IRC log started Fri Apr 21 10:44:39 2000
[msg(tunes)] permlog 2000.0421
-:- Mode change [-s] for user TUNES
-:- Mode change [-w] for user TUNES
-:- clintonr [clintonr@net255ip95.parklink.com] has left #tunes []
-:- SignOff dortmunder: #TUNES (Ping timeout for dortmunder[niantic0319.mohawk.net])
-:- XeF4 [xef4@79.ppp1-7.worldonline.dk] has joined #tunes
-:- Downix [unknown@dialup-209.245.97.97.Manchester1.Level3.net] has joined #tunes
<Downix> hello all
* Downix/#tunes has decided to make something
-:- SignOff Downix: #TUNES (Ping timeout for Downix[dialup-209.245.97.97.Manchester1.Level3.net])
01:10pm
-:- Downix [unknown@dialup-209.245.97.97.Manchester1.Level3.net] has joined #tunes
-:- SignOff Downix: #TUNES (Downix has no reason)
-:- SignOff XeF4: #TUNES (Leaving)
-:- _out [ANORMAL@200-191-39-168-as.acessonet.com.br] has joined #tunes
-:- Ghyll [karltk@msx-osl-16-52.ppp.cybercity.no] has joined #tunes
<eihrul> hey
-:- ult [ult@user-38lcn76.dialup.mindspring.com] has joined #Tunes
-:- hcf [nef@me-portland-us735.javanet.com] has joined #tunes
02:20pm
-:- SignOff eihrul: #TUNES (Ping timeout for eihrul[usr5-ppp135.lvdi.net])
-:- eihrul [lee@usr5-ppp194.lvdi.net] has joined #tunes
-:- SignOff ult: #TUNES (ircII EPIC4pre2 -- Accept no limitations)
<Ghyll> 'lo troops
02:40pm
-:- hcf has changed the topic on channel #tunes to: Tunes - Free, Reflective Computing System: http://www.tunes.org/ || Slate Tutorial: http://www.tunes.org/~water/slate-tutorial.html
-:- ult [ult@user-38lc60h.dialup.mindspring.com] has joined #Tunes
-:- Fare [rideaufr@esmeralda.enst.fr] has joined #Tunes
-:- SignOff ult: #TUNES (Ping timeout for ult[user-38lc60h.dialup.mindspring.com])
-:- SignOff _out: #TUNES (Leaving)
-:- tcn [Tom@cci-209150250051.clarityconnect.net] has joined #tunes
<Fare> t c n !
<tcn> hey fare
04:00pm
-:- SignOff tcn: #TUNES (ircII EPIC4-2000 -- Accept no limitations)
-:- nate37 [nate@ppp-207-151-70-35.dialup.pcmagic.net] has joined #tunes
-:- SignOff nate37: #TUNES (nate37 has no reason)
-:- rares [rares@nwhn-sh1-port59.snet.net] has joined #tunes
<hcf> hohum
<Ghyll> good evening, treebeard
06:10pm
-:- SignOff hcf: #TUNES (Leaving)
-:- hcf [nef@me-portland-us732.javanet.com] has joined #tunes
-:- water [water@tnt-10-71.tscnet.net] has joined #tunes
<water> hey all
06:30pm
<water> ping
06:40pm
<eihrul> hey
<water> hey hey
<water> thx for ..
<water> how'd you manage it? (sorry haven't looked at the code for it yet)
<eihrul> hmm, is cora working for you?
<eihrul> well, i wrote a new a lexer
<water> oh ok
<water> hm cora won't load
<eihrul> otherwise, when an '/' operation is done, i create a new '..' method that when invoked will put you back where you came from
<water> i haven't dl'd papers in a week
06:50pm
<water> hm
<water> that'll do for now i think
<water> does message lookup already know to delegate to 'parent' if not found?
<eihrul> yes
<water> then you shouldn't have to allocate the ..
<water> i.e. keep it in root as a primitive that acts on the receiver
<water> hm that would require some mb
<eihrul> well, i'd have to store where to return to somewhere
<eihrul> and it was just convenient at the time to store it in the closure itself :)
<water> ok
<water> well the system just has to work like a file system, roughly
<water> oh wait using hash-tables makes that a bit prickly i imagine
-:- SignOff Ghyll: #TUNES (Ping timeout for Ghyll[msx-osl-16-52.ppp.cybercity.no])
<water> well we'll just make concessions for lisp, then ;)
<water> since we don't care about performance right now
<water> btw, i'll link (from slate docs) to the code in your tunes account so you can update the online version whenever you want
<water> if you want, i'll link to some html page you write up to document it and whatever else you write
<eihrul> dern it... i need cora :/ hard to find papers otherwise
<eihrul> what was the other research paper search engine again?
<water> yes cora is quite useful. i don't recall if it has a mirror
<water> abi ri?
<abi> ri is researchindex at http://citeseer.nj.nec.com/cs?dd=2
07:00pm
-:- ult [ult@user-37kbakn.dialup.mindspring.com] has joined #Tunes
* water/#tunes considers getting a new squeak-powered handheld
<ult> Don't you already have a posh one?
<water> it was stolen
<water> :(
<ult> Dohhh
<water> yeah it was my favorite computer since the apple 2, though i hate winCE
<ult> still
<ult> Those are pricey
-:- Ghyll [karltk@msx-osl-16-26.ppp.cybercity.no] has joined #tunes
<water> yeah but it's the best thing i can use while at work
<water> which right now is about 60-70 hours per week
* water/#tunes pictures demo'ing his database software for the military bigwigs on a handheld computer hooked into a display projector :)
07:10pm
* eihrul/#tunes ponders why PhDs have such an affinity for one letter variable names.
* Ghyll/#tunes just rebooted his (crashed) computer by logging on ttyS1 using his wince.
<water> nice
<ult> eihrul: I've seen code that used 20 totally unrelated variables -- named x1 - x20
<water> yeah i like the terminal app for it
<ult> (The ScrollZ irc client patch is perhaps the most incredibly pathetic source code on the planet)
<eihrul> ult: yes, but "why?" is the question here...
<Ghyll> water: I even tried killing the offending process, but it seems my video card was wedged, so there wasn't much else to do. at least it saved me from fscking 40gigs..
<Ghyll> eihrul: because readability is inversely proportional to the amount of education of the author.
-:- Downix [unknown@dialup-209.245.103.104.Manchester1.Level3.net] has joined #tunes
<rares> Downix
<rares> hey
<Ghyll> eihrul: this is especially true in mathematics, and as we all know, informatics grew out of maths..
<eihrul> Ghyll: so... one might conjecture that there is some secret cabal of educators teaching students to obfuscate their ideas...
<Ghyll> ult: if they'd been r0-r20, I'd understand better :)
<Ghyll> eihrul: that's one theory. my theory is that obfuscation in coding comes naturally, and only through good thinking and good teaching can one learn to write readable code. 
<Ghyll> eihrul: which means that most educators don't teach programming well enough. 
<water> yes and academics don't think for other people
<Ghyll> eihrul: at least that should explain why some students write 300 line heap sort algorithms.
<Ghyll> water: some hardly think for themselves :(
<Downix> hmm
07:20pm
-:- dfsdf [root@210-55-47-123.dialup.xtra.co.nz] has joined #tunes
<water> hi root :)
<water> *ahem* i mean dfsdf
* Ghyll/#tunes haxx0rs through dfsdf's irc client.
<Downix> hrm?
<water> hey dx, what do you think of the "pocket pc" handhelds? which would you recommend?
<dfsdf> just checking out xchat 1.5.0
<dfsdf> yeah yeah, i'm ircing as root...
<dfsdf> so? windows users do the same
<water> heh
<Downix> I would recommend... the Palm or that new one by the guys that made the Palm
<water> bah.. i like running squeak at work
<rares> Pocket PC pfffffffffff
<dfsdf> pocket pc has quiet a few features tho
<water> well, i'm only looking at performance basically
<rares> I want one of Crusoe based ones
<water> yeah that would be ideal
-:- SignOff Downix: #TUNES (Ping timeout for Downix[dialup-209.245.103.104.Manchester1.Level3.net])
-:- Downix [unknown@dialup-209.245.103.104.Manchester1.Level3.net] has joined #tunes
<water> especially if they let you bypass the code morphing
<rares> the code morphing is not an extra layer
<rares> it's kind of an emergent property
<Downix> you can by-pass the code morphing
<Ghyll> Downix: can you exchange it with a new target ?
07:30pm
<Downix> yes
<Downix> I know one firm that's working on a PDP-11 code morph
<water> heh
<Ghyll> Downix: wow.. for fun ?
<Downix> imagine the power of the PDP-11 once again unleashed onto the world...
<water> proof that coding is still far too low-level
<Downix> Ghyll:  no, they need to replace one of their aging servers, and they don't want to get rid of all of their software
<ult> PDP-11? That gives a new meaning to the term aging
<Ghyll> Downix: wtf do they run that's so ancient ?
<Downix> Ghyll:  it works.
<Downix> Ghyll:  dunno
* water/#tunes pictures a specialized pda replacing a mainframe
* Downix/#tunes nods, funny ain't it?
<Ghyll> "what's that lcd device hanging in that picture frame over there ?" "it's our mainframe replacement"
<Downix> lol
<Downix> Ghyll:  thing is, a lot of firms think that a PC can replace a mainframe
<Ghyll> Downix: that's probably because a pc is cheaper, and people are full of hope :)
<Downix> yup
<Ghyll> Downix: I only know of one local firm that has run a mainframe.. but I'm not exactly well-versed in the computer installations around here. 
* water/#tunes thinks their full of something else
<Ghyll> Downix: most seem to use a network of workstations. 
<water> er they're
<Ghyll> water: well, that too.
* Downix/#tunes nods
<Downix> I like the mainframe concept
<Downix> can't afford it tho
<water> hm squeakers are offering money for bug fixes now
<water> and it looks like squeak will be fully modular in another month or so
<Downix> nice
07:40pm
<water> yeah every project and workspace is a new namespace where you can "lift" and "push" changesets between projects and the set of squeak modules
<water> and the image is just a set of modules that get dynamically loaded (this is the last feature to implement fully)
* Ghyll/#tunes notices that it's getting early. time for a nap.
-:- SignOff Ghyll: #TUNES (yawn)
07:50pm
<hcf> hohum
<water> hum ho
<water> anyone got something to discuss?
<Downix> I just got word from Amiga, Inc that they'd accepted our proposal
<ult> Cooool!
<ult> Congratulations
<water> yeah
08:00pm
<Downix> So it looks like Amie will be the OS on Eddas
<Downix> or, one of the OS's
<eihrul> ugh, it is a really bad thing when a supposedly-terminating calculation prints messages supposedly-forever on the screen
<Downix> multi
<Downix> annoying
08:10pm
-:- ESE [Maxx@204.134.104.32] has joined #tunes
<ESE> re's
-:- ESE is now known as Morpheus[Matrix]
<Morpheus[Matrix]> much better :)
<Downix> hey morph
<water> hm
<Morpheus[Matrix]> Ahh.. someone is here ! 
-:- water is now known as Neo
-:- Neo is now known as water
<Morpheus[Matrix]> Free Your Mind!
<water> heh
<ult> Get Creative!
<Morpheus[Matrix]> Any new OS' lately?
<water> you mean tunes stuff or otherwise?
<Downix> loads
<Morpheus[Matrix]> yeah
<Morpheus[Matrix]> Any in DJGPP C?
-:- SignOff dfsdf: #TUNES ()
<water> "yeah"?
<Morpheus[Matrix]> I'm working on SkyOS and ReactOS right now... But I'm always looking to help other groups...  Preferably ones that want to write server grade operating systems... desktops are nice too :)
<Downix> you're the guy from ReactOS?
<Morpheus[Matrix]> I'm on the team
* Downix/#tunes nods
<Downix> ok
<Morpheus[Matrix]> Kernel team and GUI that is...
<water> hm perhaps #osdev or #dolphin would be better for this
<Downix> I used to be on the team years ago
<Morpheus[Matrix]> Down: kewl :)
<ult> Morph: Tunes is vaporware -- theory -- more than likely will never exist.
<Morpheus[Matrix]> Mind telling me the basics...
<ult> Morph: However...there is a -large- quantity of theory =)
<water> maude and squeak are not vapor
<Morpheus[Matrix]> :)
<ult> maude and squeak are maude and squeak, not tunes
<water> and a large quantity of idiocy
* Morpheus[Matrix]/#tunes grins
<water> maude is pretty damned close to tunes... as close as scheme
08:20pm
-:- SignOff Downix: #TUNES (Ping timeout for Downix[dialup-209.245.103.104.Manchester1.Level3.net])
-:- mjm [grr@AC86AE5A.ipt.aol.com] has joined #tunes
<mjm>  Http://bbScriptz.VirtualAve.Net/IRC.htm
<mjm>  Http://bbScriptz.VirtualAve.Net/IRC.htm
<mjm>  Http://bbScriptz.VirtualAve.Net/IRC.htm
<mjm>  Http://bbScriptz.VirtualAve.Net/IRC.htm
<mjm>  Http://bbScriptz.VirtualAve.Net/IRC.htm
-:- mjm [grr@AC86AE5A.ipt.aol.com] has left #tunes []
-:- mjm [grr@AC86AE5A.ipt.aol.com] has joined #tunes
-:- mjm is now known as jr
-:- jr [grr@AC86AE5A.ipt.aol.com] has left #tunes []
<water> now there's a person to ban
<Morpheus[Matrix]> water: :)
* water/#tunes just finished a rather unrewarding tour of EFNet
<Morpheus[Matrix]> Stupid #freedows had me banned from EFNet.... it sux
<Morpheus[Matrix]> :)
<Morpheus[Matrix]> I had #trax ruff `em up
* water/#tunes is not surprised
-:- water [water@tnt-10-71.tscnet.net] has left #tunes []
-GlobalMessage(s@GlobalMessage)- WARNING: if you were just spammed by a user called "mlm" telling you to go to a url, do not go to it, as it is designed to have you enter your password for their use.
09:20pm
-:- SignOff Morpheus[Matrix]: #TUNES (Leaving)
<ult> Those freedows kids are pretty fscking stupid
-:- rares [rares@nwhn-sh1-port59.snet.net] has left #tunes []
* hcf/#tunes returns to hohum'ing
09:30pm
-:- water [water@tnt-10-71.tscnet.net] has joined #tunes
-:- SignOff ult: #TUNES (Leaving)
-:- SignOff hcf: #TUNES (Leaving)
-:- SignOff water: #TUNES (The Tao went that-a-way!)
-:- Kyle_L [kyle@cr168790-a.nmkt1.on.wave.home.com] has joined #Tunes
-:- water [water@tnt-10-71.tscnet.net] has joined #tunes
<eihrul> re
<water> hey
<water> sorry, i have little tunes steam tonight
-:- SignOff Kyle_L: #TUNES (Leaving)
11:00pm
* eihrul/#tunes is just playing around with graphs.
<water> why graphs?
<eihrul> compiler? :)
<water> oic
<water> play on, then :)
11:10pm
<eihrul> haskell is really nice for that
<water> i assume so
<water> i found a #haskell on efnet, but no one there seemed very experienced
<eihrul> 3 lines of code finds all the dominators in a control flow graph :)
<water> wow
-:- hcf [nef@me-portland-us813.javanet.com] has joined #tunes
<water> re
<hcf> water: didnt enjoy efnet? ;)
<water> oh yeah meeting someone who worked on the L4 uK and witnessing yet another researcher with an attitude due to funding is sooo much fun
<eihrul> eeg, liedtke?
<water> no someone else
<water> in .au
<water> she works on Mungi in particular
<eihrul> oh
11:20pm
<water> damn it i can't seem to shake this "crackpot" feeling
<water> it's been too long since i got any respect for what i do from people i can talk to face-to-face
<eihrul> well, that is not uncommon
-:- nate37 [nate@ppp-206-117-27-42.dialup.pcmagic.net] has joined #tunes
<eihrul> so just take comfort that it is their flaw and not yours >:)
<water> yeah well it's not like they treat me like a 2nd-class citizen at work, which is where i spend most of my time
<water> although this programming i'm doing will likely get me a very early promotion
11:30pm
<eihrul> okay, i can now find back edges and natural loops in graphs based on the dominator tree :)
<water> cool
<water> how is the process of adapting techniques for static languages to a dynamic language without explicit typing?
-:- NetSplit: fontana.openprojects.net split from varley.openprojects.net [11:47pm]
-:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [fontana.openprojects.net]
-:- Netjoined: fontana.openprojects.net varley.openprojects.net
-:- zarq [zarq@10dyn219.delft.casema.net] has joined #tunes
11:50pm
<eihrul> well, that's going to be the tricky part
<eihrul> but for the most part, the "trick" is determining when do to the optimizations
<eihrul> it is very easy to apply them otherwise
<eihrul> though another prospect is that more information can be gathered to facilitate traditionally static optimizations
<eihrul> like branch prediction and loop unrolling
<water> the trick is also to not waste time either calculating whether or not the optimization should be made or applying the optimization needlessly
<water> yes
<eihrul> well, yes :)
<eihrul> i like Self's threshold counting scheme for now
<eihrul> though, i also will probably have the compiler decide before hand if a piece of code is an inner loop that looks critical
<eihrul> if it is, instrumentation will be inserted (say in an *outer* loop that has tons of nested inner loops)
<eihrul> so that it is not intrusive, but still gives a good estimate of whether the inner loops are entered frequently
<water> hm
<water> i suppose you can highly-optimize the compiler using the compiler itself?
<eihrul> that i'm not too sure of yet
<water> actually, here's the interesting question:
<eihrul> but i'd like that to happen actually
<water> could you take object frameworks (like bmo's) that describe a static execution of an object system, and just convert it into a binary essentially?
<water> well yeah it'd be good for tunes
<eihrul> well, that's needed to make the system self-hosting :)
<eihrul> what i was planing was to have the compiler take a specification of certain objects to be saved into a self-contained image
<water> hm i guess my motives there were pretty transparent ;)
<eihrul> and then the compiler would just compile it into a static executable that implemented as much of the needed environment functionality in it to get it running
<water> "certain" objects?
<water> hm
<eihrul> well, more or less an entry point
<water> ok
<eihrul> then determining all reachable objects from there
<eihrul> or perhaps something more fine grained :)
<water> right, according to method call
[msg(TUNES)] newlog 2000.0422
IRC log ended Sat Apr 22 00:00:00 2000