IRC log started Sun Oct 17 00:00:00 1999 [msg(TUNES)] permlog 1999.1017 -:- Fare [fare@d026.paris-121.cybercable.fr] has joined #Tunes good morning, Fare * Fare/#Tunes goes watch star wars, ep.I k -:- SignOff Fare: #TUNES (Ping timeout for Fare[d026.paris-121.cybercable.fr]) 12:40am -:- SignOff Syntax: #TUNES (Ping timeout for Syntax[08-165.006.popsite.net]) -:- Syntax [dma@08-165.006.popsite.net] has joined #Tunes -:- water [water@tnt-9-166.tscnet.net] has left #tunes [] -:- SignOff air: #TUNES (BRiX [http://www.qzx.com/brix] :: sleep) -:- SignOff Syntax: #TUNES (Ping timeout for Syntax[08-165.006.popsite.net]) -:- Fare [fare@d026.paris-121.cybercable.fr] has joined #Tunes -:- SignOff Fare: #TUNES (Connection reset by pear) -:- NetSplit: sterling.openprojects.net split from varley.openprojects.net [08:23am] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [sterling.openprojects.net] -:- Netjoined: sterling.openprojects.net varley.openprojects.net -:- Tril [dem@bespin.dhs.org] has joined #tunes -:- ServerMode/#tunes [+o Tril] by sterling.openprojects.net -:- smkl [sami@glubimox.yok.utu.fi] has joined #tunes -:- Syntax [dma@06-108.006.popsite.net] has joined #Tunes -:- eihrul [eihrul@usr5-ppp49.lvdi.net] has joined #tunes -:- SignOff Syntax: #TUNES (Ping timeout for Syntax[06-108.006.popsite.net]) -:- Syntax [dma@08-175.006.popsite.net] has joined #Tunes -:- air [brand@p0wer.qzx.com] has joined #tunes hey air * smkl/#tunes is compiling glibc2.1.2 it will probably break everything 10:10am what do u have now? glibc2.0.6 oh i have 2.1.1 it doesn't run all binaries 206 doesnt or 211 doesnt? 2.0.6 i'm just compiling 2.1.1 hmm -:- binEng [Anders@j141.ryd.student.liu.se] has joined #tunes 10:20am -:- fire [no@209-68-229-150.dialup.cust.tfb.com] has joined #tunes hey binEng long time no see hullo it's me hickserv agreed... ah ic you just can't keep to the same nick, can you? ;) guess not whatcha been up to lately? I've been working on AKOS the language/object part 10:30am i see im not familier with AKOS ignorance is bliss ;) at lest until we have some code hrm i've been trying to push fare to get some stuff done on tunes? yes any result, then? no i just emailed him last night not yet was there anything special you had in mind? like the HLL no space left I didn't know you had any concrete plans regarding that at all? well we gotta at least make the plans! hmm, I wonder what comes first, plans or ideas? maybe it doesn't matter we certainly have enough ideas but concrete such? I can't remember I've seen any concrete ideas 10:40am my mind has just gone blank i can't remember the word that means to postpone in order to delay which is what fare is doing * binEng/#tunes tries to think of any pun on 'fire' but fails you can try to catch me but never will you mean intentially? grrr.. yes nah not quite what I was looking for :) * fire/#tunes thinks and thinks eihrul: you awake? -:- hcf [nef@me-portland-us124.javanet.com] has joined #tunes hi hcf hi re hi hcf 10:50am binEng: you good with c? intermediate, I guess. in my interpreter you define a word and i need to store it so that when the programmer enters the object name it can access that object on the stack so i need to store that name and be able to retrieve it quickly hashing, maybe? yes well that was my first idea fire; making food, back in a few :( but it is not like i know exactly how many objects the programmer will define it would suck to have to rehash it every time is speed much important? important enough does the interpreter read raw text? yes 11:00am I'd make a list with struct { *name , *value } and deploy extra data structures to make lookup faster, I think how big is the list though? that's the problem maybe you could have the programmer enter an estimation of the number of variables, as to make the initial list size approximately the right size heh no way to increase performance only, ofc. ok im gonna try the rehash and see how that goes speed wise then i'll post the code on my homepage and let people hack it :) :) if you really think about it, the programmer will be accessing objects more than defining them whatever -:- SignOff Syntax: #TUNES (Excess Flood) -:- Syntax [dma@08-175.006.popsite.net] has joined #Tunes 11:10am -:- SignOff Syntax: #TUNES (Excess Flood) -:- Syntax [dma@08-175.006.popsite.net] has joined #Tunes -:- SignOff Syntax: #TUNES (Excess Flood) -:- Syntax [dma@08-175.006.popsite.net] has joined #Tunes -:- SignOff Syntax: #TUNES (Excess Flood) -:- Syntax [dma@08-175.006.popsite.net] has joined #Tunes What's the matter with Syntax? * eihrul/#tunes ponders. all he does is sit in here idle ack, he's in every channel heh perhaps he is flooding or something good guess Sherlock ;) eihrul: what do you think about my object name problem? hashing... either that or perhaps a binary tree mabye a binary tree is better its more dynamic like i said i don't want to rehash not quite as fast as hashing the thing is though, the number of objects is static why are you interpreting raw text in the first place if speed is so important?! like i said before binEng: i want to later make this into a compiler where speed is actually important 11:20am so... don't hash until you've read in all objects if you don't want to rehash hmm i'll send GNU a bug report with glibc2 dir attached or something well geee... um it doesn't really work that way oh well i've never implemented a hash befor e i hope this works well, i have gcc hash function somewhere core spouted it one day er wait i think thats wrong one wait, that's right :) lucky you i had to help someone write a hash table so you benefit :) :D gee thanks actually there is a bug in string_hash () :) need to advance the character after the loop i forgot what << does 11:30am remember your direction... >> right, << left er directions -:- water [water@tnt-9-187.tscnet.net] has joined #tunes hi water hey 11:40am what's up? not much right now... been thinking and working with the computers in the house hmm soon i'll have huge amounts of diskspace free 11:50am archiving mp3's? no, compiling glibc2 ah does anyone here have a bt848 card? 12:00pm -:- xunil [user9474@98CB3C33.ipt.aol.com] has joined #tunes hello, xunil hi, whats this room about? abi: tunes? i think tunes is http://www.tunes.org, a free reflective computing system or for programming languages what the internet is to networks or like flypaper for hallucinogenic programmers or discombobulated or obnubilated -:- water has changed the topic on channel #tunes to: Free Reflective Computing System (http://www.tunes.org/) got questions? :) not yet:) k im gonna go since im so bored here:) as ive done with the other rooms:)... sorry its ok. bye -:- xunil [user9474@98CB3C33.ipt.aol.com] has left #tunes [] eek. aol heh 12:10pm gee how discouraging i am heart broken ;) ? that xunil would leave us so so so, I'm sure another AOLler comes by sooner or later heh 12:40pm hmm this is cute i can access web pages and telnet but i cant download files from ftp or http within brix? no, in win98 and linux oh ahhhh, they are .exe files. i gzip'd em and they download fine linux netscape, win98 ie4, linux ncftp and win98 ftp wouldnt download them as .exe that is stupid but you can (at least i can) it may be a security setting, or the default action might be to execute (guised as open) im talking about ftp too what ftp clients? i just told u what's "win98 ftp", though? just ie4? 01:00pm u open a dos box and type 'ftp' its a console ftp app oh. that IS odd, then and ncftp in linux wouldnt get it either what about the generic linux client? this is whats odd, i never have in the past, and dont right now, have problems ftping those .exe files between my linux and win98 machines -:- NetSplit: lackey.openprojects.net split from devlin.openprojects.net [01:02pm] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [lackey.openprojects.net] i used lynx on my isp to download the .exe files then i gzip'd em and downloaded to linux, gunzip'd and ftp'd to win98 blame it on M$'s huge bureaucracy for not making ftp work consistentlym, and netscape for following M$ :) but neither linux nor win98 could download the raw .exe file -:- Netjoined: lackey.openprojects.net devlin.openprojects.net -:- Syntax [dma@08-175.006.popsite.net] has joined #Tunes air: you remember to download in binary mode? eihrul: im not stupid, and ncftp defaults to binary never said you were stupid i think the phone line is why i cant download .exe files maybe .exe cant be downloaded over string lol 01:10pm u guys ever see tv in linux on a bt848? -:- SignOff fire: #TUNES (Ping timeout for fire[209-68-229-150.dialup.cust.tfb.com]) no, why? -:- SignOff hcf: #TUNES (Leaving) just wondering what its like i would like it very much if it would display in the root window and look good like my ati wonder does in win98 btw, do you know what's wrong if my screen occasionally goes blank in x? 229537 glibc-2.1.2 * smkl/#tunes gives up * water/#tunes goes to #linpeople water: screensaver? i thought it was a wrong refresh rate setting smkl: whats wrong? water: does it blank while yer using it? yes for half a sec, then returns to normal thats weird does yer monitor do anything? air: it's taking 230M disk space smkl: ouch well, it might be because i had to enter manually my monitor settings without the user's guide 01:20pm but i've tried all sorts of settings, and the ones i have had worked best so far ahh maybe there are docs on the net * water/#tunes checks cool... found it hmm... exactly what i thought, 85hz refresh darn well, it seems i got a good deal on it, anyway -:- hcf [nef@me-portland-us742.javanet.com] has joined #tunes why is hcf hopping? just ignore me k 01:30pm well, i will be playing a game or two. bbl -:- water [water@tnt-9-187.tscnet.net] has left #tunes [] 01:50pm -:- Kaufmann [Kaufmann@200.224.105.146] has joined #tunes Whee! Anyone? ... -:- SignOff Kaufmann: #TUNES (Ping timeout for Kaufmann[200.224.105.146]) 02:00pm -:- water [water@tnt-10-238.tscnet.net] has joined #tunes -:- water [water@tnt-10-238.tscnet.net] has left #tunes [] -:- SignOff binEng: #TUNES ( ) -:- water [water@tnt-9-54.tscnet.net] has joined #tunes -:- SignOff water: #TUNES (Leaving) -:- Kaufmann [Kaufmann@dial752.infolink.com.br] has joined #tunes Muahahahahaha! 03:40pm -:- SignOff hcf: #TUNES (Leaving) -:- SignOff Kaufmann: #TUNES (Read error to Kaufmann[dial752.infolink.com.br]: Connection reset by peer) abi: seen core core was last seen on IRC 19 days, 2 hours, 5 minutes and 36 seconds ago, saying: what's up? did you write more of EMK? [Tue Sep 28 15:20:31 1999] 05:30pm -:- SignOff Syntax: #TUNES (Ping timeout for Syntax[08-175.006.popsite.net]) -:- Synner [dma@04-174.006.popsite.net] has joined #Tunes -:- water [water@tnt-9-248.tscnet.net] has joined #Tunes -:- SignOff air: #TUNES (BRiX [http://www.qzx.com/brix] :: reboot) water: evening... s/ev/g'ev/g so it is * eihrul/#tunes nods. :) i had a realization just a little while ago, slightly related to arrow which is? that the idea of a modality, especially when you understand the logical formalism, really *could* have some serious benefit for human thought processes 06:50pm but that needs explanation for you, right? i'll look modality up :) save you the trouble well you'll have to look up modal logic research to find the formalism okay... the next important question is, why? :) why what? why does it benefit human thought processes? ah well, the modality formalism takes the idea of a relation over a set of objects and places the person the space of that relation again, why is this beneficial? :) so, if i'm at 'x' in the world under that relation 'R', and 'xRy', then the logic of modalities says that 'y' is possible for me, expressed as y i'm getting to that * eihrul/#tunes shushes himself. heh the point is that any language (formal or otherwise) offers a set of possibilities, that while not limited to just one relation, is certainly a similar idea and that "what's possible for me to do" is a modality which can be restricted or molded into "what i want to do" or "what i must do" ... so it is beneficial as a way of describing many languages and what there function is? or generalizing that anyway kind of i think it gives people a way to *really* identify where their worlds and others' worlds intersect 07:00pm to express transitions/relations from one state to another? sure it also gives a new way to interpret the world * eihrul/#tunes wows... Water did not say 'kind of,' but rather 'sure' since any given object doesn't support just one modality heh it at least makes me ask the question of what do my different modalities do to me, and how i deal with them and it also shows how subtle influences in a person's life can be veery powerful, but now in a formal way s/veery/very how so, in that a transition doesn't have a 'magnitude'? hmm well, i suppose a 'fuzzy graph' (weighting the transitions) would help that well, wouldn't such be just many atomic (minimal) transitions? or maybe allowing the graph to be a continuous thing, like a vector field yeah, it probably would so magnitude becomes the quantity of them well, you could allow 'micro-transitions', too eh? and quantify based on the "number" of micro-transitions i mean that a fine enough level of distinguishing of states might do the same as providing magnitudes * eihrul/#tunes nods. just didn't know what you were inferring by 'micro-transitions' ok 07:10pm well, that's my thought for today :) i should really be focusing on the arrow code well, wasn't that your original theory for you language in the first place? er your yep but it wasn't very well formalized at all which is why i wanted to code it, so as to step back and observe its formal properties and to allow others to do the same, which they couldn't with what i had for the lingo yah, thoughts are not necessarily conducive to sharing hehe... thoughts in *english* or any other usual language what do you think? bs or something to build on? i wouldn't quite call it bs but i wouldn't necessarily call it profound as a way of possibly modelling languages it could be useful 07:20pm hmm maybe "language" is the wrong word i of course *mean* the same thing as a formal language, but that's not what we think of for our languages * water/#Tunes ponders ontology? :) maybe feelings or maybe expressions or hmm * eihrul/#tunes ponders... okay, so maybe as a way of modelling what languages try to model :) heh. whatever *that* is !Hyrlaris:*! q3 game going @ quake3.roundtablepizza.com you're all invited hell, let's just call it "stuff" :) stuff isn't exactly formal yeah, but language will be mis-interpreted by the average person even the avg programmer won't get it fully i think that i'll stick with "ontology" yeah best to not ambiguate concepts by giving them commonly known names if you don't want all the baggage that comes with them of course, my ontology idea isn't what everyone else calls it, not exactly there's a small amount of baggage, but it won't be too hard to take care of uh oh i just had a really twisted thought modalities model modalities i mean, ontologies for me are language models, so... 07:30pm * eihrul/#tunes ponders. as a graph, an ontology (lang model) could represent the allowable transitions between expressions transitions by the one using the language... er * eihrul/#tunes hmms. that's pretty complicated to think about the relation between the transition and the initial state? the relation between the allowable expression types you know, for even programming languages, you can have expressions of primitives or ceertain kinds of combinations of them ah i was thinking the other direction :) what direction is that? was thinking vertically rather than horizontally explain well, i think i was thinking of a nonsense concept :) whereas i think you meant the relation between the final states i was thinking of how a given transition related to the initial state the initial state of *what*, though? i don't know :) * eihrul/#tunes ponders. maybe of a given object just hard to decide where 'what' comes in or rather, originates well, i'll ponder that one for now, i'd like to consider how to make the arrow code really work 07:40pm like right now, i have an arrow class that can be operated on to support the identity, reverse, and composition ideas * eihrul/#tunes had an idea. eh? dunno if its bs or what :) as a graph, an ontology (lang model) could represent the allowable transitions between expressions now... re: how does the user interact with the system by perhaps causing transitions? yeah, basically 'picking' the transitions from those allowed but also causing them what about it in particular? well, that was the idea itself :) oh yeah, i've been thinking about that and then, how would the user instantiate models within the system perhaps... that's a good question by causing a transition wherein the model exists in the system yeah, i follow that but what does the smalltalk code need to be to support that? -:- _QZ [brand@p0wer.qzx.com] has joined #tunes well, you'd need some initial model that allowed such (wow. he's using his old nick) <_QZ> im not in X (ah) <_QZ> stupid gdm wont start E or something well, you might need a model that also wouldn't restrict it eh? well, the arrow idea is supposed to support any infinitary construction via a finite spec * eihrul/#tunes ponders. what if the system itself is infinitary :) er... via any *constructed* spec it *is* then all models already exist in the system :) heh yeah, theoretically or atleast you could pretend they do and instantiate them as they as traversed 'pretending' via specifying a graph that contains them chicken vs. egg => the question is what kind of egg to lay 07:50pm -:- fire [no@209-68-229-172.dialup.cust.tfb.com] has joined #tunes * eihrul/#tunes tries to puncture this nasty recursive loop. *sigh* 5:54 minute mile it just has to be simple for the system to do ok for a freshman fire: not bad thanks i finally broke 6 for instance, i know that every system will have a graph for 'car' and 'cdr' (lisp notation for head and tail functions) er.. cancle that about the lisp notation water: well, hrmm, if the graph specifies allowed transitions, then what about a graph created by observing transitions that have happened (which have never happened before) it's sort of like car and cdr, except you reify references as arrows er... explain explaining the arrow system to eihrul? well, it almost boiling down to a graph just being instantiated in the system by the user the user specifies transitions fire: actually, we're explaining it to each other in order to figure out the code and the system, upon observing the specified transitions, instantiates the graph to model that (which is being modeled by the user) yeah, that has to happen all the time, but what does it mean for bootstrapping? that means reading the logs would be a very educational experience :) water: user needs to be able to specify transitions, system needs to instantiate them fire: sounds like a project for you :) * water/#Tunes watches eihrul back into that dreaded loop okay, so the problem becomes, how to do that :) eihrul: do you have the arrow code in a squeak browser? yes k, so do i well, the reference class is trivial the arrow class works, but it doesn't fit into anything the graph and world classes are very minimal 08:00pm the wierd thing about the code is that world inherits from graph, and graph from arrow, so that you can trivially represent a graph as an arrow (as, say, the parent node of the meta-graph) and a world (whatever that is) is both a graph and the graph's head arrow a bad thing is that the 'intension' for a graph is supposed to be some 'arrow statement' that specifies the graph which i can't figure out what to do with * eihrul/#tunes ponders. if the user just draws a graph, it's easy to just send all the arrows to the 'cache' but i might have problems if the 'intension' were empty in that case hmm * eihrul/#tunes got lost somewhere. about 'intension' for the graph