00:23:49 witten joined #tunes
00:28:03 <water> hey witten
00:28:48 <witten> hey
00:28:56 <witten> what's up?
00:29:26 <water> the usual
00:29:50 <water> except adding on to that, trying to get a cable modem set up and buying an ibook
00:30:45 <witten> ooh
00:31:04 <witten> you gonna run tunes on it? :)
00:31:15 <water> heh
00:31:25 <water> slate at least
00:31:30 <water> also squeak and self
00:31:48 <water> i've needed a new laptop
00:32:05 <witten> I think I need one too
00:32:15 <witten> I have to give mine back to work
01:09:13 [QUIT] witten quit: bye
02:44:46 [QUIT] nate37 quit: Ping timeout for nate37[cx83983-d.irvn1.occa.home.com]
02:47:37 nate37 joined #tunes
03:23:29 [QUIT] MysticOne quit: Ping timeout for MysticOne[porpoise33.panama.gulf.net]
03:28:10 [QUIT] Ghyll quit: Ping timeout for Ghyll[janus.prosalg.no]
03:46:34 [QUIT] smkl quit: Ping timeout for smkl[glubimox.yok.utu.fi]
04:03:08 smkl joined #tunes
04:28:52 TheDebugger joined #tunes
04:50:21 [QUIT] ult quit: Excess Flood
04:50:24 ult joined #tunes
04:50:45 [QUIT] ult quit: Excess Flood
04:50:45 ult joined #tunes
04:59:07 _101 joined #tunes
04:59:23 <_101> hi
05:04:06 [QUIT] TheDebugger quit: KVirc 2.0.0 'Phoenix'
05:30:00 [QUIT] water quit: Ping timeout for water[tnt-9-59.tscnet.net]
05:33:13 water joined #tunes
06:15:59 hcf joined #tunes
06:19:26 abi joined #tunes
07:05:37 eihrul joined #tunes
07:07:01 [QUIT] hcf quit: Ping timeout for hcf[207-172-225-188.s188.tnt1.pld.me.dialup.rcn.com]
07:15:18 hcf joined #tunes
07:18:52 [QUIT] eihrul quit: Leaving
07:42:08 [QUIT] water quit: The Tao went that-a-way!
08:02:37 _101 left #tunes
08:35:56 coreyr joined #tunes
08:41:56 [QUIT] gREMLiNs quit: Ping timeout for gREMLiNs[195.252.64.44]
08:49:38 [QUIT] JALH quit: LAG!
08:59:47 arbit joined #tunes
09:40:59 [NICK] coreyr changed nick to: cor[class]
09:45:08 eihrul joined #tunes
10:21:54 [QUIT] eihrul quit: Leaving
10:28:42 arbit left #tunes
10:38:38 arbit joined #tunes
10:50:02 [QUIT] Sk2 quit: bfn
11:16:12 [QUIT] hcf quit: Leaving
11:33:43 eihrul joined #tunes
11:33:46 eihrul left #tunes
11:34:03 eihrul joined #tunes
12:13:49 [NICK] ult changed nick to: ult[home]
12:15:21 [QUIT] eihrul quit: Leaving
12:15:48 arbit left #tunes
12:29:36 hcf joined #tunes
12:49:02 <hcf> abi: bullant is a PL, related to eiffel? at www.bullant.com
13:13:19 JALH joined #tunes
13:13:26 <JALH> humm
13:13:35 <JALH> has anyone coded any code for tunes yet?
13:31:07 [QUIT] morton quit: changing servers
13:51:49 eihrul joined #tunes
14:06:27 [QUIT] MysticOne quit: I'm outta here!  Catch everyone later!
14:12:53 nojityi joined #tunes
14:18:20 water joined #tunes
14:18:28 <water> hey all
14:18:55 <water> hi nojityi
14:19:10 <nojityi> hay
14:19:37 <water> have you looked out our (motley) website?
14:19:49 <nojityi> yah.. this is 'arbit'
14:19:55 <water> oh ok
14:20:10 <water> couldn't tell for sure
14:20:45 <nojityi> im the one with the unwieldy hostname
14:23:26 water checks logs
14:23:56 <water> wow. nothing
14:24:08 <water> i never seem to run into this JALH fellow
14:24:14 <nojityi> jalh?
14:24:26 <water> look at the logs, arbit
14:24:33 <nojityi> ermm where are they
14:24:46 <water> abi tell nojityi about #tunes logs
14:46:49 smkl joined #tunes
14:46:53 ult[home] joined #tunes
14:47:46 cor[class] joined #tunes
14:52:42 eihrul joined #tunes
14:52:51 <water> re eih
14:52:56 <eihrul> re
14:54:24 <water> eih: this is my last weekend for a few... and since i can't get you to use email, i would like it if you would try this time
14:54:34 JALH joined #tunes
14:54:40 <water> wow
14:54:43 <water> hi jalh
14:54:44 <JALH> hey!
14:54:50 JALH dances around
14:54:56 <water> heh
14:55:03 <water> you seem to miss everyone here
14:55:05 <JALH> WOWHOO!
14:55:05 <JALH> )
14:55:11 <water> (i read the logs)
14:55:14 JALH got ZealOS to boot
14:55:32 <water> oh ok
14:55:52 <eihrul> what's "ZealOS"?
14:56:34 <eihrul> ugh
14:56:41 <eihrul> okay, why should we care about it, then?
14:56:57 <JALH> duno
14:56:59 <eihrul> what good is an operating system that's like every other one, but... in assembly?
14:57:01 JALH is happy
14:57:13 <JALH> dunno?
14:57:27 <water> JALH, no offense, but we have different goals
14:57:37 <water> unless you want to discuss tunes?
14:58:03 <JALH> ya
14:58:16 <JALH> has any code been written for 'tunes' yet?
14:58:28 <water> yes but it doesn't meet the specs at all
14:58:36 <water> we are still researching
14:58:40 <JALH> bahh
14:58:53 <JALH> been doin that for... uhh 2 years iirc
14:58:54 <water> jalh: i doubt you could code tunes in 50 years
14:58:54 JALH left #tunes
14:59:05 <water> asshole
14:59:12 <water> oh well
14:59:32 <water> eih: how can i help you with slate.lisp or pidgin?
15:00:23 <water> abi tell JALH about squeak
15:00:42 <water> abi tell JALH about squeakos
15:00:54 <water> abi tell JALH about squeaknos
15:01:08 <water> there, that'll get his attention
15:02:52 <water> yeah, i got his attention
15:03:14 <water> pointed out why tunes is not something to criticize by indicating how different squeak is
15:03:58 <eihrul> water: the whole object framework from the ground all the way to the user? :)
15:04:15 <water> heh ok first we have the object '1' :)
15:04:24 <water> then '2' :)
15:04:37 <water> hum
15:04:58 <water> is there any way you can break up that question into some more specific ones?
15:05:23 <eihrul> yes and no :)
15:05:28 <water> actually, just make it work and make the source code malleable
15:05:31 <water> actually, just make it work and make the source code malleable
15:05:32 <water> actually, just make it work and make the source code malleable
15:05:45 <water> does that help?
15:06:28 <eihrul> is it just me, or am i seeing triples?
15:06:44 <water> that's my cheap attempt at emphasis
15:06:58 <water> yeah ok maybe i should use markup tags
15:07:13 [QUIT] JALH quit: brb, lag
15:07:37 <water> but does it help?
15:08:25 <eihrul> not really :)
15:08:31 <water> damn it
15:09:14 <water> well, i'll work on the docs, but it'll be a while before the docs answer enough for you
15:09:31 <water> docs are really not effective for specifying code *before* it's written
15:09:53 <water> have you ever heard the phrase "it's easier to ask forgiveness than permission"?
15:10:06 <eihrul> yep
15:10:26 <eihrul> docs are relatively good for *planning* it though
15:10:33 <water> so do whatever you want within the loose boundaries of "it's still slate"
15:11:03 <water> yeah but we have no compelling evidence in favor of one design decision or another in many cases
15:11:09 <water> NOT YET, anyway
15:11:23 <water> which is what the "malleable" part is for
15:11:49 <eihrul> well, of course, because design decisions are arbitrary
15:12:17 <water> well, i mean things like "what namespace structure to use for the system-wide components"
15:12:38 <water> or how to represent MOs, which isn't even an issue yet
15:13:08 [QUIT] JALH quit: brb, restarting to test ZealOS
15:13:09 <water> since all you have to do is enumerate the MOs' functions that are absolutely required
15:13:32 <water> e.g. the one MO per the natural numbers
15:13:58 <water> we can port the objects to more clear MOs (probably wrapping the originals) later
15:16:22 <water> grok?
15:16:50 <eihrul> sure
15:16:59 <eihrul> in the end, its just a free-for-all on my part :P
15:17:19 <water> yeah well the whole point is for me to crack at your code myself
15:17:55 [QUIT] hcf quit: Ping timeout for hcf[207-172-225-15.s15.tnt1.pld.me.dialup.rcn.com]
15:18:46 <water> which i will do with my trusty new notebook computer =)
15:18:49 hcf joined #tunes
15:19:33 <water> i have been sorely missing the ability to work on code while not at home (which is very often)
15:19:49 <water> s/very often/almost all the time/
15:20:47 eihrul won the sparcbook.
15:20:55 <water> awesome!
15:21:00 <eihrul> i just need to get all these funky credit card/money order issues worked out
15:21:08 <water> now we can both run Self on the go! =)
15:23:08 <eihrul> that'd be nice
15:24:26 eihrul thinks he found a bug in Hugs98.
15:25:11 <eihrul> rrr, n'm :)
15:26:36 <nojityi> self doesnt run on linux?
15:26:49 <water> not just yet
15:27:24 <water> it will run on sparc linux first, of course
15:27:57 <nojityi> is it an interpreter or does it compile to native?
15:27:58 water is on the self-interest mlist on egroups
15:28:22 <water> it's a bytecode vm with dynamic compilation and run-time optimization
15:28:38 <water> iow HotSpot is a cheap knock-off of the self vm
15:28:53 <nojityi> u think hotspots inferior?
15:29:04 <water> no, i know that it's inferior
15:29:10 <nojityi> why so
15:29:27 <water> Sun started Java after the Self project was abandoned
15:29:34 <water> which it also funded
15:29:46 <water> it borrowed self technology to build hotspot
15:30:06 <water> and some of the more advanced and useful optimization features did not make it into the hotspot vm
15:30:41 <water> i believe that there is a paper on this, probablyon sun's website
15:30:59 <nojityi> is self useful for anything besides a nice vm?
15:31:43 <water> heh
15:31:55 <nojityi> not to disparage it, i just dont know much about it
15:31:57 <water> it's mostly a research system... it has a nice gui environment that is self-hosted
15:32:06 <water> abi self?
15:32:06 <abi> self is a prototype-based object system at http://www.sun.com/research/self/ or old mirror at http://www.cs.ucsb.edu/oocsb/self/
15:32:26 <water> self is similar to squeak as an environmnet
15:32:33 <water> *similar*
15:35:27 <nojityi> too bad i dont have a sparc
15:35:30 nojityi bbl dinner
15:35:37 <eihrul> well, there's lots of stuff Self didn't do
15:35:46 <eihrul> as good as it was, there was much room for improvement :)
15:36:04 <water> sure
15:36:07 <eihrul> although i believe Self's contribution was that it even allowed for these improvements i speak of :)
15:36:24 <water> probably most hampered by the fact that no one ported any of it for 7 years at least
15:38:10 eihrul hmms.
15:38:31 <water> which pissed me off when i first heard of it in '94
15:40:02 <water> heh i think i discovered self, tunes, and merlin all in the first couple of weeks after i got net access at the university for the first time
15:41:16 <eihrul> well, if i could figure out a good way to do register allocation, i could actually make some progress on the way low-level end :)
15:41:44 <eihrul> i got to hierarchical instruction scheduling, but no luck past that :)
15:42:03 [QUIT] JALH quit: g'night bb tomorrow
15:42:10 <water> uh sure
15:42:32 <eihrul> yes, without register allocation, you can't generate code :)
15:42:39 <eihrul> it is that simple :P
15:43:04 <water> yeah but super-optimal register allocation is not on my mind at this point
15:43:22 <eihrul> i'm not talking about optimal
15:43:33 <eihrul> the most "optimal" method of doing it is graph coloring
15:43:39 <eihrul> and the thought of that makes me want to vomit
15:43:40 <water> well then why not re-use some other back-end?
15:43:45 <eihrul> i'm looking for *simple* :)
15:43:51 [NICK] lar[game] changed nick to: lar1
15:43:59 <eihrul> what other back-end do you speak of?
15:44:15 <water> hm
15:44:29 <water> doesn't kaffe have a JIT compiler?
15:44:50 <eihrul> yep
15:45:05 <eihrul> i'm pretty sure they'd use graph coloring
15:45:10 <water> wouldn't that be re-usable to some degree?
15:45:26 <water> what's the big deal about graph-coloring?
15:45:51 <eihrul> mainly that it sucks :)
15:45:59 <eihrul> very slow, very imprecise
15:46:20 <eihrul> linear scan is much faster, and just as imprecise :)
15:46:30 <water> abi: free compilers?
15:46:30 <abi> water: i don't know
15:47:23 <hcf> abi: catalog of free compilers and interpreters?
15:47:30 <abi> catalog of free compilers and interpreters is at http://www.idiom.com/free-compilers/
15:47:54 <water> oh yeah, why didn't i think of that? ;)
15:49:46 p0seid0n joined #tunes
15:49:55 <water> hi poseidon
15:50:36 <p0seid0n> hiya
15:50:53 <water> are you interested in Tunes?
15:51:09 <water> or have we met?
15:51:37 <p0seid0n> No.
15:52:46 <water> have you looked at our website?
15:53:04 <p0seid0n> No :)
15:53:15 <water> abi tunes?
15:53:15 <abi> tunes is a free reflective computing system at http://www.tunes.org or to programming languages what the internet is to networks
15:54:44 <water> i can answer most of the questions you might have
15:55:12 <water> and there are other people here who can answer the questions i either can't or am not good at answering
15:56:01 <water> damn it, gotta reboot
15:56:11 <water> (more TS&Ring of the cable modem
15:56:12 <water> )
15:56:16 [QUIT] water quit: The Tao went that-a-way!
15:56:27 p0seid0n left #tunes
16:18:55 lar1 joined #tunes
16:19:07 lar1 left #tunes
16:38:00 Kyle_L joined #tunes
16:42:09 Fare joined #tunes
16:43:24 water joined #tunes
16:43:33 <water> re
16:43:35 <water> fey Fare
16:43:44 <water> hey Fare
16:44:58 <Fare> Gakuk
16:45:02 <Fare> not here for long
16:45:16 <Fare> back by train, to go by train
16:45:30 <water> as it is with me, only with a big ship
16:45:44 <Fare> have just replied to a tunes message
16:45:55 <water> saw it
16:46:04 <Fare> Jacques Pitrat truly rocks. Too bad his reflective system is autistic.
16:46:22 <water> which system is that? that KR/AI one?
16:48:58 <Fare> MACISTE
16:49:20 <water> i've never looked at that
16:50:04 <Fare> Also, the kind of problems he tries to solve with MACISTE really show his being an AI old-timer (like, math quizz for advanced high-school kids)
16:50:18 <water> heh yeah that sounds right
16:50:43 <Fare> but he truly develops great techniques nonetheless
16:51:06 <water> got a url for it?
16:51:30 eihrul returns from the evils of homework.
16:55:24 <Fare> hum; not really. Somewhere on www-poleia.lip6.fr, you should find stuff;
16:56:04 <Fare> but most of what he wrote is in conference articles or french technical reports (or his disciples' PhDs).
16:56:35 [QUIT] Kyle_L quit: Leaving
16:56:47 <Fare> You may also read JL(?) Laurière's book about AI, which was translated in english
16:58:08 <Fare> Laurière was a whizz programmer, disciple of Pitrat who in many ways surpassed the master, but who unfortunately ended his career in PH
16:58:12 <water> yeah, "if it's still in print or available"
16:58:37 <Fare> not available in France anymore (but in some old tech libraries)
16:58:46 <Fare> never to be in print again
16:58:56 <Fare> (at least the original french version)
16:58:57 <water> iow not available except to the obsessed if in america
16:59:35 <Fare> it was such a success that it was translated into english; listed on amazon (dunno about availability)
16:59:39 <water> anyway, i'm getting a new laptop to do my owrk on away from home
17:00:12 <water> so expect some more productivity from me
17:00:21 <water> at the very least i'll be able to write again
17:02:31 <water> it's an ibook btw, save us all your lame mac-bashing comments
17:14:50 <Fare> damn, over 200 late e-mail to cope up with just for the last 3 days (and hundreds more in the old pipe)
17:15:11 <water> yeah i get the same volume
17:15:41 <Fare> I hate macs, but as far as laptops go, they have the edge, thanks to their not using these stupid power-eating PC CPUs.
17:16:09 <water> yep
17:16:31 <water> plus mac's run self, squeak, gnu tools, plus other decent stuff
17:16:38 <Fare> However, if you want the screen, memory, HD, to be on-par with a PC laptop, it costs _a lot_.
17:16:43 <water> and i don't need games
17:16:54 <water> Fare: i have the cash
17:17:01 <water> i have a lot of cash
17:17:14 <Fare> are you well paid in the navy?
17:17:15 <water> which reminds me, i'm about to invest in some stock
17:17:21 Fare whistles "In the navy..."
17:17:22 <water> no but i save
17:17:32 <Fare> do you restore?
17:17:34 water slaps Fare
17:17:40 <water> "resotre"?
17:17:45 <water> "restore"?
17:17:53 <Fare> stupid pun.
17:18:33 <Fare> water: do you run System 8, MacOS X, and/or Linux?
17:18:50 <Fare> (or even, NetBSD or OpenBSD?)
17:19:05 <water> it runs mac os 9 for now
17:19:19 <water> i don't know which bsd or *nix i'll run yet
17:19:46 <water> but for now i'll just get familiar with it
17:21:01 <water> Fare: at some point i want to work with you point by point on the HLL specs/requirements
17:21:16 <water> right now, there's not a simple way to do that
17:21:31 <water> mostly because it would require a lot of interactivity
17:21:48 <water> which is often difficult or impossible for both of us
17:22:48 <Fare> water: good.
17:23:06 <water> i have a print out of the web site
17:23:10 <Fare> water: will you be there on tuesday night (evening for you) ?
17:23:14 <water> (apart from review)
17:23:21 <water> absolutely not
17:23:30 <Fare> when will you be online?
17:23:33 <water> i will be in the pacific ocean on an aircraft carrier
17:23:53 Tril joined #tunes
17:23:53 [MODE] ChanServ set mode: +o Tril
17:23:58 <Fare> Tril!!!!!!
17:24:00 <water> a couple of weeks from now
17:24:09 <Fare> water: have a nice trip!
17:24:10 <water> heh
17:24:12 <water> hi tril
17:24:16 <Tril> yo
17:24:17 <Fare> water: are laptops allowed on board?
17:24:27 <water> Fare: no sarcasm please
17:24:31 <water> Fare: yes
17:24:37 <Fare> sarcasm?
17:24:38 <water> that's why i got it
17:24:45 <water> "have a nice trip!"
17:24:58 dwelch joined #tunes
17:25:09 <water> hi dwelch
17:25:15 <Fare> no sarcasm intended. Shit happens, but there might still be things to rejoice about.
17:25:31 <dwelch> hi
17:25:39 [TOPIC] hcf: TUNES: Free Reflective Computing System http://www.tunes.org || Slate Programming Language http://www.tunes.org/~water/slate-home.html || http://lambda.weblogs.com
17:25:51 <water> Fare: forget i mentioned it... just please don't say something like that again
17:26:28 <Fare> Hey, the last book I read was by Chalamov about his life in the Kolyma's hell (were Stalin had his worst extermination camps). Even there, there could be rejoicing (although never a sane one)
17:26:46 <Fare> s/were/where/
17:26:51 <water> exactly, sanity is precious
17:27:04 <water> dwelch: anything we can do for you? :)
17:27:45 <dwelch> Not really. I just came by to see what was going on
17:27:49 <water> ok
17:28:16 Fare is back from a meta-knowledge resourcing half-week
17:28:45 <Fare> Tril: could you find time to hack the lispm?
17:29:29 <Tril> Fare: do you want me to put VNCserver on it so you can do so remotely?
17:29:42 <Fare> hey, does VNCserver work on a LispM ????
17:30:01 <Tril> It remains in Mac OS all the time as far as I know. So VNC should work. Right?
17:30:15 <Fare> I have not much time for it right now. If I hack anything, it must be tunes.
17:30:31 <water> *ahem* slate.lisp? :)
17:30:36 <Fare> I'm not sure whether Mac OS actually runs or not while the LispM is active.
17:30:43 <Tril> Fare: Well, it sure looked like it to me.
17:31:06 <Fare> that might be a swell trick, although I wonder about the keyboard input...
17:31:26 <water> hey fare, would you consider helping to code the new slate.lisp?
17:32:18 <Fare> water: where is it (both URLwise and advancementwise) ?
17:32:35 <water> brb
17:35:44 <water> damn it, thte cable modem works for my roommate but not for me!
17:35:55 <water> sorry, i got side-tracked
17:36:12 <water> Fare: slate docs need sorely to be updated
17:36:23 <water> which is something i can work on while at sea
17:36:33 <water> but eihrul knows most everything you would need to know
17:36:50 <Fare> hum
17:36:54 <water> as for conceptual development
17:37:11 <Fare> I improved my architectural model for Tunes bootstrap, this week.
17:37:16 <water> most of that is still in my head
17:37:34 <water> but i need a basic slate system to hack and to hack upon
17:37:40 <water> Fare: how so?
17:38:43 <water> Fare: what improvements have you devised?
17:38:44 <Fare> water: clearer design choices with better rationale.
17:38:53 <water> such as...?
17:41:16 <Fare> Well, the layering of the bootstrap into 1) compiler bootstrap (such as in slate.lisp?), 2) basic runtime compiled with said compiler, 3) persistence and interaction developed and debugged, 4) integrating the whole dev channel (including CVS and make equivalent) into the system
17:41:30 <Fare> and then more and more details.
17:41:44 <Fare> Gotta write all that into a message to be posted on tunes@
17:42:00 <water> uh sure
17:42:25 <water> Fare, that makes no sense to me
17:42:34 <Fare> what makes no sense?
17:42:55 <Tril> Fare: I told him to write the list about it too, and he said they won't listen??
17:43:04 <water> in what sense are you using the term bootstrap, and how is this layer system you describe meaningful?
17:43:24 <Fare> Tril: told whom about what?
17:43:39 <Fare> "bootstrap", as in, make the system self-sufficient
17:43:50 <Tril> nevermind i'm not paying attention to this conversation :) 
17:43:52 <water> Fare: who cares about that right now?
17:44:08 <Tril> helping eihrul with his email for a sec
17:44:10 <water> Fare: we don't even have an HLL that does what we need
17:44:41 <water> changing the HLL later would most likely require us to re-perform that bootstrap anyway
17:45:04 <water> and if we don't have the HLL first, then we'll have to change it later, ergo wasted bootstrapping
17:45:13 <Fare> I have clear ideas on lots of the HLL, now. Gotta compare w/ slate, if you have up-to-date docs.
17:45:32 <water> Fare: i seriously doubt you grok how to make the HLL
17:45:49 <Fare> No, we don't need change everything w/ the HLL, if we take a massively source-transforming approach.
17:45:58 <water> heheh
17:46:05 <water> Fare: which would be complex
17:46:26 <water> much more complex than a smallt group of developers like we can handle
17:46:34 <Fare> Which is complex, but not of a complexity > than the intrinsic complexity of what is needed to bootstrap, anyway
17:46:46 <water> it's still wasted
17:46:49 <Fare> it might still be on the least complexity path to bootstrap
17:47:00 <water> you're multiplying the efforts required by your own admission
17:47:15 <Fare> No. Only the efforts needed at this one step.
17:47:28 <Fare> But if you look at the bootstrap process as a whole, it's a win.
17:47:49 <water> you're head is in the clouds
17:47:54 <water> s/you're/your/
17:48:09 <Fare> Microkernels told me to always consider whole-system complexity, not just show-subsystem complexity
17:48:26 <water> so what?
17:48:43 <water> i doubt you really know what that means anyway
17:48:54 <Fare> rewrite rules are in the shortest path
17:48:58 <water> you probably are still stuck on tunes having primitive types
17:49:09 <Fare> so is pattern-matching in general and code-matching in particular
17:49:16 <water> no, they *can* be in the shortest path
17:49:48 <water> pattern-matching requires a static typesystem Fare
17:49:54 <Fare> They ARE in the shortest path, all the more since they are 1) part of the goal 2) way to eat our own dog food ASAP
17:50:08 <Fare> no, pat-matching doesn't (see Prolog, Erlang, etc)
17:50:36 <water> hahahahahaha
17:50:46 <water> that's no proof at all
17:50:56 <Fare> (even "OO" languages have kind of lame pat-matching through object dispatch).
17:51:02 <water> yes i know this
17:51:09 <eihrul> maybe i'm naive, but i don't necessarily see why pattern matching needs static types :)
17:51:38 <water> Fare seems to think he can avoid the literal types
17:51:43 <Fare> eih: neither do I, in front of Prolog, Erlang, and more.
17:51:50 <water> and maude?
17:52:03 <eihrul> water: literals, or literal types? :)
17:52:08 <Fare> water: how do literal types interfere with pattern matching???
17:52:34 <water> Fare: the don't, they make it possible
17:52:41 <Fare> uh???
17:52:53 <water> and in tunes, no types are supposed to be primitive as a matter of course
17:53:13 <water> Fare: what are expression symbols made of? :)
17:53:23 <Fare> You just need primitive/library-provided patterns.
17:53:44 <water> oh ok so we rely on a HACK, is that it?
17:53:58 <eihrul> lets keep it civil :)
17:54:10 <water> don't even think about keeping the language clean?
17:54:18 <Fare> Lack of "absolute" primitives doesn't preclude availability of "relative" primitives
17:54:36 <water> so what?
17:54:54 <water> i'd bet you can't dynamically change what's primitive and what's not with what you currently grok
17:55:24 <water> unless you can actually explain it to me without your usual references to papers without explanations
17:55:25 <Fare> Of course, I can.
17:55:40 <water> how? what system currently does this?
17:56:14 <Fare> Dunno which does, except perhaps kind of what ML functors do.
17:56:38 <water> lol
17:56:48 <water> that's a joke, isn't it?
17:56:51 <Fare> (except the functors don't work in matching mode, only in building mode)
17:57:26 <Fare> No. You declare an interface. Then, how this interface is implemented is hidden away by the language.
17:57:52 <water> but how are declarations part of a pure language of objects?
17:57:55 <Fare> It requires that you declare (or import) patterns before you use them, but allows these patterns to not be primitive.
17:58:04 <water> (in concrete terms)
17:58:09 <water> what does that mean?
17:58:20 <Fare> Why pure language? At least the meta-language is not pure, since it imports declarations.
17:58:21 <water> show me a pattern reified as an object
17:58:45 <Fare> The only languages that are truly pure are pure combinatorial languages (unlambda, Joy, etc)
17:58:45 <water> heh but the meta-language has to be a subset of the language
17:59:00 <water> Fare: don't insult me
17:59:08 <Fare> uh?
17:59:16 <water> combinatorial language are not pure
17:59:29 <Fare> unlambda is pure
17:59:29 <water> languages
17:59:36 <water> um no
17:59:39 <Fare> (well, ok, except for its print primitive)
17:59:49 <water> you see?
18:00:03 <Fare> he could very well have done I/O in monadic style
18:00:04 <water> you'll always have excuses like that
18:00:09 <Fare> the rest of the language is pure
18:00:19 <water> yeah right
18:00:25 <Fare> I told him to do it monadic style, but he wouldn't.
18:00:33 <water> oh of course
18:00:35 <Fare> and I was too lazy to split a dialect.
18:00:37 Kyle_L joined #tunes
18:01:16 <Fare> the author is a friend. A truly nice guy. Too bad I'm not gay like him, else I could court him :)
18:01:22 <water> so what are the differences between slate as you understand it and the HLL ideas you have?
18:01:42 <Fare> If slate pretends to be pure, a lot.
18:01:57 <water> and why does your HLL spec and requirements embarrass me when i show it to CS people? :)
18:02:05 <Fare> If slate hasn't pattern-matching and source-matching, some more.
18:02:23 <Fare> water: because I wrote it looong ago and never truly revised it.
18:02:27 <water> we haven't gotten around to working on that
18:02:43 <Fare> because you don't dare rewrite it
18:02:46 <water> we've been concerned with the literal type system
18:03:05 <Fare> what literal type system?
18:03:11 <water> you're right, because i don't grok half the obscure references you made in it
18:03:23 <Fare> well, write another one, clean room.
18:03:35 <water> that's what diktuon was supposed to be for
18:03:41 <Fare> my documents are not the be-all end-all, you know!
18:03:41 <water> but diktuon crashed hard
18:03:49 <Fare> crashed hard?
18:03:51 <water> Fare: what else is there?
18:03:58 <water> yes i lost my database
18:04:03 <Fare> how so??
18:04:07 <water> well, corey lost it
18:04:23 <water> and i didn't get a backup like an idiot
18:04:33 <water> and i WAS rewriting your site
18:04:52 <Fare> backups are your friends!
18:04:52 <water> right now, i'm working too long of hours to do any serious work
18:05:04 water gives Fare the finger
18:05:14 <water> i know that
18:05:15 <Fare> Which reminds me that it's about time I backup all my systems on CD-ROM (CVS, /etc, ...)
18:05:23 Tril looks the other way not having backups of a lot
18:05:24 <water> all of corey's backups failed
18:05:30 Fare eats the finger
18:05:34 <water> Tril: what?!?!?!
18:05:46 <water> geez, one more reason not to use bespin
18:05:51 <Kyle_L> That reminds me.....
18:05:51 <Tril> water: the only one backing up anything on bespin regularly is Fare.
18:06:02 <water> what use are you?
18:06:06 <Tril> water: if something should be added to his list we should tell him.
18:06:07 <Fare> water: use bespin, but just you back up stuff you put on it!
18:06:31 <Tril> water: I don't even want to know the other reasons.
18:06:34 <water> yeah why don't i maintain the whole thing while i work >70 hours per week????
18:06:49 Fare has a script to do backup of parts of bespin I need
18:07:00 <Fare> if you want me to add stuff, just you say so.
18:07:09 <water> heh oh yeah time for me to learn Yet Another scripting language
18:07:10 <Tril> Fare: add the postgres databases
18:07:18 <Fare> Tril: where?
18:07:41 <Tril> Fare: su - postgres -c "pg_dumpall -a > db.out"  ; go backup db.out
18:07:46 <Fare> Tril: you know, you can add stuff to the script, too
18:08:05 <Fare> Tril: is such db.out sufficient to restore?
18:08:08 <Tril> actually...actualy that is wrong just pg_dumpall > db.out should work
18:08:14 <Tril> Fare: yes
18:08:16 <water> Fare: for an "open" person, people sure don't trust you very much
18:08:28 <water> Fare: ever wonder why that is?
18:10:28 <Tril> no -a option
18:11:05 <water> heh the discussion has been reduced to unix commands again :)
18:11:07 <Fare> water: uh?
18:11:19 <Fare> ./backup:33: command not found: pg_dumpall
18:11:45 <Fare> still need the su postgres?
18:11:52 <Tril> Fare: "su - postgres" should get the PATH from ~postgres/.profile
18:11:54 <water> Fare: if you'd actually take my input on things and ask my advice, you might get some itneresting help
18:11:58 <Fare> water: how that, "people don't trust me" ?
18:12:14 <water> Fare: e.g. no one uses CVS
18:12:14 <Fare> water: your advice on what?
18:12:22 <Fare> water: I do.
18:12:23 <water> on the HLL or Tunes!
18:12:28 <Tril> My command was exact except for -a
18:12:33 <water> Fare: exactly, only you do
18:12:37 <Fare> Tril does
18:12:54 <water> big deal, he administers it
18:13:08 <water> s/it/bespin/
18:13:10 <Tril> water: you think I don't use CVS because I don't trust Fare?
18:13:25 <Tril> it's actually because I have bad discipline, but whatever you want to think.
18:13:28 <water> no i think 100 other people are in that situation
18:13:35 <Fare> core did.
18:13:50 <Fare> 100 people don't hack, because there's no critical mass to hack
18:13:59 <Tril> could we turn this conversation into something productive? like get coreyr here I need to ask him what diktuon2 is going to be like.
18:14:08 <water> Farte: then why are the web pages CVS'd?
18:14:14 <Fare> fsck your non-constructive criticism. Tell us about slate.
18:14:26 <water> ok
18:14:38 <water> but eihrul better get in on this
18:14:45 <Fare> eih: still here?
18:14:48 <eihrul> yep
18:14:50 <water> yeah he's here
18:14:58 <water> talking off-channel of course
18:15:05 <Fare> ok, so what should I know about slate that's misrepresented in the old docs?
18:15:21 <water> well are you aware of the new syntax?
18:15:25 <Fare> water: don't want us to chat about slate in the logged channel??
18:15:36 <water> no, i do
18:15:37 <Fare> I'm not aware of much anything.
18:15:38 <Tril> Fare: he meant he knows eihrul is here due to msg
18:16:14 <water> hum, okay time to go to square-one about slate, then
18:16:29 <Fare> I'm all ears.
18:16:48 <water> slate expressions are a sequence of object-name lookups and evaluation handles
18:17:04 <water> there are many ways to approach the same evaluation
18:17:39 <water> because each evaluation handle looks to the MO to find the rewrite rules to apply to the object graph that is the Slate object system
18:17:42 <Fare> object-name lookups meant to be done at runtime?
18:18:00 <water> it doesn't matter semantically, but yes
18:18:09 <Fare> Tril: tar: var/named: Cannot stat: No such file or directory
18:18:13 <water> er...wait
18:18:16 <water> just yes
18:18:21 <Tril> fare I'll msg you about it
18:19:19 <water> the rewrite rules is much like the linear graph rewrite of bawden's paper, but in this case, linearity or nonlinearity is a property of each link (lookup path) between objects
18:21:09 <Fare> what's an evaluation handle? just the same as "slot" or "method" in usual object languages?
18:21:15 <water> now, i'm hoping your roughly familiar with discussions i've had with you about "return" (^), "parent", "meta", and "<" (formerly "..")
18:21:23 <water> no
18:22:16 <water> each object has a meta-object which defines a rewrite rule that is applied when the evaluation handle is mentioned
18:22:18 <water> e.g.
18:22:57 <water> myPoint rotate by : degrees 90 ....
18:23:14 <water> the periods on the end are evaluation handles
18:23:37 <water> it depends on how the situation is implemented whether multiple periods are needed or not
18:24:02 <eihrul> evaluation handle => token/command/word/whatever that says "do an apply of the top of the stack to the object below the top"
18:24:10 <water> because the rewrite rules can be factored into smaller steps called iteratively, which is more flexible
18:24:34 <Fare> reminds me of POP-11's stack language
18:24:36 <water> thx eih
18:24:47 <water> it may not have to be a stack, though
18:25:07 <water> but the semantics would be similar
18:25:07 <eihrul> practically, its a stack
18:25:39 <Fare> so the "...." here means that you have 90 degrees by: and rotate to apply?
18:26:11 <water> each token besides the "...." is a namespace lookup
18:26:48 <water> ":" actually inherits from a non-local namespace to provide values for the "by" object
18:27:42 <water> and this process of "assignment" is done by rewrite
18:28:19 <water> from there, the "by" provides a parameter to the rotate object
18:28:36 <water> which in turn uses rewrite to return a new object based on myPoint
18:29:48 <Fare> so : does some job that makes a . unnecessary for it?
18:29:51 <water> that remote namespace where the values came from would be named, say, "rotations", with "degrees" within
18:30:23 <water> the : makes the rewrite more modular
18:30:42 <Fare> how so?
18:30:51 <water> you can change what namespace the : behaves like without affecting the rewrite rules
18:31:06 <Fare> ?
18:31:27 <water> ok the : basically inherits from another namespace
18:31:28 <Fare> I'm missing something deep about your evaluation model
18:31:32 <water> which contains values
18:31:58 <water> perhaps it's the aspect of inheritance?
18:32:21 <water> (meta-level rewrite of a sort)
18:32:28 <eihrul> basic idea:
18:32:41 <eihrul> each symbol, other than the one designated to denote application
18:32:54 <eihrul> does a symbol lookup which pushes the result of the lookup on the evaluation stack
18:33:07 <eihrul> so lookups just grow this stack with new lookup results
18:33:11 <Fare> ok
18:33:30 <eihrul> an apply just reduces this by replacing the top two elements of the stack with the result of application
18:33:30 <water> yes
18:33:36 <Fare> ok
18:33:49 <water> well, that's one of the main kinds of rewrite methods
18:33:52 <Fare> eih: what about namespace switch fore and back?
18:34:08 <eihrul> eh?
18:34:36 <Fare> lookups don't evaluate; they just push on the stack? So you need evaluate (with "." ?) looked up things to change namespaces?
18:34:55 <water> you can lookup without evaluating
18:35:10 <eihrul> no
18:35:17 <eihrul> lookups go to the object at the top of stack
18:35:33 <eihrul> so you "change namespaces" by virtue of changing the top of the stack
18:35:40 <water> in most cases the "<" will reverse a lookup
18:36:25 <Fare> eih: ok
18:36:41 <Fare> water: what's a reverse lookup?
18:37:07 <eihrul> oh, water has this notion that all references should be linear
18:37:12 <eihrul> )
18:37:12 <water> "<" isn't a reverse lookup, it just pushes the enclosing object
18:37:18 <water> eih: noo
18:37:50 <Fare> water: as in DUP or OVER ?
18:38:00 <water> as in over
18:38:27 <water> but it's not always like over
18:38:34 <water> "<" is like ".." in a file system
18:39:01 <eihrul> well, is "<" one of those strange little operators like lookup and apply now or is it still a slot? :P
18:39:08 <water> it works to show what object contains the object currently at the top of the stack
18:39:17 <water> eihrul: it's still a slot
18:39:33 <Fare> water: so that foo/.. is not ., if foo is a symlink instead of a real dir?
18:39:45 <eihrul> water: well, containment only makes sense in terms of the stack :)
18:40:19 <water> eih: why?
18:40:25 <water> Fare: hm
18:40:48 <eihrul> water: we went over this already :)
18:40:50 <water> Fare: right
18:41:10 <water> eih: when? i'm not connecting discussions correctly then
18:41:20 <Fare> why do you need a "<" for?
18:41:42 <Fare> do the "<" constitute a reverse hierarchy?
18:41:50 <water> so you can manipulate an object without evaluating it
18:42:12 <Fare> I thought that you never evaluated by default, just pushed on the stack?
18:42:42 <water> right, but there needs to be a way to leave that context without rewriting sometimes
18:42:55 <Fare> and what if the TOS doesn't contain the looked up word?
18:43:12 <water> TOS?
18:43:13 <Fare> leave to where?
18:43:16 <Fare> Top Of Stack
18:43:31 <water> then it leaves to wherever the < slot is set
18:43:43 <water> which also pushes that object onto the stack
18:44:09 [QUIT] hcf quit: Leaving
18:44:39 <water> hum
18:44:50 <water> i'm not sure where to proceed
18:45:30 <Fare> looks quite ugly. Why not just a DROP ?
18:45:37 <Fare> or a ROTn ?
18:45:41 <eihrul> well, meta-objects would be a good place to go :)
18:45:46 <Fare> or a PICK ?
18:45:53 <water> it can be changed
18:45:57 <Fare> or a NDROP ?
18:46:09 <water> yes meta-objects would be good
18:46:44 <water> the scheme i prefer right now is that each object has exactly one meta-object unique to it
18:46:52 <eihrul> Fare: well, the stack is more an implicit part of slate evaluation, it isn't necessarily meant to be the programming model as in Forth :)
18:46:53 <Fare> currently, looks like a bizarre FORTH dialect that quotes everything by default.
18:47:14 <water> which would share its multiple parts by delegation and inheritance
18:47:30 <water> yep
18:47:42 <eihrul> water: rmm, why can't many objects share a meta-object?
18:48:35 <water> eihrul: well, the issue i was thinking of is that MO's would be simpler to use if the object they implement were uniformly referrable
18:48:41 [QUIT] _ruiner_broodwar quit: destroy what destroys you
18:48:48 <Fare> It's getting late, here
18:48:59 <eihrul> water: then what's the point of MO's at all?
18:49:00 <water> which if they had multiple one's, you would have to give the MO some way to know one from the other
18:49:12 <eihrul> why shouldn't objects just be self-representing in every way then, including meta-behavior?
18:49:16 <water> eihrul: mo's can still be shared, man
18:49:42 <Fare> gotta go.
18:49:47 <water> Fare:cya
18:49:52 <Fare> bye!
18:49:54 <water> what do you think?
18:50:08 <water> Fare, what do you think?
18:50:09 <water> gah
18:50:33 <Fare> From the operational view I was given, I don't understand the point yet.
18:50:35 <water> eih: because changing an object's own lookup procedure is circular :)
18:50:54 <eihrul> well, lookup consults the meta-object of the object on how to do the lookup
18:50:59 <eihrul> and apply does likewise
18:51:01 <water> especially the lookup to the part that modifies the lookup
18:51:07 <Fare> looks like dropping is difficult; or do you have a block syntax for mass-drop?
18:51:12 <eihrul> the net effect, you can simulate a lot of stuff
18:51:21 <water> "dropping"?
18:51:28 <Fare> stack save/restore
18:51:28 <water> eih: sure
18:51:31 <eihrul> water: going back to some object earlier in the stack
18:51:37 <eihrul> water: vs. writing ..........
18:51:46 <eihrul> that's a sugar issue, though :)
18:52:13 <water> Fare: no special syntax for it, it has to be supported primitively until we bootstrap
18:52:19 <water> Fare: afaik
18:52:41 <water> Fare: which is why there are other features we didn't mention
18:53:27 <water> Fare: but anyway, i'll see you in a couple weeks, and you'll ignore what you learned here and make your own HLL thingie anyway
18:53:55 <Fare> sure. Bye!
18:53:57 [QUIT] Fare quit: Connection reset by pear
18:54:07 <water> eihrul: where were we?
18:54:42 <cor[class]> worse than a soap opera
18:54:47 [NICK] cor[class] changed nick to: corey
18:55:27 <eihrul> meta-objects
18:55:33 <corey> `cept i cant tell who's sleeping with whom :)
18:55:46 <water> heh
18:55:59 <water> eihrul: yeah but what were you saying about "writing"?
18:56:24 <eihrul> oh, just like in the myPoint example
18:56:35 <eihrul> unless you make an extensible syntax
18:56:42 <eihrul> )
18:56:59 <water> um extending the syntax how?
18:57:16 <water> adding slots is supposed to be part of extending htte syntax you know
18:57:17 <eihrul> as in mapping some other syntax to core slate
18:57:25 <water> oh
18:58:14 <water> well, what kind of "macro" system do you propose?
18:58:45 <eihrul> at the moment, i can't name any particular details
18:58:59 <eihrul> but i can predict its utility, regardless of what form it comes in :)
18:59:36 <water> ok explain how you would want to use it
19:00:27 <eihrul> well, for instance, what if you wanted to make Slate syntax look exactly like Smalltalk? :)
19:00:49 <water> i think i've already demosntrated that
19:01:04 <eihrul> that is an approximation
19:01:09 <eihrul> we want EXACT :)
19:01:18 water shrugs
19:01:32 <eihrul> okay, how about perl syntax?
19:01:39 <water> that would just be implementing a reader system over the basic evaluator
19:01:55 <water> which means a parse afaik
19:01:57 <eihrul> well, sure
19:01:59 <water> er... parser
19:02:20 <eihrul> but some modular way to handle all that, and allow them to interact
19:02:29 <water> and building slate objects that perform what the target language does
19:02:52 <eihrul> i.e. escaping to the lisp parser from within the smalltalk parser, etc. :)
19:02:59 <water> hum
19:03:35 <eihrul> this suggests a "parser stack" of sorts :)
19:03:45 <water> well since we are just discussing a graph of interconnected objects upon which arbitrary rewrite rules may be invoked...
19:03:46 <eihrul> where you can enter into a different parser module from another
19:03:51 <eihrul> and then leave back, and such :)
19:04:07 <water> i don't see much of a problem, at least conceptually :)
19:04:23 <eihrul> yeah, i know
19:04:33 <eihrul> the thing is we just need a system in slate for doing something like that :)
19:04:37 <water> hum modularity of parsers
19:05:11 <water> ok i have one way to approach it at the top of my head
19:05:15 <eihrul> or perhaps this goes back to the quoting issue? :)
19:05:28 <water> no, i don't think so
19:06:08 <water> basically, you're suggesting that the top-level of each supported language environment be able to switch to another parser?
19:06:27 <water> or maybe not just top-level? (which sounds complicated)
19:06:31 <eihrul> it doesn't have to be able to :)
19:06:48 <eihrul> but would allow for "interoperability", for what it is worth :)
19:06:50 <water> pardon? it doesn't?
19:07:26 <eihrul> i was referring to not just the top-level :)
19:07:52 <water> well, it would be pretty easy to build trees of slate objects underlying the environment used as a "lingua franca"
19:08:01 <water> eih: oh ok
19:08:38 <eihrul> isn't that what tunes wants? :)
19:08:42 <water> and each object added to the tree would have a type that corresponded to the language grammar
19:08:57 <water> right, i'm trying to envision this in slate
19:09:43 <water> so, interactively speaking, the parser would fill in the slots for the grammar currently selected
19:10:45 <water> hm
19:10:57 <water> if you think about EBNF notation
19:11:22 <water> that's obviously the range of expressiveness we need
19:12:03 <water> now for each term that we know an object in a parse tree must be, we can annotate it with the case of the term that the user selects dynamically
19:12:50 <water> is anyone following this?
19:14:05 <water> geez
19:14:09 <water> *ping*
19:15:13 <eihrul> eh?
19:15:29 <water> did what i just say make sense?
19:15:49 <eihrul> no
19:15:59 <water> hum
19:16:10 <water> why not?
19:17:00 <eihrul> water-speak
19:17:24 <water> oh ok
19:18:01 water tries to re-formulate it concretely
19:19:22 <water> ok when parsing one of the usual languages, we have to tokenize, then analyze the tokens, usually on a context-sensitive basis, to recursively describe the parse tree
19:19:32 <water> trivial example... lisp
19:20:17 <water> hm
19:20:39 <water> each node will either be a list cell or an atom
19:21:38 <water> when building the tree of slate objects representing a syntax structure, if we do a recursive-descent system, we have unset slots before analyzing the next symbol
19:22:11 <water> this "unset slot" idea is what i'm talking about
19:22:38 <water> what it can be is required to be determined by the grammar
19:23:22 <water> i'm suggesting we could have a type namespace for things like this
19:23:49 <water> which would have both atoms and lists in it for this example
19:24:06 <water> eih: does that make sense?
19:24:22 <eihrul> i s'pose so
19:24:27 <eihrul> i'll come up with something :)
19:24:39 <water> heh
19:24:49 <eihrul> since i need to have this for pidgin if it is to be usable
19:25:05 <water> well this would be a straightforward translation from EBNF rules,right?
19:25:17 <water> bah :)
19:26:45 <water> is anyone else here?
19:27:57 [QUIT] eihrul quit: Leaving
19:28:52 <corey> i am kinda
19:29:13 <water> great, one person is "kinda here"
19:29:21 <water> what's up for you?
19:29:38 <corey> its friday
19:38:10 <Tril> hey...corey
19:38:15 <Tril> tell me about diktuon2
19:38:23 <corey> ok
19:38:29 <corey> first you need an account.
19:38:30 <Tril> what's it going to be like?
19:38:49 <water> yes i'll need one too of course
19:39:01 <corey> water: you already have one!
19:39:51 <corey> Tril: add yourself into the users table in the diktuon2 db
19:40:10 [NICK] lar1 changed nick to: lar[life]
19:40:23 <Tril> corey: you're running the wrong psql...does that work?
19:40:30 <corey> yep
19:40:41 <Tril> it should segfault if you create a table?
19:40:41 <corey> which psql should i be using?
19:40:51 <Tril>  /usr/local/pgsql/bin
19:40:57 <corey> kay.
19:41:02 <Tril> at least for now.
19:41:28 <Tril> PGLIB=/usr/local/pgsql/lib ...etc
19:41:44 <corey> well im not very academic (yet) so ill explain things oversimply
19:41:53 <water> corey: oh heh
19:42:11 <corey> diktuon2 consists of nodes
19:42:19 <corey> nodes can hold other nodes
19:42:28 <corey> this creates a tree
19:42:42 <corey> root is at the top
19:43:01 <corey> btw url is diktuon.tunes.org/diktuon2
19:43:14 <water> oh?
19:43:55 <water> oh ok mapped from your account's space
19:44:03 <water> (which was what i was using)
19:44:08 <water> but continue
19:45:15 <corey> ok
19:45:27 <corey> nodes have a format
19:45:54 <corey> in xml, because php has an xml parser that makes it easy for me to add links and other stuff
19:46:27 <corey> im working on the dtd
19:46:43 <water> ok where's a good xml tool that will read in dtd's?
19:46:59 <corey> when you find a free one tell me. :)
19:47:05 <water> oh damn, hcf ain't here ;)
19:47:16 water looks at SF
19:47:49 <corey> <a> taking only a href arg
19:48:20 <corey> <node> to link within diktuon
19:48:36 <corey> <node name=root_blah> to be exact
19:48:54 <corey> any suggestions on these are welcome
19:49:03 <corey> <list> and <item> for lists
19:49:22 <corey> <p>, <b>, <i> like html
19:49:27 <water> hm
19:49:32 <corey> and remember to close all tags
19:49:59 <corey> i need a validator
19:50:05 <corey> php doesnt have one
19:50:16 <corey> but ill work something out
19:50:24 <corey> just login and play with it
19:50:33 <water> corey: if i find a good editor, i shouldn't have to do that
19:50:42 <corey> you mean login?
19:50:56 <Tril> Defining XML as an application profile of SGML means that any fully conformant SGML system will be able
19:50:56 <Tril>                           to read XML documents. However, using and understanding XML documents does not require a system that is capable of
19:50:57 <Tril>                           understanding the full generality of SGML. XML is, roughly speaking, a restricted form of SGML.
19:51:21 <water> corey: i mean closing tags
19:51:28 <corey> oh
19:51:34 <corey> well by using xml
19:51:50 <corey> you could use any tool
19:52:10 <corey> and you dont even have to use diktuon as a frontend
19:52:15 <corey> you could just put it in the db
19:52:26 <corey> and also
19:52:29 <water> heh yeah but i'd like a tool that's *useful*
19:52:46 <water> i can't telnet to the db from the ship
19:52:57 <water> or ssh or whatever
19:53:01 <corey> put you could make nodes and then submit them when you got back
19:53:04 <corey> but
19:53:16 <water> yeah a 2-week lag time?
19:53:17 <corey> and also its easily convertable to other forms
19:53:23 <water> i'd rather email them
19:53:27 <corey> that too
19:54:24 <corey> so it could be converted to flat html, sgml, and from there anything just about
19:55:11 <corey> look at root_tunes_todo for what i have plans for
19:55:17 <water> http://www.conglomerate.org/shots.html
19:55:24 <corey> email me any suggestions,complaints,insults
19:55:27 <water> hey this looks pretty good
19:56:04 <Tril> you meant root_diktuon_todo
19:56:18 <corey> yeah
19:56:20 <corey> sorry
19:56:20 <corey> )
19:56:44 <corey> i proposed a project redesign
19:56:52 <corey> because when you start thinking nodes... :)
19:57:08 <corey> a documentation subproject
19:57:11 <corey> from tunes
19:57:15 <corey> with review under that
19:57:36 <corey> and review's siblings being glossary and members publishings
19:58:10 <corey> more tree like
19:58:43 <water> whatever's intuitive
19:59:28 <water> damn 8pm
19:59:34 <corey> i dont think i have the authority to decide whats intuitive
19:59:38 <corey> 11pm here :P
19:59:50 <water> i still haven't eaten
20:00:20 <corey> well, again comments coreyr@tunes.org
20:00:25 <corey> farewell tonight
20:00:31 <Tril> ok, thanks for sharing
20:00:33 <water> /
20:00:38 <water> yeah thx
20:00:47 <corey> ill keep on working this weekend
20:00:55 <corey> later
20:01:07 <Tril> Is there a lot of stuff in the diktuon db that you want me to convert?
20:01:11 <Tril> (either of you)
20:01:24 <corey> from tunes.org -> diktuon?
20:01:27 <corey> oh
20:01:29 <Tril> from diktuon1 -> dikton2
20:01:34 <corey> hmm
20:01:37 <corey> not really
20:01:45 <corey> it just a couple more glos nodes
20:01:56 <corey> which could be gotten from tunes.org
20:02:18 <corey> i was thinking of replacing diktuon1 this weekend
20:02:26 <corey> diktuon2 is much more stable
20:02:42 <water> go for it
20:02:54 <corey> ok goodnight. :)
20:03:09 <corey> water: hope something turns up tonight.
20:03:16 [QUIT] Kyle_L quit: Leaving
20:03:17 [NICK] corey changed nick to: cor[afk]
20:03:36 <water> hm what to do
20:06:05 [QUIT] water quit: Read error to water[tnt-10-62.tscnet.net]: Connection reset by peer
20:06:07 water` joined #tunes
20:06:22 [NICK] water` changed nick to: water
20:11:42 hcf joined #tunes
20:11:54 <water> re hcf
20:12:05 <hcf> re water
20:12:28 <water> http://www.conglomerate.org seems to make a promising xml/dtd tool
20:12:49 <hcf> btw, a mod'd abi the other day, u shouldnt see any diff, if u do, tell me
20:12:54 <water> there was another that seemed better, but it was proprietary
20:13:06 <water> ok
20:14:01 <hcf> do u want poop put in d2 like u did w/ d1?
20:14:33 <water> sure, but the dtd will be a factor now
20:14:51 <water> which seems fairly simple
20:15:25 <hcf> yeah, and corey hasnt worked out node to node linking yet
20:15:43 <water> implementation you mean?
20:15:50 <hcf> yeah
20:15:55 <water> oh ok
20:16:02 <water> because he already discussed the syntax
20:16:23 <hcf> smth else is, tunes.org's content should be frozen at some point
20:16:50 <water> when does it ever change?
20:17:12 <hcf> it changed while we were doing d1 (pre-loss)
20:17:27 <water> what changed?
20:17:34 <hcf> see the cvs log
20:17:40 <water> um no thanks
20:18:30 <Tril> hcf: that point being "when none of the site is static any longer"
20:18:31 <hcf> recently fare changed most mention of 'commercial' to 'proprietary'
20:18:43 <hcf> s/ion/ions/
20:19:04 <water> oh wow, glad to see he's doing something important with his time :P
20:19:52 <Tril> hcf: recently = august?
20:20:00 <hcf> perhaps
20:20:04 <Tril> is that what those checkins he didnt describe were ?  :)
20:20:14 <Tril> now I know why.
20:20:42 water thinks that is rather pathetic
20:20:58 <water> oh well
20:21:06 <Tril> I have to go now...
20:21:16 <Tril> I'll try to work on tunes-stuff all weekend
20:21:29 [QUIT] Tril quit: bye
20:22:02 <water> i hope he gets that backup script modified to include diktuon db's
20:32:04 <hcf> abi: bnf is Backus-Naur Form at http://www.stud.ifi.uio.no/~lmariusg/download/artikler/bnf.html
20:32:06 <hcf> abi: ebnf is Extended Backus-Naur Form at http://www.stud.ifi.uio.no/~lmariusg/download/artikler/bnf.html
20:34:19 <hcf> abi: bnf =~ s/ at/, see/
20:34:20 <abi> OK, hcf
20:34:21 <hcf> abi: ebnf =~ s/ at/, see/
20:34:21 <abi> OK, hcf
20:35:57 <hcf> water: here?
21:13:46 <water> back
21:14:12 <hcf> it seems by the bit of poop i'v read that DTDs for xml are depreciated in favor of XSDs (xml schema definitions).
21:14:16 <hcf> w3c links to a dtd2xsd tool, but it seems to be a shitty perl hack.
21:14:24 <water> heh
21:14:32 <hcf> looking for anything particular wrt [e]bnf?
21:14:39 <water> damn standards
21:14:50 <water> not really, just ebnf's for various languages
21:15:11 <hcf> k, 1sec
21:15:35 <water> thx, ur the coolest
21:17:39 <hcf> http://www.cs.man.ac.uk/~pjj/bnf/bnf.html
21:17:55 <water> ah there it is
21:18:27 <hcf> theres abnf that the RFCs use, standardized, but isnt used for PLs
21:19:05 <water> why? is it not expressive enough or is it a cultural thing?
21:19:12 <hcf> damn, cant reach http://www.csci.csusb.edu/www/dick/maths/welcome.html
21:19:19 <hcf> damn 'dick'
21:19:26 <hcf> he has several PL bnfs
21:19:40 <water> ok
21:20:14 <hcf> theres probly some bnfs on cats eye
21:26:32 <water> huh what's cats eye?
21:27:14 <hcf> abi: cet?
21:27:15 <abi> cet is Cat's Eye Technologies, esoteric proglangs at http://www.catseye.mb.ca/
21:27:32 <water> geez too much info to track
21:28:19 <hcf> sorry
21:31:15 <water> abi stats?
21:31:16 <abi> Since Fri Sep 15 06:19:25 2000, there have been 8 modifications and 17 questions.  I have been awake for 15 hours, 11 minutes, 50 seconds this session, and currently reference 1462 factoids.
21:32:37 <water> mdpn on cat's eye is good
21:33:21 <water> unfortunately, too ebnf-like
21:34:13 <hcf> ooh, new poop on cet
21:35:22 <hcf> btw, dont expect to find any bnfs on the other weird lang sites
21:35:44 <water> "weird" as in?
21:35:45 <hcf> unless theyre buried in docs
21:36:00 <hcf> the links from cet, like the tarpit and such
21:36:09 <water> oh
21:36:24 [NICK] lar[life] changed nick to: lar[sleep]
21:37:10 <hcf> abi: mdpn is Multi-Directional Pattern Notation at http://www.catseye.mb.ca/esoteric/mdpn/index.html
21:38:58 <hcf> if its needed,
21:39:00 <hcf> abi: dopl?
21:39:00 <abi> dopl is Dictionary of Programming Languages at http://cgibin.erols.com/ziring/cgi-bin/cep/cep.pl
21:39:31 <hcf> has a rather large collection of source exaples for a large number of PLs, not bnf afaik tho
21:41:24 <hcf> as i recall,
21:41:27 <hcf> abi: slim?
21:41:27 <abi> slim is a setl relative at http://birch.eecs.lehigh.edu/slim/ or quite free and small
21:41:48 <hcf> had some bnf poop online and/or in the archive
21:42:32 <water> probably in the tarball and the papers
21:43:11 <water> sunday will be quite busy for me
21:44:07 <hcf> the slim online docs use ebnf
21:44:42 <hcf> u consider ebnf to bnf enhanced w/ {} and [] type things right?
21:44:58 <water> right
21:47:37 <water> http://birch.eecs.lehigh.edu/slim/papers/compiler-generator/default.htm
21:49:53 <water> bbiab
21:49:54 [QUIT] morton quit: morton has no reason
21:52:54 [QUIT] nate37 quit: Ping timeout for nate37[cx83983-d.irvn1.occa.home.com]
21:53:28 nate37 joined #tunes
22:11:39 <water> yep, sunday's gonna be ridiculously busy
22:12:20 water crosses his fingers in hopes that the ibook will show up then
22:12:50 <water> monday works too, but then the time squeeze to configure it will be high
00:00:44 [QUIT] hcf quit: Leaving