IRC log started Wed Jun 23 00:00:00 1999 [msg(TUNES)] permlog 1999.0623 -:- SignOff water: #TUNES (Ping timeout for water[ppp-tnt-231.tscnet.net]) -:- Manoj [srivasta@tiamat.ametro.net] has joined #tunes -:- Manoj [srivasta@tiamat.ametro.net] has left #tunes [] -lilo(lilo@varley.openprojects.net)- [GlobalNotice] Hi all. A few server moves will occur in a moment....thanks in advance for your patience! !lackey.openprojects.net!! Remote CONNECT lewis.openprojects.net 8004 from lilo !lackey.openprojects.net!! Remote CONNECT lewis.openprojects.net 8005 from lilo -:- SignOff _QZ: #TUNES (BRiX [http://www.qzx.com/brix] :: sleep) !lewis.openprojects.net!! Remote CONNECT carter.openprojects.net 8004 from lilo !lewis.openprojects.net!! Remote CONNECT carter.openprojects.net 8005 from lilo -:- NetSplit: koontz.openprojects.net split from clarke.openprojects.net [12:33am] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [koontz.openprojects.net] -:- Netjoined: koontz.openprojects.net clarke.openprojects.net -:- Fare [rideau@quatramaran.ens.fr] has joined #tunes -:- ServerMode/#tunes [+o Fare] by sterling.openprojects.net -:- NetSplit: sterling.openprojects.net split from clarke.openprojects.net [12:33am] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [sterling.openprojects.net] -:- Netjoined: sterling.openprojects.net clarke.openprojects.net -:- Fare [rideau@quatramaran.ens.fr] has joined #tunes -:- ServerMode/#tunes [+o Fare] by lewis.openprojects.net !lewis.openprojects.net!! Remote CONNECT norton.openprojects.net 8005 from lilo !lewis.openprojects.net!! Remote CONNECT norton.openprojects.net 8005 from lilo -lilo(lilo@varley.openprojects.net)- [GlobalNotice] all done for now. It's surely an accident that lewis and clarke are connected. :) Thanks. -:- water [water@ppp-tnt-156.tscnet.net] has joined #tunes -:- _water [water@ppp-tnt-160.tscnet.net] has joined #tunes -:- SignOff water: #TUNES (Ping timeout for water[ppp-tnt-156.tscnet.net]) -:- _water is now known as water -:- SignOff water: #TUNES (Ping timeout for water[ppp-tnt-160.tscnet.net]) !aphzen:*! Linux.Com's IRC server is being upgraded. There will be a brief service interruption in ten minutes. Our apologies, but we're just trying to make it all work better =) !HyrlikW:*! irc.linux.com should be cnamed to the roundrobin... :) no downtime... !aphzen:*! Any day now ;) !aphzen:*! irc.linux.com, being shut down now. -:- AlonzoTG [Alonzo@209-122-205-152.s152.tnt4.lnh.md.dialup.rcn.com] has joined #tunes -:- SignOff AlonzoTG: #TUNES (Have Nice Day :)) -:- SignOff Tril: #TUNES (Ping timeout for Tril[sloth.wcug.wwu.edu]) -:- Tril [dem@sloth.wcug.wwu.edu] has joined #TUNES -:- mode/#tunes [+o Tril] by ChanServ Tril: hi! 07:40am -:- Beam [nobody3@d95.nnb2.interaccess.com] has joined #tunes hey guys 09:40am Corey! hey fare how's it goin? -:- AlonzoTG [Alonzo@216-164-225-140.s394.tnt6.lnh.md.dialup.rcn.com] has joined #tunes BEAM ON!!! 09:50am -:- billyboof [hatefull@hrfr-sh1-port32.snet.net] has joined #tunes alonzo, you here? 10:00am B00F!!!! * AlonzoTG/#tunes slaps billyboof around a bit with a large trout 10:10am -:- ProGuy [Paul@p401-059.ppp.get2net.dk] has joined #tunes -:- beholder [beholder@ppp-076.m2-2.sub.ican.net] has joined #tunes Tril? Tril is a security person. for tunes, though ,not unix ;) 10:30am Anyone here? Ho La Fucking SHIT!!! Well, I am ;) ino Bjoccara (Editor) / Hardcover / Date Published: January 1993 Our Price: $285.50 :) THATS CRIMINAL!!!!!!!!!! NOBODY SOHULD BE ALLOWED TO CHARGE THAT MUCH FOR A BOOK!!! hehe... You'r right! Alon: You haven't been to college or univeristy yet have you! :) All the books are very expensive I've seen them go up to $100 but 285 is BEYOND COMPREHENSION!!! Alon: Yep, it's certianly the most expensive I've seen... but I have seen some reach $198 I saw an IEEE text for $400 =\ MCSE (Ms Zombie) books are around $500 here * AlonzoTG/#tunes 's jaw hits the ground People actually PAY that much? Alon: How many MCSE's do you know... I know about 9 or 10 now... and they ALL paid for the MS books! * AlonzoTG/#tunes has no contacts. =\ 10:40am Alon: Socialize... Socialize with MS people, yuk! ;) * AlonzoTG/#tunes lives in a concrete bunker.. the door hasn't opened in nearly a decade... =\ AlonzoTG: That's not good for health ;) Alon: You need perspective... 10:50am -:- hcf [nef@me-portland-us105.javanet.com] has joined #tunes Hey hcf hoy beh sup?\ hcf: Notin much. On vacation at the moment doing anything useful? ;) hcf: Thankfully nothing! :) 11:00am -:- SignOff AlonzoTG: #TUNES (ADVENTURE!!!! =P) -:- SignOff billyboof: #TUNES (Leaving) -:- s_r [s_rr@phila-dialup110.nni.com] has joined #tunes hey s_r hey hi, s_r UniOS wiz sr: UniOS wiz? hehe... wish I has a UniOS :) beh well the best way to get it is to just start coding * Fare/#Tunes is back for a minute Hey Fare Fare: anythiing good in those links? -:- eStormy [stormy@rain.futuresouth.com] has joined #Tunes hilo estormy: /nick ethunder -:- SignOff Fare: #TUNES (Ping timeout for Fare[quatramaran.ens.fr]) * eStormy/#Tunes likes eStormy, thanks. are there lightning storms where you are frequently as much as anywhere else i reckon so you got your name from the concept of an "E-Storm"? Electronic Storm? my real life name is stormy, but that's always taken on irc, so i used iStormy, but then i started hanging on two channels on linpeople.org, so i needed two nicks, so in here, i'm eStormy (e=earth, the os i wanna write) 11:50am do you people ever comment on your real life name, like "what kind of name is that?" yes i haven't come up with a good answer yet "well my parents really liked marijuana..." afk a mintue -:- Fare [rideau@quatramaran.ens.fr] has joined #Tunes which links? 12:00pm Fare: those links i gave u recently Oh, only visited some of them I knew about half of them. -:- s_r [s_rr@phila-dialup110.nni.com] has left #tunes [] 12:10pm -:- AlonzoTG [Alonzo@216-164-229-182.s436.tnt8.lnh.md.dialup.rcn.com] has joined #tunes -:- SignOff eStormy: #TUNES (Ping timeout for eStormy[rain.futuresouth.com]) -:- SignOff beholder: #TUNES (Read error to beholder[ppp-076.m2-2.sub.ican.net]: EOF from client) -:- SignOff ProGuy: #TUNES (Ping timeout for ProGuy[p401-059.ppp.get2net.dk]) -:- s_r [s_r@phila-dialup256.nni.com] has joined #tunes -:- SignOff AlonzoTG: #TUNES (Ping timeout for AlonzoTG[216-164-229-182.s436.tnt8.lnh.md.dialup.rcn.com]) -:- NetSplit: sterling.openprojects.net split from koontz.openprojects.net [02:28pm] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [sterling.openprojects.net] -:- Netjoined: sterling.openprojects.net koontz.openprojects.net -:- Beam [nobody3@d95.nnb2.interaccess.com] has joined #tunes -:- Fare [rideau@quatramaran.ens.fr] has joined #tunes -lilo(lilo@varley.openprojects.net)- [GlobalNotice] Hi all. We just experienced a brief network outage to sterling, one of our major hubs. Apologies for the inconvenience. -:- tcn [tcn@cci-209150250127.clarityconnect.net] has joined #tunes -:- AlonzoTG [Alonzo@216-164-229-182.s436.tnt8.lnh.md.dialup.rcn.com] has joined #tunes hi tcn BEAM ON!!! hey ATG OFF! beam me up scotty be back tomorrow.. bye! later why are some ASM functions and C functions prefixed with an underscore? like: __syscall and such? Is tehre some rule for this? i looked through the ASM HOWTO and it doesn't mention this it seems to be a C thing I think each underscore indicates a lower level the idea is that the high-level version (the one you use in normal programs) doesn't have any underscores tcn can i write some forth objects for your kernel :) 02:40pm sure, maybe.. what sort of things? a video server that would be neat and a video game dev kit you called asm games a lost art gfx routines would be cool what would a 'video server' do? abi: seen Tril Tril was last seen saying something on IRC Mon Jun 21 13:46:45 1999 provide access to video ram and do graphics routines sort of like an X server ack or is this impinging on retro design standards? that's not what I had in mind for video.. just direct memory access direct memory access for trusted apps right? 02:50pm untrusted programs would have to submit the pixmaps to be put into the vram no, for any apps.. (well actually windowed apps will directly access an intermediate buffer) -:- SignOff hcf: #TUNES (Leaving) -:- hcf [nef@me-portland-us318.javanet.com] has joined #tunes and as far as X's networkability.. I don't see a need for special networked video support when the whole OS is a distributed environment >>> tcn [tcn@cci-209150250127.clarityconnect.net] requested PING 930174969 628390 from #tunes >>> s_r [s_r@phila-dialup256.nni.com] requested PING 930186094 from #tunes i'm still here, are you? >>> s_r [s_r@phila-dialup256.nni.com] requested PING 930186109 from #tunes heh tcn what have you most recently added to Retro? it's not what I've added, it's what I've deleted I'm rearranging making it more modular ? more forth-based 03:00pm maybe you should change forth to fit your needs i was sorta following two standards, forth and C enhance it make a variant of forth if you aren't making a new language forth is good for getting shit done so when is the kernel going to be written in forth? anyway I'm making forth more integral to the os, not just another module i'm not writing the whole thing in forth.. sometimes assembly is better i want to make a language that is just as good as assembly in its respects but also high level what Forth should be oh yeah irc 03:10pm hey, maybe the languages we have are fine, and we just need better OSes man, all week I've been wishing we had Unix at work even unix is good, compared to windows well, I wanna get off the net bye -:- SignOff tcn: #TUNES (tcn has no reason) 03:20pm -:- beholder [beholder@ppp-028.m2-1.sub.ican.net] has joined #tunes -:- SignOff Beam: #TUNES (Leaving) Fare: here?\ -:- SignOff s_r: #TUNES (Leaving) 03:50pm -:- beholder [beholder@ppp-028.m2-1.sub.ican.net] has left #tunes [] zergzer * Fare/#Tunes is back for a few minutes do u care about reflective sql? uh? reflective sql? what's that? http://www.cs.indiana.edu/l/www/hyplan/dalkilic/ I'm not sure about sql. Surely, we may add database-like structure in the language, queriable from the language, w/o the need of a separate SQL language (unrelated) also checkout http://pauillac.inria.fr/~gimenez/papers.html 04:10pm hcf: you send me so much material I hardly have time to scrutinize any! hcf: that's good... want more? ;) well, sure. Do you actually read the material? maybe YOU could produce short reviews... most of it is over my head i dont feel qualified to do reviews s/head/head or to time consuming to read. hum. givme keywords reflective SQL looks like a kluge as compared to a real persistent programing language (such as STAPLE or other british projects) hcf: implementation, logic, formal, systems, concurrency, etc but at least reflection is getting out there, u know sure. hmm i guess i should say keyphrases hum. Like? hcf: what search engines are you using? deja, google, alta, yahoo, meta, alltheweb, InferenceFind, odp, ftpsearch but that's reflection from the easy and dirty end nothing new I dunno meta, alltheweb, InferenceFind, odp. You don't use FermiVista? fermivista.math.jussieu.fr 04:20pm duz that come in english? sure has water tried this? brb back -:- _QZ [brand@p0wer.qzx.com] has joined #tunes 04:30pm hoy _QZ -:- s_r [s_rr@phila-dialup164.nni.com] has joined #tunes when underscores are used with function names and data in C, what does that mean? Is this some standard way of interfacing with asm? I see things like ___syscall, what do the underscores mean? grrrr! * Fare/#Tunes 's ISP disconnected me s_r: that's a standard way to produce non-standard identifiers can you explain or point me to someplace where i can understand it? s_r: standard programs should never use __identifiers, so that libraries may use them for internal purposes. or course, libraries may have __clashes, but this is no perfect world when you have a one global namespace. i see the ___vars all the time in asm objects that are part of a program or kernerl the kernel sources reference ___vars all the time i wonder why so that its exported structures/symbols do not clash with POSIX namespace see how I had to do it in byteorder/*.h include/linux/byteorder/*.h, that is 04:40pm but why do kernel sources have things like ___syscall I told you already. Because syscall would be a POSIX nameclash when someone #include whereas ___syscall is guaranteed not to be. ok but doesn't it have something to do with asm itself i think i remember from somewhere nope no no and no that things accessed in asm have to be prefixed with a _ because C compiler changes all symbols to have a _ in front of them you're confusing the issue with C vs asm identifiers. Many systems used to prepend and underscore to C identifiers when making them visible to the assembler, to avoid conflict with, e.g. register names. but gas resolves the conflict by using % to denote a register. -:- iepos [iepos@d14.t1-7.tecinfo.com] has joined #TUNES -:- s_rr [s_rr@phila-dialup497.nni.com] has joined #tunes ok what did i miss, fare so that such convention is no more needed, and ELF doesn't have this convention, whereas a.out did iepos! hey! s_r: you should read the .info for gcc and binutils. so there's no reason to do the ___ thing other than to make sure it doesn't clash with something else? s_r: isn't that enough of a reason???? s_r: "is there any other reason for that but to get the thing working?" -:- SignOff s_r: #TUNES (Ping timeout for s_r[phila-dialup164.nni.com]) "and why is the variable named counter instead of _237 ?" -:- SignOff hcf: #TUNES (Ping timeout for hcf[me-portland-us318.javanet.com]) hmmm........ mmmmh! so, what's on your mind, Fare? 04:50pm iepos what's on your mind? oh... still thinking about the paradoxes.... thinking about ontologies? ummm... yeah... sure whatever that is :-) Internet Explorer Position s_rr, have you played with combinators? no combinator? a combinator is a kind of function it takes a few parameters and returns a result that is just the parameters applied to each other... sort of... like the S combinator, symbolized sometimes by "x.y.z.(x z (y z))" hmmm it takes 3 parameters, x, y, and z, and returns 'x z (y z)' just a type of function in C-ish syntax, that's x(z)(y(z)) iepos what are your projects? any OSes? sort of... i'm trying to come up with a general-purpose logic system right now that doesn't have any inconsistencies... like a natural language parser? but also does not rely on bound variables... iepos: what about the one from my Master's Thesis ;-> well a parser is hardly a logic system... again, what you want is Curry's illative logic oh... i finally got a copy of his book. Fare would you consider Curry a brilliant man? and I gave a consistent interpretation of it in my masters' thesis & lambdaND s_r: more than brilliant s_r: why? just curious my problem is with this paradox... considering this statement: "Y \x.x -> Z" 05:00pm oops i never knew about this world of logic until i learned about TUNES. for me everything associated with computers was always straight code before that time i mean Y as the combinator ... and Z as a false statement if my system as substitution then I get '(Y \x.x -> Z) -> ((Y \x.x -> Z) -> Z)" ... by reducing the application of Y... then by natural deduction... you get '(Y \x.x -> Z) -> Z' just by getting rid of the duplication... iepos: the solution is simple then that is the same as the original '(Y \x.x -> Z)' when un-substituted... iepos: it's curry's E combinator, and my interpretation as call-by-value then ... from that you can prove Z... anything... looks like anything can be proven ... you get a sentence that *if it were true* would be false. Happily, it isn't true, and neither is its negation. so the solution is intuitionnistic logic yeah... another way of phrasing the problem statement is "if this statement is true, then "... iepos: only the sentence isn't true iepos: so that's no problem. indeed. i don't assume it is... suppose the sentence is '(Y \x.x -> Z)' where Z is the falsehood and Y is the combinator... that's FOO's paradox, for some suitable value of FOO (whose paradox is it, already?) you can prove '(Y \x.x -> Z) -> ((Y \x.x -> Z) -> Z)' by expanding the Y, right? iepos: Y (K \bot) iepos: no you can't hmm... why not? iepos: because can only deduce (P A A) from (E A), iepos: and obviously you haven't E (Y (K \bot)) umm... wait ... would you accept this statement 'S x y z -> x z (y z)' for all x, y, and z ? iepos: are you familiar with Curry's illative logic? no... not familiar with it ... would you accept that statement? iepos: no, only if E x, E y, E z "E x" basically means "x is well-formed". hmm... really... what is E then? hmmm one way is to consider it as "term x converges to some value v" the same goes with 'Y x -> x (Y x)' then... (in a call-by-value mind-set) so is (\x.x -> Z) not well-formed then? yes it is, but Y (K \bot) is not. i don't see the connection ... i'm not even using K... and what is \bot? That is, we lose P (E x) (P (E y) (E (x y))) 05:10pm sometimes, we have E x and E y but not E (x y) wait... you say that 'Y x -> x (Y x)' holds if 'E x' right? maybe that's what Curry did wrong. I think he wanted that E x and E y imply E (x y) but maybe I remember wrongly. and you say that 'E \x.x -> Z'... right + ? No, it holds if E (Y x) aaaaaaaaaaaaaaahhhhhhhhhhh! or rather, if E (x (Y x)) hmm.... i figured that a limitation of that sort would be necessary... but.... again, think (E x) as "evaluation of x converges to a value v" I don't remember if Curry requires E x => E y => E (x y). If you have access to his book, please have a peek and do tell me... all right... just a sec If he did, then there's the opportunity for writing a paper... i haven't read the illative logic section yet you definitely should I skipped most of his book (knew lambda-calculus from other means already), and jumped to that chapter He modifies it in volume II but the basic idea is in volume I already maybe he modified the E statement in vII... don't remember -:- s_rr [s_rr@phila-dialup497.nni.com] has left #tunes [] ... Rule E. EX, EY |- E(XY). ok, he got it wrong, then. Does he do the same in vII ? i don't have vII or maybe I'm mistaken about the intended meaning of E? so, what is wrong with that? iepos: the Y paradox is wrong with that iepos: is EX the well-formedness of X? not sure... maybe I got the letters wrong. rule E is evil, in my book. he uses another squiggly-like E... maybe that is what you are really talking about... that one with the rule is bold normal E. -:- SignOff _QZ: #TUNES (BRiX [http://www.qzx.com/brix] :: testing new modem) 05:20pm hmm... actually he mentions that requirement of convergence as being one way of avoiding the paradox suggested by Church... but he mentions a case when it prevents useful valid reasoning .... which caser? s/r// he says the statement 'QY(NY)' is valid but yet has no normal form so he mentions the solution I gave. Good. not sure what Q is here... what's Q? N is negation and Y is the combinator i think Q is equality I think. isn't Y a blind variable in thisi context? ah... i think you're right not bold... Y is something that doesn't have a normal form yes, Q is equality, I remember now. But Q Y (N Y) is obviously false! oh, if neither Y, nor (N Y) have normal forms, then it is true! well.. he says its true 8) (or at least, it is consistent) i don't think Q is equality I don't remember in which notebook I wrote down the meaning of all his combinators... He has a table near the end of the volume... oh nevermind.. it is equality yeah, if neither Y nor (N Y) have normal form, then it's true so we have expressions that converge despite that their subexpressions don't. maybe that's something he wanted to avoid. -:- ruben [user5219@148.236.3.89] has joined #Tunes maybe he wanted both rule E and the inverse rule. but both these rules are evil. 05:30pm Hello, could some body help me whit a problem? what problem... ? problem is I don't have a ppc... just an o-l-d performa 476 (68040) abi: no problem is you don't have a clue! okay, Fare. I want to run a program compiled in SCO, in a linux machine ruben: what you want is named iBCS ruben: search HOWTOs for it. every seems all rigth, the program runs , but the terminal type is "sensi" gotta go bbiafm i have the sensi terminfo file in /usr/lib/terminfo/s/sensi i can see the information that i couldn't see whit any other terminal type i tried before, but sensi terminal type, don't let me use ... keys I need use then. hmmm.... well i don't know how to fix it. -:- _QZ [brand@p0wer.qzx.com] has joined #tunes do you know about any other channel or group ? not really.... bye, thanks ... :-) -:- ruben [user5219@148.236.3.89] has left #Tunes [] fare, are you back... ? -:- SignOff _QZ: #TUNES (Ping timeout for _QZ[p0wer.qzx.com]) 05:40pm -:- _QZ [brand@p0wer.qzx.com] has joined #tunes hey, QZ... what's going on with Brix these days? <_QZ> any1 know anything about serial ports? not sure... used to... they're not too hard... used to have a weird doc... told hardware and software parts... you could always look at the linux source... -:- SignOff _QZ: #TUNES (Ping timeout for _QZ[p0wer.qzx.com]) -:- _QZ [brand@p0wer.qzx.com] has joined #tunes 05:50pm -:- SignOff iepos: #TUNES (iepos) -:- SignOff _QZ: #TUNES (Ping timeout for _QZ[p0wer.qzx.com]) -:- Kaufmann [Kaufmann@200.224.105.192] has joined #Tunes Anyone in the US mind doing me a favor? -:- _QZ [brand@p0wer.qzx.com] has joined #tunes 06:20pm <_QZ> im doing 37333bps right now :) Unbelievable, that's actually worse than mine <_QZ> hey its better than the 31200 i have been doing for years <_QZ> i have tried several 56k modems and none could do over 26400. my 33.6 courier is the only one that could do 31200 <_QZ> but my friend just flashed 2 of his couriers to v90 and let me try one <_QZ> now im gonna have to pay $60 to flash my courier QZ, my nominal speed is always around 44000 to 48000, but it never really gets beyond 9600bps because of the shitty phone lines. <_QZ> yer getting 9600bps connections? <_QZ> wow 600Bps increase in speed is noticeable :) 06:30pm heh 06:40pm -:- hcf [nef@me-portland-us831.javanet.com] has joined #tunes BRB -:- SignOff Kaufmann: #TUNES ("Nationalism is an infantile disease. It is the measles of mankind." - Albert Einstein) * Fare/#Tunes is back (not for long) -:- iepos [iepos@d22.t1-1.tecinfo.com] has joined #TUNES gakuk, iepos! gakuk! oh, in that "QY(NY)" .... Y is the "this statement is false"... he defines it as YN, where Y is the real combinator had to look back a few pages :-) ok, but then it's a particular case of the equality of statements Y such that neither Y nor (N Y) has a normal form which is perfectly ok and non-contradictory it just means we have an intuitionnistic logic. looks like he's just expanding the "YN" to "N(YN)"... Q(YN)(N(YN)) where Y is the combinator 07:00pm there are many consistent ways to interpret this i think what he says is that expansion of combinators in this way is okay... the problem with paradoxes happens when you extend axioms about implications (like modus ponens and deductive theorum) to apply to everything... i mean propositions, not implications. i'm curious to see how he defines "proposition" though... in a way that excludes self-referencing statements.... also, Curry wanted a deterministic interpretation of his combinators; I give them a non-deterministic interpretation instead so actually it looks to me like my original '(Y \x.x->Z) -> (Y \x.x->Z) -> Z' is okay it's just that removing the duplication from that is not okay because the things are not propositions hmmm so I use the \lhd combinator to say that a form may be reduced into another which doesn't imply equality yeah, but say that "X a b" (where X is some nondeterministic "combinator") reduces to either 'a' or 'b', then reduction no longer implies equality... exactly i thought the whole point of reduction was finding a normal form for the same thing. not but your "reduction" can lead to something different than what you started with a different meaning, not just a different form, even under the same interpretation that's the point of normal form in a *confluent* calculus I sacrifice confluence to higher goals. yeah... i guess that f oops i mean... do you intend to use this non-determinism for just computation... or are you trying to base a logic system on it as well? or else, you introduce a different kind of reduction that preserves semantics, at the expense of not having a normal form ever. I'm basing the logic on it, as well.