IRC log started Wed Jul 7 00:00:01 1999 Hmm... but it's a sad fact that there is SOME part of the system which can't be changed without destroying the system. For me, it's my meta-metamodel, which is why I'm spending so much time trying to get it right... and to make it so minimal and flexible (to thus minimize the affect of my mistakes). [msg(TUNES)] permlog 1999.0707 On the other hand... jdl: sounds familiar I have realized that, at some level, there is a "Turing-completeness"-like thing for knowledge systems. So even if I do make a mistake with my meta-metamodel, I can probably create a new one and write one metaprogram to translate from the old meta-metamodel to the new one. hcf: ? i saw this interview w/ the owner of amazon, he called it regret minimization hcf: I haven't heard that term before, but the basic concept permeates modern software engineering techniques. hcf: oh, i thought you were talking about jdl's use of the phrase "on the other hand" :) Tril: I think I've talked myself into your point of view. :) jdl: a system designed to support an infrastructure of OTHER systems that continuously evolve, applied to itself. Some "good enough" set of human concepts WAS how I was going to approach reflection with Prism. Still, there's a lot of questions, for me anyway. And Brian wants to represent ALL information. jdl you need to change your global heap into a persistent store, and then start thinking about dynamism. you always go on about data, and structure, but you never mention any action Tril: The spec says it can be persistent. I hesitate to REQUIRE it to be persistent because Prism has other uses than for TUNES. ok, forget the persistence idea for now Tril: You mean like metaprograms? or ANY programs? * jdl/#tunes isn't sure what Tril means by action. 12:10am well, it was the question in my message, that asked about the programs that prism is outputting. But I should go so you can include an answer with that reply do you know what i mean? if not, include that in the reply too :) -:- SignOff jdl: #TUNES (Ping timeout for jdl[ultra1.inconnect.com]) we'll talk more later. * Tril/#TUNES is away: (afk) [BX-MsgLog Off] -:- jdl [jiml@ultra1.inconnect.com] has joined #tunes jdl: we'll talk more later. * Tril is away: (afk) [BX-MsgLog Off] Oh, oops. I didn't realize I was lagged until I got disconnected. I wrote a cool reply, but it's out of the buffer now... Damn... oh well. I should go to bed anyway. Goodnight! -:- SignOff jdl: #TUNES (Leaving) -:- sobit [sobit@207.203.38.27] has joined #Tunes hello all 12:20am hoy sobit how are you fine what's the point of this room read the 1st part of the topic free reflective computing system? see the site for more info i figured there would be a bot that would sing song :*) but i'll check the site -:- sobit [sobit@207.203.38.27] has left #Tunes [] >>> hcf [nef@me-portland-us1043.javanet.com] requested PING 931330089 from TUNES 12:30am !lilo:*! Hi all. Just to answer any questions, yes, I did take a job with LinuxCare. They're going to give me some time to work at Open Projects, and they're not taking over the domain in any way. I start work 9 August in SF. I'm on #lilostatus !lilo:*! for a few minutes if you have questions -:- SignOff hcf: #TUNES (Leaving) -:- smoke [smoke@7dyn62.delft.casema.net] has joined #tunes -:- SignOff Fare: #TUNES (Ping timeout for Fare[quatramaran.ens.fr]) -:- SignOff smoke: #TUNES (Ping timeout for smoke[7dyn62.delft.casema.net]) -:- Fare [rideau@quatramaran.ens.fr] has joined #Tunes * Fare/#Tunes finds it much more difficult to share the same /etc infrastructure on debian as it was on redhat -:- smoke [smoke@14dyn153.delft.casema.net] has joined #tunes hi smoke 07:10am hi fare 07:40am ls 08:40am * Tril/#TUNES is back from the dead. Gone 8 hrs 38 min 7 secs foo Fare -:- smkl [sami@MCDXL.rdyn.saunalahti.fi] has joined #tunes bar Tril i learned Linuxdoc DTD (i wrote a sgml file for my zelda specs, which aren't being released yet) 09:00am i need to learn Make better though ouch Make sucks. there is Cons, written in Perl... so why do you use it for tunes CVS btw, if you use linuxdoc for your HOWTO you are supposed to switch to DocBook. ... 09:10am hey look at this quote from distributed.net: "Thanks to everyone for making the upgrade necessary." Tril: I use it because I know it, because it's there, and because if I'm to move to something else, it should be tunes... ok Hum, I've a french friend who's also working on a reflective OS for his thesis I recently convinced him that free software was the right way to go he's looking for a server for his stuff (mostly in french, hum) I feel lame and useless why is he looking? gimme something to code because he wants one in france? why is he looking??? i mean why didnt you tell him to use bespin already I proposed that bespin host his server I did propose, but 1) I wanted your confirmation, since you're here confirm and 2) he's looking for a DNS, too dhs.org :) yeah, I proposed that, too! (or eu.org, or tunes.org, or openprojects.net, or whatever he likes) actually you can send him to www.on.openprojects.net if he doesnt like us. They do pretty much the same thing (except for the shell accounts and DNS administration). and they have much faster servers. (VA Research pays for it) I'll direct him to them... but if he uses us then he gets you as his root, which you can't beat 09:20am (if he lives near you or whatever) any ways, he should join the tunes list and come to IRC we definitely need more researchers from the academic community he leaves in the south of France. around 700 km away and of course, he's in the margin of the research community :( even worse than I! of course, it's more classy to say "fringe" or "cutting-edge" than "margin" :) where's that "thank for making the upgrade necessary" statement? hmmm. so you are saying most researchers spend time researching well-known things? That sounds wrong. it's on the announce@lists.distributed.net list not only does it sound wrong, it IS wrong. well, do they? yes and no let's say they'remostly looking at new refinements to well-known things Rocky's Lemma of Innovation Prevention: "pgrade necessary" statement? hmmm. so you are saying most researchers spend time researching well-known things? That sounds wrong. it's on the announce@lists.distributed.net list not only does it sound wrong, it IS wrong. well, do they? GGGGGrrrrrr! -:- mode/#tunes [+o Fare] by ChanServ Thanks in part to our rapid growth and expansion, we've once again felt the need to upgrade the keymaster hardware. This past week saw us adding new drives, a cd-r unit for archiving logfiles, a new CPU, and a larger case. Thanks to everyone for making the upgrade necessary. -:- Fare was kicked off #tunes by Fare (flood!!!!) -:- Fare [rideau@quatramaran.ens.fr] has joined #Tunes rocky's lemma? paste the wrong clipboard? Rocky's Lemma of Innovation Prevention: "Unless the results are known in advance, funding agencies will reject the proposal." oh my results are known in advance but the funding agencies wouldnt like them :) no, the mouse has 2 buttons, and pasting is with both at the same time, 2-button mouse not work well in X? Well, my 3-button doesnt work well in GPM, so there! which is not reliable, and often leads to cutting random stuff just before pasting (it's not a mouse, but a fscking DELL "touchpad") It leads to pasting the same text twice in a row, or not pasting when i hit the button, so i hit it 3 times and then it pastes extra * Fare/#Tunes goes edit his Xconfig to change the mouse button bindings... 09:30am do you know how to do that? there's an Emulate3Buttons option , I am not sure what you want to change, disabling that will turn off cut-paste! no, i dont know how to do it -:- SignOff Fare: #TUNES (Ping timeout for Fare[quatramaran.ens.fr]) ok i told gpm to force 3 buttons now it might work better -:- Fare [rideau@quatramaran.ens.fr] has joined #Tunes -:- SignOff Fare: #TUNES (Connection reset by pear) -:- Fare [rideau@quatramaran.ens.fr] has joined #Tunes Hum. ok, there I am I hope this new mouse binding will prove less fault-prone >>> Tril [dem@sloth.wcug.wwu.edu] requested PING 931365688 146855 from #TUNES 09:40am is the freeos server on a 28.8 link? Dunno. Anyway, it sucks -:- Fare has changed the topic on channel #tunes to: Reflective Computing System || visit #tpp freeos is the typical YAFOP very freedows-like It will be free to download and distribute, but its source code will not be freely available as with Linux. Hum. I tried to install CMUCL and failed :( well, even if it didn't suck, I'd still ignore it due to that not even a true FOP! much like Brix? they actually meant free as in no cost? ha. see my message on gnu.misc.discuss (cross-posted to cybernethics@) you mean the thread on tunes@tunes that RMS was in recently? there's another fork of the thread on the newsgroup?? no, a recent message I cross-posted (today) then i probably weon't see it on my news server today? what do i search for perhaps not. But it must be in the cybernethics archive (not on the web yet, tho) so what do you think of tpp's AKOS game project hum, not much, currently I've always had a games project for tunes (even before it was tunes), but I don't think much of multi-player games do you know what it is? only in as much as it is described on their page 09:50am AKOSintro.html? something like that did you read FAQ 1.9 ? no i have 1.8 cvs update (or cvsweb)... -:- hcf [nef@me-portland-us747.javanet.com] has joined #tunes have you seen pliant's builtin webserver? hmm well, i know it works, beacuse it is used on their site * Fare/#Tunes still hasn't found a browser to satisfactorily replace netscape. i reported a bug in it when their site didn't work in lynx lynx will make do _most_ of the time, but not always... tried Mozilla M7? no, wasn't a .deb -- besides, M6 was scary the engine may be fast, but they redraw so much that any speed gain is lost 10 times. plus it is triggers a race-condition bug in my XF86's bitblit routines... triggering bugs in other stuff is good how else would they be fixed? sure. But I don't follow XF86 development, and haven't reported the bug. Maybe I should... oh, and to be sure it's XF86's fault, I should test with another server... ok I have read the diff for 1.9 it is good eto horosho eto horosho? 10:00am abi: Translate Russian hum. Mozilla is _so_ slow. it means "it's good" Too bad closure isn't multithreaded... Fare: maybe you look forward to IE port? why not, if it's opensource? actually, since my current netscape isn't opensource, why not IE at all? i think it works in wine Tril: abi doesnt have translate ... yet, she'll have a babelfish thing when she's upgraded hcf: i was joking anyway i know, just mentioning it * Tril/#TUNES is away: (afk) [BX-MsgLog Off] Fare: here?\ somehow Hum, I've a french friend who's also working on a reflective OS for his thesis <-- who is he? maybe I should try wine and IE, someday -- when I buy a few Gigs or additional disk space... hcf: Henri LESOURD any web presence right now? * Fare/#Tunes is not used to his new mouse bindings -- but already they provide with less horrible mistakes hcf: tiny one: http://www-poleia.lip6.fr/~lesourd 10:10am abi: lesourd is Henri.LESOURD@lip6.fr at http://www-poleia.lip6.fr/~lesourd/ is babelfish the best available french->english by url ? he sent me source&code for a reflective system he built atop M$-DOS * Fare/#Tunes still needs write an anti-OO FAQ and an anti-XML FAQ... and anti- faqs Fare: is babelfish the best available french->english by url ? * hcf/#tunes assumes it isnt anti-YAFOP for instance anti-MicroKernel hum, has the new OS FAQ maintainer produced anything? not seen anything as yet perhaps u could collaborate w/ him I sent him mail w/ no response refer to it as the comp.os.research faq 10:20am Fare: whats hammourabi? his rule-based system hammerabi ;) poor abi! go ahead, i like it rough * Fare/#Tunes hammers abi oh oh oooh Tril: something bugs me... Tril: when is cybernethics@ reindexed? 10:30am Can't open perl script "/var/www/htdocs/tunes-bin/wilma_reindex": No such file or directory grrr! sh: exec: /var/www/htdocs/tunes-bin/wilma_striphtml: cannot execute: No such file or directory grrr! -rwxr-xr-x 1 tunes tunes 3132 Mar 22 15:37 wilma_hiliter* -rw-r--r-- 1 tunes tunes 5710 Mar 22 15:37 wilma_reindex hcf: yes, I know, but the scripts don't shouldnt that be +x'd? no, the problem is change of path... oh Fare: icuc, http://www.parc.xerox.com/spl/projects/oi/ yeah, it's in the review page damnit Hum, where's Tril, when you need him!? ;-> afk some script, somewhere, invokes stuff in tunes-bin, that doesn't exist anymore How to find the culprit, but by greping the whole of bespin? 10:40am Ok, found the culprit Now, bespin must reindex everything that was misfiltered since the move of tunes-bin! ouch how to do that? (culprit var symlink to /etc/glimpsefilter* well, glimpse*filter*, that wasn't updated) this'll fix the review list too, right? -:- SignOff smkl: #TUNES (Ping timeout for smkl[MCDXL.rdyn.saunalahti.fi]) 10:50am I hope so but not retroactively, unless I understand how to force reindexing with the right parameters... I'd rather Tril be here, he knows better Fare: changes to rev-re, the CSLI url is 404 thus MIA, brian cantwell smith bio and horizon day url fixes: http://www.cs.indiana.edu/horizon/960112/bcsbio.html, http://www.cs.indiana.edu/horizon/960112/index.html rev-re? -:- smkl [sami@MCDXL.rdyn.saunalahti.fi] has joined #tunes review: reflection it has been said that reflection is a property of a system that can refer to itself and manipulate its state or rough on my brain 11:00am abi: shh Someone oughto maintain that reflection page... * Fare/#Tunes is having bespin reindex everything. Hope I didn't do a BIG MISTAKE(TM) with rm -rf ... abi: abi? somebody said i was the little girl borg with a big brain or #tunes' personal little whore or gaklosmontic or flurivostuginuous or xyvarestoplik or an Artificial Idiot or braindead or a crack addict infoslut or stubborn abi is reflective! ...but abi is the little girl borg with a big brain or #tunes' personal little whore or gaklosmontic or flurivostuginuous or xyvarestoplik or an Artificial Idiot or braindead or a crack addict infoslut or stubborn... well, not an interesting kind of reflection, anyway 11:10am Fare: whats industrious evaluation? never heard about it something like eager evaluation? mentioned on http://www.parc.xerox.com/pubs-hst.html related to lazy i'd suppose 11:20am eager evaluation is opposite to lazy icuc, http://www.cs.york.ac.uk/~stuart/research_links.htm hum, the reindexing isn't done yet :) icuc, http://www-cat.ncsa.uiuc.edu/~yoder/papers/patterns/Evolution/metamorphosis.html 11:30am Fare: here? hcf: i don't know abi: neither do i because you're both ignorant! oh? how so? if you don't know, you're ignorant. not that there's anything wrong about being ignorant the more we know, the more we know that we ignore... ok u were wrong about the open impl link oh, was I? you mean, it wasn't on the Languages page? hold in rev-pl, change: http://www.parc.xerox.com/PARC/spl/eca/oi.html -to-> http://www.parc.xerox.com/spl/projects/oi/ and s/MIA// the reference oh, ok tracking MIA links? not really 11:40am http://www.cs.indiana.edu/hyplan/jsobel/rop.html http://www.hec.unil.ch/people/pbonzon/aaai95.htm Fare: ur current bookmarks rnt online r they? not the latestest version, but they are ~fare/pointers.html why? to see if u already have urls 11:50am well, I just updated them, in case you care... hum. Maybe I should make a read-only copy of my cvs ~/etc and ~/fare world-accessible? perhaps well, well, well. I should add a crontab on bespin to update my stuff, then. 12:00pm http://www.cs.rice.edu/CS/PLT/Publications/index.shtml 12:30pm isn't it in Languages.html yet? oops hum, not Publications/ but still they are a great lab if you can go there, go! finding urls that are already in the review does nothing for tunes, but it does show i'm on a right track especially this one :) wow, the rev-pl todo list is huge and there's more in my pointers.html that I haven't migrated there yet... wow, my link to bespin is _slow_ 12:40pm should i keep adding new urls for ppl to review? where? or wait for the current load to be processed -:- eStormy [stormy@rain.futuresouth.com] has joined #Tunes hum. reindexing went fine, it looks. my selective rm -r seems to have worked correctly :) luckily where what? 12:50pm pre-htmlized mailing-list archives should i keep adding new urls for ppl to review or wait for the current load to be processed? 01:00pm well, it's not a priority, but if you hit good stuff, do post it! speaking of priority, prioritizing all tunes' todos would be nice you're welcome! yet another "feel free"? yup only u or tril could properly do such 01:10pm -:- _ruiner_ [nate@t-73-211-40.dialup.wisc.edu] has joined #tunes http://www.cs.rice.edu/CS/Verification/ hoy _ruiner_ <_ruiner_> hi hcf.....whats up? _ruiner_: not much, u? -:- SignOff eStormy: #TUNES (eStormy has no reason) <_ruiner_> nada has anyone read the reiserfs paper? 01:20pm not i a great paper about disk management. * Fare/#Tunes re-subscribes to gclist... -:- binEng [Anders@dialup83-2-24.swipnet.se] has joined #tunes 01:30pm hoy binEng hoy hcf! 01:40pm -:- SignOff hcf: #TUNES (Leaving) -:- SignOff binEng: #TUNES ( ) I'm laaagged to bespin 03:30pm Hum. I put that crontab, so that my files are visible through cvs. -:- SignOff _ruiner_: #TUNES (Leaving) No risk of modification, since they are rsync'ed from quatramaran.ens.fr daily... I just hope nothing goes wrong... 03:40pm -:- hcf [nef@me-portland-us815.javanet.com] has joined #tunes hi, hcf! hoy Fare hcf: in case you care, you can cvs -d nef@bespin:/cvs checkout fare ok everything you ever wanted to know about me that you didn't dare ask :-) -:- _QZ [brand@p0wer.qzx.com] has joined #tunes hoy _QZ QZ: if BRiX isn't free, you should mention it in the review... <_QZ> brix is free <_QZ> and it is opensource to an extent _QZ: ok, where's the src? "to an extent"??? what does that mean? I don't mean free of charge I mean free <_QZ> well u get the source, and u can modify that source, but u cant distribute the source or yer modified source 04:10pm then it's not free, dammit! <_QZ> yes it is no it ain't Fare: the reindex'ing of review@ looks good, thx for fixing that hcf: you're welcome! hcf: since when was it br0ken? Fare: i told u when i noticed it the ~6th QZ: re-read the FSF's definition of free, or the OSI's DFSG... <_QZ> Fare: it isnt free to the point where u can take my work and make money off it hcf: then the cause is well-understood you can't take it and use it at all <_QZ> yes u can no one's gonna use something they cannot trust to be able to use <_QZ> u can change it in any way u want that's not enough if you can't share the changes unless your only activity is OS development, and you hate and despise people so much that you wouldn't share anything with them but then, you won't take BRiX, to begin with -:- SignOff smkl: #TUNES (sleeping ...) <_QZ> u can put the changes back into my source no, you can't because you can't be sure that that QZ guy will be there <_QZ> ya u send me yer changes and i add them in that he will accept the patch or whatever <_QZ> if i die then the source goes gpl nah that won't do it's even worse than Sun's community license except for the case when you die <_QZ> right i might not accept the patch <_QZ> that is why i want all patches sent to me hello big brother <_QZ> so i can maintain rules B&D guy? <_QZ> it keeps the os consitent <_QZ> b&d? * hcf/#tunes images freebrix projects popping up left and right if you want to maintain rules, do it by deserving authority, not by imposing it! s/images/imagines/ b&d! i guess b&d is bondage and discipline approach to programming imposing it makes you despicable abi: _QZ is also a tyrant okay, hcf. 04:20pm only if you expect to not deserve respect can it help only nothing can help you if you do not deserve respect! <_QZ> ok how bout this did you read the ending comments of the Tunes charter? <_QZ> u can distribute the source but u cant charge for it, not even media costs did you read the Tunes Charter? "not even media cost"? Are you going to sue AT&T for charging communication while downloading BRiX? <_QZ> ya that basically means u can have it on the net but u cant sell cd's unless u give them away free you could make the media cost restriction only for binaries... but even then, that'd be stupid just make it free! <_QZ> oh and i forgot to mention that internal distribution is also ok <_QZ> so a company can modify it and use it internalling the more people use it, the more people know you, the more your work and opinion and support is valuable on the market <_QZ> i dont want ppl to know me <_QZ> or respect me <_QZ> or give a flying fuck about me what do you want, then? _QZ: they'll try anyway, they're dumb like that don't you understand that it is respect that earns you money? <_QZ> i dont want money <_QZ> i want to watch bill gates fall promise; mindshare; respect; trust; that's the most precious good you can have and sell! I don't BRiX will help much, unless it's free software. hackers are no more stupid enough to believe in proprietary software. Soon, neither will be businessmen s/BRiX/think BRiX/ <_QZ> see the reason i dont want brix being distributed is to prevent flavors like linux has why prevent flavors? flavors are the OPPORTUNITY to provide with a better one <_QZ> well it prevents those little "mine is better than yers" wars they are OPPORTUNITY on the trust market, to sell _your_ flavor, that other people will trust <_QZ> and it keeps everything consitent <_QZ> consiStent damnit why prevent such war, yet make a bigger "BRiX is better than foo" war?