IRC log started Mon May 29 00:00:01 2000
[msg(TUNES)] permlog 2000.0529
-:- water [water@tnt-10-59.tscnet.net] has joined #tunes
<water> eih: i figured out what you meant about putting the structural modifiers in the MO
<water> it definitely makes the layering from base to meta level more modular and intuitive
<coreyr> water: tril was here for 18 minutes.
<water> oh?
<eihrul> water: yep, that's why i suggested it oh-so-long-ago
<water> heh
01:00am
<water> maybe i should go see tril sometime again
-:- SignOff eihrul: #TUNES ([x]chat)
<hcf> bbl
-:- SignOff hcf: #TUNES (Leaving)
01:10am
<water> note: webdb discussion is in #{}
01:50am
-:- SignOff water: #TUNES (The Tao went that-a-way!)
-:- coreyr [coreyr@net255ip95.parklink.com] has left #tunes []
-:- SignOff Fare: #TUNES (Leaving)
-:- Kyle_L [kyle@cr168790-a.nmkt1.on.wave.home.com] has joined #tunes
-:- SignOff Kyle_L: #TUNES (Leaving)
-:- Kyle_L [kyle@cr168790-a.nmkt1.on.wave.home.com] has joined #tunes
-:- eihrul [lee@usr5-ppp201.lvdi.net] has joined #tunes
-:- Netjoined: adams.openprojects.net sterling.openprojects.net
-:- I440r [mark4@purplecoder.com] has joined #tunes
-:- coreyr [coreyr@net255ip95.parklink.com] has joined #tunes
-:- hcf [nef@207-172-225-104.s104.tnt1.pld.me.dialup.rcn.com] has joined #tunes
-:- water_ [water@tnt-10-9.tscnet.net] has joined #tunes
<water_> hey all
-:- water_ is now known as water
<coreyr> hey
10:30am
-:- Fare [rideaufr@esmeralda.enst.fr] has joined #tunes
<water> Fare
<Fare> gakuk
<water> hey
<water> there's some news as for Slate stuff
<water> first of all, a formulation of the namespace access system as a graph
* eihrul/#tunes kicks his Indigo2 paperweight.
<water> heh SGI not working out for you?
<eihrul> i can't get a damned compiler cross-compiled for it :/
<water> Fare: also, the object-metaobject distinction is getting refactored
<water> blah
<Fare> able
<Fare> on the slate page?
<water> anyway, there is still the issue of /'s semantics as well as finalizing the slate object model
<water> not yet
<water> corey is working out a database system Right Now to re-organize the current docs
<Fare> good
<water> i'll port everything of tunes that i can myself, along with anyone who will help
<water> at any rate, i'm going for allowing cyclic graphs in slate for namespace access
<water> s/going for//
<water> although implementational shortcuts are certainly favored ;)
<water> so essentially, formulating everything in graph-rewrite terms is my current direction for the spec
<Fare> what remains of / ?
<Fare> water: good
<water> it's really bothersome
<water> i'm even considering making result applications explicit
<water> so that / would be unnecessary
11:40am
<water> unfortunately, this makes 'result' a permanent part of the grammar afaik, which is pretty non-uniform
<water> an alternative would be to make ( and ) usage a little different semantically
<water> btw, someone implemented MetaObjects in smalltalk last year some time
<water> i'm looking for the implementation online
<Fare> result application?
<water> 'myFunction result'
<water> like forcing a lazy function to yield its value
<Fare> aren't you overusing the object/message concept?
<water> heh
<water> it's not like that
<Fare> why not (force (delay ...))
<water> gr
<water> because result is a slot of myFunction
<water> it's the actual value, not a comman
<water> command
<Fare> and accessing value implicitly forces it?
<water> that's the idea
<water> "/" delays it
<Fare> hum. ML has explicitly lazy constructors.
<water> so? they're constructors
<Fare> ain't it the same here?
11:50am
<water> it depends... do ML's "constructors" fit into a uniform system for the whole language?
<Fare> hum. Anyone knows about majordomo moderation?
<coreyr> moderation, yeah just dont use it as much
12:00pm
-:- Kyle_L [kyle@cr168790-a.nmkt1.on.wave.home.com] has joined #tunes
<water> hey kyle
-:- NataLieR [asdasd@node-64-248-41-156.dslspeed.zyan.com] has joined #tunes
<water> hey natalier
<Kyle_L> hi
-:- NataLieR [asdasd@node-64-248-41-156.dslspeed.zyan.com] has left #tunes []
<Kyle_L> I was asking about strings in slate.  Let me see if I can refresh my memory...
<water> looking up the relevant logs?
<Kyle_L> yes.
<Kyle_L> found it
<water> what day and what line?
<Kyle_L> may 26, 1:10pm
12:10pm
<water> thx
<Kyle_L> and up a couple of lines
<Kyle_L> <water> do we really need strings in slate?
<water> ya i'm there
<water> what about it?
<Kyle_L> I can not think of any use for strings, except for communication with humans.
<water> god damn it
<water> haven't you read the ./Interfaces stuff on tunes.org?
<Kyle_L> The number of lines you have written; are thay all the same thought?
<water> roughly yes
<water> in fact, it's definitely from the same thread of thought
<Kyle_L> Let me see if I read interfaces...
<water> i'm wondering if you mean low-level ascii strings or the human notion of text.
<Kyle_L> I mean the human notion.
<water> well then don't use the term "strings" for that
<water> because that term doesn't ever mean anything but the usual notion that a programmer deals with
<Kyle_L> ok.  
<water> infact, there are other programming notions of text, like "ropes" and "threads" and "cords"
<water> not to mention parse trees :)
<water> that's why i objected to the sentence
<Kyle_L> MMMmmm.  But I still do not see what you meant by  "do we really need strings in slate?"
<water> but yes, "text" is needed because reading and writing is an important part of HCI
<water> *strings*
<water> in every programming language, "Hello" usually means exactly one thing
<water> an ASCII string
* Fare/#tunes is back from 61453
<Kyle_L> OK.  Then when you question if strings are needed, then you are asking if they are only needed to represent human text?
<water> some have subtle differences that must be taken into account (for low-level purposes) when you port the string to other programs
<water> sigh
<Kyle_L> OK.  I can see that.
<water> i'm asking if they must be part of the language itself
<water> as opposed to being a kind of loadable module
<Fare> even if not, you need to somehow name objects
<water> Fare: shut up
<Fare> nameing might be purely delegated to the interface
<Fare> I guess that's mostly a bootstrapping issue
<water> yes, now shush
<Fare> kiss my ...
12:20pm
<water> anyway, what i was considering was how that could be achieved in the slate system
<water> since quoting relies heavily on it
<Kyle_L> water: if you ask if they should be part of the language, do you meamn the interface with humans, or the machine's view of the programmers intent?
<water> and lisp's notion of quoting is lame
<hcf> Fare: dont bother getting involved w/ the discussion since ur just gonna go idle w/o reason in the middle of it anyway
<water> Kyle: both
<water> Kyle: they are two aspects of the same thing
<water> hcf: he's already done so ;)
<Kyle_L> I would imagine the former needs strings, but is also just an aspect of the IDE, whereas the latter does not need strings, and is more fundamental
<Fare> water: quoting is good. The details might be improved, but even then, lisp quoting was a breakthrough in programming at the time
-:- tcn [Tom@cci-209150250034.clarityconnect.net] has joined #tunes
<water> (it's hysterical that there's no point in even discussing it with him because he just leaves while you point it out)
<Fare> t c n !
<water> hey tom
<water> "at the time" is 40 years ago
<Fare> so what?
<water> so get a new perspective, please
<Fare> the alphabet was a breakthrough at the time it was invented
<water> for Tunes' sake
<water> exactly and it's irrelevant now
<Kyle_L> water: Your next point on quoting had only to do with strings?  Or with functions too?  Or is that a dumb question?
<water> Kyle: hm
<water> Kyle: er... which point? today's or that from the 26th?
<Kyle_L> water: from the 26th
<water> which statement? (just give the first few words)
<Kyle_L> <water> i keep playing with the notion of quoting, and it's pretty troublesome 
<water> it had to do with functions, not just strings
<Kyle_L> Does quoting exist because functions, and strings both exist in two mutually disjoint idea-spaces?
<Fare> uh?
<water> it's because you have to designate which function you want in order to invoke it
<Fare> what has quoting to do with functions and strings?
<Kyle_L> I mean that a function exists as its definition, and as action
<Fare> or do you equate string with source and functions with object?
<water> Fare: s/strings/symbols/
<Kyle_L> A string exists as a series of characters, and as an idea it represents.
<water> Fare: i'm talking in Kyle's terms to help him relate
<water> kyle: yes, but a string is also an abstract object
<water> whose identity is unique, and that is what makes it suitable for a UI
<Kyle_L> Absract object as a string, or what it represents?
<water> otherwise, when i call "foo(bar)", i don't know what foo is
12:30pm
<water> no abstract object being a view of something at the level where you don't care about it's features
<water> s/it's/its/
<water> i don't care whether "foo" is a string or not when i use it to call the function it designates, iow
<water> "foo" could be an object on a graphical screen
<water> or it could be something inferred by an interface that uses natural language and anaphora (ie "this" or "those")
<Kyle_L> I will think for a moment...
<water> tunes mlist has had many pointless discussion about unique identifiers because of confusion about this
<water> s/discussion/discussions/
<water> damn, i need to leave soon
-:- tcn [Tom@cci-209150250034.clarityconnect.net] has left #tunes []
<water> heh
<Kyle_L> Each of those other representations of a function act the same.  Strings, appear not-special.
<water> exactly
<water> but in programming languages they are usually special
<Kyle_L> OK.  
<water> heh
<Kyle_L> The quoting framwork should not use strings?
<water> it shouldn't *depend* on strings
<Kyle_L> Is quoting only an issue with the IDE with the human?
<water> the problem is that when you have this lack of dependency, you have this issue where multiple interfaces might not be able to access *every* identifier in other interfaces
<water> no, because the "IDE with a human" is what the language uses when doing reflective things
<Fare> water: funny, basile said something very like what you said, yesterday, when a-talking about symbols
12:40pm
<water> ok
<Kyle_L> So, reflection is quoting?
<water> no, but it relies on quoting
<Kyle_L> ok.
<Fare> no but it's more like kwote-ing
<water> thx for the pointless statement, Fare
<Fare> not pointless
<Kyle_L> Quoting is the act of refering to a first order object in zeroth order terms? 
<water> ok what was your current obscure point?
<Fare> see J Goubault's eval-quote calculus
<water> Fare: no, asshole, explain
<water> kyle: i suppose, but that doesn't make much sense to me
<water> Fare: or give a URL
<Fare> kwote is a function, not a macro. (kwote 'foo) => 'foo
<Fare> http://hypatia.dcs.qmw.ac.uk/data/G/GoubaultJ/S4/
<water> "access forbidden"
<Kyle_L> ...but THE zeroth order terms are purly abstract and do not contain meta info.
<water> kyle: that's vague bullshit
<Kyle_L> I will give an example
<water> no
<Kyle_L> yes
<Kyle_L> :)
<water> i have to leave soon
<Kyle_L> ok.
<Fare> not to me. Maybe you lack proper clearance.
<Kyle_L> I will save you
<water> Fare: hehehehehe
<water> it's been real, everyone
-:- SignOff water: #TUNES (The Tao went that-a-way!)
<Kyle_L> Fare: did I have some nomenclature problems?
12:50pm
<Fare> nomenclature?
<Kyle_L> was I using terms that were not standard?
<Fare> water says so.
<Kyle_L> I am asking you.
<Fare> I haven't seen the beginning of the conversation
<Kyle_L> Oh, well.
<Fare> to me, quoting is a syntactical function from source code to objects representing source code.
<Kyle_L> Is there a name for the more-general "quoting" I was refering to?
<Kyle_L> I was using quoting as an operation that takes zeroth order description of an object to identify the object.
<Kyle_L> This could be as simple as a string, or ID.  Or could be mor complex by using a few object slots to find a unique object.
<Kyle_L> Maybe FirstName="Kyle", LastName="Lahankoski" is a biquote to identify me.
<Kyle_L> Any one of the above are insufficient, but as a pair distingush me from every other human on earth.
01:00pm
<Fare> I'm not sure what you mean. An identifier?
<Kyle_L> identifier?  what sentence?
<Fare> In lisp, (quote ...) is a special form that returns objects representing the source code that follows.
<Kyle_L> Yes, but lisp is only using string quoting, there can be so many other ways to identify a piece of source code.
<Fare> an identifier (symbol, in lisp) is some sequence of character in the source that identifies a binding
<Kyle_L> ...(but maybe not in lisp)
<Fare> such as int a; in C declaring an integer variable bound to the identifier "a"
<Fare> string quoting?
<Kyle_L> yes, using context (scope), and the string together we identify the variable.
<Kyle_L> It is my mistake to be talking about lisp without knowing lisp.
<Kyle_L> What I mean above is that the "..." in (quote ...) is text, yes?
<Kyle_L> It must be to be visible on a character terminal.
<Kyle_L> I am trying to say that the attributes of an object can be used to identify it uniquly.  This act of identification is called quoting?
<Kyle_L> Fare?
<coreyr> heh
<Fare> yes, it's source text, such as in (quote (lambda (foo) stuff)
<Fare> )
<Kyle_L> So, is my idea clear?  Does it have a name?  I was calling it quoting.
<Fare> for instance (quote (+ 1 2)) returns the same thing as (list '+ 1 2), which prints as (+ 1 2)
<Fare> no, I wouldn't call this identification quoting
<Kyle_L> Does it have a name?
<Fare> if you're considering an identifier, then you identify an object (or maybe "reference" it, at a meta-level)
01:10pm
<Kyle_L> OK, quoting is a special case of object identification?
<Fare> Are you trying to say that a (source) object identifies the object it evaluates to?
<Kyle_L> "evaluate" seems to be a bad word.  There may exist an object that refers to another by using only the attributes of the latter.
-:- SignOff hcf: #TUNES (Ping timeout for hcf[207-172-225-104.s104.tnt1.pld.me.dialup.rcn.com])
<Kyle_L> This is like pointing, 
<Kyle_L> Oh!  OK, maybe the word I am looking for is "referencing"
<Kyle_L> Oh! Then quoting, and evauating are two methods of "referencing" objects.
<Fare> hum
<Fare> are you using the lispy "quoting", or your thing?
<Fare> Maybe you mean that by listing the fields and values of a (pure) object, you can identify the object?
-:- hcf [nef@207-172-225-86.s86.tnt1.pld.me.dialup.rcn.com] has joined #tunes
<Kyle_L> Quoting is the act of using a string to ID a specific object
<Fare> I'd say referencing, then
<Kyle_L> Evaluating, I beleive, is the lispy "quoting".  Again I am not sure.
<Fare> referencing is using any abstract type with uniqueness property to identify objects by association
01:20pm
<Kyle_L> excellent!  Thanks Fare.
<Fare> you're welcome
01:30pm
-:- SignOff hcf: #TUNES (Leaving)
-:- SignOff Kyle_L: #TUNES (Leaving)
-:- xanax [bigbill@loves.gettin.head.from.your-mother.net] has joined #tunes
-:- xanax [bigbill@loves.gettin.head.from.your-mother.net] has left #tunes []
-:- hcf [nef@207-172-225-224.s224.tnt1.pld.me.dialup.rcn.com] has joined #tunes
-:- Kyle_L [kyle@cr168790-a.nmkt1.on.wave.home.com] has joined #tunes
-:- SignOff Kyle_L: #TUNES (Leaving)
-:- SignOff I440r: #TUNES ([x]chat)
-:- SignOff eihrul: #TUNES (Ping timeout for eihrul[usr5-ppp201.lvdi.net])
-:- eihrul [lee@usr5-ppp68.lvdi.net] has joined #tunes
-:- I440r [mark4@purplecoder.com] has joined #tunes
-:- NataLieR [asdasd@node-64-248-41-156.dslspeed.zyan.com] has joined #tunes
<coreyr> re all.
-:- NataLieR [asdasd@node-64-248-41-156.dslspeed.zyan.com] has left #tunes []
07:50pm
-:- SignOff hcf: #TUNES (Leaving)
-:- hcf [nef@207-172-225-239.s239.tnt1.pld.me.dialup.rcn.com] has joined #tunes
-:- washort [washort@d145.narrowgate.net] has joined #tunes
<hcf> hello washort
<washort> hello.
09:00pm
-:- lispbliss [lispbliss@cont022p08.ont.micron.net] has joined #tunes
<I440r> washort hi :)
<hcf> the #lisp'ers are coming, the #lisp'ers are coming
* lispbliss/#tunes swears he didn't know washort was in here :)
-:- Peidran [peidran@qdslppp8.sttl.uswest.net] has joined #tunes
<Peidran> hoy
<lispbliss> hola
<Peidran> in case someone who messes with the webpages is here, the glossary Distributed OS entry points to a non-existent 'OS Review' page
09:20pm
<hcf> Peidran: yeah the site is full of bad links
<hcf> we're working on a new site
<coreyr> these things are being addressed.
<hcf> Peidran: do u mind if i ask u a question via /msg?
<Peidran> hcf, ok, and go for it
<Peidran> sorry, I'm multitasking :)
09:30pm
<Peidran> I like the glossary entry for 'man' :)
<Peidran> is TUNES anything more than theory right now?
<coreyr> i dont know what your talking about
<washort> coreyr: is there code?
<Peidran> me either
* Peidran/#tunes points at washort 
09:40pm
<lispbliss> I don't think there is substantial code
<coreyr> look at retro and slate
09:50pm
<Peidran> good god, FORTH
<washort> Forth is.... interesting :)
<Peidran> no kidding
<Peidran> my firmware uses forth
<washort> pity that most Forth users are more interested in writing Forth implementations rather than useful stuff. :)
<Peidran> hehe
<Peidran> sounds familiar
<washort> from where?
<Peidran> when I look at the tunes page, I don't understand what ties it all together
<Peidran> washort, oh, just as a general CS type of thing
<Peidran> spending more time on theoretical things than actually making something worthwhile
* Peidran/#tunes shrugs
<I440r> peid! what firmware ?
<washort> well, the problem is, writing a forth implementation is the easiest thing to do in forth at this point :)
<Peidran> i440r, powermac
<I440r> aha
<Peidran> washort, ah :)
<Peidran> doesn't anyone like smalltalk any more?
<I440r> washort. im writing a linux x86 forth in ASM
<Peidran> i440r, Open Firmware
<I440r> and its easy :P
<I440r> cool
<washort> I440r: in assembler? that's silly.
<I440r> nope
<I440r> just do macros to create headers
<I440r> and : defs
<I440r> and it keeps head and code space seperated too
-:- SignOff lispbliss: #TUNES (ircII EPIC4pre2.508 -- Accept no limitations)
10:00pm
-:- SignOff Peidran: #TUNES (Ping timeout for Peidran[qdslppp8.sttl.uswest.net])
<I440r> washort u scared em off :)
10:10pm
-:- Peidran [peidran@qdslppp8.sttl.uswest.net] has joined #tunes
-:- SignOff Peidran: #TUNES (changing servers)
-:- Peidran [peidran@qdslppp8.sttl.uswest.net] has joined #tunes
<Peidran> mmph
10:20pm
-:- SignOff eihrul: #TUNES ([x]chat)
-:- SignOff Peidran: #TUNES (sleep)
[msg(TUNES)] newlog 2000.0530
IRC log ended Tue May 30 00:00:02 2000