users are customers
07:40pm
-:- SignOff hcf: #TUNES (Ping timeout for hcf[me-portland-us834.javanet.com])
-:- hcf [nef@me-portland-us820.javanet.com] has joined #tunes
<ruiner> anybody check out slashdot today? about the keyboards confusing newbies?
<ruiner> soooooooo lame
<water> yep.
<ruiner> I can't believe how stupid people are
<water> so what?
<ruiner> imho if you can't use it as is, and you can't understand it enough to change it yourself, you don't belong in front of one
<water> qwerty keyboards are designed to be hard to use from the start
<ruiner> true
<ruiner> know why?
<ruiner> heh heh
<water> yes
<ruiner> dates back to typewriters
<water> fast typing caused the key arms to jam
<ruiner> but thats not the point of the article...the article is about scroll lock, num lock, print screen, pause/break....
<water> they don't make sense, though
<ruiner> if you don't know what it does, find out
<water> not even apple's keyboards make sense
<ruiner> yeah they do
<ruiner> apples....I dunno about them
<water> they're similar
<ruiner> but my keyboard makes sense....
<ruiner> I love the pause/break button
<ruiner> insert and delete too
<water> no, form definitely does not relate to function
<ruiner> page up, page down
<ruiner> bah!
08:00pm
<water> zero engineering goes into keyboard design. they merely copy the ibm standard
<ruiner> why is it that lately I've been disagreeing with everybody in here?
<ruiner> because the ibm standard works
<water> i disagree, too, though :)
<water> so?
<ruiner> so, if it works, why change it?
<water> because computing is more diverse now
<water> 3d games all have different layouts, and most aren't modifiable
<ruiner> yeah, I'm so glad that compaq is putting volume controls and easy internet keys on their keyboards
<ruiner> just what the world needs
<ruiner> bah!
<water> ok i agree about that. it is lame
<ruiner> sounds like a software deficiency to me
<water> in fact, extremely lame
<water> yes
<ruiner> if you can't write code thats going to work with the standard keyboard, then you're not very good
<water> bah
<water> :)
<ruiner> my friend's parents bought a compaq, much to my dismay
<water> only tunes is good code
<water> :)
<ruiner> after I told them not to get one too....tsk tsk
<ruiner> I said, "just don't get an acer, hewlett packard, packard bell, compaq, or anything else they sell at best buy"
<water> why would they listen to you?
<ruiner> I told them to either let me and my friend custom build it, or get a micron
<ruiner> yeah, I'm just the guy they call any time anything goes wrong with their computer....no reason to listen to me
<water> are you some kind of control freak?
<ruiner> yes I am
<water> of course
<ruiner> plus I don't really like going out to their house every two days because they can't figure out why their computer keeps freezing or why they can't get their e-mail
<water> oh
<ruiner> I could set up a system for them that would be so cool....
<water> :P yes, i empathize there
<water> who cares if it's cool? many specially-built systems are buggy
<ruiner> in the span of time that my grandpa has had the computer I built for him, he's had to reformat, reinstall windows, or call for tech support 0 times
<ruiner> and thats with WINDOWS
<water> case in point is the one i'm using now
<ruiner> in the month that they've had their compaq, they've taken it in twice and had me over there almost every 3 days fixing the damned thing
<water> if you use the 3dfx card while the usb modem is activated, the computer freezes up within ten minutes
<ruiner> what os?
<water> win98r2
<ruiner> heh heh heh
<water> it's a game machine
<ruiner> I don't think I need to say anything...
<ruiner> yeah, just don't play any 3d games online
<ruiner> lol
<water> and it can't run the decent games
<water> you're an idiot to build such machines and then criticize my friend for buying one
<ruiner> pardon?
<ruiner> the difference is I make mine work
<water> like you know what you're doing. bah!
08:10pm
<water> the guys who built this machine have been hacking hardware for a few decades
<ruiner> hey, I'm batting 1000 so far
<water> yeah, with machines that don't get upgraded i bet
<ruiner> they did a great job too I might add
<water> the usb modem was an add-on
<ruiner> hasn't needed to be upgraded yet
<water> your customers will be back to you soon
<ruiner> I'm skeptical
<water> saying "why won't my cable-modem work?"
<ruiner> ah, I might be getting a job installing cable modems for the cable service here
<ruiner> so I might be dealing with that one quite a bit
<water> yes, and possibly regretting it
<ruiner> bah!
<water> blind confidence doesn't prevent mistakes
<ruiner> I've made mistakes, I've been able to fix them though
<ruiner> like my original video card in this machine...
<ruiner> was fuct...kept overheating and totally freezing the machine
* water/#tunes doesn't care
<ruiner> alright....point is, I figured it out, and fixed it
08:20pm
-:- SignOff water: #TUNES (Ping timeout for water[ppp-tnt-7.tscnet.net])
-:- water [water@ppp-tnt-39.tscnet.net] has joined #tunes
<water> darn isp
<ruiner> heh heh
<ruiner> your friends must work for your isp
<water> bah
<water> actually, an entirely different set of people
<ruiner> lol lol lol
<ruiner> just joking around...
08:30pm
-:- overfien [foobie@m227.telcomplus.net] has joined #Tunes
<overfien> y0h whats up
<water> over-fiend? hi
08:40pm
<water> overfien: what brings you here? are you a tuneser?
<eihrul> is it faster to use the pushad instruction or manually push the registers?
<overfien> no a few peeps from my college told me to come here an wait for them
<water> pushad, i think.
<overfien> from western wwu.edu
<water> ok
* overfien/#Tunes is away coding on diff. box "asm for vms"
<ruiner> where?
<eihrul> hrmm, but pushad pushes one extra register
<eihrul> since call gates/interrupt gates already push the stack pointer
<water> is that a problem?
<eihrul> just wondering...
<eihrul> it's not a problem
08:50pm
-:- SignOff water: #TUNES (Ping timeout for water[ppp-tnt-39.tscnet.net])
<liar> eihrul: yer thinking about it wrong
<liar> why do u wanna push all regs?
09:30pm
-:- _QZ [brand@borg.qzx.com] has joined #tunes
<_QZ> ok i know that if enlightenment crashes it just leaves my windows window-less
<_QZ> and i can restart it
<_QZ> but if E is locked up and i kill it will it do the same thing?
<_QZ> or if i kill it will it close X
<_QZ> anyone?
<ruiner> I'm no expert, but I'd guess it'd close X.....hmmmmm....actually....
<ruiner> you can't change it to a different window manager without killing E first?
<_QZ> eihrul:  no
<_QZ> eihrul:  only save the regs u use
<eihrul> ah...
<eihrul> used for parameter passing?
<eihrul> or used in general
<_QZ> i dont save regs used for parameters
<eihrul> then which registers do you save?
<_QZ> i only save regs that params arent passed in and that i change
<_QZ> i also dont save regs if i usethem to return a parameter
<eihrul> but i'm not making separate entry points for each syscall
<eihrul> so i'd just have to save them all anyway
<_QZ> u that is slower?
<_QZ> u have to push all and make an extra jmp inside the syscall
<eihrul> i'm writing things in C
<_QZ> and u have to lookup the jmp address
<eihrul> that's a single lookup dude
<_QZ> what u think C has a magical way of doing it with those 3 extra steps?
09:40pm
<eihrul> grr... don't turn this into a language war
<_QZ> pushall, lookup and jmp <-- 3 extra steps
<_QZ> bbiaf
<eihrul> i'll live
<eihrul> it's not going to kill me
09:50pm
!Hyrlik:*! Gee, I can spam from here
!spice:*! Me too :)
-:- duxpond [email@208.243.170.5] has joined #Tunes
-:- duxpond has changed the topic on channel #Tunes to: http://www.electrolab.net/webcam.htm <-- This site is the shit and is being flushed!
-:- duxpond [email@208.243.170.5] has left #Tunes []
!Hyrlik:*! I just got word that by making these 'pointless' messages I am wasting precious bandwith, for this I appologize and will my my pointless messages have special inspirations for you all.
-:- ThaReaper [Tha@98CBD7A3.ipt.aol.com] has joined #tunes
<ThaReaper> hey bastards :-)
!spice:*! settle Hyrlik :)
-:- ThaReaper [Tha@98CBD7A3.ipt.aol.com] has left #tunes []
!al-Kader:*! /mode al-kader -w :)
-:- ThaReaper [Tha@98CBD7A3.ipt.aol.com] has joined #tunes
-:- ThaReaper [Tha@98CBD7A3.ipt.aol.com] has left #tunes []
-:- hcf has changed the topic on channel #Tunes to: Reflective Computing System <http://www.tunes.org> | http://www.dilbert.com/comics/dilbert/archive/images/dilbert1993043090804.gif | http://osdevnew.listbot.com | http://www.researchindex.com | http://www.tsia.org
10:10pm
-:- _QZ [brand@borg.qzx.com] has left #tunes []
<ruiner> bastards?
<ruiner> a guy using aol was calling me a bastard?
<ruiner> if that wasn't so inherently stupid I'd be offended
<liar> ha
10:20pm
<liar> i cant read could someone tell me what the tsia page talks about?
<ruiner> yeah, hold on a sec
<ruiner> read this while I'm reading tsia
<ruiner> http://users.erols.com/oddities-inc/news0899/jar.html
<liar> i already read that page :)
<liar> er, i mean someone read it to me
!spice:*! ewps
!Bhaal:*! Where was he coursing problems this time?
!Hyrlik:*! do a list of all channels and you'll see
<ruiner> are you serious?
!Bhaal:*! Hrm.. my channel list is somewhat short for some reason, but Ill believe u :)
!Hyrlik:*! I'm just curious why the ircd didn't kill him
!spice:*! so you could have the pleasure ?
!Bhaal:*! *shrug* when u did the gline?
!Bhaal:*! haha
!Hyrlik:*! maybe the gline didn't propogate to irc.linux.com
!Bhaal:*! it should have...
!ltd:*! it did
10:30pm
<liar> serious?
<liar> im serious about not being able to read
!spice:*! wrong syntax buddy
!Hyrlik:*! whats the right syntax?
!Hyrlik:*! *@ip right?
!ltd:*! no nick in a gline
!ltd:*! G *!*@208.243.170.5 934525467
!al-Kader:*! I dont meant to rain on your /wallop parade, but we should keep this in #openprojects.
!Hyrlik:*! oh, so it would be always *@IPADDR.* <TIME> : <REASON>
!al-Kader:*! no offense...
!ltd:*! /umode -w
10:40pm
<ruiner> uhhuh....so you tell the chimpanzee what to say....and he types it?
<ruiner> banana
<liar> yup
11:00pm
-:- SignOff ruiner: #TUNES (Leaving)
-:- SignOff hcf: #TUNES (Leaving)
-:- SignOff liar: #TUNES (BRiX [http://www.qzx.com/brix] :: sleep)
[msg(TUNES)] newlog 1999.0813
IRC log ended Fri Aug 13 00:00:00 1999