IRC log started Mon Jun 21 00:00:00 1999 [msg(TUNES)] permlog 1999.0621 -:- pooh0 [pooh0@un-152-2.university.indiana.edu] has left #Tunes [] -:- BlackPhoenix [black-phoe@adm.univd.kharkov.ua] has joined #tunes -:- BlackPhoenix [black-phoe@adm.univd.kharkov.ua] has left #tunes [] -:- NetSplit: sterling.openprojects.net split from clarke.openprojects.net [01:14am] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [sterling.openprojects.net] -:- Netjoined: sterling.openprojects.net clarke.openprojects.net -:- Fare [rideau@quatramaran.ens.fr] has joined #tunes -:- s_r [s_rr@phila-dialup089.nni.com] has joined #tunes -:- water [water@ppp-tnt-74.tscnet.net] has joined #tunes lo! hey fare i'm trying to complete my review of all significant reflection papers linked to by Tunes. 02:50am -:- SignOff Fare: #TUNES (Ping timeout for Fare[quatramaran.ens.fr]) -:- Fare [rideau@quatramaran.ens.fr] has joined #Tunes wb Fare: wb grrr what? the internet is br0ken what's the address for the bofh apology server tell me about it dunno i haven't been there in a year at least I'm looking for a bofh-class apology for a nasty thing I did (post as root on a forum) ok, couldn't help you. wouldn't know where to begin 03:20am fare: care to continue the discussion from earlier? sure, but must have lunch first will you be there in 15 minutes or so? yep no prob 03:30am I'm here (with a bowl&chopsticks in hand) ok where were we? * water/#tunes is getting the necessary docs for intro to ontologies. umm you last said that proofs were absolute 03:50am i'd like to disagree once again, my stance is to favor an inconsistent system with context-management proofs are finite, checkable, objects (it's in the paper :) that's relative. they're constructions from a vocabulary proofs are to be managed in-context, too, and are consistent within a consistent context, yes, they are consistent a same proof can be checked in many contexts maybe. the context changes, the proof stays depends on what level we call context-space "many" the proof is what you use to communicate confidence perhaps context is an imprecise word confidence is obscure i want a "geometrical" abstraction my father is a geometrist I don't understand what *you* mean by "geometrical" confidence is a statement derived or no formally achieved or intuitively posited confidence is deliberately vague in either case i'd rather have a development intensionally refer to the context in which it was constructed hmm. that doesn't sound right so? what do you mean, so? by defaulting a proof's binding to the construction's origin, the system must actively translate the proof's meaning to any other place. 04:00am intension is always ultimately out of the system, by def what? why? by definition of intention if it defineable within the system, it's extesion the intension of _what_ extension of context thanks for talking in circles. you've demonstrated that intension is relative. whereas extension is absolute bs intension is only relative for a first-order system i.e. equivalent to first-order logic if you make everything available within the same space, the distinction becomes quite blurred correction: intension is only absolute for a first-order system or extension, if you will -:- SignOff s_r: #TUNES (Ping timeout for s_r[phila-dialup089.nni.com]) have you looked at RbCl? let's go beyond these philosophical considerations ok RbCl? I've read a paper on it the "no kernel" one? yup is it good? i have it in front of me good idea, but not formalized enough ok anyway... continue, please 04:10am don't want to? fare: still there? well, they oughta have formalized the idea more what do you want me to say? fare: n/m the no kernel stuff. just a side question "let's go beyong these philsophical considerations" beyond well?? ok. i see that i'll have to lead this discussion. beyond philosophy, inconsistency management is easily a benefit as a framework for providing, say, multiple user support without "implementation hacks" as is done today. what are the basic entities in arrow? umm. arrows. :) consistency management is a benefit but restricting a system to a single consistent state is not a benefit what are primitives to deal with arrows? primitives? at what level? who talks about a *single* consistent state? 04:20am primitives: what logical sentences can you assert about arrows? fare: if you can't have inconsistency, then you can't manage consistency can you give a grammar for such sentences? fare: arrow logic!!!!!!!!!!! give a formal grammar, please. fare: read the f***ing paper geez i don't have to, because it's already been done by logicians besides, my reflective relativised arrow logic is the very thing that doesn't fit into a first-order grammar! you do, so that we understand better fare: are you listening? there is no grammar for it, unless it is reflective! and very reflective, at that! well, you can describe your logical connectors incosistent, i believe, too (or just say you use the usual ones) fare: they're in the paper under the heading "logic" and describe your atomic properties (in categorical logic, application of arrows, composability and composition of arrows, equality of arrows, domain and codomain of arrows, pertainance to a category, etc) 1) f*** categories 2) the atomic properties are _relative_ 3) equality is hence _relative_ domain and codomain? arrows have those? 04:30am it's obvious that you have no idea what fundamental concept the arrow construct is for! so your arrows are untyped? so far, so good. at the system level, yes at other levels, no how do they differ from relational logic, as used in languages such as prolog? where the hell did you assume a type system? (or equivalently from a non-deterministic lambda-calculus) arrows aren't relations they're binary selectors uh? hehe what's the difference? the difference is not mainly technical, but mainly political. i've confused the mighty Fare arrows are selections themselves arrows have no inherent conceptual structure except for the relations between them that define their reference structure so an arrow is just a couple (a,b)? or rather, a triplet (R,a,b) ? no, just the couple and not the triplet so you can't distinguish among arrows with same (a,b) ? the triplet is added by a conceptual layer either the difference is there, or it isn't at the system level, only the binary interpretation exists, but that is not the only level triplet-based knowledge systems have existed for decades. the difference is relative! Jacques Pitrat has built reflective systems on top of them for decades! no kidding thanks for the info :P you're welcome JL Lauriere has written a few books on that, and his well-known system SNARK it still isn't the same as mine maybe they have been translated in english; dunno what's the difference, then? there you go, assuming you've seen my stuff in a book 04:40am you always do this, and you still have no idea what my system is about binary selectors allow arbitrary interpretation they remove the overhead of triplet frames no, I'm trying to understand your stuff by comparing to things I know, since I find your explanations too confusing to be self-standing the truly different theories are always difficult to understand so one of the elements of the triple is implicit, and the other two explicit? no as long as you mean that implicit = absolute, no implicit is relative, and constructed by arrows, for me I don't mean and never have meant implicit = absolute implicit = implicit, absolute = absolute to me, you do because i've looked at your work and the stuff you've learned, for the most part don't the wish to be different overcome the wish to be useful s/don't/let not/ i don't care if i'm not useful to the establishment, because then they'll confine me i have other desires besides this project and they would restrict my societal freedom from doing those things anyway...... fare: for the actual system level, nothing at all is implicit nothing! that's impossible no, it isn't. it allows me to distinguish ontologies cleanly it also doesn't have to be a language to be useful just because it doesn't have a "bottom", doesn't make it impossible if you want a language, i can give you an ontology but that's not the system it has to be a language to be communicable it has to be communicable to be useful at large that's what ontologies are for or else, it'll stay forever in your head and never in another one's. ha! i've intuitively explained this to many non-programmers 04:50am to feel != to understand they understand it better than programmers can, very often at least they don't deny its existence to feel != to understand is expressed in a first-order sentence! :P i guess i should re-emphasize that i consider current human languages an evolutionary backwater You think you know when you can learn, are more sure when you can write, even more when you can teach, but certain when you can program. -- Alan Perlis i can think of a great many concepts that programmers often outright reject! i remember reading about a logic course where the majority of programmers dropped out the prof asked the students to frame sentences as logical statements the programmers found it the most difficult! even the "HLL" ones i'll be the first to claim that programmer's view of the world is very restricted and destructive and consistently argue it I can't say. I don't program enough to be called a "programmer", and don't unprogram enough to be called a "non-programmer" so I'm a non-non-programmer, which, in intuitionnistic logic, is not the same as a programmer 05:00am yeah, i know you're still a mathematical formalist anyway, i have to go soon see you tomorrow -:- water [water@ppp-tnt-74.tscnet.net] has left #tunes [] 05:10am -:- s_r [s_rr@phila-dialup282.nni.com] has joined #tunes -:- SignOff s_r: #TUNES (Ping timeout for s_r[phila-dialup282.nni.com]) -:- hcf [nef@me-portland-us1006.javanet.com] has joined #tunes -:- water [water@ppp-tnt-45.tscnet.net] has joined #tunes hi lo what's up? hmmm... up is the opposite of down hehe not much ok, i'll wait then got a question do u know the term for commonness and uncommonness of words? hmm vulgarity? i mean like get a web page (english) statistically? run it thru this filter get the least common words, then search for them to find similar pages oh i read some paper about that... keywords? no i'm stumped damn u should keep a history of urls to papers uv read it'd be helpful in cases like this well, i doubt there's a term for it anyway i think that they just described it 10:20am how would this be described? they talked about using word frequencies, and then sorting out common words from the least frequent maybe i still have a copy of it... * water/#tunes grep's information retrieval nope. it's gone. darn ok thx anyway btw just look up a research group for "info retrieval" u still reading papers, writing, thinking too much to need more links? well, kinda, but links might still help especially if they're good any 404s that u need found? umm. you should comb through the tunes site for those. you'll find plenty any specific ones? that u want none come to mind, mostly review, of course well i was gonna do this sort of thing b4 but im waiting for the site to "fixed" as in, the review to be proper set up s/per/perly/ yes, as am i plus, last time i did link checking for review, fare hated my output 10:30am really? why? cuz i didnt go ahead and fix everything i just made a list of link changes / MIAs oh i can see his point but it was work that NOONE else was doing or about to do true also 10:40am hum 10:50am hey fare hey hcf! water? water is an expression of the Tao i found that book oh good hold on, i'll look up that term abi: hcf is an expression of the D'oh ...but hcf is owner of abi... abi: no, hcf is an expression of the D'oh okay, hcf. * water/#tunes likes having electronic free books hehe Fare: know what happened to the janus project mentioned in the reflection rev? nope hcf: the book has a chapter addressing Luhn's theory for Zipf's law. water: uh? fare: hcf asked about word frequency terms Fare: do ppl have updates (link corrections) to the rev pages, they havnt added yet? The Janus guys have left the place. Maybe they still do their stuff in another french university hcf: not afaik hcf: the paper is basically mathematical, and not very technical hcf: it doesn't have a word for what you were thinking of. as far as i can tell... Fare: what were the head janus ppl's names? hcf: they just discuss the indexing of keywords Jean-Pierre Regourd; Jean-Christophe Pazzaglia; water: thx for looking no prob 11:00am well i found http://mediatech.didael.it/ita/staff/Pazzaglia.html has some papers on reflection water: http://igd.rz-berlin.mpg.de/~www/prolog/formonto.htm probly useless thanks i'll check it out -:- billyboof [hatefull@nrwc-sh8-port153.snet.net] has joined #tunes hello que tal, billyboof wow. the first page has a paper on multi-agent systems hi bill * billyboof/#tunes just got an upgrade upgrade of what? water: u into agent stuff too? hcf: yeah cyrix 6x86 166mhz w/ 32mb -> amd k6-2 450mhz w/ 128mb 166MHz or PR166? hcf: that ontology page has abstract stuff! particularly about graphical representations! (using guess what) billy: cool Fare: no .. it's that old that they still called it P-166 then it's better than my 150MHz P200+ ? hcf: actually, it's sorta hokey, but oh well Fare: actually it's a P200, but the motherboard i have says that the cyrix 200's were peaces of shit, so only up to 166mhz cyrix was supported brb wierd. 11:10am * billyboof/#tunes will bbiab... putting my cards in the new motherboard -:- melux [melonbar@yoda.wcis.com] has joined #Tunes -:- melux [melonbar@yoda.wcis.com] has left #Tunes [] back water: "using guess what", what? hmm? oh, using arrows. :) heh but it's just a structured inheritance and aggregation structure oops. too much "structure" :) so u want other pages w/ keywords: ontologies, graphical representation or visualization ? no thanks i've got enough idea already http://www.cs.umd.edu/users/ericp/ http://www.isi.edu/~valente/papers/banff-onto/doc.html * water/#tunes can't resist hcf: you're a better search engine than google and altavista! hell yeah hcf: what can you find on non-deterministic call-by-value lambda-calculus? ;-> or on logical reflection? hehe or on the notion of implementation? well, i've figured out the ontology problem water: i was pretty sure u'd like the 2 above urls, since the search returned them and ur arrow paper as the only hits my arrow paper's there? who put my paper on a search site? tunes is indexed by alta and thus ur paper is 11:30am oh, and my paper's only a link away hmm its a good indicater that i'm using good keywords yeah that means that i'd better improve my output if someone should stumble upon it the world is watching ;) hehe -:- jdl [jiml@ultra1.inconnect.com] has joined #tunes * billyboof/#tunes is back hello, all hi jim -:- SignOff hcf: #TUNES (Ping timeout for hcf[me-portland-us1006.javanet.com]) Anybody seen Tril recently? I wanted to talk with him about his type system not recently oh well... what's up? hcf was helping fare and i with research links -:- hcf [nef@209.94.148.175] has joined #tunes there he is! wb "There he is. One of God's own prototypes. Some kind of high powered mutant never even considered for mass production. Too weird to live, and too rare to die." 11:40am oookkkaayyy hmm no comment jim: maybe hcf can help you with any research you're doing. water: I'm not doing any research at the moment :) ...so much to do, so little time.... ok water: actually, last night I was plyaing around with visualizing an Arrow graph cool -:- SignOff billyboof: #TUNES (hatefull@antisocial.com) If my understanding of Arrow is correct, then it's harder than it sounds :) yep ftp://ftp.forum.swarthmore.edu/threads/geometry.research/3d.ontology An Arrow is a binary selector of two other Arrows, correct? hcf is a search-engine machine jdl: yes water: so when you draw an arrow graph, arrow heads and tails point to the arcs of other arrows, right? (not the heads or tails) jdl: yeah water: ok, well at least I was doing it right. :) jdl: an arrow is like a Lisp Cons cell. ok water: that I understand water: You and Fare will have to forgive me for being a heretic, but I never found Lisp to be that interesting. :) sure. no prob water: How do _you_ visualize the following arrow constructs? (MORE) water: (I'm using syntax as follows -- arrow_name: head, tail) (MORE) water: Construct #1: a: a, a abi: tunes is also a merry band of heretics i'm sorry, but that's too long, hcf water: Construct #2: a: a, b b: b, b water: Construct #3: a: b, a b: b, b (DONE) geez i try not to, actually :) :) i do have some drawings on scratch paper 11:50am of those things water: I'm a visual person; I tend to understand things better when I can visualize them. jdl: how do u visualize reflection? the way i draw them is by making a straight segment for the arrow arc, drawing a light circle around that part, and extending the ends until they reach the "light circle" of the target arrow. hcf: I don't; but if I did, it would depend on the context. "Reflection" means different things in different systems. once the drawing is done, i re-shape it into something more intuitive jdl: tunes reflection hcf: tunes reflection is still a little too vague for me to understand :) yeah, it's sort of vague to me too. :) water: can you say again in different words?