IRC log started Fri Jun  9 00:00:01 2000
-:- smoke [smoke@vengeance.et.tudelft.nl] has joined #tunes
[msg(TUNES)] permlog 2000.0609
-:- SignOff coreyr: #TUNES (Ping timeout for coreyr[net255ip95.parklink.com])
-:- coreyr [coreyr@net255ip95.parklink.com] has joined #tunes
-:- SignOff smoke: #TUNES (fucking fuck windows)
-:- billh [billh@cx739861-a.dt1.sdca.home.com] has joined #tunes
-:- SignOff abi: #TUNES (Ping timeout for abi[bespin.dhs.org])
-:- abi [nef@bespin.dhs.org] has joined #tunes
<hcf> wb abi
<abi> its good to be bad
<hcf> thats nice
<abi> s/bad/back/
-:- SignOff abi: #TUNES (Ping timeout for abi[bespin.dhs.org])
01:10am
-:- abi [nef@bespin.dhs.org] has joined #tunes
-:- _Luke [luke@203.46.39.102] has joined #tunes
<hcf> abi: OScat is a catalog of operating systems, large classified list w/ small nuggets of info about past and present(?) OSes at http://www.cs.wvu.edu/~jdm/classes/cs258/OScat/catalog.html
02:20am
-:- SignOff water: #TUNES (Ping timeout for water[tnt-10-148.tscnet.net])
-:- SignOff hcf: #TUNES (Ping timeout for hcf[207-172-225-144.s144.tnt1.pld.me.dialup.rcn.com])
-:- SignOff billh: #TUNES ([BX] We drink more beers than Norm on Cheers!)
-:- hcf [nef@207-172-225-125.s125.tnt1.pld.me.dialup.rcn.com] has joined #tunes
<hcf> Fare: when u de-idle i got a question for u
04:50am
-:- Kyle_L [kyle@cr168790-a.nmkt1.on.wave.home.com] has joined #Tunes
<hcf> re Kyle_L
<Kyle_L> hi
05:00am
-:- ult [ult@user-38lcdhf.dialup.mindspring.com] has joined #Tunes
-:- SignOff ult: #TUNES (Leaving)
-:- ult [ult@user-38lcdhf.dialup.mindspring.com] has joined #Tunes
-:- SignOff hcf: #TUNES (Leaving)
* Fare/#Tunes unidles
* Fare/#Tunes has ADSL, at last!
<Kyle_L> hi Fare
<Fare> lo KL
<Fare> wassup?
<Kyle_L> Defered BitField implementation 'till today.
<Fare> defered?
<Kyle_L> Deferred
<Kyle_L> (delayed)
07:40am
-:- water [water@tnt-10-0.tscnet.net] has joined #tunes
-:- eihrul [lee@usr5-ppp197.lvdi.net] has joined #tunes
<water> hey
<Fare> yeah
<water> the online slate docs are pretty much obsolete now
<water> but that means i'm that much closer to a full spec
08:50am
<Fare> good
-:- smoke [smoke@15dyn238.delft.casema.net] has joined #tunes
09:00am
<water> i guess i should mention the syntax shift, unless you already heard of it
<water> it's been determined that ('s convey redundant information. so we shifted from using () to encoding message-application points with single symbols.... "." for now
<water> "/" is also redundant, and was removed
<water> however, "result" now has to be explicitly called, so we shortened it to a single character, "^" for now
<Fare> what was /, already?
<Fare> why must result be called explicitly?
<Fare> couldn't it have been implicit?
09:10am
<coreyr> water: wasnt it ambigious what result meant?
<Fare> coreyr!
<water> corey: why?
<coreyr> i dunno
<coreyr> thats what i got from the convo.
<water> it's not ambiguous except where you have to call multiple ones in a sequence... iow just hard to read sometimes
<water> Fare: we made result explicit since "/" was semantically troublesome
<Fare> can you restate examples of old, previous, and lisp syntax?
<water> Fare: "/" was just the operator for shifting the evaluator namespace
<Fare> s/previous/current/
<Fare> replaced by . ?
<water> yeah i'll give an example
09:20am
<water> "(my Point (x (: 4)))" would now be "myPoint x : 4 . . ."
<water> which implies....
-:- SignOff smoke: #TUNES (One day sheep will rule the world)
<water> "((myPoint) x)" is now "myPoint . x^"
<water> (which would both return 4 now)
<Fare> uh? wouldn't that have been ((myPoint x)) ?
<water> hm
<Fare> sorry
<water> np you might be right.... there's subtle differences
<Fare> I mean, to lookup an object slot in a usual LISP object system, that is (x Point) or (Point 'x) or such
<water> oh
<water> slate doesn't have to quote selector names explicitly to look up slots
<water> since everything works that way
<water> gah brb
<Fare> ok
-:- smoke [smoke@15dyn238.delft.casema.net] has joined #tunes
09:30am
<water> back
<Fare> fore
<Fare> so a.x ==> (a 'x) ?
<water> anyway, evaluation order is specified by placement of "." and "^"
<Fare> and a^ ==> (a) ?
<water> actually, i think it's a x . ==> (a 'x) 
<Fare> is . postfix?
<water> yes a^ ==> (a)
<water> pretty much
<water> i didn't intend it that way, though :)
<water> although remember that ^ is an actual slot
<water> not just a syntactical operator
<water> because "a x" only just puts you in a position to inspect x's contents
09:40am
<water> once you're there, though, "." applies x and dumps you back to the original namespace
<water> heh it really *is* stack-based
<water> but the namespace system is not necessarily hierarchical
<water> ".." is not the mechanism for returning through the namespaces after evaluation... that's a result of "."
<water> ".." is a separate explicit slot that can be invoked any time
<water> perhaps there should be another name for it, because it looks a lot like ". ." which is 2 applies
-:- morton [morton@AC94F2E0.ipt.aol.com] has joined #tunes
<coreyr> , !
<water> eh?
<water> a comma?
<Fare> ok, so a^ ==> (a 'result)
<Fare> or is is ((a) 'result) ?
<water> damn it, i made a mistake earlier
<water> a^. ==> ((a) 'result)
09:50am
<water> otherwise it will place you in the ^ slot of a
<water> crazy, ain't it? :)
<morton> water: make any breakthroughs w/ your language lately?
<water> morton: yes
<Fare> a^. ?
<water> fare: what about it? yes, that's the correct expression
* Fare/#Tunes is confused
<water> heh
<water> ok let me explain
<Fare> ok, I'll read your specs with interest
<water> when you enter a symbol, it gets looked up in the current namespace and that reference is followed to a new one
<water> the next symbol you enter is looked up in *that* namespace
<water> "^" is a slot and a namespace in this sense
<Fare> "followed to a new one"?
<morton> hey, does anyone know what's going to happen to that "Brix" os? has 'air' lost interest or what?
<water> when you enter "." the current namespace evaluates and returns (based on the current namespace stack)
<Fare> "a slot and a namespace"?
<Fare> morton: maybe got a job?
<water> hm
<water> Fare: consider a graph
<water> Fare: namespaces are nodes, selectors are directed edges
<Fare> ok
<water> invoking the selector just pushes the evaluator along a path and modifies its internal namespace stack
<Fare> that's for selection. What about application?
<Fare> or can you only ever apply the current context?
<water> application backtracks along the path (that's the default anyway) which pops namespaces off the stack
-:- smklsmkl [sami@ppp96.dial-in.verkkotieto.fi] has joined #tunes
<water> yes you can only ever apply in the current context
<water> but shifting contexts and cloning them helps to make that more versatile
<water> since every use of a symbol besides "." changes the context, that's not so difficult :)
10:00am
<water> one thing we noticed as analyzing ( and ) was that they were (usually) very closely linked with uses of the "lookup" and "apply" phases of evaluation
<water> but it turned out that lookup was totally implicit to every slot-invocation
<morton> doers air ever come to #tunes anymore?
10:10am
<water> no
<morton> judging from the logs, #tunes seemed much livelier + hopeful a year ago
<morton> :-(
10:20am
<coreyr> morton: dont bring me down bruce.
* Fare/#Tunes thoght that the namespace stuff would be resolved statically
<Fare> (so there would be no context stack  at runtime)
<morton> coreyr: bruce?
<Fare> and that "apply" would be context-free
<Fare> "bruce?" ?
<morton> it's like (except for the static members) every few months an old group of OS hopefuls gets flushed out and a new group comes in and so on :-)
<coreyr> morton: isnt that how all projects work?
<morton> coreyr: i guess so
<morton> the logs are funny heh
<morton> <CodingHea> Yeas! :)  <CodingHea> How TUNES goes???
-:- SignOff morton: #TUNES (<>)
10:30am
<Fare> scifi channel? in France?
<ult> whaaa?
<Fare> wrong channel, sorry :)
10:40am
<water> hm
<water> Fare: the namespace stuff is static as long as you don't change it :)
<Fare> uh?
<water> partial-evaluation applies to this case, right?
<Fare> sure.
<water> there you have it, then
<Fare> I meant, it is usually a good idea to be able to distinguish normal applications from "macro" stuff that has special application semantics
<water> besides, note that whatever namespace you're at is just a compiled-away context stack
<water> such as?
<water> i'm not sure what you're referring to exactly, so i can't tell how one would address it in slate
<Fare> such as (a b) meaning that I apply a with parameter b, regardless of the value of a and b
<Fare> except if I syntactically know that a is a special form
<water> huh?
<water> you know how slate objects get values, right?
-:- smoke is now known as smoke`
-:- smoke` is now known as smoke
<water> e.g. "myObject x : 4 . ." requires that ":" have the numbers in it
<water> so that it can pass a clone of 4 to the "^" of "x"
<ult> slate is ugly 
<water> basically, you make ":" inherit from another node on the namespace graph that has access to those natural numbers (and whatever else you want)
<water> ult: got any suggestions for the syntax that work?
-:- I440r [mark4@purplecoder.com] has joined #tunes
<water> hey
* ult/#Tunes ponders
<water> eih: what do you think?
<ult> well, things like ^ and . aren't terribly intuitive, unfortunately.
<water> they are for people who use self and smalltalk
<ult> and why are you putting two periods after eachother?
<water> "." used to be ")"
10:50am
<water> but "(" was totally redundant
<ult>  One of the nice things about lisp is that it has a very simple syntax
<water> no kidding
<ult> It doesn't appear that slate does
<water> lol
<water> there's only one special operator in slate
<water> everything else is slot-names
<ult> so its just a "command stream"?
<water> kind of... it's a lookup-chain
<water> the "."s are applies
<water> the lookup-stream is also a path among the namespaces to find the desired object
<ult> ok, so its almost like forth?
<water> it's concatenative, yes
<water> beyond that, there's not a lot of resemblance
<ult> Well, that's what I meant.
<water> ok
<ult> I had thought you were trying to a tree-like syntax though.
<water> i was... but it turned out that the semantics don't need it
<water> like i said, the "("s turned out to be totally redundant
<water> i think this language isn't really best suited for text-representation
<water> probably some sort of visual metaphor would help in reading and writing code
<ult> what does myObject x refer to?
<water> it refers to the namespace accessed by "x"
<water> within the myObject namespace
<ult> hmm ok
* water/#tunes ponders on a related issue
<ult> so foo bar [some operation] .. baz [ some other operation ]
<ult> bar in foo namespace, some operation, baz in foo namespace, some other operation
<water> yep
<ult> ok, that makes a lot more sense than what I first thought it was.
<water> ok
11:00am
<water> it's just odd that in trying to strip down the syntax, i turned a language with explicit cons cells to one with implicit cons cells for syntax
<water> ult: just to know for the tutorial, what did you first think that the slate syntax meant?
11:10am
<ult> heh
<ult> I was showering
<ult> water: Well, the myObject x: 4 looked like you were cloning myObject into x, and sending x the message '4', but the extra period confoozed me.
<I440r> ug theres a disgusting thought hehe
<I440r> ult neked :P
<ult> i440r: The girlies don't think so...
<I440r> sure... sure...
<I440r> i believe you :P
<water> hm
11:30am
<water> ok thanks
<ult> is there unlimited namespace nesting in slate?
<water> yes of course
<ult> hmm
<ult> what are the implicit cons cells for then?
<water> (object message) pairs
<water> iow what "." operates on
<ult> hmm
-:- SignOff coreyr: #TUNES ([x]chat)
11:40am
-:- SignOff eihrul: #TUNES ([x]chat)
<ult> gotta love the xemacs source
<ult> function declared noreturn does return.
<Fare> ult: ouch. How come?
12:00pm
<ult> Fare: It's just lame code. Stuff illegal according to ANSI standard all over the place.
<ult> Fare: It doesn't look harmful, just sloppy.
<ult>  /* This highly dubious kludge ... shut up Jamie, I'm tired of your slagging. */
12:10pm
-:- morton [morton@AC90781F.ipt.aol.com] has joined #tunes
-:- SignOff morton: #TUNES (Leaving)
-:- hcf [nef@207-172-225-163.s163.tnt1.pld.me.dialup.rcn.com] has joined #tunes
-:- SignOff ult: #TUNES (Leaving)
<water> re
<hcf> who r u re'ing?
<water> u
<hcf> u dont seem to use re <who> so it gets confusing
<water> ic
<water> slate being concatenative in syntax is pretty wierd
<water> if i renamed "..", do you think it would help?
<water> it looks too much like two "."s
<hcf> concatenative means? spaceless?
<water> it means like forth
* Fare/#Tunes is back from 3053
<water> you place words side-by-side instead of using grouping operators
<water> such as the parentheses of lisp
12:40pm
<water> at least now the syntax really works
-:- water has changed the topic on channel #tunes to: TUNES, Free Reflective Computing System: http://www.tunes.org/ || Slate Programming Language
12:50pm
<water> hm
<hcf> water: hows the documenting going?
<water> working on it... there's a lot to change and fix
<water> any suggestions for ".."?
-:- NetSplit: zsoldos.openprojects.net split from hogan.openprojects.net [12:58pm]
-:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [zsoldos.openprojects.net]
01:00pm
<hcf> hmm
-:- Netjoined: zsoldos.openprojects.net hogan.openprojects.net
-:- smoke [smoke@15dyn238.delft.casema.net] has joined #tunes
<hcf> A X going into A's X then X .. to get back into A, right?
<water> yep
<hcf> can it just be .. w/o the X part?
<water> but "." does a similar thing
<water> oh the X part isn't needed
<water> "a x" puts you in x
<water> once you're there, typing ".." puts you back in a
01:10pm
<hcf> if ur in A, does plain X move u into X?
<water> yep
<water> that's the effect of concatenative syntax... every expression extends into the next
<Fare> water: is it X having a link to A, or is it the compilation context remembering the upper lexical scope?
<water> upper lexical scope?
<water> no, there's no scoping going on
<water> you can achieve the same thing by inheritance
<Fare> so X can be only pointed to by one upper context?
<water> yes... X is actually the link itself
<hcf> brb
<water> the only other way to get at it is by cloning
<water> unfortunately, cloning can get out of hand, in terms of resolving object identity in the way the programmer wants
<water> mostly because we have mutation going on in various spots
01:20pm
<hcf> water: given A X moves u into X, whats does X A do?
<water> here's the reason for the syntax:
<water> you always start in a namespace...
<hcf> X isnt accessable w/o going thru A n/m
<water> every name entered is looked up in the current namespace
<water> if the name is there, it gets followed into the namespace it refers to
<water> then the next name gets parsed, etc
<water> does that help explain?
* water/#tunes needs to make some diagrams of the namespace system idea
<hcf> i /think/ i got it
<water> namespaces are nodes of a graph, names are directed edges between the nodes
<water> the lookup process makes a path to the node desired from the current one
<water> evaluation includes a back-tracking process (usually)
<hcf> how do normal graphs handle the .. issue?
<hcf> syntax wise
<water> for now, what i'm considering is that every expression maintains the path of lookup via a stack
<water> ".." involves putting another name onto the stack like the usual names do
<hcf> $CWD and $OLDCWD? ;)
<water> it just happens to backtrack without the evaluator doing it
01:30pm
<water> uh
<water> something like that, except ".." is in userland
<water> in other words, not part of the evaluator (or meta-object system)
<Fare> the pliant guys will publish 40-bit restricted crypto on their site.
<water> good for them
<Fare> of course, just changing one line will give you as much crypto as you want
<water> even better
<Fare> and since pliant is mostly source, no static binary, well, ...
<hcf> water: how does one get to the root?
<water> either by entering "." multiple times, or by ".." multiple times
<water> but they do subtly different things of course
<hcf> but thats relative aint it?
-:- eihrul [lee@usr5-ppp67.lvdi.net] has joined #tunes
<water> relative to what?
<hcf> to ur current space
<water> right
<water> you can always make new links by cloning root
<water> and placing the clone in the current namespace
<water> but things like that have to be done carefully
<hcf> in one's homedir, `cd ../..` goes to root (usually), do u have a equiv to `cd /`?
<water> to get anywhere?
<hcf> to get to the root w/o using the relativeness of ..
<water> it can be done, but i'd rather keep it in meta-level, not userland
<water> having access to root everywhere for now is not a good idea
<water> until i develop the meta-system to handle cross-module security
<water> because userland security right now is entirely based on each message-send
<water> which means security is based on single links only
<hcf> got an equiv to `cd -`?
<water> what does that do?
<Fare> access to root everywhere is evil.
<hcf> goes to previous working dir
<Fare> although dynamically bound stuff is good
<water> yes i have that
<water> "."
<hcf> cd ~/1/2 ; cd ~/3/4 ; cd - goes to ~/1/2
<Fare> well, actually, access to root can be a security feature for some components
<water> but if you don't want evaluations, you have to back out of your work first
<water> Fare: i know, i'm dealing with it
<Fare> like, a component that demands access to the "real" console (SAK)
01:40pm
<water> Fare: besides, this is the root *namespace* not root *user*
<Fare> water: same difference
<water> that's what you think
<Fare> or rather, users are special cases of namespaces.
<water> eihrul: you there?
<water> eihrul: if we replace ".." with some other symbol, we can allow multiple "."s to be typed as a single symbol
<water> which would ease coding by a bit
<hcf> how about / instead of .. ?
<water> hm
<hcf> or << like rewind ;)
<water> any single symbol would be good.... picking one that's intuitive would be the best
<water> maybe
* hcf/#tunes is liking <<
<water> the easier it is to read the better
<water> yes it makes more sense than anything i've considered so far
01:50pm
<water> does anyone have other ideas?
<water> not that i don't like <<, but i want to make sure we pick something really good
<hcf> << seemingly denotes more direction than .. does
<water> yes
<water> and it's very visible
<water> damn, almost everything related to syntax has to get scrapped
<water> bbiaf
<hcf> if theres no need to refer directly to the current space, then << can easily be <
<hcf> making multiple <'s more readable
02:00pm
-:- coreyr [coreyr@net255ip95.parklink.com] has joined #tunes
<water> yeah
02:20pm
<hcf> what now?
<water> hm?
<coreyr> hcf: 2nd star to the right
<hcf> huh?
<coreyr> nm.
<water> i'm still re-arranging the tutorial
<water> anyway, it took about 8 years to make smalltalk a mature language, during which time its syntax and libraries changed quite a bit, so don't complain about development speed of Tunes HLL crap
02:30pm
<hcf> complaining i'm not
02:40pm
-:- SignOff smklsmkl: #TUNES (back tomorrow)
<hcf> abi: omega is an open source implementation of M-Technology (MUMPS) at http://home.clara.net/finch/
<abi> ...but omega is a self-like language with static typing at http://infosoft.soft.uni-linz.ac.at/Info/Omega.html...
<hcf> abi: omega is also an open source implementation of M-Technology (MUMPS) at http://home.clara.net/finch/
<abi> okay, hcf.
<water> what is that, exactly?
<hcf> dunno, just saw it announced on fm
<hcf> its probly crap
03:30pm
<hcf> bbl
-:- SignOff hcf: #TUNES (Leaving)
-:- SignOff water: #TUNES (The Tao went that-a-way!)
03:40pm
-:- tcn [Tom@cci-209150250063.clarityconnect.net] has joined #tunes
-:- water [water@tnt-10-136.tscnet.net] has joined #tunes
<water> re all
<water> brb (again)
04:10pm
-:- sig_9_4u [amiranda@208.35.215.250] has joined #tunes
-:- sig_9_4u [amiranda@208.35.215.250] has left #tunes []
-:- tcn [Tom@cci-209150250063.clarityconnect.net] has left #tunes []
-:- morton [morton@AC94C256.ipt.aol.com] has joined #TuNeS
<water> hi morton
<morton> hi mr. rice
<water> :P
<morton> hmm... i'd better get back to the AOL chatrooms
<water> please don't use that name
<morton> private room 'theology' awaits heh
<water> have fun
<morton> water: want to discuss philosphical matters?
<water> sure. but not on this channel
<water> care to make a new one?
04:30pm
<morton> goin to be away for next 30 min.
<water> ok
04:40pm
<morton> back.
<morton> water: /join #philosophes
<morton> if u want to
05:00pm
-:- hcf [nef@207-172-225-204.s204.tnt1.pld.me.dialup.rcn.com] has joined #tunes
<morton> what does "hcf" stand for?
<hcf> halt and catch fire, highest common factor, and a bunch of other things
<morton> halt and catch fire??
<hcf> abi: foldoc for hcf
<abi> hcf may be sought in foldoc at http://wombat.doc.ic.ac.uk/foldoc/foldoc.cgi?query=hcf
<water> bbl
-:- SignOff water: #TUNES (The Tao went that-a-way!)
05:10pm
-:- water [water@tnt-10-136.tscnet.net] has joined #tunes
-:- billh [billh@cx739861-a.dt1.sdca.home.com] has joined #tunes
<water> hi bill
<water> forget it, i need to get out of the house
<water> cya
-:- SignOff water: #TUNES (The Tao went that-a-way!)
-:- lmaxson [lmaxson@adsl-63-194-24-207.dsl.lsan03.pacbell.net] has joined #tunes
<morton> hi dude
<morton> lmaxson: changed your ideas about SL/I?
05:30pm
-:- ult [ult@user-38lcdpn.dialup.mindspring.com] has joined #Tunes
<morton> so you're not going to talk at all now?
<ult> hahahahahahahaha
<ult> morton your a lamer.
<morton> wtf do you keep saying that?
<ult> Why not?
<hcf> ult: someone had to fill the void left by atg
<ult> hcf :)
<lmaxson> morton, no, just doing what I was asked to do: submit a description of the language.
<lmaxson> Sorry for the slow response.  I was on another channel.
<lmaxson> I have been reading the material on slate.
<lmaxson> I think SL/I will hold up very well indeed.
05:40pm
<lmaxson> If you prefer, I will return to passive mode.
-:- lmaxson [lmaxson@adsl-63-194-24-207.dsl.lsan03.pacbell.net] has left #tunes []
05:50pm
-:- SignOff ult: #TUNES (Read error to ult[user-38lcdpn.dialup.mindspring.com]: Connection reset by peer)
-:- SignOff morton: #TUNES (.)
-:- gus [user5952@usr39-dialup22.mix1.WillowSprings.cw.net] has joined #tunes
-:- gus [user5952@usr39-dialup22.mix1.WillowSprings.cw.net] has left #tunes []
-:- nate37 [nate@cx83983-d.irvn1.occa.home.com] has joined #tunes
-:- ult [ult@user-38lcc8m.dialup.mindspring.com] has joined #Tunes
<I440r> ult look in ftp.purplecoder.com/pub/isforth
06:40pm
-:- SignOff smoke: #TUNES (zzzee you tomorow)
-:- SignOff billh: #TUNES (BitchX-1.0c16 by panasync)
-:- SignOff hcf: #TUNES (Leaving)
-:- hcf [nef@207-172-225-4.s4.tnt1.pld.me.dialup.rcn.com] has joined #tunes
-:- SignOff hcf: #TUNES (Leaving)
-:- SignOff nate37: #TUNES (changing servers)
-:- lar1 [larman@adsl-63-204-134-1.dsl.snfc21.pacbell.net] has joined #tunes
-:- nate37 [nate@cx83983-d.irvn1.occa.home.com] has joined #tunes
-:- SignOff lar1: #TUNES (Leaving)
-:- SignOff nate37: #TUNES (Read error: 3.1415926535 (Excessive pi))
[msg(TUNES)] newlog 2000.0610
IRC log ended Sat Jun 10 00:00:01 2000