IRC log started Mon Jan 3 00:00:01 2000 Is that not the timer with the lowest value? i think i need to compare N-C and X-C [msg(TUNES)] permlog 2000.0103 ya i think thats right no, it wouldnt be the timer with the lowest value if C=2^32-5, N=2^32-2 and N=1 then N would have the lowest value but X is still next N=2^32 and 1? the counter will tick 5 times and then roll over X is 3 ticks away and N is 6 i think comparing N-C to X-C will work Ok N-C=3 and X-C=6 unsigned 32bit number no i need the absolute value * eihrul/#tunes yawns. :) eihrul: am i right? eihrul: feel free to step in at any time i just got back what's the problem? well, the problem is , as usualy, explaining what tunes is :) abi: air's problem abi: the problem is, as usal, explaining what tunes is i haven't a clue, eihrul abi: the problem is, as usual, explaining what tunes is eihrul: bugger all, i dunno abi: problem is, as usual, explaining what tunes is eihrul: bugger all, i dunno 12:10am hmmm abi doesn't like sentences with what i don't know, eihrul when using 'what' in a factoid's defn u need to use a \ thus, blah \what blah abi: the problem is, as usual, explaining \what tunes is what is the problem? the problem is , as usualy, explaining what tunes is * eihrul/#tunes ponders. air: i have to go pretend to do homework soon, so hurry up eihrul: Homework? u want me to repeat it? abi: no, problem is the problem is, as usualy, explaining \what tunes is okay, hcf. air: the problem? the problem is, as usualy, explaining what tunes is air: sure, because you did a horrible job of explaining it the first time heh ok i have a 32bit counter that inc's every 10ms eihrul: Yes :) when i add a timer i store C+time where time is number of 10ms intervals before timer expires C=counter a pointer to the next timer to expire is stored as X and when i create a new timer, N, i must compare it to X to see if it will replace X as the next timer do i just do "cmp |X-C|,|N-C|"? how long the timers wait before they go off is irrelevent eh? the timer with the nearest event deadline is closer in the list or rather closer to the head of the list what is so hard about that? -:- Viper [Viper@T8Hv9n25.midsouth.rr.com] has joined #tunes is that the right comparison? the counter can roll over from 2^32-1 to 0 so if C=2^32-5, X=2^32-2 and N=1 then X is still the next timer eh you should never have a problem where 2^32 overflows.... there will never be a problem with it overflowing, but it will overflow and that is expected 12:20am not easily the counter should get kicked back to zero before any roll over can happen (unless you can say that things are waiting for an event at all times) -:- SignOff Viper: #TUNES ([x]chat) if the comp has been up for a day and u set a timer to expire in 497 days then that timer value will be less than the current counter value nope that counter can be reset to zero as soon as your timer queue is empty no that counter is the system uptime counter i wouldn't use that for scheduling timers yes i'd just use a delta from the time when the first timer is inserted into the queue i am based on the counter just like in dos when u save the current bios time of day counter, add some value to it and compare and loop until they are equal Emm equals delta Ekcs over delta why well, its impossible to have a deadline less than the counter yes it is its also impossible for you to overflow and then go above the counter from the value of the counter if the comp has been running for 490 days and u set a timer to expire in 2 weeks than... so.... any value less than the counter should be moved into the place in the list any value above the counter will expire before any value below it where X[0] > X [1] or put at the end of the list if no such place exists, thus creating that place very simple :) could otherwise even have two alternating queues... what the hell is X[0] and X[1] X [0] is a given timer deadline in the list of timers X [1] the next timer following it.... normally, barring overflowing X [0] < X [1] so that sorting the list and doing quick checks for expiration actually works.... -:- smoke [smoke@15dyn78.delft.casema.net] has joined #tunes 12:30am now... its impossible that N < C, unless there's overflow if there is you either put it at the end of the list or find where X [0] > X [1] (overflow) two alternating queues would probably be faster the timers are are stored in a sorted list, my problem is comparing the new timer to the timers in the list to find its place okay.... explaining again, you have two alternating queues :) we'll call them Current and Next if adding the timer's length to C does not overflow you sort it into Current if it does overflow you sort it into Next ok i understand now... tricky part when C itself overflows.... you make swap the roles of the queues -make but you only need to check the first timer in Current oh crap, that will slow down my scheduler bno it won't because you always know that timers in Next expire after the last timer in Current a single list would be better no it wouldn't! yes it would trust me.... it wouldn't the reason i'm suggesting two queues is because scanning through the list for the first such overflow and then doing another insertion sort is WAAAAAAAY slower i dont wanna add overflow checking of the counter into my scheduler dude.... all you have to do is a jump-if-carry in insertion and a jump-if-carry on increment of the counter it WON'T kill you 2-3 cycles! bah! it will what you gain there will be lost in insertion time many fold and then every 497 days i will have a couple more cycles as it switches queues * eihrul/#tunes laughs. if ever you give me cause to dare call you naive, you do now even if i have 2 lists i still have to insert a new timer yes... but you have to do a linear search over one list to find the start of the next and if you just put a single pointer to the start of the first overflow you have my scheme, basically and actually... that would work without having to swap queues hmm oh well im going to bed. thanks for yer help you'd only have to swap on removal of a task er timer actually, nope same thing cya boils down to swapping on overflow (C) -:- SignOff air: #TUNES (BRiX [http://www.qzx.com/brix] :: sleep) gah! 12:40am -:- SignOff eihrul: #TUNES ([x]chat) -:- _ruiner_ [DIY@ppp255.wi.centurytel.net] has joined #tunes -:- SignOff lar1: #TUNES (Leaving) -:- SignOff _ruiner_: #TUNES (destroy what destroys you) -:- Josh [josh@pool-209-138-63-227.dlls.grid.net] has joined #tunes Hello Any one home? Guess not... -:- Josh [josh@pool-209-138-63-227.dlls.grid.net] has left #tunes [] -:- SignOff hcf: #TUNES (Leaving) 02:00am -:- SignOff Fare: #TUNES (Ping timeout for Fare[quatramaran.ens.fr]) -:- iefbr14 [eric@gw.capgemini.fr] has joined #tunes -:- AlonzoTG [Alonzo@216-164-131-186.s186.tnt2.lnhva.md.dialup.rcn.com] has joined #tunes -:- SignOff AlonzoTG: #TUNES (Have Nice Day :)) -:- SignOff smoke: #TUNES (Ping timeout for smoke[15dyn78.delft.casema.net]) -:- AlonzoTG [Alonzo@216-164-133-149.s149.tnt3.lnhva.md.dialup.rcn.com] has joined #tunes -:- AlonzoTG has changed the topic on channel #tunes to: [Jan 03 17:48:42 ZTZ] The keyserver says: "Why is a cow ? Mu. (Ommmmmmmmmm)" -- www.distributed.net -:- SignOff AlonzoTG: #TUNES (Ping timeout for AlonzoTG[216-164-133-149.s149.tnt3.lnhva.md.dialup.rcn.com]) -:- iefbr14 [eric@gw.capgemini.fr] has left #tunes [] -:- iefbr14 [eric@gw.capgemini.fr] has joined #tunes -:- SignOff iefbr14: #TUNES (Ping timeout for iefbr14[gw.capgemini.fr]) -:- tmf [thomas@193.217.63.152] has joined #tunes hi happy new year! 07:10am happy new year :) 07:20am -:- smoke [smoke@15dyn78.delft.casema.net] has joined #tunes -:- Fare [fare@quatramaran.ens.fr] has joined #Tunes -:- AlonzoTG [Alonzo@207-172-184-62.s62.tnt6.lnhva.md.dialup.rcn.com] has joined #tunes -:- SignOff smoke: #TUNES (One day sheep will rule the world) -:- SignOff AlonzoTG: #TUNES (Ping timeout for AlonzoTG[207-172-184-62.s62.tnt6.lnhva.md.dialup.rcn.com]) -:- AlonzoTG [Alonzo@207-172-184-62.s62.tnt6.lnhva.md.dialup.rcn.com] has joined #tunes * AlonzoTG/#tunes plants charges around the room and sets the detonators... hmm we need Kaufmann or Downix to nominate TUNES for slashdot award which award is that? "most improved" ... it has most money 10:10am url? just press "awards" from the main page yes and you get the awards /. has won the lower awards and there is a story too Slashdot is Giving Away $100,000 found it it will probably be given for "GNOKDE for win32" whatever 10:20am -is(ejb@master.debian.org)- Please remember that it is not polite to /msg people who you don't know at random, thanks. -GlobalMessage(s@GlobalMessage)- Sorry all, that was a globalnotice. nothing personal :> fuck I told the dnetc to work on RC5 because CSC crashes my machine but then when it goes to load packets it selects CSC FUCK them!! -:- mibin [mibin@62.11.105.52] has joined #tunes * zarq/#tunes is back from being away: trug 10:50am -:- SignOff mibin: #TUNES (Ping timeout for mibin[62.11.105.52]) -:- mibin [mibin@62.11.104.85] has joined #tunes -:- SignOff Fufie: #TUNES (Ping timeout for Fufie[tunnel-44-18.vpn.uib.no]) -:- SignOff smkl: #TUNES (Ping timeout for smkl[glubimox.yok.utu.fi]) -:- smkl [sami@glubimox.yok.utu.fi] has joined #tunes * AlonzoTG/#tunes hurls a laser guided flaming woodchuck at smkl * smkl/#tunes dies, then raises from the grave as zombie and rapes ATG. ATG goes to jail because he is a necrophil. * AlonzoTG/#tunes smashes smkl with an Iron Butterfly -:- SignOff mibin: #TUNES (main(){fork();main();}) -:- SignOff AlonzoTG: #TUNES (Have Nice Day :)) -:- hcf [nef@me-portland-us203.javanet.com] has joined #tunes -:- AlonzoTG [Alonzo@207-172-50-66.s320.tnt7.lnhva.md.dialup.rcn.com] has joined #tunes ho hum AlonzoTG: u here? 01:20pm -:- hcf has changed the topic on channel #tunes to: TUNES, Free Reflective Computing System -:- hcf has changed the topic on channel #tunes to: TUNES, Free Reflective Computing System || meta machine om 01:40pm AlonzoTG: abi: lal? lal is probably _linkers and loaders_ at http://linker.iecc.com -:- SignOff smkl: #TUNES (Ping timeout for smkl[glubimox.yok.utu.fi]) om 01:50pm -:- Fufie [stig@tunnel-44-22.vpn.uib.no] has joined #tunes -:- lambda [lambda@cogeco-88-83.cgocable.net] has joined #tunes -:- SignOff Fare: #TUNES (Ping timeout for Fare[quatramaran.ens.fr]) -:- smoke [smoke@15dyn78.delft.casema.net] has joined #tunes -:- smkl [sami@glubimox.yok.utu.fi] has joined #tunes -:- Fare [fare@quatramaran.ens.fr] has joined #Tunes -:- SignOff lambda: #TUNES ((lambda (x) (exit x)) irc) -:- SignOff Fufie: #TUNES (out) -:- SignOff smoke: #TUNES (epic4-2000) -:- zarq is now known as zarq__ * AlonzoTG/#tunes smashes zarq__ with an Iron Butterfly -:- zarq__ is now known as zarq_ ever seen hellraiser? yea, I finkso. lousy movie, but they had some nice gadgets :) my 'doze shell reports some of my gnu files were written in the yarn 100 =\ 02:50pm -:- imarkov [pervert@212.116.150.203] has joined #tunes * AlonzoTG/#tunes works imarkov over with a 2x4 -:- SignOff zarq_: #TUNES (User pressed ^C five times.) hi, do you know channels discussing logic programming and prolog? -:- zarq [zarq@9dyn105.delft.casema.net] has joined #tunes Thats it, for bringing up that topick, you deserve to die!!!! why do you think so? I just recieved a book today, well I thought it was a book when I ordered it, but it turned out to be a fucking collection of papers. $40 down the drain. =((((((((((((((((((((((((((((((((((((( =) how is that book entitled? nope, but there is one channel about functional programming Algebraic and logic programming... =\ There was no clue on the website that it wasn't a book. whats the name of this channel, smkl? well this is #tunes =P #lambda at ircnet and then of course there is #tunes hm i thought #tunes is a channel concerned with specific project or at least thats what i saw at www.tunes.org a few minutes ago ack i'm lagged whats this 'ircnet' the project has very large interests 03:00pm irc.stealth.net is one server .. ircnet is one of the four large irc networks efnet, ircnet, dalnet, undernet om * zarq/#tunes is away: z thanks, i entered ther imarkov: welcome hm it seems the people interested in these areas are not quite many :) * Fare/#Tunes is the original irresponsible for tunes.org imarkov: how did you learn about tunes.org? 03:10pm -:- SignOff imarkov: #TUNES (Read error to imarkov[212.116.150.203]: Connection reset by peer) damn peer! -:- imarkov [pervert@212.116.150.206] has joined #tunes -:- eihrul [lee@usr5-ppp10.lvdi.net] has joined #tunes wb imarkov hi again,smkl imarkov: if you like rewrite logic, there's Maude, or vigneron's page at http://www.loria.fr/~vigneron/RewritingHP/ abi: Maude? Maude is a reflective rewriting logic language at http://maude.csl.sri.com/ actually im a newcomer in the are, what im interested in is not rewrite logic but the incorporation of destructive assignment in prolog without disturbing the logical interpretation of the language 03:20pm side effects in prolog? they've been used for a long time without disturbing the logical interpretation of the language? then you're looking either for monads (a la Mercury), or for linear logic * AlonzoTG/#tunes hurls a laser guided flaming woodchuck at Fare FARE MUST DIE!!!! ATG: must I? ATG: why particularly so today? I just spent $45 on a book that turned out to be a collection of papers. which book? which book is probably that? abi: forget which book Fare: I forgot which book Algebraic and logic programming. is it any good? I was trying to learn the arcane secret of prolog... Well it says on the cover, but not on the website, that it is a collection of lecture notes. =((((( i found some sources on linear logic, loli llp etc. i have downloaded the mercury ref manual but i dont like them both -:- Fufie [stig@tunnel-44-24.vpn.uib.no] has joined #tunes the last book on prolog I ordered was out of print =(((( whats monad, fare? imarkov: I don't know anything else than linear logic and monads to achieve side-effects in a pure way theres something called 'Actor Prolog' at www.cplire.ru/Lab144/index.html imarkov: other communicating prologs (cc/prolog, etc) can be reduced to linear logic -:- thusyanth [user7464@pop74.hyperlinx.net] has joined #tunes side-effects can be useful though -:- thusyanth [user7464@pop74.hyperlinx.net] has left #tunes [] it's hard to write evil programs if you can't rely on side-effects :) stack overflow, memory exhaustion etc are side effects that can be achieved within the constraints of a purely functional language :) evil without the evil.... Fufie: linear logic can be seen as a logical formalization of side-effects. * AlonzoTG/#tunes wants B00KS!!!! BOOKS suck, because there's OO in them. whats cc/prolog, fare some prolog enriched with communication stuff; for better successors, see Oz or Goedel. 03:30pm i desperately tried for a week to compile goedel on nt/sicstutus3 and i failed i didnt know goedel is enriched this way, though hum. Well, maybe try Oz (or Erlang?) instead Oz definitely has side-effects (and distributed programming) so has Erlang (but I dunno what's left of logic programming in it) does oz have oop :) imarkov: yes really? Iph B00kz suck How the fuck am I supposed to learn shit that you were born knowing???? if ATG: by attending courses at Uni, of course! ;-> ATG: or by reading papers and proceedings scattered around the internet abi: mozart? mozart is probably a concurrent constraint programming language at http://mozart.ps.uni-sb.de/ abi: mozart is also an implementation of Oz okay, Fare. No right-thinking university would admitt me. I wouldn't be able to pass the fucking co-requs to take those courses that I would be interested in. I coulnd't afford them anyway. =\ ATG: you're an untermench designed to obey the superior spirits who go to university and a lot of losers go to univ. Only 1% of universities are cluefull enough to realize CS != C++/ASM only 1% of students are clueful enough to realize and care that only 1% of universities are clueful. hey, L.Jospin (then minister of education, now prime) wanted to bring 80% of same-aged citizens to university 03:40pm In the good old days Us were for the 1337 and they treated students like the 1337 =\ umm.. there is a big difference between regular students and students who are discovered early and helped along by professors what about students who are discovered early and not helped along by professors? -:- SignOff imarkov: #TUNES (Ping timeout for imarkov[212.116.150.206]) poor professors then -:- ult [pope@user-38lc6ad.dialup.mindspring.com] has joined #Tunes -:- imarkov [pervert@212.116.150.203] has joined #tunes 03:50pm -:- SignOff eihrul: #TUNES (Ping timeout for eihrul[usr5-ppp10.lvdi.net]) -:- SignOff hcf: #TUNES (Ping timeout for hcf[me-portland-us203.javanet.com]) -:- ult_ [pope@user-37kba0o.dialup.mindspring.com] has joined #Tunes -:- SignOff ult: #TUNES (Ping timeout for ult[user-38lc6ad.dialup.mindspring.com]) -:- ult_ is now known as ult -:- SignOff Fufie: #TUNES (natt) -:- SignOff ult: #TUNES (Ping timeout for ult[user-37kba0o.dialup.mindspring.com]) -:- SignOff imarkov: #TUNES (installing mozart) -:- lambda [lambda@cogeco-88-83.cgocable.net] has joined #tunes -:- SignOff lambda: #TUNES (Read error to lambda[cogeco-88-83.cgocable.net]: Connection reset by peer) -:- Tril [dem@bespin.dhs.org] has joined #TUNES -:- mode/#tunes [+o Tril] by ChanServ Tril! 05:50pm hello 06:00pm bbl * Tril/#TUNES is away: (afk) [BX-MsgLog Off] -:- air [brand@p0wer.qzx.com] has joined #tunes 06:10pm -:- lar1 [larman@dialup-209.245.136.176.SanJose1.Level3.net] has joined #tunes Hey 0m -:- lar1 has changed the topic on channel #tunes to: TUNES, Free Reflective Computing System: || Meta Machine: You here tril? 06:20pm lar1: have u seen bubble stinky boy by joe cartoon? www.joecartoon.com 06:40pm -:- ult [pope@user-38lc648.dialup.mindspring.com] has joined #Tunes -:- ult is now known as ultima -:- ultima is now known as ult Hey ult om -:- washort [washort@d146.narrowgate.net] has joined #tunes 07:20pm anyone here? om? wussup? 07:30pm -:- lar1 is now known as lar_eating -:- ult is now known as iNcOgNiTo -:- SignOff washort: #TUNES (Ping timeout for washort[d146.narrowgate.net]) -:- lar_eating is now known as lar2 -:- lar2 is now known as lar1 -:- SignOff lar1: #TUNES (Leaving) -:- hcf [nef@me-portland-us1022.javanet.com] has joined #tunes -:- SignOff iNcOgNiTo: #TUNES (Read error to iNcOgNiTo[user-38lc648.dialup.mindspring.com]: Connection reset by peer) -:- lar1 [larman@dialup-209.245.133.6.SanJose1.Level3.net] has joined #tunes -:- carlito [sabanmr@137.28.142.93] has joined #tunes abi: seen _ruiner_ _ruiner_ was last seen on IRC 19 hours, 37 minutes and 45 seconds ago, saying: lar1: u should sleep now [Mon Jan 3 01:18:29 2000] Fare: you have any info on _ruiner_? What info are you looking for? when he's coming back online in the next couple of days carlito: why dont u email him? hcf: :) good idea! just wondering if I could catch him online * hcf/#tunes is sick of typing that * hcf/#tunes is more sick of getting that response * hcf/#tunes is away: (afk) 09:00pm bye all -:- SignOff carlito: #TUNES (Leaving) 09:20pm next time carlito drops by suggest to him that he stay on irc 24/7 and maybe he can catch ruiner -:- SignOff air: #TUNES (BRiX [http://www.qzx.com/brix] :: sleep) 09:30pm -:- SignOff AlonzoTG: #TUNES (Ping timeout for AlonzoTG[207-172-50-66.s320.tnt7.lnhva.md.dialup.rcn.com]) -:- Downix [down@d-gnaps-143.ici.net] has joined #tunes -:- eihrul [lee@usr5-ppp187.lvdi.net] has joined #tunes Hey eihrul lo meta machine? i think meta machine is at http://homepages.paradise.net.nz/~tonybeve/mm/The%20Meta%20Machine.htm I don't know, hcf added that, I just reformatted the topic 10:40pm gah, lisp is the meta machine! :) Heh abi: seen core core was last seen on IRC 6 days, 7 hours, 23 minutes and 30 seconds ago, saying: changes the density of matter [Tue Dec 28 15:23:44 1999] hmm 10:50pm ok, I need to get to bed 11:00pm -:- SignOff Downix: #TUNES (BitchX: born to raise hell) Hmm, good night all -:- SignOff lar1: #TUNES (Leaving) 11:10pm * eihrul/#tunes coughs. [msg(TUNES)] newlog 2000.0104 IRC log ended Tue Jan 4 00:00:01 2000