IRC log started Sun Apr 2 00:00:03 2000 [msg(TUNES)] permlog 2000.0402 -:- water [water@tnt-9-207.tscnet.net] has joined #tunes Water!! heh hi lar Whats up? slogging through the finishing details on slate nifty oh good, nothing like nate37 spewing garbage in the tunes logs Heh with the generous help of many others, of course Accually, if you would be willing to adopt a taging system, I'm coding a program to strip logs for relivent topics revdiable: ooops.. thought hello water wow how the hell did come out? that would require some irc client mods, though I just typeed in 'hello water' 01:30am hower ya doing water? water: No, all you have to do is type =Slatesomthing.open and =SlateSomthing.end lar1: i'd rather just use #modtunes water: Ok. What ever you perfer. hm the spiritual machines 'symposium's mere existence answers its own question Hmm, thats an intresting look on that I suppose its true it's not Oh? Oh :) Are you the only one working on slate? for now, yes Hmm. Is it to the stage where implemtation can occur? although eihrul claims he will help once his work with dolphin is done almost all of it can be implemented now, but there are important details which i am still working on mostly meta-behavior stuff Must one grok the wole language to impelment parts? 01:40am i believe so Ok there are important but subtle differences between slate and any other programming language btw, i am re-factoring the slate web docs this weekend in order to prepare for finalizing them Nifty yeah whatever To grok slate one must grok many other languages, right? hm grokking slate infers grokking many other languages, so yes but you don't have to learn them first theoretically Hows the learning curve for slate? how should i know? You made the thing ;) i've explained as much as i can on the mlist and to eihrul yes i did make it, but that doesn't mean i can tell you how hard it will be for you to learn it especially as all the docs have been grown incrementally so far which is why i'm making an explicit effort to rewrite them thoroughly When do you plan to have that done? well i'll post the re-factored site by tomorrow night but after that i still must add a lot of things Ok -:- nate37 [nate37@ppp-207-151-70-36.dialup.pcmagic.net] has left #tunes [] mostly the additions will be for didactic purposes, though so the learning curve, whatever its current level, will become shallower Good Tunes intends to have a LLL? not *a* LLL *many* LLL's read the site docs 01:50am the LLL is just the currently-used mapping from HLL to hardware -:- nate37 [nate@ppp-206-117-3-152.dialup.pcmagic.net] has joined #tunes Is acd&i a must for somone intrested in slate? heh definitely not -:- hcf [nef@me-portland-us100.javanet.com] has joined #tunes that's a book for learning how to make an optimizing compiler, particularly for a statically-compiled and linked language hi hcf hi water hey hcf Hmm You don't have any documentation on Slate's specification on the page? water: how goes the battle against the plague called html? of course i do, lar hcf: it's not the html that's the problem, it's the content :) water: Hmm, where? under 'details' look for 'semantics' and 'syntax' its not a formal spec per se, but it's the best info you'll get you could also look at the 'references' page to find papers on an object model that slate uses for its bmo's Ok 03:00am What language is Slate going to be implemted in? Forth? whatever anyone decides to implement it in eihrul prefers either lisp or c++ tcn could probably code a simple environment up in forth Ok Will the revised docs contain more of a 'spec'? i don't care, as long as it supports dynamic compilation and helps to lead to self-hosting yes they'll resemble the r5rs a little bit more or hm * water/#tunes tries to think of a better example of a spec/tutorial for a language well, somewhere between the r5rs and a less structured spec would seem appropriate pcp would be a good example of the 'less structured' influence that i'd like Hmm Is slate so unconventional to the point that you no longer have commands such as "if, then, and switch"? lol not even smalltalk has those let alone self or slate What is used in its place? or forth, for that matter message-passing (or function-application if you prefer) in smalltalk you write 'myBoolean ifTrue: thenBlock ifFalse: elseBlock' for a simple conditional Ah! Example code! This I can grok! that's not slate code, though Right Perhaps you could include some sample slate code in the revised documentation? you should probably try using smalltalk a bit to get the general feel of slate ideas of course i will Ok, cool, cool. 03:10am * lar1/#tunes searches for smalltalk resources i haven't had a clear enough idea of how the language should be optimally-used before to write lots of good examples Hmm. After the revised documentation are you going to start implementation? i suppose, but i'm not looking that far ahead yet if i did anything, it would be to write it in smalltalk squeak == smalltalk? squeak is a smalltalk-80 environment, yes although its gradually becoming something else What envrioment would you suguest I learn? s/learn/use? to use? squeak of course Ok it has programmer's features you can't get anywhere else Squeak is a VM, right? although it's experimental still, and easy to break (and fix, thankfully) yes its vm-based but the vm source code comes from smalltalk code in the squeak distro so squeak is reflective, to a point Hey wow... this sutff is nifty... there are some very good morphic tutorials now if you want good docs, get on the squeak swiki and do keyword searches Yeah, these tutorials look really good. I am downloading it now Thanks goodness for DSL ;) :) 03:20am heh * water/#tunes crosses his fingers for cable modems within the next 2 months * nate37/#tunes crosses his fingers that the crappy NIC can _handle_ a cable modem -:- XeF4 [xef4@105.ppp1-20.worldonline.dk] has joined #tunes how is XeF4 made, anyway? water: with Xe and F under very high pressure they obviously excite the Xenon using particular frequencies to set up its electron orbitals in the right config just high pressure? that's it? possibly some excitement, but afaik, no though.. I started using this nick before forgetting so much about chem hm i suppose that it *does* affect the ion balance equation water: Must programs written in squeak be run in a smalltalk vm, or can they be natively compiled? hah of course they can be natively compiled, if you have a smalltalk-to-native compiler actually, there's a jit vm in the works right now Cool can't anything be conveted to native machine code? yeah they'd have made it jit a while ago, but it has to be cross-platform 03:30am well, squeak already can compile to ansi c if u tied in the vm, and had it automaticllt run the program... that would count at native machine code squeak is not the end-all-be-all of smalltalk's either i thinks nate: that's already been done there's even a project to modularize the vm to make that method require less overhead water: I'm saying with any vm... not just sqeak.. doesn't matter the language... as long as the vm can run nativly but that's a doh anyways whatever it doesn't count as native code It would be sweet if squeak was implemented in hardware it depends ur definition of native code. lar: that's been done as well nate: what are you smoking? if there are smalltalk bytecodes in your program, it's not native water: At reasonable costs? lar: wtf are *reasonable costs* in open source? but it will still bumble down to it anyway no it won't without jit get a clue a bytecode vm is an interpreter i'm saying if u intertwinded the vm with the data nevermind i know what you mean, it's been done many times, and everyone knows it's not native code I guess the question is not 'can it be compiled nativly' but 'does a compiler exist' trust me, i see lots of discussions on this on the squeak mlist nate37: that's like saying "every computer simulation is physical reality, since there are physical electrons moving around" hmmm I must be tired... I strayed from my point. have you ever had a point? the point was that of couse it can be compiled... just u might halfta right the compiler your self... it just came out wrong s/right/write/ wow thanks for the insight water: I know... duh.. as I said I'm tired. I wasn't thinking about my point and yes I know that was stupid as hell... again because I strayed. 03:40am water: so what are the goals of slate? fulfill hll specs and provide an ideal basis for implementing arrow what is ur market? lol (if any) let me put it this way.. better question: what isn't my market what type of programmer would use it? and everything i do is free of course programmers wouldn't use it very much ;) because they wouldn't have to ;) ummm.. then I don't understand what it is. I really must go -- I've been on too long as it is. cya lar1 water: common users? nate: do you grok why tunes ideas are what they are? water: Thanks for the input. I am most defentily learning this Squeak stuff! Later. -:- SignOff lar1: #TUNES (Sleep) no because I don't know what tunes ideas there are if i did.. I would ask err.. grok ok so the question is where to start with tunes or with slate? what's the most high-level programming language you've ever used? both hmmmm slate is a tunes hll candidate what would count as high level? java scipt? that's pretty remote from the hardware heh but java script has a specific kind of api set i was thinking of a language with more general support BASIC? :) have you played with lisp at all? that's easy ;) ummm nope i'll pretend you didn't say that 03:50am heh ok not that I remember BASIC commands at all so i assume basic, pascal, c, c++, java, jscript? horribable. yes nothing else? (ok asm) command.com batch? ;) :) perl? :P it's disgusting the way new programmers in the '90s learn crap languages MFC? I know it's a API... that's not a language but it is creating a different type of programmers (if i should even call them that) ok go look on the tunes language review page take a look at some of the entries there, particularly the ones that aren't legacy languages ok I assume it is lynx friendly. try out a lisp or forth or squeak it should be why do you use lynx? netscape is slow with HSP baah hsp? can't wait till I get my NIC to work forget it, i don't care water: do you recommend a forth tutorial? * nate37/#tunes nods his head, "winmodem in linux" zef4: a particular one? yes I'm kinda learning forth anything from http://www.forth.org/ is pretty good water: tunes-rev is textual, why's it matter what browser he uses? hcf: i've seen lynx screw up on simple pages water: and regular browsers dont? regular browsers at least are obvious when they make mistakes, as opposed to text terminals -:- SignOff XeF4: #TUNES (Leaving) anythings else? 04:00am hm yes, Self k there are more esoteric languages that are also good, but you should learn the ones i already mentioned first i'm learning forth... and then i'll look into lisp, smalltalk (squeak), and self hmmm ok -:- SignOff hcf: #TUNES (Ping timeout for hcf[me-portland-us100.javanet.com]) e.g. haskell and maude for self? huh? haskell and maude are implementations of self i presume hell no oh sorry self is still one implementation I thought u were refering to what I just said no, sorry to confuse my fault u were referinning to esoteric languages, right? yes ok haskell is a purely functional language maude works on the idea of solving equations using rewriting rules both are reflective and powerful cool but they can't be learned without some base knowledge about those former languages, afaik they are also research languages, so they don't have commercially-appealing environments per se i should also learn the former languages to learn slate then hm so am i so "pitafull"? * nate37/#tunes wishes he could spell :) only when u are in ur bad mood wow. more crap hits the tunes mailing list clue: i wasn't in a bad mood 04:10am ummmm... then when I asked u "do u know anything about home brewed computers" in #{} did u respond "I don't f***ing care!!" and leave? don't pretend to understand my psychology i am motivated by signal and noise ummm.. it was a noise is signal room i can be intensely bothered by the 3rd bessel function of the color patterns on a brick wrong and don't pretend that my words actually reflect what i really mean then what is it? (lar1 told me it was a noise is signal room) the english language is stone-age technology and lar1 is god of irc, of course on IRC the only reflection of a person is words you're annoying me again this stuff is in the logs, do you understand that? i use the logs to keep track of conversations and development on tunes hmm, how is haskell reflective? with monads? yes there are various evaluators in haskell, many built on monads 04:20am -:- SignOff nate37: #TUNES (nate37 has no reason) what a snotty little brat how is it that a 14-year old with little programming experience comes on to the #tunes channel and expects us to coddle him and think of his ideas as special? he has ideas? yeah just like every other teeny-bopper coder who comes here 04:30am and just like the new addition to the tunes mlist, kyle, loves to spread his good news to us all 04:40am is haskell purely functional? monads "remove" some side-effects, but is it a pure functional language? (the ads say so) oh well it's a matter of definition 06:20am -:- bineng [Anders@j141.ryd.student.liu.se] has joined #tunes -:- smoke [smoke@16dyn1.delft.casema.net] has joined #tunes the idea is to construct an imperative program with monads 07:40am some people consider the clean way where you pass the state of world as an argument to be more pure 07:50am -:- Connection closed from irc.us.openprojects.net: Success -:- 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 lackey.openprojects.net) -:- Your host is lackey.openprojects.net, running version u2.10.05.18.(ipcheck4-5) (from lackey.openprojects.net) -:- This server was cobbled together Wed Apr 28 1999 at 12 02:19 EDT(from lackey.openprojects.net) -:- lackey.openprojects.net u2.10.05.18.(ipcheck4-5) dioswkfcg biklmnopstv -:- [local users on irc(75)] 9% -:- [global users on irc(389)] 44% -:- [invisible users on irc(490)] 56% -:- [ircops on irc(21)] 2% -:- [total users on irc(879)] -:- [unknown connections(1)] -:- [total servers on irc(30)] (avg. 29 users per server) -:- [total channels created(328)] (avg. 2 users per channel) !lackey.openprojects.net Highest connection count: 86 (85 clients) !lackey.openprojects.net Welcome to Open Projects! You are on 2 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 -:- Topic for #TUNES: Tunes http://www.tunes.org || Slate http://www.tunes.org/~water/slate-home.html || Interview w/ water http://www.linux.com/interviews/20000322/46/ -:- topic set by hcf [Sat Apr 1 06:59:39 2000] -:- [Users(#tunes:8)] [ TUNES ] [ smoke ] [ bineng ] [ water ] [ eihrul ] [ smkl ] [ Fare ] [ Fufie ] -:- Channel #tunes was created at Sun Feb 28 08:48:06 1999 -:- BitchX+Deb1an: Join to #tunes was synced in 6.108 secs!! -:- abi [nef@bespin.dhs.org] has joined #tunes -:- Mode change [-ws] for user TUNES -:- ult [ult@user-37kbatu.dialup.mindspring.com] has joined #Tunes -:- SignOff Fufie: #TUNES (brb) -:- SignOff smoke: #TUNES (Ping timeout for smoke[16dyn1.delft.casema.net]) -:- tcn [r@cci-209150250051.clarityconnect.net] has joined #tunes -:- SignOff ult: #TUNES (Read error to ult[user-37kbatu.dialup.mindspring.com]: Connection reset by peer) -:- ult [ult@user-37kbas2.dialup.mindspring.com] has joined #tunes -:- smoke [smoke@16dyn1.delft.casema.net] has joined #tunes -:- SignOff ult: #TUNES (Read error to ult[user-37kbas2.dialup.mindspring.com]: Connection reset by peer) -:- ult [ult@user-37kbago.dialup.mindspring.com] has joined #Tunes -:- SignOff bineng: #TUNES ( ) -:- SignOff water: #TUNES (Ping timeout for water[tnt-9-207.tscnet.net]) -:- water [water@tnt-9-75.tscnet.net] has joined #tunes yes haskell is purely functional. monads pass around the state of the world eh? delayed reaction :) well, i was napping :) why did you respond to that crap about databases on the mlist? 11:10am -:- SignOff eihrul: #TUNES (Ping timeout for eihrul[usr5-ppp93.lvdi.net]) It was after midnight that's your excuse? :) was it that bad? heh dunno it's just that this db theory stuff is coming from a total newbie I tried to be nice yeah, i prefer to ignore such threads so that they die on their own you know set theory pretty well right? how's that compare to db theory? e.g. prism, random newbie musings, etc I might have to unsubscribe from that list db theory is 99.9% about a static implementation and 0.1% about actual semantics and there's a lot more to set theory than most people like kyle really grok as a small example, hypersets and classes does it deal with hierarchies? heh not sure what you mean umm, let's say "{a,b,c} are part of {d}" hypersets are sets which have definitions which are not constructive. call them co-inductive, cyclic, or self-referential as you wish huh? ok how can {a,b,c} be part of {d} ? unless d={a,b,c} or something -:- eihrul [lee@usr5-ppp189.lvdi.net] has joined #tunes set theory also has modalities, which are formal versions of queries say you're modeling assemblies and subassemblies.. typical db stuff. anyway 11:20am hm? you stopped any software based on set theory, afayk? there are programming languages based on the ordinary versions of set theory SETL is one there are a few others, but most programming languages that can handle sets and classes and such don't do so explicitly for example, maude can handle a lot using equational rewrite but maude is rewrite-focused, not set-focused escher has support for sets of course, escher still hasn't been implemented what kind of sets? finite lame ones? heh oh just about every programming language has those stupid "call the bit-field a set" constructs escher has better support than that i think .. it has something to do with logic programming hm abi's link and google/dmoz's link to escher return nothing umm.. it's not practical to switch languages just because the ones you're using don't inherently support all the features you need. 11:30am www.cs.bris.ac.uk seems to be down tcn: huh? i see in other words: find an extensible language and work with it ah well then i'd say to work with maude :) if you want functional-logic programming with real power -:- nate37 [nate@ppp-207-151-70-52.dialup.pcmagic.net] has joined #tunes hey water, het tcn s/het/hey/ hey 11:40am jesus, 3 more messages heh, for what? yeah it's the lame thread that never ends well i just sent my own response out about slate hopefully it will counter-balance the other thread what's 'the other thread' about? db "theory" ehhe nate: why don't you join the tunes mailing list, too, and share your vast knowledge and experience in language design issues? ;) everyone else who's a newbie seems to think themselves an expert, so i don't see why younger teenagers can't join in ;) ok i'm done with this 'db theory' thread.. we're coming from different dimensions.. heh yeah well they don't seem to grok how a reflective language *can* handle meta-information maybe i should educate them fixing a broken theory by adding more complication who's doing that? them or me? 12:10pm -:- rares [rares@wtrb-sh9-port71.snet.net] has joined #tunes read the last message oic water: I don't argue unless either I'm a expert to the field, or it realates to me directly somehow (and, of course, I have a reason to argue). but I rarly think I'm a expert at a field * water/#tunes is debating on how far he has to go to get the message across without coming across as rude he says many RDBMS optimize joins.. but that doesn't spare me from typing the damn things in. I usally listen on mailing lists (as u can see, not on irc ;) yep i'll read them on the web.. this is getting ridiculous :) hey.. the 'wilma' program has a y2k bug :) umm, it's still not fixed? it lists 2000 at the bottom heh yeah, it's a known bug mailto: dem@tunes.org or the wilma bug-report address, whatever that is