IRC log started Mon Jan 10 00:00:01 2000 [msg(TUNES)] permlog 2000.0110 bleh.. it doesn't give you much of a suggestion of where to start beta? hmmm... beta is an advanced object-oriented language that integrates objects and functions into patterns, or at http://www.daimi.au.dk/~beta/ anyway, ideally, i could make a slate version of the squak design er... squeak yes, beta * eihrul/#tunes would like to play around with giving lambda's the notion of self. :) hopefully, the slate version could be kept cleaner that also solves another problem almost indirectly that lambda's normally can't reenter into themselves what problem is that? oh unless they have an explicit binding in lisp hm so it also gives them a way to reenter themselves wow though, i wonder how this fits in with tail recursion optimizations :) heh because entering into the self would infer cloning which tail recursion seeks to avoid well, cloning can be lazy actually its totally in accord with tail recursion :) how's that? because, in the case the lambda does not return self you may safely place it on the stack and when you have tail recursion you're usually not returning the self but some other value you're accumulating k but, in any case... even if the last step returns the self it's been too long since i worked with these issues if the lambda tail recurses into itself its technically not returning itself :) my head's been stuck in model theory and set theory and logics for too long so tail recursion still works well, make sure to add that to my docs btw, if i made this into a wiki, would you help out? because if you have for instance: (lambda (x) (if (null x) self (self (car x)))) (self (car x)) doesn't return the self that gets invoked hm it returns some other later self its only when it returns the current self that the current self needs to be free that's a really odd expression so tail recursion still works nicely :) 12:10am yes i was just doing a little experiment a nice one though it should have been (cdr) :) heh boy i'm really out of the lisp mentality say, as long as we're talking about 'self', why not use 'head' and 'tail' instead of 'car' and 'cdr'? or maybe 0 and 1 * eihrul/#tunes wonders how'd you access x, given the object self. you'd need some way to hash the symbol 'x to the index in the object well, smalltalk stuff already addresses that and then you basically have smalltalk :) yep but, that's of no consequence what's more interesting is you can derive an object from a procedure :) indeed i'd like to establish a habit form the beginning in slate of reflecting a lot on the environment model so that you'd wind up with a whole family of object system environments instead of building up a big monolithic hierarchy where everyone gets comfortable you think that's reasonable? sure ok i don't quite like large hierarchies myself because the objects within the hierarchy usually interact in non-obvious ways and to understand a given part of the hierarchy you have to understand almost the rest of it which is not fun well, it's just that getting used to one code system can inhibit proper use of a reflective system (like lisp or forth) and i'd like all the code to be simple * eihrul/#tunes ponders how to derive prototyping. yet powerful, of course :) well, the self papers go into that well, i mean from my lambda-with-self model :) oh -:- SignOff air: #TUNES (BRiX [http://www.qzx.com/brix] :: sleep) you basically include implicit parent pointers in object representations so it's part of the environment more interesting would be how you associated with the parent and it'd be simple to have methods that modify cloned objects -:- hcf [nef@me-portland-us735.javanet.com] has joined #tunes hcf: discussing slate in detail hi hcf hi cool the trick is to have the method look-up work through objects from self to parent 12:20am that reminds me of one of my pondrances from a few months ago of having semantics like method lookup btw, self also handles multiple inheritance in oddd ways being defined within the language (since they can be expressed outside of it) sure well, squeak's interpreter is written in smalltalk you could probably just do it with lisp macros of sorts :) also, self implements variable access as a primitive method which doesn't work like assignment it's strictly a message that i remember well, it raises some interesting points self allows inheritance by parts, which allows you to get data from one parent and behaviour from another given (method object paramters)...., a macro can be produced to send the selector 'method' to some specific closure in a specific slot that knows how to do the hash lookup :) i'm not sure how that meshes with beta or lisp, though hm (or could be redefined to work in whatever manner you wish) well ok but this would also allow you to modify how a selector is sent to a given object :) well, it should be possible by reflection but not the default the default semantics should be a little 'safer' i.e. it should be clear to the user when they're poking around with message-send semantics * eihrul/#tunes feels tempted to make a small object system in CL. :) yuck well just to play with some of these ideas :) that's all alright but i'd like it to get made independent of cl hmm, oddly enough 12:30am i have to represent frames with objects to make this work :) yes, that makes sense, actually so, you work with lisp a lot? that's good, i've never been able to stomach it http://c2.com/cgi/wiki?CodeSmells http://linus.socs.uts.edu.au/~cotar/proto/programme.html water: i'm rather new to it, but i'm loving it quite much -:- Closing Link: TUNES[bespin.dhs.org] by asimov.openprojects.net (Ping timeout for TUNES[bespin.dhs.org]) -:- Connection closed from irc.us.openprojects.net: Success -:- Use /Server to connect to a server -:- Connecting to port 6667 of server irc.us.openprojects.net [refnum 0] -:- BitchX+Deb1an: For more information about BitchX type /about -:- Welcome to the Internet Relay Network TUNES (from clarke.openprojects.net) -:- Your host is clarke.openprojects.net, running version u2.10.05.18.(ipcheck4-5) (from clarke.openprojects.net) -:- This server was cobbled together Wed Apr 28 1999 at 13 24:47 EDT(from clarke.openprojects.net) -:- clarke.openprojects.net u2.10.05.18.(ipcheck4-5) dioswkfcg biklmnopstv -:- [local users on irc(9)] 1% -:- [global users on irc(283)] 46% -:- [invisible users on irc(337)] 54% -:- [ircops on irc(20)] 3% -:- [total users on irc(620)] -:- [unknown connections(0)] -:- [total servers on irc(32)] (avg. 19 users per server) -:- [total channels created(195)] (avg. 3 users per channel) !clarke.openprojects.net Highest connection count: 18 (17 clients) !clarke.openprojects.net Welcome to Open Projects! You are on 4 ca 1(2) ft 14(14) tr. -:- Mode change [+f] for user TUNES -:- Mode change [+iws] for user TUNES -:- JOIN activated by "TUNES #tunes tunes@bespin.dhs.org " -:- TUNES [tunes@bespin.dhs.org] has joined #tunes -:- [Users(#tunes:10)] [ TUNES ] [ hcf ] [ eihrul ] [ water ] [ zarq ] [ abi ] [ Fufie ] [ thomas ] [ Fare ] [ smkl ] -:- Channel #tunes was created at Sun Feb 28 08:48:06 1999 (eihrul/#tunes) mainly because of the macro system and closures -:- BitchX+Deb1an: Join to #tunes was synced in 6.968 secs!! -:- Mode change [-ws] for user TUNES well, i understands the facts of its benefits, but i can't bring myself to get use to it hcf: lol why not? not sure maybe i like having objects as primitives as tactile things that i manipulate though, i think i've unintentionally shown objects with constructors are more expressive than lambda expressions :) heh well, you could definitely help to explain slate to lispers thanks, hcf btw, eihrul, the second url has some good stuff on prototyping water: that was actually a link from cetus' self page hm k i do forget things over time especially with the amount of stuff i look at well, hcf, it seems i got some good conversation going on the tunes mlist again * eihrul/#tunes ponders how to represent a lambda with an object. :) heh 12:40am well you have slot one: a tuple of symbols yes, i have that... and slot two: an expression with those symbols maybe that's too vague no tis not, actually water: has there been any merging of self and lisp b4? hcf: well, i know that lisp objects handle a lot of oo concepts but no one has ever changed the lisp language into something truly oo i.e. no real oo language has been based *directly* on lisp ast's and functional semantics heh. just think... we're making computer history ;) what of lisp does newtonscript have? but seriously *someone* had to think of this before hcf: you have any live links on it's documentation? hcf: i have a compiler for it well, it's like dylan is 12:50am end-user syntax etc http://www.cc.gatech.edu/grads/s/Arno.Schoedl/projects/NewtonScript/index.html damn you're good! Functions are treated as objects as well, and can be assigned to variables or passed as function parameters. This plus the fact that all statements return a value gives NewtonScript a little bit of a functional aspect. The influence of LISP is obvious here. hm i hadn't noticed that aspect of it yes, i jsut read that statement i guess it's a good sign that newtonscript runs in low memory situations nicely :) yep of course, it would probably do well with type-inference etc in a larger environment for performance btw, squeak will finally be getting it's major speed upgrades in a month or so where'd eihrul go? and hopefully i can leverage this for creating a slate environment he was toying with cl i'm working on my (thing) macro :) (thing)? i think i inspired him too well :) (or is that possible?) yes cool it's too bad i didn't get the same response when i told the tunes mlist about the idea a year ago 01:00am how r (things) coming w/ that? getting close * water/#tunes studies this NS doc for lessons ftp://ftp.apple.com/developer/Newton_Development/ yeah i got that thx okay, i have variables wrapped so they query from self k hmm NS has lisp quoting 01:10am wow. this language would be really nice for teaching programming better than basic or pascal anyway how's it going, eihrul? cmucl is not liking the code :) hm why not? 01:20am ACL works though, thankfully :) well, i'm going to code up some preliminary html for this stuff and properly update my homepage for PR purposes that's good then wow. my project lists still have generic names! :) 01:30am hm. is there a way to put a diamond shape into html? a hollow diamond, that is? -:- smoke [smoke@15dyn101.delft.casema.net] has joined #tunes hi hey smoke i inspired eihrul to write up a small object system in cl i think he's still working on it * eihrul/#tunes nods. getting close :) is it common to do that? - isn't clos useful enough? we were considering making an oo base for a self/beta/lisp union er.. hybrid union => feature accumulation cool! the squeak type-inference system just got released okay it compiled :) all right! * water/#tunes high-fives eihrul now to see if it works like it should :) indeed how much of it did you implement? 01:40am http://www.mactech.com/articles/frameworks/7_6/Prototype-based_OOLs_Evins.html from nov '93 ;) water: it allows you to defines things that have a notion of "self" that can outlast their invocation yeah, self started in 87 i think, and beta in 78 eihrul: hmm ok hcf: cool 01:50am hmmm wazzup? 02:00am bugs :) -:- AriB [arielb@209-122-226-70.s324.tnt1.nyw.ny.dialup.rcn.com] has joined #tunes uh oh :) hello, ari hi we're working on a programming language idea what kind of bugs, eih? strange ones Error: Attempt to access the name field of 460800 which is not a symbol. oh well, you're treating a symbol as a lisp symbol you have to either make the symbol a number or something, or properly add a quote somewhere darn it, i really am not the lisp hacker that i should be 02:10am sigh... damn win32 brb rebooting -:- water [water@tnt-10-134.tscnet.net] has left #tunes [] 02:20am -:- water [water@tnt-10-11.tscnet.net] has joined #tunes it, almost... works it works -:- billh_ [billh@cx739861-a.dt1.sdca.home.com] has joined #tunes Hello. hi it works, eih? everything except for assignment :) hm well, it's a start dcc me the code i'll update my homepage, and add the link to the code in a few minutes wait, its still broken :) sigh... ok well, i'll make up some preliminary ebnf * billh_/#tunes is wondering if the claims that was made last night CL is true.;-) uh oh ok my new links work fine hm... but i forgot to add a few billh_: no, i just suck :) Yeah, I've heard alot of about it from various folks, including ones that I've known for a while on #freebsd. But I don't really know CL in the way that you guys do 02:30am well, we're just testing out a design what is the latex command for the letter Z (all the integers)? :) I was talking to someone about OS's and out of the blue mentions tunes and how it makes more sense to have no kernel. he also mentioned the exokernel design (finishing this report and no latex manual handy :(() ari: who was that? Yeah, I know about that technology. smoke: don't recall The Exokernel stuff has alot of implementation problems. don't know his real name but his nick was ree there's also slk abi slk? slk is the safe language no-kernel research project at http://www.cs.cornell.edu/slk/ Those are actually conflicting ideas OS theory. yes, there are a lot of problems with various research ideas A friend of mine named Yoav turned me onto that stuff, but he's dead now. and I don't really track that community. which happens we're trying to iron out the bugs from our current discussion the aegis exokernel that's what it's called Yeah, it's an MIT product. \def\Z{{\kern0.3em\sf Z\kern-0.8em Z\kern0.2em}} how could i forget :) heh. hardly a product Tripping over a chair and smashing into the door: I highly recommend it! smoke: ouch. where'd you find the doc on it? 02:40am water; tried google with some weird keywords ah it's a very ugly Z though :( hm i don't think i have any latex source with Z in it it's all compiled ps and dvi, iirc eihrul: still there? hm. i'd like to see the code whether it works or not bill: you're relatively quiet tonight :) water, you remember us talking about MIT and how bummed you were ? yes, but i got over it yeah, I actually know a bunch of high powered nerds from that scene. including Cal Tech. i have to watch myself, because i have sucha big chip on my shoulder about what ? mostly not being where i can get peer review and job satisfaction Yeah, you're a loud mouth, but I'm use to since I hang with New Yorkers and stuff. lol I'm kind classifyable as a "performance artist". i'm pretty quiet in person i'm from NY but no way can you call me a loud mouth you freaking loser :) So my actions pretty wild and often twisted driven by a strong sense of morality from my philosophy/political science training. I've largely forgotten that part of myself since I've been a hard core production nerd these last few years. i don't really care for morality in general Yeah, read Nietzsche, he's got an interesting twist on the entire topics. i ahve which reminds me i'll look for some only heidegger Nietzsche is often best read with secondary literature, like Prophets of Extremity. (for your interesest, if any, the best way to do it was {\mathbin}{AMSb}{"5A}) s/only/online 02:50am smoke: thanks The standard text for Heidegger is "Being and Time". i've only heard overviews of heidegger But it's pretty much the most inaccessible book in philosophy and it's very rarely taught correctly. how do i tell abi complex things like ``the way you write blackbord bold characters in latex is explained on http://www.mcs.kent.edu/mcsinfo/compsys/faq/latex/bbold.html briefly'' ? sorry, i don't get to read everything in the known universe water: okie * Fufie/#tunes yawns.. nice nap thanks, eih eih: what bugs are left? if you can pretty much grok that, everything in French Postmodernism derives from it. and is boring. water: Error: Bad symbol-expansion pair in SYMBOL-MACROLET: well, i don't see postmodernism being a problem to grok Baudrillard is something Nietzsched derived. eih: i think you're screwing up the symbol notion abi: latex is also see http://www.mcs.kent.edu/mcsinfo/compsys/faq/latex/bbold.html okay, hcf. water I'm talking about the acedemic expression of those philosophical investigations. i understand that But pretty much nothing beats Heidegger. iyo and much of it is derived from Heideggers work. well, logical positivism has its merits, but no one likes it neither do i, of course most folks don't even know what logical positivism is. That's impressive. well, i study logic pretty thoroughly water: almost works now, just isn't initializing things properly abi: Heidegger is see http://dmoz.org/Society/Philosophy/Philosophers/Heidegger,_Martin/ hcf: is that necessary? yeah, if you can understand Heidegger from his pure view point then it'll be and impressive feat. water: no, but oh well water I've already done the logic thing already. It's fundamentally misguided. too bad for you :) there's some interesting new work 80's and 90's... that recent water, I think you should read Heidegger like I said. i will but i doubt i will ever endorse any philosophical system outright i take my relativism seriously :P It's pretty muct the death of that kind of philosophy, since it criticizes it from a fundamental conceptual view of point of what language is. relativism is bullshit too. bullshit is valid too bullshit is bullshit. and i don't mean std relativism That's the only logical identity is deserves. sorry, i misuse the term Yeah, you really need to get a broader view of continental philosophy. 03:00am yeah, so i can disagree with every thing i agree with :) I use to have these conversation with friends of mine but I grew out it when I got into other things. trust me, i'm quite the non-positivist thinker Yeah, I know the type, and I can tell were you pretty much fit, but you lack some important philosophy of language background to complete it. so I can't really complete thoughts and have it mean anything. i've studied language issues what do you think i'm missing? Yeah, same here, but Heideggers a completely different animal. It'll be clear when you work past chapter 3. water: what's wrong with positivism as an ideal? well, let's delay the continuation of this until i do Ok, that'll be fine. fufie: let's not go there, please You can probably hit me up for reference and explanation. water: ack.. so many people who discard positivism for the wrong reasons I still remember it pretty well, and reading that book is what made me finally dump school. fufie: i don't discard it, i just don't accept it Fufie yeah, there's too much factioning, it's not as clear a problem as some would like folks to think. but then i don't really accept anything not even as an ideal? ideals are terrible ideals are important bleh. no online doc of heidegger Fufie try to rethink the problem in a more abstract space. othwerwise all science slumps into the dullness and gray world of non-positivist post-modern (wanker) research Think of the role of science and what it does in relation to nature. Fufie Hold it. and just read what I wrote. I did and think of the techniques use to "enframe" (Heidegger term) the problem. bill: differing educations here will make this conversation difficult water Yeah, some what, but this topic is accessible if I do it correctly. why the hell hasn't someone scanned heidegger's works? bill: that is not important.. I am talking about science as a social project and my experience with the social sciences who have benn diluted by non-positivist thinking and a mistaken post-modernism thinking was depressing Think of the claims of authority in philosophy, Empirical, Positivistic and the other. The "other" here is something I'm trying to relay. i really need to get a scanner and solve this dearth of electronic philosophy