IRC log started Sun Jun 20 00:00:00 1999
[msg(TUNES)] permlog 1999.0620
-:- SignOff Beam: #TUNES (Ping timeout for Beam[d87.nnb2.interaccess.com])
-:- _QZx [brand@p0wer.qzx.com] has left #tunes []
-:- SignOff Fare: #TUNES (Connection reset by pear)
-:- Fare [rideau@quatramaran.ens.fr] has joined #Tunes
-:- SignOff Fare: #TUNES (Connection reset by pear)
-:- Fare [rideau@quatramaran.ens.fr] has joined #Tunes
-:- Beam [nobody3@d99.nnb2.interaccess.com] has joined #tunes
<Fare> Corey!
04:40am
-:- smkl [sami@MCCII.rdyn.saunalahti.fi] has joined #tunes
<Beam> hey are
<Beam> hey fare
07:30am
-:- billyboof [hatefull@nrwc-sh9-port84.snet.net] has joined #tunes
<billyboof> hello
-:- SignOff billyboof: #TUNES (hatefull@antisocial.com)
08:20am
-:- s_r [s_r@phila-dialup447.nni.com] has joined #tunes
-:- water [water@ppp-tnt-164.tscnet.net] has joined #tunes
<water> hola!
09:00am
-:- SignOff s_r: #TUNES (Ping timeout for s_r[phila-dialup447.nni.com])
-:- NetSplit: tolkien.openprojects.net split from sterling.openprojects.net [09:49am]
-:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [tolkien.openprojects.net]
-:- Netjoined: tolkien.openprojects.net sterling.openprojects.net
-:- Fare [rideau@quatramaran.ens.fr] has joined #Tunes
<water> wb?
<abi> wb is probably welcome back
<water> yeah yeah
-:- Beam [nobody3@d99.nnb2.interaccess.com] has joined #tunes
10:00am
<water> wb beam
-:- SignOff water: #TUNES (Ping timeout for water[ppp-tnt-164.tscnet.net])
10:10am
-:- hcf [nef@me-portland-us945.javanet.com] has joined #tunes
-:- SignOff hcf: #TUNES (Ping timeout for hcf[me-portland-us945.javanet.com])
-:- water [water@ppp-tnt-58.tscnet.net] has joined #tunes
-:- Reject [reject@ts45l14.pathcom.com] has joined #tunes
<water> hi
-:- SignOff Reject: #TUNES (Leaving)
10:50am
-:- hcf [nef@me-portland-us347.javanet.com] has joined #tunes
<water> hi hcf
<hcf> lo water
<water> what's new?
<abi> i think new is failing at ftp://ftp.gte.com/pub/m3/notes/When-NEW-fails.txt
<hcf> nuttin much
<hcf> how bout u?
<water> working on some ideas for tunes
<hcf> and arrow, or just for tunes?
<water> mop stuff, and ontologies
<water> well, i use arrow as my hypothetical tunes construct, so yes
<hcf> do u know of any images representing reflection?
<water> well, there's this page that tunes has
<water> i think it's in review
<water> there's also the pcp website. they should have a few pics
<hcf> http://www.parc.xerox.com/spl/projects/reflection96/overview.html has one but its cartoony and queer
<water> ok. i'll look
<water> oh yeah, i've got the papers from that one. yeah, i remember that pic.
<hcf> i'm just wondering what the tunes logo would look like if we had one
11:00am
<water> dunno. i've been playing around with naming the first product we release
<water> we're not ready, anyway
* hcf/#tunes was toying w/ the idea
<hcf> i know fare doesnt want a logo
<hcf> right now at least
<water> i was thinking of naming it after a mythological figure
<hcf> yeah, which gods reign over reflection?
<water> well, my arrow project needs a wrapper, anyway, for public presentation
<water> i was thinking about athena or something
<hcf> and thus minerva
<water> yes, but minerva doesn't sound very good
<hcf> both those names have been used too much already
<water> yes they have
<hcf> know the nordic version of athena?
<water> i've considered indian and aztec gods as well
<hcf> hmm
* hcf/#tunes will look for a trait<->god site
<water> there is an oddball that i've thought of... some ancient hero story from mesopatamia
<water> i can't remember the name though.  agamemnon? no.
<smkl> gilgamesh
<water> yes, that's the one
<smkl> agamemnon is from illias
<water> ahh
<water> i haven't looked into that for quite some time
<water> i'm considering naming the project Zen or something, but that seems way too pretentious and inappropriate
<smkl> naming after some ancient being is worn-out
<water> true
<smkl> SteffyGraph
<water> huh?
<smkl> more original name
<water> yeah, but what the hell does it mean?
<water> sounds like "staph", which is bad
11:10am
<hcf> water: dont u know some good tao related names?
<smkl> Steffi Graf, a famous tennis player
<water> hehe, chinese words wouldn't do
<water> and the "unknowing" project just sounds weird
<water> oh well
* water/#tunes 's thoughts return to formalizing his theory
<water> does anyone think that i could write up a good synopsis of tunes philosophy?
<hcf> why not?
<water> why not what?
11:20am
<hcf> why wouldnt u be able to do one?
<water> i don't know. i seem to speak a different language from the rest of the group
<hcf> colaborate w/ fare or tril on it
<hcf> to insure understandablility
<water> brb
<water> collaboration seems impossible in tunes.
-:- water [water@ppp-tnt-58.tscnet.net] has left #tunes []
-:- water [water@ppp-ip87-b22.tscnet.net] has joined #tunes
<water> back
11:30am
<water> anyone here?
<hcf> .
<water> well, my point was that collaboration, even on papers, has been terrible in tunes
<water> i have first and multiple second had experience with that
<water> hand
<hcf> beyond correction?
<water> what correction?
<water> correction doesn't exist in tunes
<hcf> beyond possibly making it non-terrible
<water> people don't correct, they inject a plug for their own system, which never fits within the document at all
<water> i am as sick of tunes as i predicted i would be before i joined
<water> we are resource-poor on the most important aspect of development - the human mind
<water> which is a result of cultivating a fringe-project mentality
11:40am
<water> hcf: still there?
<hcf> yep
<hcf> so whats next?
<hcf> what can be done?
<water> i'm not sure. for me, a strong possibility is to just quit sharing development with Tunes altogether
<water> there just isn't any feedback for what i've tried to put in
<water> like right now!
<hcf> sorry
<hcf> ok
<water> i understand if you're busy...
<hcf> have u voiced this to tril/fare?
<water> yes, but i'm not sure if they really understand or just nod
<water> fare is good with theory, but has limited vision for lambdas, which is why i stick with my arrow idea
<water> his limit of vision may help him write papers, but it's no good for tunes.
<hcf> did u get any worthy feedback from your arrow paper?
<hcf> most of what i saw was, "i dont get this, dumb it down for me"
<hcf> brb
-:- SignOff water: #TUNES (Read error to water[ppp-ip87-b22.tscnet.net]: No route to host)
11:50am
-:- SignOff smkl: #TUNES (sdddsdsddsdccacd)
-:- water [water@ppp-tnt-98.tscnet.net] has joined #tunes
<water> my isp is not impressing me with its reliability
<hcf> ppl who pay except mediocrity
<water> everyone expects mediocrity
<hcf> not everyone, but its unwinable battle
<water> hcf: what's your take on tunes?
<hcf> how do u mean?
-:- _QZ [brand@p0wer.qzx.com] has joined #tunes
-:- SignOff water: #TUNES (Ping timeout for water[ppp-tnt-98.tscnet.net])
-:- _water [water@ppp-tnt-155.tscnet.net] has joined #tunes
<_water> it seems that this discussion cannot continue
<_water> hcf: what's your take on tunes?
<_water> (i repreat)
-:- SignOff _water: #TUNES (The Tao has left the building!)
12:00pm
-:- water [water@ppp-ip67-b2.tscnet.net] has joined #tunes
<water> hcf: ?
<hcf> what do u mean?
<water> hcf: what's your take on tunes?
* hcf/#tunes wont play this game
<water> sorry...
<_QZ> heh
<water> what do you want to talk about then?
<water> hi qz
<water> ok. apparently none of you want to talk with me
<_QZ> thats cuz u bash everyone :)
<water> i wouldn't bash if i didn't find people with no minds of their own
<water> :P
12:10pm
<water> i mean, do you guys have any new ideas at all for tunes?
<_QZ> well bashing them is my job
<hcf> _QZ: uv been slacking on that
<_QZ> hcf: sorry
<water> have you guys ever thought about what _exactly_ is the difference between every software product on the market and Tunes ideas?
<_QZ> wter: they dont like to think about tunes much
<water> qz: you're damn right, they don't!
<_QZ> no one rally cares if it ever gets finished because they know that current software technology is better than anything that they could produce with tunes
<_QZ> s/rally/really/
<water> qz: you could be more help to me than the rest of the Tunes members together
<water> qz: well, that's because they (and, i'm sorry to say, you) have terrible imaginations.
<_QZ> no i have a goot imagination
<_QZ> damnit
<_QZ> good
<water> qz: no, because i know exactly how tunes can outperform today's technology
<hcf> i dont pretend to do/think anything important
<_QZ> hcf: well yer not really part of the tunes project so we're not bashing u
<water> well, there's obviously nothing more to say now.
<water> i'll wait for some of the usuals to show up again
<water> you guys can discuss the venomous things you call "programming" and "OSes" as much as you want
12:20pm
<_QZ> and hwat is it that u propose?
<_QZ> can u create a better system?
<water> qz: read the f***ing arrow paper
<_QZ> summarize it for me
<water> qz: no thanks, moron
<_QZ> in less than 100 words
<water> haha
<_QZ> can an idiot home user program or whatever in arrows?
<water> that's really funny
<water> they don't
<water> nobody does, idiot
<water> it's not a programming language
<_QZ> is it 100%, or close to, natural?
<water> it's not a language
<_QZ> uhh then what the hell is it
<water> not a VM
<water> not a protocol
<hcf> qz is no better than the tunesers
<water> qz: it has no correlation in any mathematical or logical research
<water> hcf: thanks, idiot
<hcf> water: no prob, genius
<water> hcf: at least i have a formalism
<water> qz: but, it can express every other formalism within it
<water> qz: the arrow system cannot be expressed by a logical theory. i believe that i may soon be able to prove this
<_QZ> somehow i dont thing arrows is the correct solution to the problem
<hcf> water: oh, ur a tuneser?
* Tril/#TUNES is back from the dead. Gone 13 hrs 9 min 10 secs
<_QZ> s/thing/think/
<Tril> _QZ i'll take 2 of those 300a's
<Tril> the pair that are the same is ok
<hcf> water: i wasnt including u in my comment
-:- _water [water@ppp-ip68-b3.tscnet.net] has joined #tunes
<_water> back
<Tril> hi everyone
<_water> hey tril!
<hcf> _QZ: how can u think anything about arrows since u havnt read the paper
<Tril> are we having an argument here?
<_water> yes
<hcf> Tril: trying
<_QZ> hcf: i can still bash it cant i? :)
<hcf> _QZ: w/o proper foundation, ur just being silly and useless
-:- SignOff water: #TUNES (Read error to water[ppp-ip67-b2.tscnet.net]: No route to host)
-:- _water is now known as water
12:30pm
<Tril> let me go read some logs and see what we're talking about
<water> not much, yet
<_QZ> where is arrows
<water> arrows?
<abi> arrows are just objects whose only attributes are two references.
<water> blah!!!
<hcf> abi: arrow?
<abi> arrow is a homo-iconic information representation 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/
<water> oh good
<_QZ> i wouldnt want the word homo associated with brix in any way :)
<Tril> _QZ: the word homophobic already is..
<hcf> _QZ: brix is mormonic
<water> hehe
<water> :)
<_QZ> hcf: i dont think so
-:- water has changed the topic on channel #tunes to: A Free Reflective Computing System (www.tunes.org)
-:- _water [water@ppp-tnt-168.tscnet.net] has joined #tunes
<_water> back
<Tril> didnt notice you were gone. in fact water is stil here.
-:- SignOff water: #TUNES (Ping timeout for water[ppp-ip68-b3.tscnet.net])
<_water> yes, it happens quite often
-:- _water is now known as water
<Tril> if you register your nick at least you can kick the other one off instantly
<hcf> Tril: will tunes-all be on tunes-cvs?
<Tril> hcf, no
<hcf> Tril: wheres the todo list for the collaboration page?
<Tril> hcf: probasbly in an HTML comment at the end of the source
12:40pm
<hcf> Tril: nope
<_QZ> water: is there a reason u picked a font for the arrow page?
<water> arrow page? what arrow page?
<_QZ> http://www.tunes.org/files/papers/Arrow_Introduction_v8.htm
<hcf> s/page/paper
<water> oh, that
<water> it was converted from MS Word format
<water> i used arial for the headers, and roman for the text
<_QZ> oh i see yer still stuck in windows :)
<water> just when i cruise the net
<water> half my hard drive is devoted to the linux experiment
<_QZ> thats too bad cuz un*x is much better doing net stuff :)
<water> who cares?
<_QZ> 100% of my hdd is devoted to linux
<water> so?
<water> and i should care because?
<_QZ> true thinkers use un*x
<water> true dumbasses use cryptic systems like un*x
<water> a profound waste of time
<_QZ> no, only a dumbass would choose an unstable OS like windows to work in
<water> i don't have the time to learn a new os
<water> i don't want to make the time, either
<_QZ> uhh ya right
<water> it took me four f***ing days to get X11 to work properly
<water> because i didn't buy a laptop with the right graphics card
<_QZ> wow if i blow my fonts upto 24pt then i can read yer abstract
<Tril> yeah linux laptop support is not great
<_QZ> arrow would gain more popularity if ppl didnt have to go outta their way to read the paper
<water> qz: it's not ready for public release!
<water> i never claimed it was!
<water> for instance, the most likely result is a modularization of the paper, not a final release
<_QZ> so how far will i have to read before the techno-babble stops?
<water> qz: a un*x user is not one to talk about techno-babble
<_QZ> u ever heard of the term KISS?
12:50pm
<water> neither have you, since you use un*x
<water> qz: yes, but my ideas are still forming
<_QZ> unix is very simple
<water> hahahahahahaha
<water> :D
<water> :D
-:- s_r [s_rr@phila-dialup620.nni.com] has joined #tunes
<water> there's a relative statement, if i ever heard one
<s_r> water
<water> hi sr
<s_r> can i ask you a question about OSes?
<water> i'm on a rampage, but sure, why no.
<water> why not
<_QZ> s_r: no but u can go read his paper
<s_r> ok, when a program does a syscall,
<s_r> let's say in Linux
<s_r> they put the arguments in the registers
<_QZ> well i bore easily and yer paper bored me to death
<s_r> and call int 0x80
<s_r> but when the interrupt is triggered
<water> qz: in other words, a fool
<s_r> it goes into kernel mode
<s_r> some arguments, though
<s_r> are pointers to memory
<s_r> like strings
<water> ok
<s_r> if i gave char *string= "dkfja";
<s_r> as an argument
<s_r> to say, execve
<s_r> how would execve have access to the memory location
<s_r> it has a pointer
<s_r> which i sent as arg to the syscall
<s_r> but it runs in kernel mode
<s_r> so it doesn't have access to the process's address space
<water> the kernel handles that (conversion of pointers)
<s_r> so how does it get the string?
<water> the kernel has access to the memory management
<s_r> does the kernel have access to each process's address space?
<s_r> qz, do you know?
<water> actually, i'm not sure about the technical details
<s_r> either the interrupt is just handled outside the kernel and the interrupt entry specifies the same address space as that which called it
<water> tril: any topics on your mind?
<s_r> or the kernel has access to each process's address space
<_QZ> s_r: in brix the kernel uses the same address space as the process
<_QZ> that called it
<s_r> qz do you know how it works in Linux?
<_QZ> no and i dont care either
<s_r> ok
<_QZ> are u programming for linux?
<water> sr: you can get the kernel sources, and there are docs on them as well
<water> there are also several books on unix architecture
<_QZ> s_r: if u are trying to figure out how linux works to make yer own OS then yer getting info from bad sources
<water> qz: i've studied os design quite a bit, and have written posix programs
<_QZ> s_r: water: so?
01:00pm
<_QZ> that was cute
<Tril> water, actually I have some ideas for tunes, I started writing an e-mail to the list last night but it became too long and I think  I might have to make it into a paper.
<Tril> (so I didn't send it :)
<water> hmm.. ok
<water> what's it about?
<s_r> i read Linux Kernel Internals and it said the kernel has access to the user data for the program
<s_r> qz why is linux a bad reference?
<Tril> it's about the difficulties of administrating UNIX mailing lists with so many different software packages with different configuration files, that are unable to share information. I point out that this illustrates the fundamental flaw with UNIX, that TUNES aims to solve. Then I tried to give examples of how TUNES would handle mailing list management, but I need to work on those examples some more.
<water> wow. that's a good _Example_, isn't it? :)
<Tril> yup
<water> cool.
<Tril> I might work on writing a new version of my typesystem spec first ,though
<water> hmm
<water> ok
<water> tril: what changes are you thinking of for the typesystem?
<Tril> but as I was writing the examples, I wandered around a lot and ended up with some justifications for the tunes project, which could allow us to find the differences between tunes and arrow
<Tril> no changes, just explaining it better.
<water> wow. that's good too.
<Tril> more formally, i think
<water> ok. i might have the first few final copies for arrow ready soon.
<Tril> how's the smalltalk prototype coming?
<water> i'm experimenting with some prototypes
<water> they're extremely simple right now
<water> relativisation of arrow worlds is mostly the concern
01:10pm
<water> i'm also trying to implement the VM in arrow as much as possible
* Tril/#TUNES learns smalltalk syntax by reading AbstractArrow.st
<water> ok
<water> it's just message passing, right to left
<Tril> looks like overloading
<water> i've overloaded a few messages
<Tril> i read the smalltalk tutorial before and looked around in squeak a little bit  , a long time ago
<water> and my messages don't collide with standard Squeak stuff
<Tril> but never mastered the user interface or tried to write anything in ST
<water> the new Squeak stuff is very usable
<water> much better than from V1
<water> but then, the new images come with lots of useless experiments
<Tril> what's a category: ?
<water> they can be stripped. my own image is only two MB or so
<water> oh
<water> it's just a way to collect classes or messages together
<Tril> ok
<water> not even an inheritance mechanism
<water> it helps separate and document classes
* water/#tunes fires up Squeak
01:20pm
<Tril> squeak has GC?
<water> sure, why not?
<water> it's a smalltalk
<Beam> smalltalk is a OO language
<Beam> has some reflection in it
<water> not much
<Beam> oops, thought you said "what's a smalltalk"
<Beam> :)
<water> hehe
<water> thanks
<Tril> wait, so the category: is like a module? which you are putting all your arrow stuff in?
<water> yeah, it gives me a handle to collect all my classes and export/import them as a unit
<water> but it's not a full module system by any means
<Tril> no dependencies, versioning, authoring ...
<Tril> submodules?
<water> except for timestamps, no
01:30pm
<water> there's a provision someone made for a six-level browser that had sub-categories
<Tril> minimal module system is better than none
<Tril> hehe
<water> yep
<water> the import function is called "fileIn", and is very low-level
<water> it takes a filestream and interprets it
<Tril> "chosen arrows" are ones that are currently reified, right?
<water> umm. yeah
<Tril> it's an interesting term
<Tril> you are the chosen one. yo have been chosen to exist, for the time being.
<water> they're the ones that have been referenced by the user.
<water> axiom of choice analogy
<water> user choice -> system deduction
<water> (to quote the paper)
<Tril> when you have served your purpose, you will no longer be chosen and you will retreat into the void of meta-addressability until you are once again needed. 
<water> right!
<water> woo hoo!
<water> i like that
<Tril> that's the same as with objects in types. But you can have extensional types too. I assume you can have graphs that are always explicit
<Tril> you can quote me on the above :)
<water> cool. thanks
<water> yes, some of my graphs will always be explicit
<water> those are the ones at the forefront of the system's progress
<Tril> some have to be, for before the system has any information about them, such as user specifed names and typed documents, etc
<Tril> and during bootstrapping, before the abstractions are all created.
<water> right
<water> i definitely agree there
<water> but for me, user choice can also be a relative thing
<water> it's hard to explain, but the "reasonability" of a choice may be incomprehensible to another system until the choice is "explained"
<Tril> that's where trust comes in. 
<water> hmm..
01:40pm
<water> i'm not into trust. i'd rather have a system where metaphors between two system's understanding of a thing may be drawn, even dynamically, to explain something to a third system
<water> but then...
<Tril> some agents are trusted , like the user or ones the user trusts.  Choices that are deduced can be trusted because there's a proof (like an explanation of why it's there)
<water> well, i want that protocol to be available for user manipulation
<water> an explanation, too, is from some theory into another one, which makes it sort of relative
<Tril> .  In this way, ArrowGraphs are an implementational hack, in that they encode t
<Tril> he information of the arrows they replace.!
<water> it's scope would be limited by the connectivity of that scope
<Tril> what is "replace"?
<water> oh
<water> the ArrowGraph represents the node and tree for a graph
<water> it's a smalltalk object that represents many arrows
<Tril> right, theorem deduced in a proof can't be trusted more than the axioms used to deduce it!
<water> ok, but i think that axiomatics are pretty complex, considering mathematicians have trouble with them
<water> especially logicians
<water> i think that abandoning the notion of scope management in favor of handling ontologies would be a good choice
<water> maybe the scopes are the nodes of my ontology graph... ?
<Tril> the definitions you choose determine everything about the theory. Implicit assumptions are trouble
<water> explicit assumptions are trouble, too
<water> even different sets of explicit assumptions can produce similar environments, while similar sets may have drastically different behaviors
<Tril> not sure what you mean by scope, but isnt an ontology a set of "settings" for the scopes of individual properties?
<Tril> ok, that may have made no sense
<water> abi: ontology?
<abi> ontology is a statement of a logical theory in some domain or what a person believes to be provable. i.e. what _can_ possibly be done in the world
<Tril> ah. nice
<water> it's an information window
<water> which i model as a composable and invertible arrow
01:50pm
<Tril> so the difficulty with axioms is they dont clearly lead to the behavior of the world they define.  Do you think starting with the behavior desired and trying to find axioms that fulfill it is better? Is that what ontologies are?
<water> hmm
<water> not sure...
* water/#tunes tries to sort out Tril's statement.
* Tril/#TUNES thinks about ontologies
<water> i understand what you're saying, but i can't find the relationship between the two ideas
<water> it's too complex
<Tril> you still need a language to describe behavior, right?
<water> we need something simple for now
<Tril> how are ontologies constructed
<Tril> what ontology is used to describe them
<Tril> an ontology is relative to another one
<water> ontologies in their simplest form are graphs...
<Tril> and as such can crosscut multiple ontologies in their definition.
<Tril> that's a pretty powerful concept
<water> "what something is" is equivalent to "what graph(s) does x belong to (in this ontology)"
<Tril> like prism 
<water> yeah
<Tril> except that prism leaves out the "fully specified" part, and leaves some of the implicit assumptions to the humans.
<Tril> like what a model is for
<water> correction:
<water> "what something is" (in that ontology) is equivalent to "what graph(s) does x belong to (in this ontology)"
<water> yeah
-:- SignOff s_r: #TUNES (Ping timeout for s_r[phila-dialup620.nni.com])
<water> prism doesn't go far enough
-:- tmf [s720@rovmaur.ii.uib.no] has joined #tunes
<water> tmf!
<water> tml man!
<water> re-write logician!
<water> what's up?
<abi> up is the opposite of down
<tmf> yeah:)
<tmf> water, is the preview on tool finnished?
<Tril> the equivalence just mentioned is called intensionality isn't it?  the intensionality in my typesystem is just like that, "what an object is" is equivalent to "what types does it belong to".  
<water> tmf: have you looked at Maude?
<water> tril: not sure any more
<Tril> we'll see when i formalize it some more
<water> tmf: not yet, it seems i've lost the docs :(
<Tril> preview = Review?
<water> tril: i would relativise "what type does ... "
<Tril> are you reviewing TOOL?
02:00pm
<water> tril: trying to
<water> tril: trying to review tool and tml
<Tril> tmf: learned a functional language yet?
<tmf> tril, I've looked at lisp
<tmf> but haven't used it for anything yet
-:- s_r [s_rr@phila-dialup089.nni.com] has joined #tunes
<Tril> i don't see how TML can be created without knowing functional languages, so you understand what you are duplicating, and can more convincingly argue the usefulness of tml
<water> tmf: your language fits in that family of tiny languages with trivial syntax and based on re-write logic. the distinguishing feature of tml is basic reflection.
<Tril> water: relativise membership in a type? what do you mean?
<water> tmf: and i/o reflection
<tmf> argumets, that I have heard before:)
<water> tril: make it relative to context
<tmf> I recon, I have a reasonable feel on rewriting and functional programming
<water> tmf: but i've looked around and compared. that's what it is.  it's not bad, either
<tmf> ok
<water> tmf: there's a review link with other small languages
<Tril> water: context is incredibly rich and variable. I think that would be like choosing particular types you were interested in to see whether the object was in it. 
<tmf> water, where can I find it?
<Tril> water: since objects could be in so many types at once, that's what a context would be useful for.  am i using the same meaning of a context that you meant?
<water> tmf: "Bla" entry at the bottom of the page (linked to other languages, particularly "false")
<water> tmf: all by the same author
<Tril> water; i dont think you answered my question of "replace"
<Tril> you started saying what an ArrowGraph was, which I understand
<water> tril: i didn't. i'm thinking
<Tril> if you want me to focus on one thread of conversation at a time, just say it
<Tril> but take your time.  i'll read on in the source
<water> tril: replace - the system should garbage-collect based on the minimization of information noise according to a heuristic.  creating a Squeak ArrowGraph object will subsume the information of the separate arrows collected to make it.
-:- binEng [Anders@dialup93-2-49.swipnet.se] has joined #tunes
<water> tril: context - an informal term
02:10pm
<water> hi bin, sr
<binEng> wow, what a crowd! hi!
<Tril> hi bin
<binEng> water: I'm glad you're here; I'm reading your paper
<water> bin: cool
<binEng> ;)
<water> bin: it may soon be obsolete, hopefully
<binEng> obsolete?? Just how obsolete?
<water> bin: i'm going to modularize it into sections and make it less confusing
<Tril> water: when you release the new version of the paper can we make a diff for people who want to locate changes?
<water> bin: the module papers would be developed to completion
<water> tril: there won't be new versions, instead, branches
<Tril> water: i could not find the parts that changed since the other draft i read earlier, so had to read everything agian (and it took  a while)
<Tril> cvs branches?
<Tril> is cvs being used ?
<water> tril: no, the paper will fragment
<water> tril: no cvs for me, thanks
<Tril> That's what CVS branches are. Parallel development 
<Tril> did you mean it will be split into smaller papers? that's not the same as a CVS branch, then.
<water> tril: i despise version numbering
<water> tril: yes
<Tril> If you meant multiple papers that are similar, but for different audiences, that would be parallel development, suitable for a cvs branch (maybe).
<binEng> water: Let me ask you some things, even thou I haven't read it all yet.
<water> tril: no, each issue will be handled by a separate paper
<water> bin: sure
<water> tril: "ontologies" will be separate from "arrow construct introduction"
<binEng> water: First, you mention, but only briefly, that functions can be formed by "specifying a left-deterministic relation". I find this subject important, could you explain?
<Tril> of course, reading the arrow paper again doesn't hurt me, probably helps.  
<Tril> water: good
<water> bin: ok, F(x)=y  <=>  xRy where forall x, xRy => not exists z: xRz and z~=y
<binEng> uh
<water> oh great
<water> bin: i'll explain
<binEng> I get the first part
<binEng> but "forall x, xRy => not exists z: xRz and z~=y"?
<water> bin: that means that the left-side argument only relates to a unique right-side argument
<binEng> are both x and y variable?
<water> bin: the left "determines" the right
<water> bin: absolutely
<binEng> so y isn't some constant...
<water> no
<binEng> I thiiink I get it
<tmf> water, how do you conclude that "it's [tml] not bad eigher" ?
<water> cool. i'll make sure to include that explanation in the construct paper
<binEng> But how does that relate to Arrows? It's defined outside the Arrows world, right?
02:20pm
<water> bin: it's a logical concept
<water> tmf: the basic reflection for a tiny language is very useful
<binEng> which as all concepts can be descibed in Arrows?
<water> tmf: just compare tml with false
<water> bin: yes
<tmf> water, is false the better of Oortmerssen's languages?
<water> bin: i don't remember.
<binEng> Yet, it must be interpreted by an outside person (agent?) to get its meaning?
<water> tmf: i just remember reading the most about it
<water> tmf: i don't remember.
<water> oops
<water> bin: yes
<binEng> Getting a little too much? ;)
<tmf> [i'll just look at false, then :) ]
<water> hehe
<binEng> And I grasped the meaning of the word 'agent' right? (I prolly was sleepy when I read that part)
<water> i have not yet begun to type! :)
<water> bin: agent's aren't just people
<binEng> yes, I understand that
<water> bin: ok, then you probably have a good idea about it
<water> tmf: ok
<binEng> next q. "The construct thereby enables arrows to contain references to graphs themselves". How?
<water> bin: by reifying the graph as an arrow
<water> bin: and the set-inclusion relation as another graph from that arrow to its members
<binEng> this'll sound silly, but since the word isn't in my dictionary, I'll ask. What does 'reify' mean?
<water> abi: reify?
<abi> i don't know, water
<water> bleh
<water> abi: reify is to make an abstract concept concrete for the system
<water> reify?
<abi> well, reify is to make an abstract concept concrete for the system
<water> abi: reify is also to make an object that represents a user's idea
<abi> okay, water.
<binEng> making concrete?
02:30pm
<water> bin: concrete things are those that the system can manipulate
<water> bin: like arrows for my system
<Tril> FAREEEEE
<Fare> Gakuk!!
<water> hehe
<water> yablutsk!!
<Fare> Who art thou who summonth me?
<hcf> water: add a defn for concrete to abi
<water> hcf: sure
<binEng> ok, how is the graph reified, then?
<hcf> abi: reify is also see concrete
<abi> okay, hcf.
<Fare> reified? abstract? contcrete? is someone talking about my implement1 paper?
<water> sorry, fare, it's about arrow. :P
<water> bin: i make an arrow and then draw arrows from that arrow to the elements of the graph
<binEng> ah, I guessed it.
<s_r> water have you implemented the arrow system?
<water> sr: i have a simple prototype that runs on Squeak
<Fare> water: oh :)
<Fare> was alaric around?
<water> sr: i have not released an enironment for it, however.
<water> fare: not recently
<s_r> arrowOS
<binEng> Next q... How do multi-arrows differ from sets of arrows with the same source?
<water> bin: multi-arrows are sets of arrows arranged as a sort of chain
<binEng> ordered?
<water> bin: ordered = arranged = whatever
<water> bin: imagine an arrow that follows a segmented path
<Fare> what's a multi-arrow?
<water> bin: its bends are the references in between "tail" and "head"
<binEng> oh, arrows of bigger length, sort of?
<water> bin: yeah
* water/#tunes can't believe how long it takes for people to get around to reading his paper.
<binEng> Are they important?
<water> bin: not necessarily, but they're an interesting construct, since they can build N-place relations
02:40pm
-:- SignOff hcf: #TUNES (Ping timeout for hcf[me-portland-us347.javanet.com])
<binEng> Would things as each natural number be represented by an arrow?
<water> bin: yes
<water> bin: definitely. that's the point
<binEng> Would they be explicitly stored? Each of them?
<water> bin: lol. of course not
<binEng> how do you do it, then?
<water> bin: indirectly, by axioms or inductive definitions, or whatever it takes to access the natural number from the particular kind of construct you want to use
<binEng> Could you mention some resonable method?
<water> bin: induction
<binEng> um... and?
<water> bin: of course, to access natural numbers themselves, you need a representation, like an unlimited string of digits
<water> bin: but you can't get an unlimited string without induction
<Fare> water: if they are N-ary relations, why call them "arrows" ?
<binEng> could you explain closer, please?
<Fare> why not call them "N-ary relations", as everyone does?
<water> fare: to extend the metaphor that a binary relation is built from arrows.
<Fare> maybe the right metaphor is that of relations?
<binEng> (think 'water:' before all my lines)
<Fare> think before all my lines
<water> fare: because i am not referring to relations themselves, but to truth values or selectors over them.
<Fare> and think after them, too
<binEng> Fare: I'll think about it
<water> bin: not sure where to begin
<Fare> what if arrows are expressible in terms of relations, but relations are not expressible in terms of arrows?
<water> fare: what??
<water> fare: a relation happens to be a set of arrows
<water> fare: an arrow can be specified by relations that apply to it
<Fare> this begs for the notion of set!
<Fare> sets can be described in terms of relations. Can they be described in terms of arrows?
<water> fare: what's the problem, exactly
<water> fare: yes, just as graphs are.
<Tril> Fare: Graphs are sets
<water> tril: graphs don't assume set-theoretic axioms
02:50pm
<binEng> Let's start simple. Say you want a function which turns a variable x into x+1. Some simple way, to demonstrate
<Tril> bags ?
<water> tril: not sure
<Tril> (the only diff between bags and sets is that bags allow redundant elements)
<water> bin: the lambda for the function would be an arrow, as well as the + application to 1 and the input
<binEng> sorry, I'm not familiar with lambda calculus yet
<water> tril: redundant elements are irrelevant
<water> tril: graphs can contain arrows that apply to other arrows in the graph or the graph itself
<binEng> speaking about that; did you say arrows were 'colored'? Can there be dupicates?
<binEng> s/dupicates/duplicates/
<water> bin: why do people bring up the word 'coloring'? did i accidentally leave it in the paper? if so, it doesn't belong.
<water> bin: what the hell is coloring?
<water> bin: my system can do the same thing as coloring
<binEng> no, somebody used that term in the mlist... I think it was meant to separate otherwise duplicate arrows from each other.
<water> bin: yes, coloring is possible but not necessary
<water> bin: the issue doesn't affect the theory
<binEng> k
<water> bin: mostly since multiple graphs may contain the same arrow
<binEng> could Arrows be implemented as the basic working model for a real machine, instead of using bits? Would it be feasible, efficient, etc?
-:- SignOff tmf: #TUNES (Leaving)
<Tril> now that would be cool
<Tril> an arrow machine
<water> bin: yes, using state-transition diagrams
<water> hehe
<water> indeed
<water> arrow machines could be infinite-state
<water> but with finite definitions
<water> and they would be first-order objects in the system
<binEng> you'd need some sort of Arrow assembler language
<water> very interesting, indeed!
<water> bin: i have it. it's called relativised reflective arrow logic.
<water> bin: it's based on arrow logic
<binEng> nice. Maybe it would explain some things?
<water> bin: some of it is in the paper
<binEng> I'm reading :)
03:00pm
<water> hehehe. my plan for world domination shall succeed!! bwa-ha-ha!!! }:)
<water> oops
<water> ok, maybe that's a little overboard. :)
<binEng> It seems to me that implementing Arrows on current machines would be quite inefficient.
<Tril> yup
<water> bin: it wouldn't have to be done directly
<binEng> that begs for an explaination
<Tril> binEng: Implementing tunes on current machines coudl be inefficient, but don't let that stop us. 
<Tril> The point is we want software a certain way, and if the hardware needs to adapt, so be it.
<binEng> Tril: oh no, I'm not proposing everyone to stop what they're doing and just die :)
<water> bin: as long as the arrow system can withdraw the information from another computation system, then the arrow system can get that information any time it wants to, like _after_ program execution
<Tril> water now you've givenm away your true goals just wait for the men in black to come and take you away!
<water> hehe
<water> did the nsa talk to you? damn them!
<water> :)
<binEng> _After_ program execution? Not even _during_?
<water> bin: well, whatever.
<water> bin: it's just that information generation does not have to be within the arrow environment itself.
<s_r> water do you think there ever will be a computing environment of arrow's properties, like arrow OS?
<water> bin: sure, but it too will not use arrows as computational objects.
<water> btw all, arrow can be implemented efficiently
<Fare> Tril: where's that 80-column perl routine that we wrote?
<binEng> But if you for example implement the arrows as a long list of pairs of number. Imagine the time to search throu all that everytime you wanted to find out if a relation existed!
<water> bin: arrows are abstractions, so that computations only have to obey their abstract restrictions.
<Tril> fare, probably in ~dem/wordwrap.pl, i'll look
<Tril> yep
<water> bin: ahh, correct, but this could be done incrementally
<binEng> like how?
<water> bin: just keep reading
<Fare> Trli: does it work?
<binEng> I will
<water> bin: it has to do with ontologies
<binEng> maybe a massively parallell computer :)
<binEng> One processor per arrow
<water> oh god, i hope not
<binEng> Maybe I'll just shut up now until I've read this throu
<water> maybe you will
<water> :)
<water> brb
<Tril> Fare: it seems to work
03:10pm
<Tril> Fare: try it out if you want, enter some long lines into the new member form
<Tril> Fare: then go read /var/www/htdocs/tunes-bin/new-members.html
<Tril> Fare: speaking of that, there's a new member but I didn't list them because our members page should be redone
<Tril> (in some kind of db)
<water> back
<binEng> are infinite numbers of objects only caused by lazy evaluation?
<water> bin: caused?
<binEng> um, 'thanks to'?
<water> bin: lazy evaluation _allows_ me to have infinite numbers of objects on a finite-state machine.
<binEng> and that is the sole reason?
<water> bin: sole reason for what?
<binEng> for the possibly infinite arrows :)
<Tril> bin, it's the other way around
<water> bin: no, the possibility of infinities is directly due to arrows modelling information atoms, which are definitely infinite
<Tril> infinite arrows is the reason we have to use lazy evaluation
<binEng> is that in theory? like, an infinite state machine?
<water> yes, we _have_ to use lazy evaluation
<water> bin: in theory?
<Fare> Tril: better the current thing than nothing
<Fare> Tril: think continuity of services
<binEng> like, in reality there _are_ no infinite state machines
<Fare> having to actually continue doing things as they are is an incentive to make them better
<water> bin: no, in reality there are many infinite state machines
<water> bin: like people :)
<Fare> binEng: in reality, you don't exist. You're just a signal that appears on my tty
<Tril> Fare: but most people leave all the fields blank.. we should take out a lot of the fields
<water> hehe
<binEng> well I'll tell you what Fare: It the same way round here
<binEng> but opposite roles, kinda...
<water> bin: even an electron is an infinite state machine
<binEng> k, k, k
<water> fare: thanks a lot :)
<Fare> binEng: well, you're a signal that simulates pretty well someone who'd think I'm a signal myself, but in reality, you're the signal, and I'm the real man.
<water> ok, enough!
<binEng> That's funny. I could say the same thing, except of course, that if I said it, the thought would have been REAL
<water> geez
03:20pm
-:- SignOff Beam: #TUNES (Ping timeout for Beam[d99.nnb2.interaccess.com])
* Tril/#TUNES thinks Fare really needs to put smileys on more of his statements..
<binEng> Fare, or you could think of it like I'm a thought for you - and you know your thoughts are real...
<Tril> bin seems to have not seen the implied smiley on fare's statement
<Fare> Tril: a machine can't put smileys on statements
<binEng> Tril: I just ignored it :)
<Fare> I'm but a lame AI, but my creator is a real genius!
<binEng> Fare: Obviously, a machine can do everything _you_ can. Or seems to...
<water> tril: maybe the tunes-cvs mailing list would be useful just for browsing, instead of subscription
<binEng> Fare, if you even admit you're an AI you *are* lame
<Tril> water, well it's being arhived so you will be able to do that.
<water> tril: ok, i just don't want you to be disappointed if people only read the archives, as i intend to.
<Tril> water: If there's enough demand, i'll set up the mozilla CVS tools, tinderbox and bonsai
<water> hehe
<water> tril: what are they?
<binEng> Maybe it wouldn't be that bad to be an AI... if you had access to your own code, that is
<Tril>        Tinderbox is for knowing when the tree is in flames. It is our our
<Tril>        home-grown web-based tool for watching the progress of the
<Tril>        continuous builds that we run on multiple platforms. It is how we
<Tril>        are able to tell when a checkin has broken the build on some
<Tril>        platform, and how we know which person needs to be viciously
<Tril>        flogged. See the Tinderbox page for more.
<Tril>        Bonsai is tree control (get it?) It is our our home-grown
<Tril>        web-based tool for watching the up-to-the-minute goings-on in our
<Tril>        CVS repository (viewing checkins and log messages, reading diffs,
<Tril>        etc.) Check out the Bonsai page for details.
<water> sorry i asked. :)
<Tril> water: don't ask what TUNES is someday you might get a couple megs of tunes source code
<water> tril: that's supposed to stop me? ;)
<Tril> nope
<water> ok, hopefully the Turing test has been abandoned for something more profitable. :)
<Tril> yeah, if the entity can win a flame war, it's considered human
<binEng> heh
<Tril> bonsai has a search engine of CVS activity, so it's the next step from the tunes-cvs, if people need more
<binEng> What exactly is the Turing test intended to test, btw?
<water> tril: yes, i got that
<water> tril: thanks
<water> bin: just stick to the paper, ok?
<binEng> hehe
<binEng> you explain that too??
<water> bin: look it up. it's on the web many times
<binEng> hey, I know approximately...
03:30pm
<binEng> So the test just assures that a computer can think if it pass? Man, that definition *sucks*
<water> yep
<water> but then it assumes an optimal game
<water> and an infinite one at that for the computer to avoid losing
<binEng> it's a very limited test... A computer could undoubtly think but fail the test. There's more to intelligence than the human version!
<water> yep
-:- FireZX [razor@207-172-132-176.s176.tnt5.col.md.dialup.rcn.com] has joined #tunes
<water> hi fire
<FireZX> hiya
<FireZX> so whats this about A free reflective computing system?
<water> tunes?
<abi> 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
<FireZX> cool
<FireZX> another infobot :-)
<water> abi: good girl!
<abi> :)
<water> anyway, that's what tunes is about
<water> roughly
<FireZX> hehe
<FireZX> i hope the site has no frames
<water> nope. no frames
<FireZX> X windows is messed up cas my card is (Yuck!) winblows - only
<water> no graphics, either
<FireZX> cool
<FireZX> thanks
<FireZX> brb
<water> no prob
03:40pm
<FireZX> cool
<water> thanks, it's mostly fare and tril's creation
<water> i'm also a big contributor
<FireZX> it seems interesting
<FireZX> Jun 20 14:58:00 sentinel in.telnetd[3845]: connect from 200.254.53.123
<FireZX> grr
<water> but you should know that it's still being formalized
<FireZX> that guy needs to stop trying to h4x0r me
<FireZX> ok
<water> we have a lot of theory to make for it
<FireZX> alright, ill check back later
<water> funny, just before you got here, everyone was talking...
<water> ok
<FireZX> hehe
<FireZX> that usually happens when im here :/
<FireZX> when im anywhere
<water> really? sorry
<FireZX> i go to efnet, join a channel, and people are talking, then go dead silent
<water> hehe, you probably intruded
<FireZX> i think i have a hypnosis affect on people :-)
<water> you gotta greet people
<FireZX> i do
<water> oh
* binEng/#tunes wasn't greeted
<water> maybe you just pick the wrong channels
-:- Revned [none@mid-tgn-neu-vty20.as.wcom.net] has joined #Tunes
<water> hehe
<water> hi rev
<FireZX> <FireZX> hiya
<FireZX> ok, brb
<water> <retaw>
<water> ok
<FireZX> tunes 
<FireZX> hmm
<water> tunes?
<abi> i heard tunes was 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
<water> hehe
<FireZX> i know that 
<water> ok
<FireZX> tunes- is <reply> Fine tuning NT is impossible, Linux is already perfect, and the rest are getting better. - http://www.tunes.org
<FireZX> tunes-
<FireZX> tunes- 
<FireZX> doh
<FireZX> tunes-?
<abi> Fine tuning NT is impossible, Linux is already perfect, and the rest are getting better. - http://www.tunes.org
<water> where'd you get that?
-:- SignOff water: #TUNES (Read error to water[ppp-tnt-168.tscnet.net]: Connection reset by peer)
-:- water [water@ppp-tnt-168.tscnet.net] has joined #tunes
<water> wtf?
<binEng> ah, the dreaded peer
<water> hmm
<water> sure
03:50pm
<water> rev's awfully silent
<water> must be busy
-:- billyboof [hatefull@nrwc-sh9-port84.snet.net] has joined #tunes
<billyboof> hello everyone
<water> hey bill
<billyboof> hey water... hows it going
<water> bill: most people are here, just reading other stuff
<billyboof> do you know of any books on vmm?
<water> virtual memory management?
<billyboof> yeah
<water> not specifically, no
<billyboof> damn...
<water> maybe a unix internals book
<billyboof> i need a book on that... it's just too much for me to figure out on my own... :(
<water> i didn't think that anyone published that kind of info
<billyboof> vmm?
<billyboof> abi: vmm?
<abi> billyboof: wish i knew
<water> bill: have you looked into the linux kernel source?
<billyboof> no... linux source==big & cluttered
<water> bill: or the *bsd's maybe?
<billyboof> hmmm... never seen the source to *bsd...
<Tril> what about VMM?
<water> neither have i
<billyboof> i want to buy a book on it...
<billyboof> something that'll explain everything about it....
<Tril> well, I have the Intel 80386 processor reference manual. It explains in detail all the features of memory management on the 386 CPU, and how to use them.  If you want to know how memory management works at the low level for a particular processor, i'd get that kind of book.
<water> he means virtual memory
<Tril> so do i
<Tril> it has a section on it..
<water> oh yeah
<water> that's right
* water/#tunes forgot
-:- SignOff Revned: #TUNES (Leaving)
<billyboof> i have the intel ppro manual... i was looking for something a little higher level <i understand all about page tables/etc>
04:00pm
<Tril> such as
<Fare> grrrr... my ISP let me down once again!
<Tril> so if you know how page tables and how virtual address spaces work then you know how fork() works
<Tril> and you know how swapping works
<Tril> and you know how dynamic shared objects work
<billyboof> yeah... i guess if i just read the chapter in the intel book like 4 more times it'll totally make sense to me...
<Tril> i wrote a program using mmap() for the last assignment i did
<Tril>  still, it wouldnt hurt to look for a unix internals book
<billyboof> hmmm... i'll pick one up tomorrow.. <there's a borders right next to where i work>
04:10pm
<water> it's been a long time since i programmed low-level stuff
<water> not long enough... :)
* water/#tunes imagines the day when he can forget how computers work entirely...
<Tril> or when they work how you want ?
<water> hehe
<water> no, i mean that it won't then take me more than a few minutes to learn an implementation method
<water> trivialize the learning process
<Fare> grrrr, again!
<Tril> you guys need new isps
<Tril> my isp won't hang up
<water> yeah, and i just got a bill notice.
<Fare> new ISP? I'm already on cable
<Tril> Fare: not using a cable modem now?
<Tril> what happens?
<water> i guess i'm not going to continue with these jokers.
<Fare> cable modem is only as reliable as the cable operator...
<water> exactly
<Fare> moreover, for some reason, I can't manage to ssh to quatramaran!
<Tril> I don't see how anyone can pay for a lack of service.
04:20pm
<water> i already paid for this month
<Tril> water: do you have a time limit?
<water> six hours, i think, but i rarely reach it
<water> not continuously, that is
<water> :)
<Tril> 6 hours per connection?
<water> yeah
<Tril> mine is 80 hours per month, after which is an additional fee
<Tril> (per hour after the 80 hours)
<water> that would not suit me
<water> i definitely use more than that
<water> i use this connection to escape my unpleasant social circumstances
<Tril> that's the GOOD isp I pay for. Then there is the university, which gives accounts to all the students. Those always disconnect after 2.5 hours, but you can have unlimited 2.5 hour sessions.
<water> hmm
<Tril> I use that after i think i've gotten close to my limit. Plus having 2 isp allows me to have a backup in case the network is down for one of them (doesnt work for when the construction workers break a fiber optic cable, which has happened twice now)
<billyboof> my isp is big and dumb, i get unlimited time...
<Tril> cutting the main fiber into town will shut down all long distance telco, ATM machines, and all internet traffic.
<water> hehe, let's hear it for the isp abusers!
<Tril> billyboof: how reliable is it? how often is it busy or slow or broken?
<Tril> they said the internet was designed to have multiple routes, but apparently that's not the case
<water> well, the internet is not about servers
<Tril> it's up to each provider to buy multiple routes if they want safety
<water> right
<billyboof> pretty reliable actually... never slow... never busy, plus i have two access numbers from two different towns...
<billyboof> they have 2 main routes <one to boston, one to new york>
<Fare> It's up to every monopolistic provider to provide crappy service so as to maximize profits
<water> i have three, but they're unreliable simultaneously
-:- Joob [pirch@dm1057.vix.zaz.com.br] has joined #Tunes
<Joob> HELLO!!!
<water> hey joob!
<Fare> hi, joob
<Joob> Hello water how are you?
<Fare> have we met before?
<water> boob: ok
<water> oops, joob
<water> :)
* water/#tunes sorry
<Joob> hello Fare
<Joob> abi 
<abi> yes, Joob?
<Joob> Hello abi
<abi> hola, Joob
<water> joob's a brazilian who doesn't know much english, but likes tunes anyway
<Tril> he likes abi, too
<Joob> :)
<Joob> Abi is my cyber  friend
<Joob> :C)
04:30pm
<water> fare: what's this add-to-faq entry in the irc archives?
<water> wow. i just caused the irc logs to reflect!
<Tril> think we need a search engine for irc logs? 
<Tril> or can everyone just login to bespin and use grep
* water/#tunes uses his text processors
<water> exactly
-:- SignOff binEng: #TUNES (Ping timeout for binEng[dialup93-2-49.swipnet.se])
<Tril> or, not having an account, ftp all the logs and use grep
<Tril> water: what add-to-faq entry?
<water> i'm just trying to figure out who thought the June 12 logs were important enough to give a name.
<Tril> oh that's me
<water> why?
<Tril> that's when we introduced CosmicRay to tunes. I think some of the stuff he asked, should be in the FAQ (with the answers I gave)
<water> ok
<Tril> but I didn't add it yet
<water> i just got to that part. it makes sense
04:40pm
<Tril> are you reading the logs? anything else you see that should be saved on  a page?
<water> tril: i can get the relevant section from the ADD-TO-FAQ for you
<Tril> get? 
<Tril> brb
<water> tril: ok n/m
<Tril> water: if you want to add it to the FAQ yourself, go ahead
<water> tril: no thanks
<Tril> i will, someday. that link is there to remind me :)
04:50pm
<water> tril: i'm not an official member, because my views don't represent Tunes views
<water> tril: ergo, not a good FAQ maintainer
<Tril> water: you could submit a patch (as any non-member)
<water> tril: same thing.
<water> tril: no thanks. find people who want to be members.
<Tril> i'm not sure what you mean by your views
<water> i know you're not sure.
<Tril> water: you seem pretty ok with the decentralized project organization, which has caused problems for people before, who have left the project
<water> so?
-:- hcf [nef@me-portland-us717.javanet.com] has joined #tunes
<Tril> water: but you also seem rather unhappy about the range of imagination of the tunes members. I don't see why.
<Joob> i`m fly.... :C<>
<Tril> bye, joob
<water> bye joob
<Joob> bye?
<Joob> who
<Joob> sorry
<Tril> joob we don't know what you mean "fly", then
<Joob> why
<Tril> Joob: do you mean "I am flying"? Why are you flying?
<hcf> ebonic's fly -> cool
>>> hcf [nef@me-portland-us717.javanet.com] requested PING 929922848 from TUNES
<Joob> thank
<Joob> I`m flying because don`t understand a dialog
05:00pm
<Tril> water: I think there's something wrong with exclusive groups. That's why tunes allows anyone to become a member. However, if someone chooses not to be a member, they usually also won't tell anyone why because they think they will offend somebody. 
<Tril> nobody will offend me. I'm not easily offendable.
<Joob> offend = ?
<Tril> to make angry
<Fare> water: could you layout a presentation of arrow in first-order logic?
<water> fare: i've tried
<Tril> water: I'd like to talk about the arrow smalltalk code if you're here
<water> fare: the predicates are atoms and vice versa. it's very disturbing
<water> tril: lets
<Tril> it says every object is in SelfGraph, what is that, and why don't I see it anywhere ?
<water> oh
<water> i forgot to update that doc
<water> SelfGraph == self
<Tril> SELF? or the smalltalk self message? I stil dont see how it relates
<water> an ArrowWorld's SelfGraph used to be a variable, but is now self, because it is implemented now as a subclass of ArrowGrpah
<water> self message
<Tril> ok
<Tril> but what is SelfGraph for?
<water> it's obsolete now. i used it to have ArrowWorld hold a representation of itself as a graph
<Joob> Ei vocês DOIS
<Joob> estou indo embora
<Joob> bye trill
<water> now, ArrowGraph can simply refer to itself
<Joob> bye water
<water> bye joob
<Joob> bye abi
<water> i mean ArrowWorld can also refer to itself
-:- Joob [pirch@dm1057.vix.zaz.com.br] has left #Tunes []
<Tril> so it was just to let ArrowWorld be a graph? and that is now done?
<water> that guy makes me translate portuguese too often
<Fare> what's disturbing about that?
<Tril> what did he say?
<Fare> it's just like lambda-terms that operate on other lambda-terms
<water> tril: it's an ongoing thing.
<water> fare: but it's not first-order, and contains inconsistencies
* billyboof/#tunes will be back later..
-:- SignOff billyboof: #TUNES (later)
<Tril> water: do you know portuguese?
<water> fare: and can lead to paradoxes everywhere
<water> tril: spanish
<Fare> water: 1) try anyway and 2) use intuitionistic logic
<water> tril: he taught me portuguese in spanish
<water> fare: i hate intuitionist logic
05:10pm
<water> fare: i mean really despise the stuff
<Tril> logic does seem to preclude that the world is made of hierarchies
<Fare> water: why so?
<Tril> (i should learn int. logic)
<Fare> it took me a while to understand it, but it fits the computational quite well
<water> fare: explanation? no thanks
<Fare> (and proof-theoretic logic, too)
<water> fare: i hate most computational models as well
<water> fare: otherwise i wouldn't be working on tunes.
<Fare> did you read lambdaND? Use of Intuitionnistic logic is instrumental in achieving a coherent reflective system
<water> yes, i did. i disagree that it is necessary, however.
<water> it may be for _your_ idea of a reflective system.
<Fare> what's yours?
<water> not sure.
<water> one where name-scopes aren't hierarchical, at least.
<Tril> ArrowWorld is a graph of all the arrows?
<water> tril: maybe more accurately a _relativised_ graph
<water> tril: since the interface is what it is
<water> correction: _all_ is relativised
<Fare> water: have you really grokked lambdaND?
<water> fare: yeah, what about it?
05:20pm
<Tril> water: ok
<Tril> I see there isn't a multi-arrow ..?
<water> tril: no, i've temporarily removed it
<water> tril: hopefully, it won't need its own class
<Tril> Does Squeak provide a graphical interface to the arrow world?
<Tril> i.e. for every class and instance?
<Tril> that you are using
<water> tril: no, i'd need a Morphic class for that.
<Tril> so it's just in the browser
<water> tril: for now, yes.
<water> tril: the complexities of a graphical browser are too much for now.
<Tril> so you can create arrow worlds (plural), add arrows and graphs to them, and add arrows to the graphs
<Tril> and do the other operations (invert compose etc)
<Tril> what is the next step?
<water> tril: actually, now that self-reference is no problem, the only difficulty with a graphical browser would be graph representation
<water> tril: i mean its actual graphical shape
<water> next step for what, tril?
<Fare> water: well, Godel theorem states the impossibility of complete reflection in a classical system
<water> fare: no kidding.
<Tril> which is something I wondered all along. How will someone interact with the arrows, and what tools do you need for examining and modifying graphs and arrows?
<water> tril: maybe they won't have to.
<Tril> you answered and said that most people won't do that.
<Tril> yes, but YOU want to because arrows are your personal ontology
<water> fare: if you want to formalize arrow within a logic, be my guest
<Tril> it seems kind of difficult to keep track of all of them. Since htey have names, will you show a name next to each one?
<water> tril: i don't want to deal with arrows any more than the next joe.
<Tril> hmm
<water> tril: they eventually shouldn't have names.
<Tril> but they're necessary for the system to obtain the structure you want?
<Tril> i'm confused
<Fare> water: *I* can't, for I don't understand it fully
<water> tril: notice the AbstractObject docs
<water> tril: n/m it seems i removed the appropriate sentence
<Tril> water: I understand the part about names being removed
<water> tril: uiName is a variable that doesn't necessarily have to be a string.
<water> tril: it's a handle for an ontology
<Fare> water: I want complete consistent reflection, hence I'm satisfied with intuitionnistic logic
<Tril> i saw that somewhere
<Fare> moreover, you can't do better than proof-theoretic intuitionnistic logic in a computable setting
<water> fare: blah, no thanks. i'll take inconsistency any day.
<s_r> Tril what are you working on now?
<Fare> so that's no loss
05:30pm
<Fare> inconsistency means you can prove 0=1, and thence, anything
<Tril> s_r I'm looking at water's AbstractArrow.st, smalltalk code
<water> fare: well, globally, yes. not so locally
<Tril> water: but that prevents the sharing between those particular components
<water> tril: not necessarily
<water> tril: you could "wire them together" in a way that transforms info without entering a context where both components' contexts apply
<Tril> maybe!
<Fare> water: if you're going to use some kluge, you might as well choose intuitionnistic logic!
<water> anyway, i can always create an inconsistent context that i refuse the system entrance
<water> fare: no kludge. arrows can be my wires
<water> fare: and a kind of ontology could manage them
<water> i hope, anyway.
<Fare> well, that's one more reason for you to formalize it
<water> and i continue to try to do so. lord knows i've read enough about other people's ideas.
<water> but most people have first-order styles of formalisms.
<Fare> one never reads enough about other people's ideas
<Fare> first-order formalisms are used for a good reason
<water> one never finds out about all the world's ideas
<Fare> else, you have proof-theoretic formalisms
<water> what reason is that? to prevent tunes from being formalized?
<Fare> I know not of other consistent formalisms.
<Fare> how do that prevent tunes from being formalized?
<Fare> I don't see the intuitionnistic logic prevents tunes from being formalized.
<water> i don't know, but it certainly seems that way
<Tril> i'm going to go work on dscribing by types.
<water> intuitionist logic prevents inconsistency
<Tril> see later
* Tril/#TUNES is away: (afk) [BX-MsgLog Off]
<Fare> Coq is a successful tools
<water> sighh...
<Fare> s/s$//
<Fare> tell me what you don't like about it
<Fare> Maybe you have interesting grieves, but then, communicate them, don't keep them for yourself!
<water> what don't i like about a consistent language?
<water> c'mon
<water> everyone has to agree on truth then?
05:40pm
<water> facts are absolute? please!
* water/#tunes grows weary of irc discussion
<water> this is enough for today
<water> i've explained arrow to varying levels of detail to several people
<water> worked out some new ideas to explore
<water> read a few papers
<_QZ> u didnt summarize it for me :)
<water> it's time to relax!!!!
<water> sorry qz, i'm spent
<water> and you, fare, i will have to meet again
<_QZ> only the weak minded tire
<water> one's mind is not weak who can stand against the world and overcome
-:- water [water@ppp-tnt-168.tscnet.net] has left #tunes []
<Fare> water: facts are not absolute, but proofs are
<Fare> that's the essence of proof-theoretic (constructive/computational) logic
<Fare> QZ: the sleepy, too
<s_r> QZ what part of BriX are you now working on
<Fare> why isn't brix free software?
* Fare/#Tunes is back from a free software conference?
05:50pm
-:- miaow [miaow@adsl-151-200-16-202.bellatlantic.net] has joined #tunes
<Fare> miaow?
<miaow> hi
<Fare> lo
<Fare> have we met before?
<miaow> i dont think so i dont come here too often
<miaow> :)
-:- miaow [miaow@adsl-151-200-16-202.bellatlantic.net] has left #tunes []
-:- Boolean [user2002@ppp30-221.gis.net] has joined #Tunes
<Boolean> Hey there
<Boolean> Whenever I try to play an MP3 on mpg123 it says it can't access /dev/dsp.  IT is there but even as root I can't access it.  Can anyone here help me?
<hcf> Boolean: try #linpeople
<Boolean> Yeah, they won't respond
<hcf> try #debian or #redhat
<Fare> modprobe sound
-:- Boolean [user2002@ppp30-221.gis.net] has left #Tunes []
06:10pm
-:- SignOff FireZX: #TUNES (Read error to FireZX[207-172-132-176.s176.tnt5.col.md.dialup.rcn.com]: Connection reset by peer)
-:- SignOff hcf: #TUNES (Leaving)
-:- billyboof [hatefull@nrwc-sh9-port84.snet.net] has joined #tunes
<billyboof> hello
<_QZ> hi
<abi> que tal, _QZ
<billyboof> qz, think you can read/insult something i wrote?
<_QZ> sure
<Fare> billyboof: I can, you silly little slob
<billyboof> ahhaha
<billyboof> http://pages.cthome.net/consultant/os_spec.htm
<s_r> i have a question about C structs
<s_r> struct i386_gate {
<s_r> 	unsigned int	offset_low:16,	/* offset 0..15 */
<s_r> 			selector:16,
<s_r> 			word_count:8,
<s_r> 			access:8,
<s_r> 			offset_high:16;	/* offset 16..31 */
<s_r> };
<s_r> say i have that
<s_r> does that mean offset_low will be at the top of the memory struct
<s_r> like dw offset_low
<s_r> at the top in asm?
<billyboof> yeah
<billyboof> as far as i know...
<billyboof> make sure your compiler doesn't do any kind of alignment inside structures...
<billyboof> try writing the structure to a file as well, to make sure it gives you what you expect..
06:30pm
* billyboof/#tunes will brb...
<s_r> i have a question
<s_r> if i had 
<s_r> struct asdf {
<s_r> char x;
<s_r> char y;
<s_r> }
<s_r> would that be the same (in asm) as:
<s_r> x:  db 0
<s_r> y:  db 0
<s_r> ?
<Fare> nope
<Fare> char x = 0; char y = 0;
* Fare/#Tunes is away (sleep)
06:40pm
* billyboof/#tunes is back
<billyboof> qz: did you read it?
<_QZ> ya
<billyboof> so what do you see wrong with it?
<s_r> QZ what part of BriX are you working on now?
<_QZ> s_r: im not
<_QZ> billyboof: well
<_QZ> what does ipc have to do with opening files?
<billyboof> all files will be memory mapped, so ipc will done by sharing a file rw access between two processes...
<billyboof> kind of a mix between semaphores and memory sharing...
06:50pm
<billyboof> err... i have to add in something about the ability to pass messages between two applications as a way to initiate a full ipc session
<_QZ> yes u do
<s_r> qz how's file access going to be done in BriX?
<_QZ> semaphores are also part of ipc and dont have anything to do with files
-:- shadow [shadow@63.65.250.141] has joined #tunes
<_QZ> semaphores lock a resource and a resource isnt always a file
<billyboof> ahhh
<billyboof> ok .. i remembered semaphores from BBS days...
<billyboof> just remembered it involved creating a file...
<s_r> billyboof are you working on an OS?
<_QZ> under ipc u need to put a "shared mem" heading and talk about yer files in there
<billyboof> s_r: yeah
<_QZ> also add messaging and semaphore headings
<_QZ> what does a semaphore have to do with creaing a file?
<billyboof> ok cool.... i'll break ipc into two catagories... message passing and shared me
<_QZ> file locking has nothing to do with semaphores
<billyboof> qz: i just remembered from bbs's that creating semaphore flies..
<_QZ> ohhhh, yer talking about lock files?
<_QZ> a file with a pid in it
<_QZ> s_r: brix uses an object store
<billyboof> qz: i wasn't really talking about anything in particular with semaphores...
<_QZ> s_r: there is an object in the store that holds info about all the objects and it has pointers to them
<_QZ> s_r: objects can query this information object to find objects by name, type, author, etc and get the pointer to the object
<_QZ> s_r: if the object keeps a pointer to the objects it needs then no lookup is needed before opening the object
<_QZ> if a document has objects in it that are not embedded then it would have pointers to the objects in it
<_QZ> the pointers are better than filenames because u dont get broken links after the user changes a filename
<_QZ> and the info object doesnt contain filenames either. it just has keywords, titles, etc that are all searchable
<billyboof> qz: did you see anything else wrong with it?
07:00pm
<_QZ> billyboof: ya the message queues
<_QZ> oh wait lets goto vm first
<_QZ> do u plan on searching the current page directory after each timer tick for dirty pages?
<billyboof> not sure yet... that's why i want a book on virt mem mgt
<billyboof> how often should it be checked?  every 100?  every 1000?
<_QZ> i dont use vm in brix
<billyboof> ahh
<_QZ> since all brix code is in very small packets u only need to load what u use
<_QZ> the os tracks how many times each packet is opened and will flush packets with zero open counts when memory is needed
<_QZ> unless the packet has been memory locked by a system component
<_QZ> if u open an image to view it then the image can be immediately flushed from memory after the view thread draws it and dies
<_QZ> if u move the window the image is in then the image can be reloaded, redrawn and flushed
<s_r> my OS is going to be somewhat like that, but enhanced and more efficient in memory usage
<_QZ> how so?
07:10pm
<s_r> hehe just kidding i haven't thought out how memory will be used yet
<_QZ> heh
<_QZ> oh and it doesnt flush right away like in that example
<_QZ> when the count reaches zero the object is placed in a fifo
<_QZ> when memory is needed it pops the first object and flushes it
<_QZ> and continues popping/flushing until it has enuf memory
-:- shadow [shadow@63.65.250.141] has left #tunes []
<_QZ> if the object is reused before it gets popped then it gets removed from the list
<billyboof> i know i have to explain vm better... i don't really know that much about it though, i have to do a lot more rereading
<billyboof> didn't sink in well enough the first time :(
<_QZ> can yer ipc span machines?
<_QZ> transparently
<billyboof> i want to make it like that <all computing distributed>
<billyboof> actually i kinda forgot about that in the info about directory structure...
<billyboof> i want to make it so that root on a machine called BOOF is /BOOF/ and root on a machine named BILL is /BILL/ ...
<billyboof> that way if you open a docuement in the dir /BOOF/mydoc.txt, it'll be in the same place no matter where you look
<billyboof> but along with SMP i want to make it so certain applications can be sent to idle machines on the lan <with the correct security settings>
07:20pm
<_QZ> god damnit, just ruined another cd
* billyboof/#tunes hates fucking up CD's
<_QZ> i think these cd's i bought are crap
* billyboof/#tunes just got a spindle of 100 cd's
07:30pm
-:- SignOff billyboof: #TUNES (hatefull@antisocial.com)
-:- change_me [user4550@200.211.46.38] has joined #tunes
<change_me> hello
<abi> hi, change_me
<change_me> do you use linux HedHat
<change_me> cls
<change_me> end
<change_me> hello
<change_me> bay
<change_me> cls
08:10pm
-:- change_me [user4550@200.211.46.38] has left #tunes []
-:- eStormy [stormy@rain.futuresouth.com] has joined #Tunes
<eStormy> hi
<_QZ> hey
08:30pm
<_QZ> anyone here have starcraft?
08:40pm
-:- SignOff eStormy: #TUNES (eStormy has no reason)
-:- _BC [bmcbrine@pine.Alberni.Net] has joined #tunes
-:- SignOff _BC: #TUNES (I toss myself out of an airlock and try to sing....)
-:- deep [black-phoe@adm.univd.kharkov.ua] has joined #tunes
-:- deep [black-phoe@adm.univd.kharkov.ua] has left #tunes []
-:- BlackPhoenix [black-phoe@adm.univd.kharkov.ua] has joined #tunes
-:- BlackPhoenix [black-phoe@adm.univd.kharkov.ua] has left #tunes []
-:- billyboof [hatefull@nrwc-sh5-port18.snet.net] has joined #tunes
-:- SignOff _QZ: #TUNES (BRiX [http://www.qzx.com/brix] :: sleep)
-:- SignOff billyboof: #TUNES (hatefull@antisocial.com)
-:- pooh0 [pooh0@un-152-2.university.indiana.edu] has joined #Tunes
[msg(TUNES)] newlog 1999.0621
IRC log ended Mon Jun 21 00:00:00 1999