IRC log started Sun Jan 9 00:00:01 2000 any mention of beta on that list? they probably wouldn't mind working on it, but i don't know if they're language-bootstrapping savvy not that i've noticed [msg(TUNES)] permlog 2000.0109 hm. squeak lacks a parser generator hmm bineng is working on smth self inspired, he'd probly be interested in slate hmm but the parser itself isn't too bad to grok yes one prickly point is to figure out how the beta philosophy will mesh with self/squeak's i.e. having methods inherit the way self objects do is not for the faint-hearted that is, it's not hard to understand, it's just really damned powerful hmm * water/#tunes is reminded of watching his father's friend at the controls of a P-38 lightning for the first time scared the hell out of him and he was an accomplished pilot, too * Downix/#tunes laughs I love the P-38 he circled the airfield several times before he felt safe about landing it in one piece good work he never even took the engine past 50% throttle no, he was too *scared* to land he never flew it again, either wow 2000+hp can't just be flown around by anyone 12:10am i forget what it's actual rating is 2830hp in the P-38l seriously!?! Seriously the L model upgraded the superchargers yikes water: http://www.hellblazer.com/personal/squeak/ that's insane it's necessary what for? if you put it into a dive for any length of time, the tail'll break water: Well, you're not supposed to do that it's to carry the bombs it does, and the fuel tank hm The P-38 is not a dive bomber did you ever hear of the Pond Racer? for dive-bombing,try the P-40 and P-39 no kidding yes i know i was just pointing out that the p-38 was very easy to break doing that I know trust me, i used to be a big aerospace freak (still am) Same whatever happened to this guy's squeak compiler project? i wonder if it's the same code as the current compiler hmm so did you ever look at the pond racer? yes, once nice plane, eh? Yes I like the B-25 the most although you couldn't pay me enough to fly(die) in it hehe i like the p-51 personally I'm a bomber fan B-25, B-56, B-52 nah, fighters are more fun A fighter can't destroy a city right :) 12:20am sorry, i don't like war Me either which is another reason I like bombers MAD is the ultimate deterrent mad? i guess mad is the ultimate deterrent abi, MAD is Mutually Assured Destruction mad? rumour has it mad is Mutually Assured Destruction there oh yuck big yuck yeah, but it has to be really fine-grained turn you into gooey spray * Downix/#tunes nods otherwise it just becomes a power-play Yup I've been working on a way to re-define war i have my own little mad scheme :) hehe level the playing field via brain-powered informatic weapons lol theoretically, you could treat the human race as an internet even without any technology considerations My idea was to simply use psychological terror the trick is to know how to get the information to propagate lol i did some undocumented research on this in college seriously, you can float information over anyone cool Iknow if done right, no one picks up on it... it's related to various existing ideas in psychology * Downix/#tunes often times has his thoughts propogate to people nearby i have a plan to set up a new language system using these methods cool it's very unfair, but i don't really care any more :) * Downix/#tunes is thinking more in basic weapons of war well, i was considering amassing a lot of tools for this information-warfare via the language system into a group of people cool * Downix/#tunes feels like such a pragmatist -:- SignOff smkl: #TUNES (Ping timeout for smkl[glubimox.yok.utu.fi]) they would be a special forces type team tht would terrorize without anyone being able to identify what was going on hehe * Downix/#tunes sighs I just want war tohave set rules again hell, even if it doesn't work, it'd make for a great book plot 12:30am to minimize civilian problems lol yeah whatever It can be done you know read "the nova police" * Downix/#tunes nods by william s burroughs I have cool! I want wars toend, but I'm pragmatic people will always be selfish i don't think so to me, it's a disease of language not something that has to do with people, but with inherited information hmm on the other hand, i find it very rare to actually talk to a person i usually find myself talking to gobs of information pouring through a body with an oblivious mind inside it * Downix/#tunes nods I keep asking where idiots come from is it education or genetic? it could be toxins Regardless, I want to know what causes it i.e. environment both physical and sensual and social * Downix/#tunes nods idiocy is a force it's a weapon against the development of the individual yes * Downix/#tunes nods it forces people into complacency and stupidity well, i consider it at a more abstract level ok i suppose you've read some nietzsche or no ? Some I keep seeing something I almost akin to a conspiracy worldwide have you ever really contemplated the eternal return idea? yes the conspiracy is that no people are involved * Downix/#tunes shrugs darn it, the beta download is taking forever hehe it's a good thing i have session-management for downloads anyway 12:40am hm... not sure where to place the conversation now I keep pondering, who controls the conspiracy tho I know people don't i postulate information beings an extension of plato's world of ideas sans causality but i consider this world of information to cross-cut our physical world I keep pondering what someone told me once which actually doesn't contradict any known theories * Downix/#tunes nods yes? He said that this world is just a single layer, with others on top of it hm i don't like the vertical analogy, but ok of course, i'd use the term "ontology" :) * Downix/#tunes nods top is simply a way to translate it One time he showed me how to peerinto the other"layers" hm how's that not that i haven't done something similar Using a common analogy, "Astral projection" water: http://x21.deja.com/=dnc/[ST_rn=ps]/getdoc.xp?AN=523370995&CONTEXT=947407458.1260847108&hitnum=0 k k I've forgotten how to do it now tho hm... he works for oracle, eh? hehe how could you forget that? water: I was so shocked over the experience. Still am what is it... some sort of arcane technique? what's your friend's email/url? he doesn't have any and it's hardly arcane we did it in a dance club oh i've done quite similar things it's almost like falling asleep, but you remain awake sure i did that the other day in my first tank session sensory-deprivation tank * Downix/#tunes nods i'm not very good yet 12:50am i am very skilled at moving through thick crowds with my eyes closed Same here using meditation techniques * Downix/#tunes nods have you ever considered constructivist or non-positivistic paradigms? Not really Meditive techniques are quite powerful hm difficult to explain I know * Downix/#tunes lost his center focus tho basically, the idea is to abandon the idea of self as a real thing, and to instead treat self as a representation * Downix/#tunes nods I've done some freaky things through that before language works very well this way it's not freaky, it's actually quite sane Well, freaky for other people but it spooks people Yes but that's irrelevant, because their fears are just weapons for info beings One night some friends and I were swapping heights * Downix/#tunes nods (which is harsh to say, but true) VBery true swapping heights? We got so in tune with each other, we began swapping how tall we were we ranged from 6'5" to 5'4" you mean actually changing vertebrae structure yes it was strange i find that hard to swallow Well, it was subtle for the most part I'm now 6'3" I was 6'1" And I know that, I was measured the day before and after Noone wants toswallow the last stage of acceptance, that even our forms are flexible well, that's not something i would ever advertise I don't water: ur aware of tgen right? tgen? a smalltalk parser generator k oh yes, where's it at? abi: tgen tgen is a smalltalk parser generator at ftp://st.cs.uiuc.edu/pub/Smalltalk/st80_r41/T-gen2.1 i completely forgot about it -:- water [water@tnt-10-181.tscnet.net] has left #tunes [] -:- water [water@tnt-10-181.tscnet.net] has joined #tunes and some other urls icuc, http://st-www.cs.uiuc.edu/users/droberts/ http://st-www.cs.uiuc.edu/~brant/ http://java.sun.com/people/gbracha/nwst.html k sorry, i hit the wrong key-combo abi tgen? rumour has it tgen is a smalltalk parser generator at ftp://st.cs.uiuc.edu/pub/Smalltalk/st80_r41/T-gen2.1 01:00am water: know what smalltalk MT is? yes http://www.ruhr.de/home/kottan/tgen21/tgen21formt.htm it's funny how all the things oo people are fond of for software engineering are just reactions to limitations in language http://www.objectlink.co.jp/T-gen/index.html wow thanks * eihrul/#tunes coughs. hey wb was teaching someone lisp :) i think i made a convert cool goodie, more slate talk now :) heh rmm, i have 5 pages of philosophy talk unrelated to slate so -more hm as long as it's not objectivism :) i mean in scrollback oh 01:10am why is there no Leak (lisp squeak equivalent)? you'd figure, given 30 years, we'd have such by now :) heh because lispers don't see the point and becuase squeak is the system smalltalk was made for sure, but lisp is practically a whole operating environment all that's left is to just port it atop the bare hardware i.e. on palo alto machines and such true, but gui's seem to require something that lisp doesn't do well being? which is top-down programming lisp seems strictly bottom-up not always true you can program top-down in lisp you're just wasting the language if you don't but that's the lisp community mentality er rather go bottom-up well, code a useful gui in lisp, then where do you start? gimme time... if i fail, you'll be the first to know :) hmm hmm tongue tied? whom? either of you? I am nope, just thinking * Fufie/#tunes yawns welcome hi foof morning 01:20am Ok, time for bed then Fufie wakes up that is -:- SignOff Downix: #TUNES (Read error: 1 + 2 = -9 (Connection reset by third grade math)) I am heading back to bed in ten min. fufie: what would you think of a self/beta hybrid ? oh I don't know self well enough to pictuure it water: what is novel of beta that you would be reaping? it's just like smalltalk why would you make ya smalltalk? beta means that all procedures would be objects er.. functions and could inherit etc no offense.. but lisp already does that water: function inheritance? elaborate fufie: it would be a prototype-based object system lisp is low-level lisp can be low-level i know lisp does it lisp would allow bypassing the object system too sometimes that is needed in real-world apps and i'd like the object system to be central real world apps are not useful :) :) they're expedient :) besides, the object system would allow for some useful optimizations not to mention that the objects would be immutable by default you want it to be object-functional? so that i'm really discussing a self/lisp/beta hybrid yes water: but doesn't the object system also impose overheads of its own? :) eih: not necesarily. self and beta are both quite fast I have some convenient macros I use in lisp to make things more o-f and to avoid side-effects but it would be nice with a rewrite of lisp where you could impose side-effect free code in some parts for different reasons, but not incompatible ones s/free/for/ (I'm tired) it's ok water: ftp://ftp.cso.uiuc.edu/pub/lang/smalltalk/Smalltalk/st80_vw/T-gen2.2/ i'm going to include a decnet reflection system to allow for side-effects when useful but the core language should be verifiable to some degree hcf: k verifiable? ack don't go down that path well, side-effects make verifying source difficult why not? everyone who goes down that path is lost they don't come back well, it's not an immediate goal what do you mean by lost, anyway? and then they make some completely unusable language which noone understands and is documented in algebra and then they smile overbearingly heh 01:30am and say that they removed all useful things because they didn't fit into the model no, this is just a toy system for a while but the new language is the new solution to all.. it doesn't need arrays, lists, closures, etc mostly an exploratory code system heh we have a few of those at the institute who did that? wow it's the guys who love verifying code i don't love it, but some people need it i'd just like it to be simpler but it is dreadfully complex.. take threading don't worry i know about threading and there *are* ways to simplify it e.g. maude it has a system for specifying concurrency mechanisms as well as gc ack.. if they only could make the language readable to humans as well that should be some help it's a logic maude looked as bad as perl who gives a s** I do i'm only saying that the semantics can be known I like elegant languages where you can read the code lisp is one of those I used to like smalltalk as well well, this language will not be like maude ok? (I probably still do, but haven't touched it for years) yes, smalltalk is quite legible anyway, i'm calling it slate for now the syntax will be sexp-based by default good which makes reflection simpler and makes things really powerful through macros which are very useful whenr efactoring the objects will be self-like however, although also sexp-based refactoring right but i want the object system to be central surprised that noone mentioned lisp macros in the mail thread (I am not done yet) with the morphic interface from self adapted refactoring is in essence bottom-up programming well, bill is no lisp fan, and the others are idiots in a lot of ways, yes i want to consider it at an abstract declarative level, however 01:40am I figured that out and it is a noble task not necessarily logic-based, but declarative vice procedural having an analyser to find patterns would be nice well, i need to have a good system for specifying arrow graphs hm woo hoo! only 4 more minutes of downloading beta the mjolnir beta? yep it's too bad they haven't opened up the source code i wonder if they have a beta-interpreter in the code library that would help beta but not mjolnir I guess definitely well, they advertise the system as a "rapid development" tool i'm not sure if that kind of user-base would care they do however have a meta-programming system, but i'm not sure how capable it is I am starting to hate the meta word it's hard to find any tuneser who doesn't use it in every sentence yeah, it's way too vague oh well morning rambling that's a bit drastic 2am here 01:50am hmm. this parser-generator in smalltalk seems pretty flexible wow. this thesis on refactoring is ~8mb too many picture :) pictures 8mb of post script? uncompressed how many pages? it's still downloading hmm... time to unpack this 33mb tarball from mjolner this should be interesting no, it seems more like the paper is 11mb wow dmoz has a lot more tunes links in it now 02:00am http://search.dmoz.org/cgi-bin/osearch?search=tunes&cat=Computers&t=b&fo=0&all=no hmm.. time to see if NewCompiler for squeak breaks my system -:- smoke [smoke@15dyn101.delft.casema.net] has joined #tunes well, the test cases work hcf: yeah, i noticed that someone likes tunes at dmoz :) (since dmoz is now about the only directory i really use) -:- water_ [water@tnt-10-72.tscnet.net] has joined #tunes -:- SignOff water: #TUNES (Ping timeout for water[tnt-10-181.tscnet.net]) -:- water_ is now known as water 127 pages in that paper nice and language-invariant, too cool! it uses predicate logic as well :P 02:10am abstract syntax-trees, dynamic refactoring... and details on the smalltalk refactoring browser oh darn it looks like my download was cut off oh good, there's a more compact pdf format for it heh. am i the only person here with stuff to say? :) *** HaHaHaHa (Hackmail@98C99A37.ipt.aol.com) has joined #f-cpu hmm, that's not good hm? wtf was that? an aol'er why is he broadcasting all over the server? he isn't oh (cut'n'paste, sir) n/m -:- air [brand@p0wer.qzx.com] has joined #tunes 02:20am air: how do you classify the brix language? meaning what? well, is it class-based for one? yes what is method code like? uhh heh i can't find any info on the language on your site air: isnt ur lang modeled after qbasic? ;) lol ha ha ha water: i sent u a file on it umm that was quite awhile ago and it said very little at dmoz.org under the TUNES heading it says that coding has begun well, it did but it stopped? i think people got fussy and/or were ignorant 02:30am there's a lot of intellectual dead-weight in tunes, if you haven't noticed ack! i hate it when someone' postscript paper isn't numbered, so you can't scroll back what is dmoz.org? sometimes they also print in reverse order netscape open directory (a real bitch to read if you can't read the document beginning to end without reloading every page) who placed those links on dmoz? whoever the dmoz editor is for that category i think the root page will tell you cuz it has brix as 1 of the 5 links on the tunes category and it says brix is allied with tunes yep well, those links got there about a year ago jerry fass what would make him think brix is allied with tunes? hell if i know maybe that you've announced brix crap on the tunes mlist nope 02:40am hm dont think i ever posted anything to the tunes mlist is brix pointed to by the lll page? actually, that's retro i think they confuse retro with brix brix is on the os review page and the description on dmoz is from the os review 02:50am it's a huge task keeping track of alliances and cooperation ho hum Fufie: eh, you make this sound like a war it is a war against poor development tools, languages, OSes and systems here here well whatever fair enough we have no problems finding enemies, just probles finding the same enemies pproblems heh 03:00am lol... it would seem slate would be anything but monolithic boy it's a tough crowd tonight it's morning, AFAIK so literla, too :) eihrul: have u been to bed yet? no then its still night no, i sleep during the day so do i, and its night right now -:- AriB [arielb@209-122-226-70.s324.tnt1.nyw.ny.dialup.rcn.com] has joined #tunes -:- SignOff water: #TUNES (Read error to water[tnt-10-72.tscnet.net]: Connection reset by peer) 03:10am hi hi just found out about the tunes OS project umm so, did you like what you found? what's the status so far? many ideas, a good review section about prior art and several other projects partly related to tunes.. now tunes is supposed to be an OS without a kernel, am I reading that right? yes -:- water [water@tnt-10-116.tscnet.net] has joined #tunes what's the advantage of that? darn win32 03:20am its like forth, where the language is the OS i can't wait for my modem to work with linux water: winmodem? water: any luck w/ the local lug? air: i believe so. US robotics hcf: haven't contacted them yet except forth isn't as well-developed for os-coding as people would like it to be forth I remember was supposed to be very fast it is rather fast is it unix compatible? and there is an os based on it... 4os lol -:- billh_ [billh@cx739861-a.dt1.sdca.home.com] has joined #tunes no not forth...tunes there are implementations of forth on linux lol well, if someone writes a posix interface, yes but that's not the point in tunes you don't just run binaries you interact with the objects that are your data and programs The Posix interface isn't important for running apps that have been built from the ground up. hi bill Hello. do I know you ? i don't know. do i know you? :) Posix only matters if you're running Unix base apps. i'm brian rice, the arrow paper guy ok ;-) such as benchmarking apache :) arrow paper guy ? bill: and you? aib: arrow? you mean as far as what I do ? abi: arrow? well, arrow is a homo-iconic information manipulation system intended to support programming and human languages in the way that Tunes should. at http://www.tunes.org/papers/Arrow/, http://www.tunes.org/~water/ doh! This you ? hm? The URL ? yes, that's my tunes home Is this something that you're involed in ? hehe John Stuart Mill ? i'm the one-man development team That's unexpected. Most computer folks are that broadly educated. Most computer folks are *not* that broadly educated. thanks 03:30am most people know who JS Mill is, or? They are philosophy folks of the western analytic kind. most people who grok tunes know who he is most people who have had any exposure to history whouls know who he is Yeah, my primary focus in college was Computer Science and Cognitive Science. would fufie: depends also I've got pretty damn strong continental philosophy background. Heidegger, Nietzche... cool! sorry, big fan Theodore Adorno, (frankfurt school marxist).... But I don't know what a close system is. closed system? This has been mentioned several times in various papers representing knowledge. btw, the proper arrow paper format is at the second url close logic/rule. I don't remember. as arrow philosophy.ps -:- SignOff eihrul: #TUNES ([x]chat) n takes knowledge to be the closure of an information structure wit That statement above is a bit over my head. -:- SignOff air: #TUNES (BRiX [http://www.qzx.com/brix] :: sleep) well, i wrote the paper while on a warship in the persian gulf over _your_ head? ok now I'm doomed Oh, hahah. I just need to understand what a formal closed system is. i didn't have anyone to proof-read a closed system is different from a closure with my use of terms ever looked at non-monotonic logic? not a programming closure I missed that when I skipped the database classes and related material in CS. no I don't really know what that is. they don't teach it in cs, not properly hm. I probably do from studying AI, but it's not within my terminology right now. well, a turing machine formalism is a closed system "ontological" ? What the fuck are you using Heidegger's stuff in this CS paper ? it's a term from knowledge-representation not the old stuff it means the symbol set of a formal model Well, it's not specific to Heidegger, but the way you use it it's like opposite of what he means.. yes i know I'd also use the term "epistemology". i have used that term since then, yes it hasn't made it into the paper properly, though Yeah, I was a pretty heavy philosophy student in college so I know. "agents" as in Intelligent Agents" ? as in the agency of computation ok. yes i should clarify that i agree I'm skippiung around in this paper so I might be missing some things. Yeah, especially in an AI paper. there's a lot i have in a queue 03:40am not quite ai damn close, but not the same thing water, are you formally trained in what ? Not AI, right ? like discrete math or something ? i'm sorry to admit that i'm mostly self-taught water, that's ba. bad luck with legal issues water, that's bad, because this doesn't open you up to peer review. kept me from getting a full scholarship to mit i know Huh ? it kills me every day what happened ? how old are you ? i have this trust fund that killed it because of technical details you'll find MIT to be the ultimate American shit hole. what details ? Fuck'm. heh. well, my sisteer is using it up in 2 years, i'll be able to apply anywhere i want to hopefully by then, i'll have a core version of arrow running how old are you ? 22 Ok, you're kind of like me. but I'm 31. well, if you could get me some peer-review, i'd appreciate it water, where do you live in california ? some professor friends of mine completely ignored me when i joined the military seattle why did you join the military? of all things bored, and i hated college also to save money and invest but still the military.. well, i do have $30k Huh ? they shunned you what you told them you were military ? that's how much i've saved so far yep well, tell them to fuck off. Where did you go to school ? well, i wasn't that rude about it college ? texas a&m i tested out of 60 credit-hours it was a f***ing joke where's that ? heh. college station, tx water, I'm a really horrible student. I couldn't have never done what you did. I remember teaching a class with some people who had been in the military.. had to teach them to relax and use their fantasy :( well, i studied like hell This paper is difficult to read. i know a hell of a lot more than just cs like what specifically ? MIT is overrated. this arrow thing is sort of a first stage of more projects like what projects ? water, go to Berkeley or UCSD. higher math, linguistics, physics, chemistry, ... Make sure that you go to a UC campus with punkers. you know math pretty well ? philosophy of all kinds yes Ok, then what's quaternion ? 03:50am Ok, then what's a quaternion ? basically a 4x4 matrix that follows a few algebraic rules assoc it's the basis for good 3-d graphics engines well, it's a good formalism for 'em Well, that's not what it is, but that's not important. it's close i don't deal with it very often at all Well, it's a 3d extension of the imaginary number plane. that's not how i learned it but ok is this supposed to be a test or something? :) It's a 4 scalar vector (1 imaginary, 3 real). really ? really AriB no, just trying to understand where his head is. ;-) WTF did they teach you in that classes ? WTF did they teach you in that class ? what classes? the Quaternions stuff. the highest math class i took was diff-eq Ok, but did they teach you quaternions there ? and i read the f**ing book before i took the class no Then how do you know it ? i never learned s*** from classes i read books I'm trying to understand it your learing process. -:- SignOff hcf: #TUNES (Ping timeout for hcf[me-portland-us912.javanet.com]) he told you...self taught i teach mysself from books Ok, I understand. But your definition of what a quaterion is pretty way off. i haven't learned anything from a class since the first grade, with the exception of 9th grade geometry water, try and get into a UC campus. Like Davis, Berkeley or San Diego. where i learned of bolyai-lobachevsky's axioms and the formal method punt on LA. sorry, i don't really like ca Seattle is better for you ? no ok. i work here because it's the best option they gave me there aren't too many navy ports with nuclear aircraft carriers where do you work at ? Navy ? bremerton across the pugett sound So you're still in the Navy, then eh ? yes, for another year or two I was under the understanding the your were out out or something. Ok. no, they like to keep nuke technicians yeah, fuck'm. Go to school. You'll meet folks like me all over the place. who say "fuck'm" every other sentence? The paper suffers from you using alot of words that are a little off-standard from normal acedemics terms. You should defined them better before you use it. ok Its ok to do so as long as you defined them first. -:- hcf [nef@me-portland-us311.javanet.com] has joined #tunes Because this paper is boardline unreadable. ok, but it's hard for me to get a feel for that and you don't define reflective before you use it. 04:00am i don't know anyone in real life who understands even a little of it really? So i'm inclined to think that you're talking about Java programming. ;-) yeah. That's incoherent. i didn't notice that that's why I brought bill here :) thanks a million yeah, it's pretty confusing writting overall. really But if you go to college you'll be over the head of most writting teachers. well, there's absolutely no peer review you really have to go to pure philosophy folks in order to get this paper clearer. yeah, I'm kind of fitting the obnoxious role of being a peer reviewer. ;-) the only phd in tunes actually insulted me and refuses to review anything i say Huh why ? don't know in tunes ? what's that ? who has got a phd? he probably thinks i'm some schmuck who only knows win32 john shapiro you have to define terms before you use it like "tunes". who's he ? lol you're on the #tunes channel see what I mean ? http://www.tunes.org Yeah, but I don't know what the fuck it is. look at the docs abi: tunes? tunes is probably a free reflective computing system at http://www.tunes.org or for programming languages what the internet is to networks or like flypaper for hallucinogenic programmers or discombobulated or obnubilated or AOTA no one knows what it is :) but then, most members are pretty idiotic really ? why's that ? the think GEBis a good book Tunes is a Useful, Nevertheless Expedient System sorry, i'm getting tired fingers are getting lazy water, when you read Heidegger you'll punt all this shit. and John Stuart Mill. well, anywa, if you look in the references section of the paper, there are works that use the terms the way i do well, i stayed away from tunes for years http://www.tunes.org/papers/WhyNewOS/index.html What section from this URL is a good overview of the project ? what about that? none probably the HLL section is the only good explanation of tunes http://www.tunes.org/Tunes-FAQ-4.html maybe here yes, that does well i don't write anything on the tunes site besides arrow and the occasional language or os review yeah, the philosophy is bullshit. you'll learn that when you get more training. i'm not sure if i'll ever appreciate training, honsetly It's based on Kant's notion of a priori knowledge and the missunderstanding of the Subject-Concept-Object relationship. yuck now that you mention it well, that's not the point 04:10am any more comments? hold on. i could sure as hell use a lot of mark-up on that paper I'm reading the glossary. you know about the language Selff ? you know about the language Selff= ? and i can definitely explain a lot of stuff you know about the language Self ? hell yes we were just discussing hybridizing self and beta based on lisp function-object ideas earlier The Centralization defintion is pretty funny. It bags on a bunch of folks. ;-) that's all Fare who'll probly be here sometime soon i'd like to think so "Functional Programming is a very powerful paradigm for programming that originates in the early works about lambda calculus: com" Bullshit. LISP must die. * Fufie/#tunes smirks heh Self's typing structure is self contained and can't really be use for other languages. It's purely dynamically typed. umm.. javascript uses basically the same system and is opposite of stuff like Dylan from my readings of the Self "concrete type inference" paper. AriB You know what we're talking about ? Dylan fell into the trap of making infix syntax oh I wasn't following the channel...it's 7 AM here :) Ok. It's pretty late here too. PST. Yeah, I'm a Java fan. yuck and purely a Java/C+ fan. EST...I slept all day smalltalk >>> Fufie [stig@tunnel-44-22.vpn.uib.no] requested PING 947420920 from #tunes well, squeak really why ? much more accessible bill and I were talking about Elate before which is what Amiga will use (yeah right) 04:20am also there is a greater platform-independence if they can actually pull it off this time-sounds pretty cool to me amiga... the great white hope tell me all about it :) if I had three weeks undisturbed time I think I'd taken the PopLog VM+Lisp and tried to make a squeak look-a-like for tunes to get started with fufie: yeah you and everyone else and then gradually replaced Poplog and language features till we found the HLL for tunes i don't care. i had a better idea than tunes when i was 15 water: I don't have those three weeks :( and i coded a tunes prototype in college that booted off a floppy into a lisp-object based gui it took me two weeks Self boardlines on being so generic that it's almost unuseable. well.. I am not so smart.. I will need three weeks :-) have you ever used it? have you even used smalltalk? genericity can be an advantage if used properly Yeah, I've use close relatives of it. like stuff stemming from common LISP. yikes common lisp is a bloated monster totally incomparable to smalltalk or lisp er.. self Yeah, after reading the paper, I'm convinced that tunes is impractical. not lisp which paper? http://www.tunes.org/Tunes-FAQ-6.html#ss6.1 heh Just the section that you posted. Yes. oh Self is about as fringe as I get. common lisp is not bloated.. it's muscle.. a decent monster needs those muscles and it's an incomplete language. Yeah, I hate LISP and related languages. What do you guys think of Eiffel ? Eiffel is a fascist language this is all horse-shit just like Ada So you hate Ada and Eiffel ? you're a coder just like the rest of the tunes idiot I said they were fascist languages.. I don't hate other languages than javascript * Fufie/#tunes gives water a bonk over the head you're basing language-preference on ridiculous metrics Haha, the please explain what a "fascist language" is ? which metrics are those? ;-) water tunes is a CS disipline. And a confused one at that, ;-) cs is a confused discipline 04:30am CS is great. tell me, what's a computation and what's not? bill: a fascist language is a language that has some idea, it might be good, but usually is not, and then they try to force it upon the people programming in the language.. and it's done brutally and with no mercy well, that's a loaded question with no clear answer. no, that's the confusion of cs Fufie Ok, then what's your language of choice then ? bill: to write elegant and good programs, programmers must have freedom and be able to reflect and work with the language water, ok, then do you think that a human thought can be quantified into a measureable quantity ? no, but a representation can bill: I like common lisp and scheme, though there are a few other languages I like but seldom use Fufie Like what ? example please, try not to be vague. Fufie You like Common LISP ? Why ? yeah why the hell would you like the language that i don't like? ;) phone.. two sec Ok. please continue... when you're ready. water, ok, then do you think that a human thought can be quantified into a measureable quantity ? I'm asking again. i answered Then you need to explain. because you leave alot of dangling concepts. representation theory vice type theory? * billh_/#tunes studied this in college. well, then, there shouldn't be much to explain Water, stop being an asshole and explain it. well, what's a human thought? yes and also a representation. who's to say what counts and what doesn't count as a human thought or a quantifiablre aspect of it? water Explain both terms before you use it. wait. ? just explain what a representation is. thanks CL provides the most powerful tools I have yet to see in a programming language, which all are pretty orthogonal compared to most other languages, and that they are configurable and possible to change by the programmer.. you can rewrite the language to suit your needs and create elegant DSL which is not even remotedly possible in other languages (except maybe scheme+addons) DSL ? it has been said that DSL is domain-specific language or digital subscriber line bullshit what is bullshit? Fufie do you do any system level programming with something like sockets ? billh_: yes do you do socket programming in Scheme ? I have yet to try to anything with sockets in scheme (or its addons, to be precise) what do you do specifically in Scheme that's useful to you ? water explain what a representation is. a representation is just an element of a formal theory are you talking machine or human representation ? mathematical logic my main use of Scheme has been to teach concepts and write simple DSL Fufie What's a DSL ? i don't give a rat's ass about machines or people 04:40am a DSL is a language specifically made for a certain domain.. e.g matlab has a language made for numerical stuff, PHP is for web+db, etc So you just build task specific software systemsx ? with ordinary languages you must go through lex+yacc or eq. tools and use a lot of effort to make a decent DSL for what you try to solve.. this is done easily and elegantly in Scheme/Lisp Kind of like what Perl is somewhat used for ? all programs are task specific.. but not all languages Yeah, but I don't know why you're using stuff like LISP and Scheme. I find them to be poor practical languages. beta has a perfectly good grammar-handling system not all languages, because it is a mess to write DSL in most languages and with a DSL you have a language to solve the problem in c/c++ :)) Fufie It's a self building language system ? * AriB/#tunes hides ;-) billh_: lisp and scheme are programmable programming languages AriB: wrong channel to bring up crappy languages in kind of like the way Forth is used ? lol Forth also shares some of the characteristics, yes Ok, then I think those languages are crap and nothing beats a good Bison/Flex combination. basing e.g Tunes on Lisp you effectively make a (re-)programmable system from top to bottom but bison/flex take work to port, maintain, extend, etc or from bottom to top some might say they take a lot of work It sounds like you guys are using these systems as a way of avoiding compiler theory. heh no no.. we use it as a way to work on the interesting part of compilers Uh, I don't see how. we just like to de-centralize the work of the compiler * billh_/#tunes shakes his head. remember that Lisp sexp is basically your parse-tree.. no need for lex/yacc sexp is the LISP interpreter written in LISP right ? sigh... sexp? i heard sexp was the LISP interpreter written in LISP right ? abi forget sexp water: I forgot sexp augh Huh ? abi: s-exp? s-exp is probably a _syntax_ for recursive lists, that are a general purpose data structure bill, you gotta learn more about lisp it's easier to hate something you don't know you *did* get a crappy cs degree So this facilitates evaluation of functions, right ? not jsut that Fufie I know LISP, but it's been about 7 years. It's the core of the whole system, right ? -:- smkl [sami@glubimox.yok.utu.fi] has joined #tunes hey smkl it's the form the code is represented as.. it is the same as the nodes in classic parse trees.. just that they're written directly I remember a LISP function that essentially is the entire evaluation core. god damned lurker! hello water ah there you are Oh, so you're saying it's grammatic representation ? Like BNF except in LISP ? we're trying to explain to bill here why tunes favors lisp i was connecting to other irc servers ... * billh_/#tunes thinks this sound fishy. http://www.tunes.org/Review/Languages.html#LISP ok 04:50am it explains sexp hehe.. actually 'sexp' is kindof funny name at least in lisp :) (sexp 'intercourse) -> t a nice predicate :) any of you guys know Java ? it used to be called SEX yes yes that's nice obviously that had to change * water/#tunes has grown weary of this ari: I think maybe I should do a: (defun sexp (obj) (consp obj)) and then replace all calls to SEXP instead will make the code interesting to read heh blah blah blah Uh, Scheme is just another kind of LISP. sort of You know they added OOP concepts to Common LISP for a reason you know. block-structuring is different from lisp, imo who many of you guys know Compiler Theory and the tools associated with it ? i have two books on it and all sorts of tools Like I said how many of you folks actually know it ? great what do you mean know? what an asshole I have taken a few courses, written a few front-ends, worked on a few front-ends and made a javascript engine but I don't know it yet what do you expect from us? water Do you know compiler theory ? * Fufie/#tunes fwaps bill with the dragon book.. I'm wondering how well you guys know the theory. i don't have a fucking phd in it, if that's your requirement water do you know what a LALR parse is ? yes who fucking cares? 05:00am and you've use tools like Flex and Bison, right ? yes and you know about syntax directed actions, right ? yes so? billh...um what's your point what's a symbol table ? * Fufie/#tunes smirks.. this is getting ridiculous you want a fucking definition of a data-structure? there's no point in talking to you, is there? Arb I'm trying to find out how much formal training that these folks have because I'm confused as to why anybody with formal CS training would use those languages. wtf?!? which languages? lisp? Yeah. * Fufie/#tunes rofl you ever heard of INRIA ? Nope. there's a good place to start why the fuck don't you know shit abou tinria AriB There's a classic waring between folks that use traditional OOP languages and folks the use functional ones like LISP. * water/#tunes feels he can curse at a curser i'm not either asshole bill: a tiny hint, most lispers have long CS educations.. water after talking to you folks, I'm kind of convince that you folks have a pretty weak computer science background. then you're a closed-minded bigot Fufie Yeah, but it doesn't quite show. I don't think so. bill: another hint: the amount of education is _much_ larger than among people choosing c++ or java OK of course you don't ever heard of aspectJ? * billh_/#tunes shakes his head. it's an extension of java you've obviously never heard of aspectJ is basically a knights of the lambda calculus project called aspect-oriented programming abi aop? rumour has it aop is at http://www.parc.xerox.com/csl/projects/aop/ or aspect-oriented programming or just a special case of re-write logic water, my speciality is DOMs and hard core OOP. hard-core?!? lol you mean crappy DOM? as in XML DOM? abi maude? rumour has it maude is a reflective rewriting logic language at http://maude.csl.sri.com/ * Fufie/#tunes is rofl Yeah, I know about Aspect. you're the fucking ignorant one here you listen to everything the big corporations tell you and your professors listened to them too Na, I know alot of about compiler optimization theory. so fucking what? and I'm a pretty radical programmer. you can compile c and clones big fucking deal can you write a garbage collector in 40k? Uh, how about writting full compilers dude ? bill: I am a whiz on optimizing compilers.. by choosing lisp I optimised away the front-end uh, how about it's trivial in any decent language? Like JIT cores ? your job is hard because you picked a language like c to do it in not really. or java Yeah, I'm still leading to you folks not really understanding alot about CS. well, a machine could do most of your job I thought the compiler pros used ML or CL, but I might be wrong So you can write a full GC in LISP within 40k ? i don't use lisp the weenies that come out of most collges think the dragon book is good 05:10am but yes, gc's can be written in that small a space yeah, but how complete an analysis can it do ? hell, it's done even in c particularly in short-lived object situations. is that 40k for the compiled gc code or 40k for max memory used by the gc? Fufie ...what water said. compiled code Like in C ? yeah You guys understand type resolution issues ? going under 40k have been done before what would be easier to parse for formatted text, xml or tex-like (or even something else) ? it runs on my handheld computer yes and Self's concrete type inference technology ? yes definitely billh...you're doing it again :) it's a very nice system, but it's hard to port because of the way they did it Yeah, but water, you really didn't understand what a Quaternion was so I'm pretty suspect about your claims, you can understand why, right ? smoke: I find xml ok, but I have a decent architecture/system for working with xml fuck you Fufie You guys do any document raster stuff ? so i don't work with quaternions every three years water, it's definitely not a matrix. certainly not a 4x4. bill: document raster stuff? Yeah. let me get a reference for you Like RTF and shit. fufie; hm you mean displaying text? smoke: the code is freely availbale though (cl, c++, java and easy to add more languages) I mean building a Quark Xpress engine. no.. I have never ever built a Quark Xpress engine huh fufie; well, it's for a linux demo scene magazine.. it's very simple formatting, text in 1 font which can be bold and coloured, and there are some inline images. that's all damn it i confused quaternion sorry bill smoke: you can probaly use a troff-like scheme then.. Fufie ever use OpenStep ? but i know cs because i can talk to people about it, uinlike math * AriB/#tunes has no idea what quaternions and Quark Xpress have to do with compilers and 40k GC's arib: me neither ari: got a point there water Yeah, not to be insulting but can you understand why I'd suspicious about that ? fufie; ah let's investigate that some more :) thanks no, honestly no AriB it's a general hacker knowledge test. you could give someone the benefit of a doubt smoke: it's basically a line-based system.. like a man-page.. see the MAN-PAGE howto for linux instead of being an asshole Fufie You did open me up to the possibilty of why LISP would be good for DOMs. "jack of all trades master of none" water more content please and less flaming, ok ? bill: look who's talking? bill: what kind of DOMs? Distributed Object Models you mean? Fufie The full shit. Fufie XML is only one expression of that technology. "technology"? fuck with DOMs was my first professional gig. *fucking* with DOMs was my first professional gig. bill: as for XML I have a full system for that, and I am expanding it to be distributed XML do you know the origins of DOMs ? and why they exist in the first place ? who cares? when talking of XML, DOM means Document Object Model.. Distributed Object Models are different please be precise fufie; is writing a simple xml-like parser a trivial job? Ok, XML technologies. 05:20am ie. "billh_ : water more content please and less flaming, ok ?" smoke: a simple one is trivial, a full xml-parser is a mess well, honestly, i find this discussion completely off the point smoke: there are already XML-parsers for most languages But both the Document and Distributed formats are tightly related in OOP. both are OO-related.. but they are still fundamentally different Fufie Like I said you know the Quake Xpress originals of DOMs right ? and how they are used, eh ? the XML DOM comes from SGML and Quark has never been SGML-aware to my knowledge and Quark is not distributed either DOMs technology is more general than this. * Fufie/#tunes sighs Fufie you familiar with OpenDoc or DCOM ? i'm suddenly far less impressed with bill OpenDoc I know some about.. DCOM I don't know enough about.. but now you're into different things again water: only just now? hcf: heh OpenDoc and DCOM are the different expression of the same technology. It's based in DOMs. hcf: yeah, he's over-educated in "practical" cs, and not enough in the "abstract, useless shit" water: this is a different kind of practical cs than I am used to no kidding bill: then your DOM idea must be really really general Nope, DOMs technologies are RTF, XML, and the Framemaker internal format. The Framemaker internal format is MIF yeah, it's a RTF like technology. it is not but all are similar. drop the technologies buzzword it's just a format water: then he has nothing to talk about lol true This is demonstration of how modern your knowledge is. i guess he hasn't heard of hyperprogramming or orthogonal persistence it's also pretty important stuff. RTF is a dumb and simple format for describing sttructure in a document.. it has _nothing_ to do with e.g DCOM which you mentioned.. except DCOM has a specification in RTF :-) to the companies you play with, yes, bill Fufie RTF is a compound document format that's tight integrated with OLD. Fufie RTF is a compound document format that's tight integrated with OLE. I thought they retired OLE and made OLE2 and then they retired it and made ActiveX The name doesn't matter, it's pretty much the same stuff. I only use the old name out of habit. how quaint indeed Did you know that RTF was specifically design to glue to ActiveX ? but you know what.. this discussion will bear no fruits because we live in different worlds Fufie not really. sigh.. oh that's right our world doesn't exist ActiveX is from 97?.. and RTF is from the 80s.. yes ;) DOMs are very tight related to some of your ideas about LISP. LISP in alot of ways is a DOM. 05:30am lol RTF, is from the 70's. lisp though *is* from the 50's * AriB/#tunes sighs Yeah, but not it's current forms. ari: sorry, not much fun for you? I msg'ed bill and he didn't respond ari: it's usually not like this.. but sometimes we have people who go to the wrong colleges here to educate us.. it is kindof futile yeah, recoding the same old shit in c, and calling it "technology" or java/c++, sorry :P I've got to find some common ground. I think java was an important step which is why I mentioned DOMs. it allowed the mainstream to go to a slightly higher level and show that gc isn't too bad yeah but it's so irrelevant :) look, we're not going to like anything you say, because it's all coded in c and therefore closed DOMs are directly compatible with data structure representation in LISP. or at least a certain part of it. uh, retract that. It's completely mappable to LISP. no user could ever decide to change something that you compile away into binary Fufie Actually GC in it's current for is pretty slowl. lisp will have no problems implementing DOM, no. almost as irrelevant as quaternions and Quark Fufie HotSpot will change that hopefully. lol borrowed from self of course AriB It's relvant because in building any OS you have to have eventually have documents. sun had to buy out a company of decent programmers to get "technology" and you have to represent them some howl. first you need a compiler :P bill: that is true.. there will be a OM OM ? bill: represent them as language objects Object Model squeak has a book format, for instance built from objects So OM is your version of DOM then ? lol no NO ari: quaternions are important for complex numbers.. a very important part of any OS you can change OM on the fly lol well then don't let water work on that part :) yeah, you have to have complex numbers in file-structure blocks or the whole thing falls to pieces :) OSes are pretty complex -:- Ghyll [karltk@mp-217-240-237.daxnet.no] has joined #tunes yes they certainly can be "quternions" useful for anything requiring 3d rotation and representation. bill: there will be an OpenDoc like syystem, rest assured bill: i know i confused the quaternion with the quaternion multiplciation so sue me :P Fufie you need to give me examples of why a LISP style system would be good to use. bill: umm.. disn't you say that DOMs were great and that Sexp were basically DOM? I can already see a good DOM example, but you haven't quite pick up on it yet, right ? 05:40am Yeah, I'm trying to see where our concepts overlap. imagine a programming language it is used to write a program in it can be good or bad, high level or low level it is however pretty constant there is usually specs and it can't change often that will break compilers ok. so programming languages are basically constants yes, some what. this is not ideal The concepts don't change though. yes they do because I need a language to write servers in and I need a language to write games in and one to allow kids to experiment with don't underestimate kids and one to make a hypermedia system Ok, I understand. and one to express xml-structures * water/#tunes envisions 9-yr olds playing with c++ Fufie Please continue then. i think we should put a vt100 in kindergarten clases around the world it would be nice to have a language to make those programming languages in in the classical way this is done with lex/yacc or similar tools this is a tough job painful and people who take compiler courses feel like kings Yes, sometimes, but I'm use to it and I like it. like it's really important and unique abilities their penises grow two inches longer lol * smoke/#tunes would like to know some more bison :( Fufie: you claim they don't ? but you need a language to do this easily in, that language is lisp because it is basically a language to write parse trees in smoke: check out south dakota :) Fufie please continue. it is the representation that lex and bison generates Fufie Yeah, I know that. Fufie: any self-respecting knight knows that whacking the dragon will not only give him all the gold. it will also automatically bestow upon him the title of Foo the Great. And we all know what part of a good knight really deserves a 'the Great' titulation. continue please... so in order to build the languages you need to express an elegant solution to your problem.. be it a server, a distributed system etc.. you need a tailoured language for it and that is why lisp makes elegant solutions for you Uh, ok. I can see a section of what you're saying. it definitely makes code easier to read now imagine this power we have in lisp to make programming languages imagine that used in the OS think of yourself to not only be able t change the language but also to be able to change the and make a tailoured OS s/the// that is what lisp is all about imho.. thank you everyone.. and rest assured.. I have taken a compiler course ;) :) bravo Ok, Then how do you do syntax directed definitions and stuff within that language ? 05:50am do I really need to do that? yes, the stuff that I do is needs that particular feature. explain why you need it it's complicated. ghyll: you recently took that course.. you translate :) but with your claims you can build stuff like infix evalutors within LISP for any purpose, right ? I actually know how to do that. you can build infix evaluators yes do you need infix? Yeah, I know that. PostFix, prefix, you guys actively use stuff like Breath-first-search and shit right ? Fufie: 'sdd' or my ravings about dragons ? ghyll: your raving about dragons i think the emphasis there was 'do you *need* infix?' I was thinking of infix syntax of a language.. that was what I thought you meant yeah, I know and that's fine. how you traverse the tree afterwards is up to you I like random traversal so you use stuff like breath-first-search to walk graphs right ? Fufie: ah. just pointing out that making compilers for todays common languages is a real bitch which WILL make your genitals grow. the dragon part was of course a reference to the Dragon Book, but Aho, Sethi, Ullman :) you *could* ghyll: it is more likely that the genitals will shring when writing a c++ front-end shrink Fufie Does Scheme handle Unicode ? bill: if you want it to Fufie: when in battle, yes. but the enormous ego boost that happens afterwards, when your creating manages not to trip on the conformance tests... (remember that it is a programmable preogramming language) so you can make it handle Unicode easily ? sigh... water shut up. I'm trying to have a conversation. ;-) ghyll: yes.. then the genitals will grow considerably.. so far noone has been able to write a bug-free c++ front-end bill: yes i'm just completely disillusioned fufie lol so true :) ghyll: it is just a theory So how would you use Unicode within Scheme ? Fufie: there's also a theory that says it's provably impossible to make a bugfree c++ front-end. This is interesting. ghyll: around 93 there was rumour of someone actually bein close to writing a bug-free front-end.. but then Bjarne and friends added more to c++.. they didn't want to have the small penises.. :) lol Fufie: that theory will eventually be proved by induction over the failed attempts :) bill: just like ordinary text Fufie So how would you use Unicode within Scheme ? just like squeak handles unicode, btw how do they handle that ? Fufie: I suspect bjarne adds to c++ as a kind of therapy to get out all of his agression and negative thoughts.. bill: I would probably use the unicode support to read files, write files, and display things to a colour monitor 06:00am ghyll: he should be empty by now Fufie: one should think so.. Fufie you mean color graphics view, then eh ? Fufie: but I expect M$ to follow in his footsteps with Cool. wow. bill's a fast learner bill: I mean that I would use the unicode support like any other char-based format was supported * billh_/#tunes thinks that water is an idiot. likewise bill: do you have microsoft certification in anything? yeah, but I've some pretty serious software engineering experience. * Fufie/#tunes rests his case heh. and all because i confused a quaternion with its multiplication Like some pretty heavy system level stuff. Yeah, you're position on being completely sold on Scheme doesn't really convince me. "pretty heavy" stuff, huh? -:- Ghyll has changed the topic on channel #tunes to: lock cmpxchg8b Macintosh reverse engineering and graphics programming (2d, 3d). amougst other things,. ooh Like occasion file system stuff too. uh oh ahhh z-buffers, s-buffers and all that? * Fufie/#tunes hated Foley or do you just call library api's? the 2d thing is within the context of DOMs so it's quite bit hotter than what most folks deal within in MS land, which why I mentioned Quark.. waters: you forgot w-buffers, a-buffers, frame, accum and stencil buffers, and back buffers, too .. heh Fufie: then you should read watt & watt.. it's a lot terser. Surface subdivision and other patch stuff. ooh magic special stuff I'm also into JIT cores. ghyll: graphics is so.. yeah.. out like PPC -> x86 translation. ascii forever unicode you mean :) Fufie Uh, you can't be serious, right ? ok.. maybe I want a lambda-symbol billh_: does that mean you work on radiosity rendering as well ? ari: umm yes :D I don't do that but I know folks that work with that stuff. So I'm a very different kind of software engineer than you folks. I am not an engineer mostly in the pure OOP tradition. billh_: that part I you probably got right. s/pure/mainstream/ s/I// Yeah, the LISP arguments don't really convince me. a lot of sedding here.. :) and I use to use that language to do certain AI base things. microsoft certification works.. it protects one from heathen ideas.. lol Fufie: perhaps you should pass them by reference, next time ? But its different than what you guy propose. * AriB/#tunes hides yeah, without MS, y2k might have been the end of it all bill: d you know why lisp is the ai language? Yeah, symbol manipulation and association, etc... I certainly do. But I also hate traditional AI. bill: good boy It's pretty juvenile shit. billh_: you're into the adolescent stuff ? Yeah, but quoting John Stuart Mill is following along that same though. bill: yes.. that is sme of the powerful tools available.. but the real reason is that lisp is as I said a programmable programming language, which let ai-researchers tailour lisp to their ai-system 06:10am yeah, I know. but that is of course not convincing.. just idiots work with ai of course they haven't even managed to create HAL all the real work is done every day where bill works and we're soon approaching win2001 Fufie: they came very close: IBM lol Yeah, I'm pretty convinced that you guys are clueless about OS theory and why stuff like the Linux kernel still exists. (HAL)++ ghyll: i hope you're kidding yeah, I definitely do real work. we know why linux exists, thanks bill: what do you think about Tanenbaum? water: (HAL)++ == IBM Tan is dated. heh. oh yes Keep in mind folks run Scheme on top of LISP for a reason. windows also exists Keep in mind folks run Scheme on top of *Linux* for a reason. huh? Thats more correct. AriB: that's because the plants should receive light.. It works. so you have Linux -> LISP -> Scheme ? rscheme doesn't run on top of lisp I misspoke. i thought so and I also corrected myself if you scroll back on that vt100 terminal. ;-) did you know that people have made chips specifically designed for Scheme? billh_: we use vt220 these days.. Well, I remember symbolics. symbolics made chips for lisp but they went out of business many years ago. open genera :) I don't track LISP derivative languages. Cray also went out of business.. Yeah, but they got bought out since they have useful technology. * Fufie/#tunes yawns yawn all you want but the LISP derived truth hurts. this is (yet again) pointless I yawned because I was tired and actually yawned Fufie: I think rms mentioned that symbolics used PDP chips. I might be a bit off; he might've just meant that PDPs were good for running Lisp.. Fufie Ok, sorry. ;-) symbolics used PDP-derived chips I think oh I think I know someone with a PDP in his apartment PDP sucks. * billh_/#tunes shakes his head. PDP's were nice because you had space in a word for extra bits which makes lisp-implementation easier Fufie: so has alpha :) yes :) hopefully 64-bit systems will make lisp more viable for raw computing again Ok, I'm out of here. i guess you're not going to help me, then? :) Huh ? helping you ? find someone to review my papers with the papers ? please yes please yeah, I'm #vmac/efnet. So you can get me there. i know they need work 06:20am i don't want you But the entire Tunes community kind of butchers language in an unconstructive way. lol yeah, c doesn't deserve butchering So you're better off hanging with pure philosophy folks. whatever We suck (tm) go away you're just an asshole waiting to happen They're more equiped to deal with it. does the mouse bring up any menus? #wrong billh_: you mean the thoughts shouldn't be thought ? Ghyll not the way they constructed it. It wouldn't fly with my circle of philosophy folks. water: you told me the same thing yesterday :) well, billh_ is a bit better than the lamer who challenged tunes mailing list to write quines it's true quines ? This is new. * Ghyll/#tunes has obviously missed a spot here.. abi quine? i think quine is at http://www.nyx.net/~gthompso/quine.htm billh_: gbe, hofstadter Oh, god you read that crap ? s/gbe/geb/ geb is definitely crap seriously hofstadter ? man that stuff is serious crap. water: it's where quining is coined.. at least we agree there ghyll: i don't ccare it's not like they're important water: they're very not. which is their point. besides, quine's stuff is obsolescent well, sure it's fun to look at water: imho, that's the entire point. nothing more. ok i just find any reference to geb to be offensive Douglas ? water: won't mention it with you around anymore, then :) on the other hand, i recall saying a year ago that the tunes docs were embarraassing Douglas Hofstadter ? bill: YES water: they're not whizz-bang fancy. that shit is really fucking old trendy cray. thanks for the info, bill that shit is really fucking old trendy crap. you're welcome. i'll keep it in mind geez that paper really isn't traditional CS material. it belong in the philosophy department. well of course it isn't i have a separate tech doc I suggest reading stuff like Patricia Churchland and her writting style. Same with Chomsky. how about Danielle Steel? read them lol and a few other from that disipline. They have the clearest writting I've ever read. bill: have you read any Wilde? bill: like it or not, i've managed to find lots of formal research from many fields that support my ideas Wilde what's he / ? -:- malc [malc@masq.redline.ru] has joined #tunes but none of them integrate the ideas, because they're from vastly differing fields water: especially Wilde ? :) water, Martin Heidegger is the real test when I ask about philosophy stuff. 06:30am ghyll: huh? no lo Fufie bill: Wilde is essential on any reading list not nietzsche? hi malc nietzche is important for different reasons. well, i happen to like those different reasons i don't study philosophy for philosophy Heidegger's stuff is some what based in Nietszche. yes, but it's not the same It's completely different. Just read it. macl: malc: have you taken any compiler courses? But you should get it from a Heidegger scholar instead of the normal philosophy folks. lol They don't teach it correctly for the most part. Fufie: no, never never did any cs related courses, why? they don't teach anything correctly you can't read any teaching on philosophy in a straightforward way billh_ : Fufie Please continue then. smoke : i think we should put a vt100 in kindergarten clases around the world Fufie : it would be nice to have a language to make those programming languages in -:- SignOff Ghyll: #TUNES (Excess Flood) malc: that's ok.. you didn't miss much Heidegger will give you another set of high specialize language skills. -:- Ghyll [karltk@mp-217-240-237.daxnet.no] has joined #tunes remind me not to play with my mouse again.. accident with the mouse? ghyll: that would have been a lot nicer in norwegian if a girl had said it :) i'm sure Fufie: i wouldt mind. though im more interested to learn about lambda calculus and constraint logic programming Fufie: 'nicer' is hardly the correct term. now if i could just fit in the time between my job and my reading of cs research papers and coding malc: lc is nice enough and irc'ing of course ok out of here. bye asshole bye theory soaked idiot. :) -:- billh_ [billh@cx739861-a.dt1.sdca.home.com] has left #tunes [] MSCE! must be, and regular efnet'er.. oh well, he wouldn't have thought it an insult anyway he is one an msce, that is Fufie: maybe, its just pisses me off that i dont understand 80% of Mozart docs, lack of proper education shows big time I was just surprised at how full of shit he was. wow. he would be worse to review my paper than atg Mozart? Mozart is a concurrent constraint programming language at http://mozart.ps.uni-sb.de/ or an implementation of Oz He seemed like an East Eurpean with a temper.. atg? he's a californian, afaik alonzoTG =alan t. grimes tunes' local weed-smoker water: even worse :) and geb afficionado malc: have you got a working url? Fufie: sec wel, i am really in need of sleep go to bed water go to water bed hey all, slate is an idea for hybridizing self, beta, and lisp-style oo abi; no, mozart is a concurrent constraint programming language at http://www.mozart-oz.org/ it's a functional oo idea with (by default) immutable objects and lisp syntax Fufie: www.mozart-oz.org abi: mozart? somebody said mozart was a concurrent constraint programming language at http://mozart.ps.uni-sb.de/ or an implementation of Oz 06:40am abi: no, mozart is a concurrent constraint programming language at http://www.mozart-oz.org/ okay, smkl. abi: mozart is also an implementation of Oz okay, smkl. smkl: is abi a self-hacked bot ? anyway, i thought slate could be implemented fairly simply and have some direct-manipulation gui features written for it as a tunes prototype Ghyll: it's just an infobot .. i don't think hcf has made any additions to it if anyone has thoughts on it, let hcf know or email me water: url? slate? there isn't one yet smkl: ah, ok. malc: it looked really interesting.. (too bad I have negative time) i made one in college as a tunes prototype but i don't have the code any more recode it water code is good it'll take time, and i really need to focus on arrow Fufie: yeah indeed, thats why im pissed malc: do you know scheme? scheme is a dialect of Lisp that stresses conceptual elegance and simplicity or ((nearly) (perfect)) all it would take is to use the squeak smalltalk compiler system and translation-to-c system to generate ansi c sources malc: because in.. "lambda : the ultimate imperative" a few practical examples of lc is given I think smkl: iv made addition but they dont really show Fufie: never coded in it, i know CL to some extent so.. hmm s/ion/ions/ hey hcf were you listening the whole time? sorta * water/#tunes 's eyelids begin to weaken Fufie: ehm.. sortof sick of english. bleh.. i didn't get around to announcing slate and i'm too tired now malc: babelfish will make it all come clear in your favourite language :) perhaps in 12 hours good night all nite thank you fufie for keeping that asshole at bay Ghyll: sorry mate babelfish doesnt support russian ;) -:- water [water@tnt-10-116.tscnet.net] has left #tunes [] malc: no matter, you read german fluently, don't you :) eins zwei drei .. Ghyll: not really :) glory days of Deutsch are over das pferd geht hinter bei.. s/p/P/ hmm.. it's been a while :) the horse goes ... uhm ich habe eine Kirche unter dem Tisch malc: don't ask.. it was all I could contruct that rhymed :) Fufie: grosse Tisch ? :) kliene Tisch kleine * Ghyll/#tunes looks for for his sharfesess. Fufie: Do you really have Church under the table? gueß what I found :) yes.. Alonzo Church Fufie: he must be starting to smell by now.. 06:50am you grow used to it Fufie: as he grows moss and tentacles ? * Fufie/#tunes breathes ok i think i finally got through to bill hehe Fufie: AriB: what surgical instrument did you use ? persistence and my razor sharp sense of logic AriB: the latter doesn't seem like an asset in this case.. let me rephrase that. a razor sharp sense of _his_ logic AriB: I won't say 'takes one to know one'.. :) Yeah, what they are saying is actually interesting for certain application that I've got, but I don't fully know the style of the programming pardigm just yet. ok there you go :) i'll have to experiment with it to get an idea of what they mean. see? he's not that hopeless :) ghyll: shaddup :) AriB: like most such people he's probably nicer in person (aka /query) hey I brought him in here in the first place. I had to bear some responsibility :) it was a bit of a cultural clash for both sides 07:00am happens all the time I mean you have to admit this stuff is controversial the other day I invited a few moslem friends for pork chops and some beers.. uh oh (joking) controversial is good lol well I'm jewish so I'd have the same problem * Ghyll/#tunes had a moslem friend over yesterday.. controversial means that there is atint of entropy ari: not with the beer I hope for dinner. he tasted quite well.. no-not with beer jewish. bah I don't like to eat pigs anyway controversy leads to new ways of thinking about things-some will be bad but sometimes it can lead to big discoveries Fufie: neither do the pigs ari: except in kansas Fufie: isn't that a given ? :) * Fufie/#tunes yawns 07:10am -:- SignOff hcf: #TUNES (Leaving) afk (I checked in the howto.ht) hmm.. this water dude... hmm.. hes twisted yes in the worst possible way 07:20am oh. that's bad coming from you dinner over here. see you -:- SignOff Ghyll: #TUNES (foo) Fufie: imo squeak kinda sucks visual wise if water is twisted I can't even begin to imagine what you think of billh marasmatic... after all 31... squeak is sluggish because it's not built using the OS/library calls 07:30am hmm.. reading #tunes logs is sure entertaining so many people in one damn place who know what sexp, inria, squeak, self is... fascinating 07:40am * smoke/#tunes looks in another direction -:- ult [noone@user-38lc644.dialup.mindspring.com] has joined #Tunes -:- ult is now known as UltOS -:- SignOff malc: #TUNES (Read error to malc[masq.redline.ru]: Connection reset by peer) -:- SignOff UltOS: #TUNES (Leaving) -:- ult [noone@user-37kba92.dialup.mindspring.com] has joined #Tunes -:- SignOff smoke: #TUNES (Ping timeout for smoke[15dyn101.delft.casema.net]) -:- SignOff AriB: #TUNES (Read error to AriB[209-122-226-70.s324.tnt1.nyw.ny.dialup.rcn.com]: Connection reset by peer) -:- AlonzoTG [Alonzo@216-164-137-196.s450.tnt4.lnhva.md.dialup.rcn.com] has joined #tunes -:- Ghyll [karltk@mp-217-230-180.daxnet.no] has joined #tunes * Fufie/#tunes yawns om 09:40am malc : reading #tunes logs is sure entertaining i already had it that way, Fufie. augh sorry 09:50am -:- hcf [nef@me-portland-us704.javanet.com] has joined #tunes -:- binEng [Anders@j141.ryd.student.liu.se] has joined #tunes * Fufie/#tunes stretches.. * binEng/#tunes reboots -:- SignOff binEng: #TUNES ( <k!14>) -:- air [brand@p0wer.qzx.com] has joined #tunes -:- binEng [Anders@j141.ryd.student.liu.se] has joined #tunes binEng: you run w2k ? nah, win98 <-- defragging :P had to be something like that.. yes -:- eihrul [lee@usr5-ppp122.lvdi.net] has joined #tunes eihrul the next version of windows will have a speech system it will be cool Fufie: huh? "start solitaire" Fufie: it already has. Fufie: os/2 had that back in 95 http://www.cuttingedge.org, http://www.balaams-ass.com "sorry, you are supposed to work, I'll open excel for you" Any Muslims here who want to blow someone up? "start linux" 94 even.... =\ * ult/#Tunes sighs but the OS/2 kernel blowed chunks. "I am sorry Dave, I am afraid I can't let you doo that" AlonzoTG: hmm. you're probably right. it will be called win2001 heh Fufie: wonder when win9000 will get here.. pretty soon basic numerical stuff 3.1 -> 95 -> 98 -> 2000 -> ... -> 9000 find the curve and you have the time it is probably exponential. 10:40am when was 3.1 released ? 91? yeah, I have the slitely updated version, 3.11 which came out in 93 =\ I let my bitchy little witchster who oiled her breaks rent my car last night... I havn't seen her since. =\ she had lost the keys to her car. G hmm.. that exponential curve was too much for my 8bit calculator.. it could be handled by a 2-bit company it's probably the fact that they don't care one bit for competition that makes this curve so hard to work out on my non-windows powered calculator 10:50am ikke-ideell #wrong -:- hcf [nef@me-portland-us704.javanet.com] has left #tunes [] -:- hcf [nef@me-portland-us704.javanet.com] has joined #tunes -:- SignOff hcf: #TUNES (Leaving) -:- SignOff AlonzoTG: #TUNES (Have Nice Day :)) 11:00am hmm.. heavenly creatures on the telly tonight err.. the weather ladies ? no.. the peter jackson movie (bad taste, braindead, meet the feebles, frighteners and soon LOTR) eh9? 11:30am -:- hcf [nef@me-portland-us704.javanet.com] has joined #tunes -:- SignOff ult: #TUNES (Ping timeout for ult[user-37kba92.dialup.mindspring.com]) -:- Downix [down@d-gnaps-130.ici.net] has joined #tunes -:- SignOff Ghyll: #TUNES (foo) -:- albero [albero@ppp-20-135-252.libero.it] has joined #tunes lock cmpxchg8b -:- albero [albero@ppp-20-135-252.libero.it] has left #tunes [] -:- hcf has changed the topic on channel #tunes to: TUNES, Free Reflective Computing System uh, what was that? what was what? albero dunno is there anyway to make NT read fat32 drives? yes, get a FAT32 driver for it so there is a fat32 driver then? why would it not come with one? 12:10pm -:- AlonzoTG [Alonzo@209-122-212-218.s599.tnt7.lnhva.md.dialup.rcn.com] has joined #tunes -:- smoke [smoke@15dyn101.delft.casema.net] has joined #tunes -:- SignOff hcf: #TUNES (Leaving) -:- SignOff Downix: #TUNES (Ping timeout for Downix[d-gnaps-130.ici.net]) -:- water [water@tnt-9-143.tscnet.net] has joined #tunes -:- water [water@tnt-9-143.tscnet.net] has left #tunes [] -:- SignOff AlonzoTG: #TUNES (Have Nice Day :)) -:- SignOff smoke: #TUNES (Ping timeout for smoke[15dyn101.delft.casema.net]) Can HLL's like Lisp access code in external DLL's and such? -:- smoke [smoke@15dyn101.delft.casema.net] has joined #tunes 02:10pm usually yes usually through an alien interface to C ic I guess C is just the common format? X is the least common denominator and most languages support it s/X/C/ :) You got me wondering there :) very close those keys 02:20pm * smoke/#tunes looks at his dvorak layout :) does anyone here happen to know a decent report on dvorak vs qwerty perhaps? haven't seen any dvorak is evil can't say I have much of an opinion :) -:- beholder [beholder@ppp-132.m2-3.sub.ican.net] has joined #tunes Hey bin My damn ICQ is really acting up tonight :( it is tempting to ditch qwerty, so dvorak must be evil.. all evil is tempting :) hi beh I don't really think dvorak would make you happier as a person binEng: So how goes your night? 02:30pm beholder: pretty well. Not much studying, though (an exam soon) :) binEng: What class is the test for? i like dvorak very much. i have the feeling it stresses my fingers way less than qwerty did beholder: Maths (integrals and stuff) smoke: well, do you think you're so much better than the rest of us, then? ;) binEng: Ahh. I'm getting a new CPU tomorrow to finish of my new server (i'm quite happy) :) beholder: what kind? binEng: Celeron 400 beholder: did you read the mail I forwarded? binEng: Yep, I didn't send a reply back yet though, I'm not sure if you wanted me to reply :) beholder: only if you feel like it binEng: Did you want me to reply to you personally or to the prisim list? * beholder/#tunes swears loudly due to lack of ICQ connectivity... umm, if you do feel you have something to say about it, maybe the others want to hear too 02:40pm Ok, will do then. If I find something to comment about that is. It sounds like a natural extension of the work you have been doing (which I have hopefully been helpfull in commenting on all along) Or maybe a better term is "natural expression" of the work you have been doing yes, you do have been of help (thanks). It might be seen as some sort of rationale to my project.. I plan to continue it. Yes, it does sound like a real rationale. Correct modeling of information is a worth cause. worth=worthy :) yup I think in the end this is the same thing that drives tunes and arrows I'm not sure on either one of them.. but could be From what I understand Perhaps the final rationale is a subject for the main tunes mlist, then binEng: So is there going to be an updated doc for your system soon? binEng: Yes, possibly. 02:50pm To be realistic, not soon. Currently, I'm not "explicitly" working on it binEng: Understood. Concentrate on that school stuff just a few more days binEng: Just a few more days of school left? just a few days until the exam :) Then I have some free time(tm). Have you been doing anything yourself lately? binEng: Working, going to job interviews... learning more unix. ic binEng: In other words nothing of any real value ;) that depends on how you see it, ofc... but in a way, yes otoh that would be a very computer-centric view binEng: Well, I like FreeBSD, I'm enjoying learning it beholder: sure, it wasn't my intention to judge how good you're doing or anything binEng: Hehe, that's right, it's my GF's job to judge me on everything ;) :) we're not quite there yet, lol binEng: So how is school, still thinking of changing courses? 03:00pm Yes. That'll have to be by the autumn semester. binEng: Heard much on the quality/validity of the alternate program A little. From what I've heard it seems equivalent. Different focus ofc, but that's what I want binEng: Ofc anyone here looked at simpla? Simpla is a basic concept language for use in teaching children and adults with no programming experience or computer knowledge whatsoever. nope have you got an url? http://sourceforge.net/project/?group_id=1057 it was announced on freshmeat today, i haven't had a look yet, so it may be completely useless oh it's a perl script eh WOW it's only 63 lines ??! 03:10pm haha either i'm mistaken or the freshmeat staff failed a little on this one :) 63 lines of perl should be simple enough :-) Bye -:- SignOff binEng: #TUNES ( <k!14>) 03:20pm -:- AlonzoTG [Alonzo@209-122-248-218.s472.tnt8.lnhva.md.dialup.rcn.com] has joined #tunes -:- washort [washort@d146.narrowgate.net] has joined #tunes -:- SignOff beholder: #TUNES ([x]chat) -:- SignOff AlonzoTG: #TUNES (Have Nice Day :)) -:- billh_ [billh@cx739861-a.dt1.sdca.home.com] has joined #tunes Hello. ;-) you're back? :) Yeah, is there a good pointer to CL ? I'm going to do a little research on the subject to see how clued in the LISP community is overall. billh_: what particularly do you want? What you guys said was very different than my training in traditional AI and LISP. I wrote the one here: http://www.tunes.org/Review/Languages.html#CLISP www.lisp.org is the main site though -:- SignOff smoke: #TUNES (zzzzzz!) Shit, I've also got to do a cross compiler build too. Ok. the review is way too biased now :P 04:00pm smkl: there is 7 cons listed.. and 10 Pros :) I used to use CLOS. But I've fogotten the name of it until now. How do you do stream programming in CL ? that depends.. ANIS CL has several stream types which are useful.. though the stream system was never CLOSified so it's not too easy to reconfigure.. most implementations support Gray Streams though (named after Gray) which are CLOSified and fully customisable "support" as in "native implementaion", ie asm/low-level C Yeah, I'll have to check this out. I've got compiler topics that might merge well with CL. how fast is CL ? I understood there was some relatively new compiler technology that made it reasonably fast ? the question is hard to answer because a language has no speed, an implementation does but some compilers generate really good asm code 04:10pm Like pretty close to plain C ? CMUCL and ACL are probably the best ones (I'll answer the C issue, let me elaborate) Speed is an important issue and if I can get CL to run at Bison speeds, etc... then I might reconsider it for various projects I do. you can probably get bison speed if you implement it like bison mainly, because I ended up writting up small languages for various projects anyway using Perl and other Unix tools you rpobably won't do that though Ok, because bison is really daft dat ? daft ? everyone I have ever spoken to which have used lex/yacc or flex/bison has said "there must be a better way" bison is designed for C not for Java/C++/etc other ways of representing the info is more natural in c++, java, etc it's like that in CL too Fufie You know that C++ has runtime binding now, right ? Yeah, I'm aware of your criticism, but I was never taught LISP derived languages from a pure computer science theory sense. you mean rtti or single polymorphism? yeah, RTTI. And the threading issue is implementation depedent. lisp is much closer to computer science than most other languages at least theoretical cs well, lisp was originally just a formalism for predicate logic, IIRC. Well, the wording is awkward, but I know these ideas from OOP. Fufie: has anyone ever tried to write a C compiler in Common Lisp, AFAYK? :) mainly, because it's the most flexible human readable language for doing data structures. Is this a correct understanding ? eih: yes.. on lisp machines the c compiler was written in lisp Anybody have a URL for a LISP implementation of C++ ? 04:20pm eih: if you check out the mozilla repository you will also find that javascript is prototyped/written in common lisp first.. :-) Hmmm, Ok, that's interesting. but what about a C++ compiler ? probably not there aren't many c++ compilers and most are written in c/c++ as the implementors know those languages best you guys heard of GNOME ? yes.. I use gnome C++? isn't that the language with no formal grammar? :) Fufie You look at GtkObject ? Fufie: well, i was looking if there is some availabe Common Lisp code for a C compiler... so i don't have to port LCC :) eih: port to what? Fufie what do you think of GtkObject ? and the entire GtkObject system ? bill: what do you mean? the struct GtkObject? Fufie: cmucl :) gtk's object system is prototype-based and is a fine example of OO in c eih: a noble task :-) Fufie: because cmucl shouldn't be dependent upon some external C compiler to compile itself! Fufie So you like it ? bill: it's nice enough.. but I am not a gui-person.. Fufie: because that kinda makes constructing a self contained system hard :) have you seen the libgnome* code ? ? and how it's used ? the lack of type safety ? eih: but cmucl is pretty limited in ability to be ported to new platforms.. a c-compiler in ANSI CL would rock though 04:30pm eih: tunes if it ever comes around, could probably use the c-compiler then :-) bill: c has no type safety bill: and c++ has limited type safety That's correct, and GtkObject lack type-safety. bill: the Gtk Object systems has a type-tag -s And therefore ? all gtk objects are typed I recall Yeah, I think you should look at the code. not as orthogonally as java/lisp/smalltalk/etc.. but it has a type and particularly the casting conventions, etc... Fufie: in what ways is it limitd in its porting ability? :) eih: it compiles to asm -:- nepenthe [DIY@ppp138.wi.centurytel.net] has joined #tunes -:- SignOff nepenthe: #TUNES (destroy what destroys you) -:- NetSplit: fontana.openprojects.net split from varley.openprojects.net [04:35pm] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [fontana.openprojects.net] Fufie: that could be fixed, eventually probably not easily -:- Netjoined: fontana.openprojects.net varley.openprojects.net -:- zarq [zarq@10dyn92.delft.casema.net] has joined #tunes but it'd be a fun summer project :) Fufie Yeah, the GTK-- community has been complaining about GtkObject and badly it's used. and they've been trying to push a language to take the role of it. eih: indeed :-) bill: I don't know gtk-- Fufie i personally agree. It's a C++ binding for GtkObject and the Gtk class library. bill: but there is one lesson to be learnt from Gtk's Object system Fufie and programming in the GtkObject is like programming within Java's virtual machine layer. It's very low level. it tries to add a lot of needed thing with it's casts, type-tags and "inheritance" hierarchies stuff which isn't in the language it is implemented in so they're trying to make themselves a language to solve their problem yeah, and ? and it follows greendpun's tenth rule of programming 04:40pm greenspun Ok. and that is ? -:- SignOff washort: #TUNES (Ping timeout for washort[d146.narrowgate.net]) "any sufficiently complicated C or Fortran program contains an ad hoc informally-specified bug-ridden slow implementation of half of Common Lisp" which is what Gtk's Object system is Well, that's difficult to say. (not half.. but maybe one tenth) and very particular to the perspective of CL. and certainly not what the GtkObject folks had in mind when they first started to implement it. hehe.. that is typical I have done that many times before I found common lisp I built a lot of fun code in c++ to hack around stuff and do "impossible" stuff.. Ok. impossible stuff ? yeah.. like in Coplien's book changing classes and inheritance runtime adding functions to classes.. dynamic type systems.. etc which was things I needed to solve a particular problem Fufie you know XML right ? what do you think of it ? -:- NetSplit: fontana.openprojects.net split from tolkien.openprojects.net [04:47pm] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [fontana.openprojects.net] -:- Netjoined: fontana.openprojects.net tolkien.openprojects.net -:- zarq [zarq@10dyn92.delft.casema.net] has joined #tunes I don't agree with myself on XML some days I like it other days I hate it Ok, I understand. and some days, like today, I see both perspectives XML is misused at this time. just remember that to use XML you need to experiment with it Fufie How do you specify a type system in CL ? you have to try out crazy stuff Ok. * billh_/#tunes is listening. Gakuk hi fare bill: the type system is inherent in CL lo, Fufmann 04:50pm bill: all objects have a type bill: and everything is an object bill: even the types Ok, So you have stuff like bitfields, etc... with various properties attached to them, righ ? yes what do you think about Dylan in contrast to CL ? and you have regular functions which also are objects and have a type Yeah, I don't know Dylan well enough.. I hope it becomes a success though.. but not at the expense of lisp How do you roughly do a hash table in CL? (make-hash-table) :) and how extensible is that table ? Dylan looks like a modern LISP with ugly syntax and lack of good macros. Fufie I'm going to get food, BBL. But I'll be monitoring. ;-) * Fufie/#tunes yawns.. soon time for a nap Fufie can I respecify the hash function within a hash table ? This is very important. yes and will the speed of this hash table be close to a C version of it using stactically allocated table ? that depends on the C implementation and the CL implementation but expect it to be just as fast and if you're unhappy with the standard hash-tables.. make your own how good is the free/open source CLs ? make my own in C, right ? in C you must make your own do you know much about hashing? it's an interesting field hard to make things spiffy Yeah, I've got a specific problem to solve, but only from a limited perspective. It's really a number theory topic to get a good spread across the entire array. 05:00pm I know number theory folks that do various kind of nutty stuff. but you don't want to mess with it in 95% of the cases or 99% if you prefer it's nice to have standard hash tables Fufie what do you think about writting a Power PC JIT core -> x8s translator in CL ? x86... sounds mostly like a C job JIT has to be low low level I'll talk to you more about it latter. I've got to get food and sunlight. ;-) I need sleep goodnight afk You can metaprogram your JIT in CL just like the ARDI people metaprogrammed part of their m68k->PC translator in perl 05:10pm Fare: i.e. write CL that writes C? 05:20pm for instance. Or CL that writes assembler (if you're writing a PC code generator, you can reuse it for your metaprogram) 05:50pm then you just need the Lisp C compiler :) uh? Fare, you mean prototype it ? back. ;-) Fare: a C compiler written in Lisp, sir eihrul i think we're talking about something that's very different than that. billh_: i know that, i wasn't trying to contribute to your discussion :) Fare Yeah, was think about doing code generation with CL, but I don't really know it's facilities in order pull that off. 06:00pm Fare how did you know that ARDI did that ? I would haven't expected that folks would know that stuff on this channel. -:- phriday [HaPpYsCrIp@cr479526-a.crdva1.bc.wave.home.com] has joined #tunes eihrul: yes, but why compile C? I thought you only wanted to compile PPC to x86? billh: you think people here don't read? -:- _1BadDude [Alonzo@216-164-139-65.s319.tnt5.lnhva.md.dialup.rcn.com] has joined #tunes Fare: eh? no, that was someone else who said that Well, most folks here I would have expected not to be interested in that stuff. <_1BadDude> om So it definitely came as a suprise. billh: well, I hope it's a happy surprise yeah, it is. ;-) It's a rare topic that folks can talk about it. erg, so it quite a suprise to hear this on this channel. 1BadDude: mu ATG: your greeting word betrayed you <_1BadDude> I couldn't get my default nick for some strange reason. hum. What's a valid fqdn that resolves to 127.0.0.1 ? <_1BadDude> I am not trying to disguise myself. (no, localhost is NOT a valid fqdn) Actually the instruction were represented using LISP constructions. any of u have ultima 9? air: try #warez on efnet or something * Fare/#Tunes stopped ultima around 7 eihrul: i have the game I tried U8, but was definitely too slow on my 386dx33 wow... but win98 reboot while it was installing and now it wont install Fare: it was quite slow on a P100 air: using loseY2K-2 is begging trouble dylans is looking like a good language to learn for this task. i deleted all registry entries, any ideas what else might be making it not install? billh: maybe. Or OCAML. I can definitely recommend OCAML. air: bad karma? air: don't you use Linux to make backups of loseY2K-2 ? OCAML ? OCAML is a language from ML family with support for objects and modules at http://caml.inria.fr/ocaml/ * Fare/#Tunes pats abi 06:10pm Fare yeah, I need something that has stronger type information. -:- SignOff zarq: #TUNES (Ping timeout for zarq[10dyn92.delft.casema.net]) OCAML might be just what you need. which is why I was looking at Dylan's compile time support. they've apparently got a C translator. C translator? you mean, d2c? Yeah, I'm just doing preliminary investigation. but that's what I read. 06:20pm -:- zarq [zarq@9dyn107.delft.casema.net] has joined #tunes Ok, I'm out of here. Thanks. ;-) -:- billh_ [billh@cx739861-a.dt1.sdca.home.com] has left #tunes [] 06:40pm -:- Chief [lol@137.28.142.238] has joined #tunes -:- Chief [lol@137.28.142.238] has left #tunes [] -:- rares [user3551@wtrb-sh4-port136.snet.net] has joined #tunes -:- rares [user3551@wtrb-sh4-port136.snet.net] has left #tunes [] -:- ult [noone@user-37kbanu.dialup.mindspring.com] has joined #Tunes -:- ruiner [DIY@ppp084.wi.centurytel.net] has joined #tunes anyone know howto remove a timer in bitch? bitchx 08:00pm -:- SignOff eihrul: #TUNES (Ping timeout for eihrul[usr5-ppp122.lvdi.net]) -:- eihrul [lee@usr5-ppp97.lvdi.net] has joined #tunes gah 08:20pm -:- washort [washort@d120.narrowgate.net] has joined #tunes * _1BadDude/#tunes left clicks on washort -:- water [water@tnt-10-134.tscnet.net] has joined #tunes last day of vacation -:- Downix [down@d-gnaps-199.ici.net] has joined #tunes water: kill a general maybe an admiral eihrul: Trying to help him get a promotion? ult: no, just another holiday heh well, maybe i could convince my dept to switch from nt to linux since i'm now the assistant sys-admin the only reason i'm not the actual sys-admin is that it would make my boss look bad :) heh suuuuuuuuuuure seriously, i actually know more than they do scary, huh? :) isn't that normal? 08:50pm you mean in the military? lol thats right washort among other places. :) yeah, i guess so, but you'd think officers with engineering degress would know something I either knew more, or had more common sense than all my bosses except one i mean, i coded up a set of vb scripts for their database in a day ro two that they were stumped on for weeks water: provided a degree equates with knowledge eih: point * ult/#Tunes knew a guy with a 2.5 gpa who got into MIT even places like that.. if a kid has rich parents and a silver tounge he go anywhere. lol GPA is not necessarily the measure of intelligence either * _1BadDude/#tunes hurls a laser guided flaming woodchuck at water what's that for, dude? :) <_1BadDude> That's for having a brain that will work better than mine ever will untill long after nanotech... uh nanotech? I have some of that in my brain. i ain't that great It's pretty cool...jacked my IQ 112 points... <_1BadDude> no, ult, thats' biotech. water just has experience. i'm jusst obsessed :) _1BadDude: No. Didn't you know? I'm an AI. intj? :) <_1BadDude> om -:- SignOff ruiner: #TUNES (destroy what destroys you) cool. i started a cool tunes mlist discussion now, trick is... to keep them going that's one less karma point that i'm in the hole :) (i think we've had more mailing list activity in the past two days than in the past two months) i can keep it going, trust me there's no end to the good topics that can be discussed <_1BadDude> correction: 6 months. What's the discussion about? refactoring in various pprogramming langiuages oic that could go on for 6months... introduce some gun control ... very tunes-relevant, imo well, it hasn't gotten hostile yet although laurent martelli displayed a bit of ignorance 09:00pm hmm i'm going to wait a day to add another comment to the list on that topic i forget: who was here when i was discussing slate (my resurrected college tunes prototype)? i know eihrul was ok, so no one cares? about? the slate idea? i believe everyone is idle nope >>> eihrul [lee@usr5-ppp97.lvdi.net] requested PING 2930690907 from #tunes 09:10pm wow. the new version of my download session manager is excellent it'll actually store url's in a queue and has some nice features for automation hmm it's almost as good as the linux tools hmm Anyone know anything about TAO? not enough -:- Kaufmann [Kaufmann@dial799.infolink.com.br] has joined #tunes hmm Shalom! * Downix/#tunes is now going to study up on it Study up on what? -:- SignOff _1BadDude: #TUNES (Gonna read some Brave Gnu World! =)) Kaufmann: Tao oh Since Amiga is using it, I wanna see what the hell it has to offer really They turned down your offer? Amiga is using a traditional Eastern philosophical system? ult: I'm making hardware, not an OS lol lol Erm. abi amiga? somebody said amiga was one example, AS/400 is another one You mean one of the Tao OS's? or am I missing something? ult: Ysa abi: no, amiga is the great white hope okay, water. That claim to be -so- great but in reality suck? Much like Scheme? =) Tao OS Amiga is a unique computer system, found at http://www.amiga.com abi: Amiga? Amiga is, like, the great white hope lol There are plenty of Tao OS's though -- I don't think that makes Amiga unique... ult: I know, that is rediculous someone enlighten me please abi: amiga is also a unique computer system, found at http://www.amiga.com okay, water. amiga? amiga is the great white hope or a unique computer system, found at http://www.amiga.com 09:20pm much better yes please oh please, speak to me huh? WTFF is Tao OS? oh Kaufmann: an OS made by Tao abi tao? tao is, like, an expression of the Water or tao, the (so-called) ultimate os at http://www8.pair.com/mnajtiv/tao.html ah thankee! wrong Tao really? the Tao Development group's OS is called ELate abi: Elate? i haven't a clue, downix k hm airgh you're confusing me http://www.tao-group.com Cool abi: tao is also at http://www.tao-group.com okay, water. -:- SignOff eihrul: #TUNES (Read error to eihrul[usr5-ppp97.lvdi.net]: EOF from client) wow. nice page -:- eihrul [lee@usr5-ppp97.lvdi.net] has joined #tunes wb -:- SignOff ult: #TUNES (Ping timeout for ult[user-37kbanu.dialup.mindspring.com]) openprojects seemed to have barfed... hm. i had trouble connecting earlier urgh I feel dizzy now why? hehe water: I'm studying Tao... holy shit! huh? water: What Elate can do is quite extraordinary oh... * water/#tunes takes a look this site is weird i think it looks pretty nice GOD DAMN SLASHDOT MODERATOR ASSHOLES!!! take it easy dude what did we do this time? Look at this post: http://slashdot.org/comments.pl?sid=00/01/09/1021200&cid=10 09:30pm Two, count 'em, _two_ "Troll" moderations. Jeez. Kaufmann: anything with the words "beowulf cluster" in them is automatically scored down by most ppl. :) -:- Downix_ [down@d-gnaps-70.ici.net] has joined #tunes yes true ok, knocked the phone off the table lol -:- SignOff Downix: #TUNES (Ping timeout for Downix[d-gnaps-199.ici.net]) hm. now t check out tech specs on elate, not just their blurb darn, all they offer freely is a glossary -:- Downix [down@d-gnaps-63.ici.net] has joined #tunes I'll bet that none of these moderator dickheads have even bothered to read the post ok, that's it Kaufmann: go become a moderator, then * Downix/#tunes throws the phone out the window Kaufmann: um.... what's your problem? any time a strange computational device gets mentioned, someone has to say "Beowulf Cluster". it's gotten old. why does tao only offer a glossary for elate? -:- SignOff Downix_: #TUNES (Ping timeout for Downix_[d-gnaps-70.ici.net]) Not to mention the fact that they've already rejected four times my submission of "Microsoft releases disclosed-source program". only registered developers are good enough for them to talk to? Kaufmann: heh. what program? water: who to talk to? tao washort, but it was a valid question, not intended as flamebait or anything. Maybe it doesn't deserve getting moderated up, but why moderate it down? Kaufmann: knee-jerk response. eihrul, I am, once in a while. can't say i disagree, though... washort, the Zone Integration Server yeah, beowulfing is a buzzword now that and "PORT LINUX TO IT!" lol heh i often hear "port squeak to it" as well washort: you mean, "Does it run Linux?" lol eihrul: eh, both :) potentially prefixed by "But" water: well, there's a little more credibility to that one. :) downix: do they say what language elate works with? i suppose c++, but... water: i've never seen anyone say "port squeak to it" on slashdot though. :) water: let me see washort: well, no, of course not C, C++, Java but whenever someone mentions a platform on the squeak mlist, it's sure to be brought up k java makes sense for it compiled to native, of course water: Of course (but the native itself is interpreted) not native jvm silly :) k 09:40pm just saying that the os's gc would clash with languages that use pointer arithmetic * Downix/#tunes nods how so? dangling pointer possibilities after an object gets gc'ed * Downix/#tunes nods it'd be hard to prevent ok I'm worried, I don't know TAO or their Elate OS water: conservative garbage collectors exist... maybe they check pointer stuff at compile time eih: true the boehm-gc afaik works very well with C++ * eihrul/#tunes had a sick that: garbage collected assembly. hm... but that's within the program, isn't it? yes Well, regardless,Elate is not for my CPU family limited to within the process the elate gc seems to be in the os itself well, tao obvious provides some nice tools for elate whoa! what? it has a binary portable format hmm Yes that gets translated entirely to native Amino/Amiga was describing that to me and everything in the os and apps is written that way * Downix/#tunes nods hm... the translator code must be thorny Smart idea Likely is I want to see it myself Since it does not support my CPU family, it would have to be translated hm... 'translate-on-load' * Downix/#tunes is one for making sure he gets as many good OS'es as he can for his chipset a bit like jit * Downix/#tunes nods and endianness-invariant code means more pointer tricks hmm The guy who' s in charge of TAO is a genious, I've met him cool At age 20 he was a veteren of the Amiga community made some of the best games well, their intermediate code acheives a compression ratio of about 1.7 to 1 on average not too bad Actually.... hmm my hardware could get even better performance for it 09:50pm -:- lar1 [larman@dialup-209.244.105.32.SanJose1.Level3.net] has joined #tunes Hey hey lar1 -:- rares [rares@wtrb-sh5-port30.snet.net] has joined #tunes Dx you called Just on here talking, no time for #f-cpu hm.. this dynamic binding idea sounds like fine-grained persistence nobody in #f-cpu just us chickens How long would the avgrage x86 box take to dump 128MB of data from ram and dump it to a harddrive? average? lol water: Home users computer what mhz? cpu and bus lar1: On my Cyrix it would take 12 minutes water: 366Mhz cpu, 66Mhz bus Downix: Heh * lar1/#tunes is trying to figure out if he can cut some corners when it comes to persistance rough guess <30 seconds lar1: hehe lol water: What if there was a UDMA/33 drive? then 3 sec, no? heh not sure exactly, but quickly, yes water: Is there a better way to persist? yeah, fine-grained only move the data that changes hmm journal the changes, if possible that'll really cut down the i/o times Journal _all_ changes that are done in memory??? that would take lots of storage and cycles! not *all* nitwit, just the data that changes hehe I said all changes, not all the stuff that has to be kept to have the store stay consistent it's not that much unless you're doing digital media editing, in which case "suck it up!" :) 10:00pm Heh ever notice swap file activity? I do hmm 8MHZ + 760MMHZ /2 720 cycles / byte * 1 second / *384MHZ * 128MB = 2 minutes -:- SignOff washort: #TUNES (Ping timeout for washort[d120.narrowgate.net]) a persistent system should be slightly less active hm well, study how the beos handles journalling Ok i've never heard of media apps losing much data when you pull the plug on a be system beos supposedlyy lags behind 2.3.latest in latencies *SMACK* That was the biggest brain fart I have ever had how much? thanks you water! :) 2.3 seconds? uh sure 2.3.latest kern-version that's not too bad, imo "2.3.latest"? oh the linux kernel hmm you could have said that no windows 4.10.222 silly of course linux it's the only one anybody can actually see the guts of sure water: You mean you think of other kernels _besideds_ linux? ;) i thought you were talking about performance numbers two timer lol :) whatever linux is just tiding me over until something better comes out water: I as well Linux belongs in the buisness and server world, no where else, there are better choices for the desktop like elate seems fairly nice you forgert potentially Destop work and kernel work are two separate projects heh rares: I still like Linux, in fact I love Linux, I just don't think my grandma should be using it "kernel work" there's a big market to shoot for :) should a win user be using KDE Gnome KDE and GNOME are clunky yeah, kde works for win users for now but give them time * Kaufmann/#tunes is not paying attention to the channel at the time kauf: hm I plan to fix that Kaufmann: You are wise actually that's probably his away message which means it's worse he's lurking No, it's not my away message silly you :) 10:10pm water: Do you like politics? gah politics lar1: not really water: Ok politics are more like a symptom, than a thing to be liked I am addicted to politics that's why I regularly take vacataions away from /. Would anyone be interested in a paper on micrometa government? micrometa what is that? hm lar1: rmm, meta government? give an overview freedom of the cell to free itself :) eihrul: Yes, a form of government where power is truly distributed evenly lar1, sure! how? forcibly distributed? water: Not done with it yet, I am just getting a feel to see if people would accually read this thing lar1 one problem lar1: "truly" is very ironic in that statement :) yes eihrul: How so? the sooner people are talking about freedom instead of exercising it the sooner they are likely to use it use = lose yep lar1: because, truth is evil :) and, it'd be very hard to distribute power completely evenly unless everyone is dead i think the general public doesn't understand the concepts of being or process oonly sight sound taste touch and smell eihrul: Well, close to evenly then ;) no sense of structure motion or cause and effect the only way it could be done is by evolution either social or biological or technological, i suppose lar1: sure, but what about fun things like corruption, stupidity, ect? lar1: how do you account for those? :) water evolution is futher down the spiral revolution is breaking out of the spiral even in the case where everyone is dead eihrul: Protections are built in to keep corruption from bringing the whole system down you can think of situations where one corpse might hold more power than another rares: i'm saying that the force required must be greater than our will for power eihrul: When you are a corpse, does it matter?? lar1: from bringing it down... but that doesn't mean they won't effect anyone even if they don't bring the system down if you're undead like most of us here yeah eihrul: Well, assuming that the entire country doesn't corrupt, the corrupted entity can be isolated and dealt with "can be" sounds theoretical lar1: unfortunately, new hoomans pop up all the time :) water: Yes, it depends on the willingness of people there is only one reason government exist (as opposed to the reason politicians exist) lar1: so you have a readily available stream of corruption... unless you sterilize people eihrul: But current systems don't handle the problem any better lar1: how are you so sure yours does handle the problem any better? well any gov't designed that tries to insure peace by eradicating evil is arrogant and in denial 10:20pm Government is a Bad Thing, but there isn't really away around it yet. In time I belive technoligy will allow for elimination of such, but until then A Better Way should be sought. the only one that works is one that makes room for evil but keeps evil too busy fighting itself rares: yes, the only reason you have gov't in the first place is because of that :) lar1: gov may be A bad thing, but it's not THE bad thing boy we're chipper tonight lol lar1: i don't think its so much the modes of government but the people who practice them :) ***change of subject*** water: True, its the things like corruption, and greed that are the reason government must exist water: Aggreed pick a subject, any technical subject lar1: any government depends upon its practioners adhering to it yeah i've flamed over this chestnut way too long and way too m,any times myself *positive* subject eihrul: Of course water: Hows li... nevermind s/practioners/practitioners/g lar1: lol thanks for not bringing it up :) * eihrul/#tunes digs around for a new topique. okay I'm looking for some strategy slate * lar1/#tunes is impressed at how cheery water is tonight! what about it?... dunno help me pls waht'dya all think? water: about what? i'll write volumes for salate if you help me about the idea of hybridizing self and beta and lisp Whats salate? wait no i'll never write for Slate (MS) *slate* Whats slate, then? wrong slate where does lisp fit in? :) yesterday it was just a self-beta hybrid well, it uses abstract syntax trees, for one Heh eihrul! Tommarow it will include forth! ;) well, the ast's are from beta, but it helps to just think of it as lisp water: ah, representation of code? yes well woise man once say: positive, negative, plus carrier, otherwise no go. but code=object under the beta system i'd like to have functional programming based on object-orientation water: could you elaborate on the code=object part? (i'm not familiar with beta) for reference: abi beta? beta is, like, an advanced object-oriented language that integrates objects and functions into patterns, or at http://www.daimi.au.dk/~beta/ anyway, they treat objects (structures with slots) and functions in the same structure interesting... that means that every object has an initialization routine that runs as a thread if only i had the time tolearn all these beautiful things and it means that an obbjects' methods can inherit from other methods or, you can use part-objects to act like interfaces water: you mean such that the 'initialization' acts as a function? right very similar to an old idea i had which acts like a thread at the os-level of using constructors of objects to represent functions so it can be blocked, synch-d, etc yes, it's a similar idea where by default the constructor returned some captured state that represented an object * water/#tunes nods that could be overridden (which is where i think the idea diverges) to act more like a function :) well, beta allows objects to run like co-routines or thread teams so that's what you're borrowing from beta? :) 10:30pm so whe nn do we build the space ship to run it on it's a nice system, actually, and beta code is fairly legible eih: yes rares: lol water i';m serious mjolner 5.0 is free and runs on a lot of platforms water: did you get a chance to check out that URL I had posted? the Beer! one? this thing needs a large playground well, the 5.0 version has a large playground water: actually, that is almost how most type systems work in general for functional programming languages, such as haskell... just haskell doesn't make the link as strongly between the two i wouldn't release this thing to an insanme society lar1: which one? lar1: oh yes water: Cool, huh? i saw that a long time ago yes eih: hm yes i recall thar that mjolner? water: Some of those were completely unreadable! mjolnir? lar1: loooked at false? abi, mjolner? bugger all, i dunno, kaufmann water: although they tend to view types as functions... rather than functions as types abi, mjolnir? no idea, kaufmann abi beta? beta is an advanced object-oriented language that integrates objects and functions into patterns, or at http://www.daimi.au.dk/~beta/ water: Nope but i suppose that's why its functional programming :) lar1: it's the shortest, iirc eih: yes, i'd like object-centered programming water: Hmm, I'll check it out boy E=MC^2 for programmers with function-centered programming provided as a well-defined service guys rares: ? functions = objects eih: no read up on Critical Theories in English Literature i didn't mean it quite that way :) objects = functions * c^2 :) this stuff comes up 99% of the time i understand that objects are function, yes but at a different level objects act as representations function is an object in motion it's a direct-manipulation interface idea which brings me to another aspect of slate i'd like it to support a direct-manipulation interface, self-style yes, that means either a gui or a graphical representation in text-mode of course there'd be a parser well yeah but it would be abstract most societies don't value educating their citizens to be able to use this kind of technology not to mention politics anyway, this idea is what i was working on for a few weeks in college rares: unfortunately true but there's ways to bypass some parts of society i mean how ghard is it to include a programming course with a video game i had a light-wieght high-level gui running over a small parser/interpreter based on a stripped lisp implementation heh that's an idea 10:40pm use a game-scripting language as the subject matter dude you KNOW i'd be the first even if I had to hide the lessons in the guise of magic spells learned while mudding * Kaufmann/#tunes will beat it now. Peace! heh oh. bye kauf cya K -:- SignOff Kaufmann: #TUNES (kathyanne gets kick'd / as leaves fall on still waters / she is here no more.) haikus re spooky well, anyway, i'd like to come up with a small implementation and set up a home page for slate (or whatever i call it) btw, alternate name is mobius :) water: what is your plan of attack? hmm salt well, i could roll the parser over the squeak smalltalk source, and generate the ansi c from that and then use the squeak vm sources to port it to various platforms hmm salt sounds like an interesting name that would be quick'n dirty salt? i don't know why i don't either :) hmm i liked mobius because it's a cool simple idea that seems to relate to reflection and of course i'd give slate lots of easy to learn reflective features hmm Dx: ? yes? basically, it's a merlin spin-off of sorts hmm Dx do that again and the whole frin channel will hmm you :) lol yeah :) hmm 8) hmm hmm hmm hmm! hmhm AH HAH! ;) LOL lar! anyway water: Have you hit the cold medicine hard agian? no, i'm not on it right now ahh water: Hmm, so you are just randomly in a good mood? it's random to you water: Hmm, well, I like this mood i have an internal happiness index generator based on complex algorithms :) no one's crack'd it yet * Downix/#tunes is just in a random mood water: Heh... perhaps some chicks could take a whack at it? ;) lar1: i sure hope they would 10:50pm i haven't found any chicks with interesting algorithms, though :( * rares/#tunes runs aeropund singing lalal lalalala MBill Gates to live to 200! Lalaslalala lalala this is the IRS lalala lalal uhh lol * lar1/#tunes smacks rares with an anvil * water/#tunes fires a few rounds into rares okay I'll never crack again * lar1/#tunes wonders what fire has to do wi... nevermind ;) well anyway who would really support slate from the tunes group? it wouldn't be too hard I don't know enough about any of those languages to commit to anything, plus I am pretty buisy already. hmm i'd love to but i have to get some small projects under my belt before I can just jump into this alright I'm going to college where I just might vbe available Downix: There you go on thinking again! Don't you know too much of that is bad for you?? rares: Chico? i'm planning on finding a person or two from the beta group, the self group, squeak, and maybe a tuneser Wporcester Polytech rares: Ah of course jecel will probably take notice lar1: nah water: perhaps a small little paper explaining what you picture as slate as a whole would be nice? :) eih: probably some html lar1 it ain't anywhere near as bad for anyone in this room as iti is for me eih: a little ebnf grammar, etc yes, that'd help Well if we all hang around water, mabye some of that smartness will rub off, eh? ;) hmm ok i'll start with the self/beta ebnf's and lisp-ify them as much as possible well todelooo ok i need to think if I get hired to do Accessyou nmay never hear of me ever again ! I'm serious I want to be done with HOME -:- rares [rares@wtrb-sh5-port30.snet.net] has left #tunes [] darn it, i keep forgetting to get a jpeg picture of myself water: Hmm, well would you like me to remind you? 11:00pm no, i'll mark it on my handheld computer that seems to do the trick i just forgot to put it there :) water: I assume you aren't in possission of that jpeg currently? i don't have a digital cam or a scanner or a picture of myself not a good pic, anyway water: well crappy pics are fine ;) not if they're 1" square water: hmm, beta is... interesting hmm eih: yes it is it compiles to some quite fast code, too there's also the meta-beta project and an extension called gbeta (gpl'd) meta-beta :) but gbeta happens to be a student's project which is very slow hmm Dx: say something Downix: And not hmm either! water: Have you seen the movie "The wall" by pink floyd? Well, I'm looking over the MIPS 3D extentions lar1: of course water: did you like it? lar1: to a point, yes water: heh. it's rock opera, not a great film but it is fairly moving It carries a good message sure, i suppose multiple messages, accually. It makes you think 11:10pm water: some parts of beta seem noticeably, umm, single-purposed :) hm... like what? virtual patterns oh still looking it over but that's the first thing to jump out at me * Downix/#tunes likes the Wall yeah those will be a special case of a more general scheme in slate good Not enough rock operas i happen to like the set of videos for "OK Computer" radiohead could probably make an excellent rock opera water: As could Trent Reznor hm wasn't that called "se7en"? water: dunno I must go i know the music was his style at the time Sleep is required heh gnite munchkin Mnuchkin? Eh? you're a minor Old fart!.. snicker.... yeah, no martini for you :) water: thinking of others. Tori Amos, Kurt Cobain could have, The Who is making anotherone Water: yah, well hmmph! ;) later all who: again? -:- SignOff lar1: #TUNES (Martitni for meeee!) wtf do they have left to sing about? water: They made "Tommy" now they're making "Lighthouse" 11:20pm darn it, the sself site is still down hehe water: I guess the drugs finally wore off, huh? i'm checking the mlist for clues about this you derived that from me saying "darn"? hehe water: "inner" is a keyword in beta i take it? this tutorial just uses it but does not specify what it is... hmm... the intel port of self is almost done i think so, yes there's a beta quick-ref that lists all keywords and syntax looks as if it calls into a derrived pattern yep wow. the self fans are porting it to everything hmm... self was taken down on Dec27 but they don't know when it'll be back up i thought you said self was proprietary sun contraption? :) well, it was i mean, it was always open-source but the code was ridiculously platform specific to the sparc hmm until david ungar got off his ass and did the work :) hmm 104 people on the self mlist 11:30pm quite a few are squeakers and a few are tunesers * eihrul/#tunes tries to figure out if beta distinguishes the notions of object instantiation and calling a procedure. i don't think it does btw, they allow virtual patterns to do some relatively generic things, iirc but it's been a while because you can consider your object a temporary stack of local variables that is cloned to upon its exit sure to result in an object so i'm just trying to find the place in the tutorial where beta reaffirms this hmm or disappoints me :) heh the beta docs aren't very high-quality probably because they've changed language features sometimes regardless of what beta does, it would still be a valid abstraction :) right and actually you need to create a new stack definitely something slate should have on calling a given procedure anyway so the only real difference is the stack persists outside of the lifetime of the procedure hm beta does deal with that well, that's how my old little pet language worked :) i remember there are ways to have the stack persist or get flushed each time although i never finished its implementation its just a matter of where the stack is placed sure which wouldn't be hard for an ast-based lang if some other value is returned, its can be on some specific structure like a stack whereas if it returns the "stack" itself then its a kind of a GC problem, you need to make sure it exists while references to it exist (since its been returned :>) so you put it on the heap yeah 11:40pm and where you put it could merely be an optimization heuristic have you looked at the beta gc ? -:- SignOff Downix: #TUNES (Ping timeout for Downix[d-gnaps-63.ici.net]) nope it's based on reachability from a chosen object basically, you label something "root" isn't that how most GC's work? (save reference counting) well, i thought a few had some alternate layouts most treat things like global variables and stack variables as roots i guess i'm no expert and just recurse down them well, beta i think allows you to change the root there's a very good paper on garbage collection i have, if you want i have a few, but more are welcome what's it titled? Uniprocessor Garbage Collection Techniques send it sounds like a good overview it even ventures dangerously close to concurrent garbage collection techniques :) heh uh oh :) have you heard of the train algorithm, then? the treadmill? i haven't heard of that one i have some papers on it on my linux partition send? er oh, doh linux partition well, i have a system to mount it, but it's really unstable well, treadmill algorithm utilized a circular list of objects hm train basically is generational and kept pointers to the start of new objects to the start of unused objects ok to the start of old and new objects and then contiuuously pulled used objects into new and alternating roles, and so on and so forth is that anything like train? hm not afaict there's a paper on gc using re-write logic that i have, but i lost the url and haven't relocated it it seems re-write logic actually handles a lot of gc strategies cleanly -:- SignOff phriday: #TUNES (Ping timeout for phriday[cr479526-a.crdva1.bc.wave.home.com]) this seems a nice paper do you remember where you found the paper? 11:50pm i thought it was on maude's site, but i checked and it wasn't there or attached to the rewrite homepage * eihrul/#tunes wonders if you could allow objects-as-functions in a lisp type setting... which is problem what you're doing :) yeah, treadmill isn't train that's what i'd like to do if you could just somehow get a hold of the current "frame" within a lambda expression ? frame => local variable stack right and then return it out of the lambda expression you have a constructor well, i'd like lambda's to be robust objects then you just need some way to access the variables or slots within that frame that could be reflected upon like any other objects hm i'll have to work that out which? accessing frame variables for lambdas etc well, the frame can be considered "self" actually maybe a reflector i.e. a closure object and this is also fits very well in the model because methods are just lambdas within the constructor evaluator+lambda=closure object right (and because of the behavior of closures, they have access to self) and other variables within self as well btw, self=this in beta this reminds me of C++ too much :) eek "this", i mean oh so i'd prefer to use "self," if you don't mind :) np * water/#tunes starts up his mjolner system [msg(TUNES)] newlog 2000.0110 IRC log ended Mon Jan 10 00:00:01 2000