IRC log started Thu Feb 10 00:00:01 2000 try it multiple times in mulitple non ~/html dirs you mean run it again? oh get a new copy from /tmp/slate [msg(TUNES)] permlog 2000.0210 it's trying (unsuccessfully) to access the template etc from your directory er navbar oops 1sec probly complaining about the declaration of $dir too reget it cool "no files to process" was that 1st or 2nd time? well i already processed the files, so 2nd k and i just tested it one one file from ../bu s/one one/on one so its working fine now yes thank you :) 12:10am put ur description and keywords in the template now oh yeah 2am here... 'nite -:- SignOff ult: #TUNES (Leaving) any special formatting for keywords' contents? -:- NetSplit: merril.openprojects.net split from adams.openprojects.net [12:16am] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [merril.openprojects.net] -:- Netjoined: merril.openprojects.net adams.openprojects.net -:- smkl [sami@glubimox.yok.utu.fi] has joined #tunes -:- abi [nef@bespin.dhs.org] has joined #tunes -:- Plundis [plundis@130.238.23.252] has joined #tunes -:- water [water@tnt-10-50.tscnet.net] has joined #tunes -:- smoke [smoke@15dyn71.delft.casema.net] has joined #tunes -:- hcf [nef@me-portland-us1011.javanet.com] has joined #tunes content="aword, a phrase, blah blah" should be fine k 12:20am -:- NetSplit: varley.openprojects.net split from adams.openprojects.net [12:23am] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [varley.openprojects.net] -:- Netjoined: varley.openprojects.net adams.openprojects.net -:- smkl [sami@glubimox.yok.utu.fi] has joined #tunes -:- abi [nef@bespin.dhs.org] has joined #tunes -:- Plundis [plundis@130.238.23.252] has joined #tunes -:- water [water@tnt-10-50.tscnet.net] has joined #tunes -:- smoke [smoke@15dyn71.delft.casema.net] has joined #tunes -:- hcf [nef@me-portland-us1011.javanet.com] has joined #tunes hm well i'm gonna concentrate on groking rewrite now, unless you have anything else reget /tmp/slate/slate-template.html i put Slate in the keywords list, in case s-engines dont parse that from elsewhere why wouldn't they parrse the f***ing title? they should but just in case -:- NetSplit: merril.openprojects.net split from hogan.openprojects.net [12:28am] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [merril.openprojects.net] -:- Netjoined: merril.openprojects.net hogan.openprojects.net -:- smoke [smoke@15dyn71.delft.casema.net] has joined #tunes refs needs in slate in the title ok, updated the site with the new template * water/#tunes thinks about this HM. it seems someone's doing all the suggesting about web-design to people better off just working on the language design :) 12:30am gimme a break. my pages already look better than those for Cecil * hcf/#tunes shuts up -:- SignOff water: #TUNES (Read error to water[tnt-10-50.tscnet.net]: Connection reset by peer) -:- NetSplit: sterling.openprojects.net split from irc.linux.com [12:36am] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [sterling.openprojects.net] -:- Netjoined: sterling.openprojects.net irc.linux.com -:- Plundis [plundis@130.238.23.252] has joined #tunes -:- NetSplit: sterling.openprojects.net split from verne.openprojects.net [12:39am] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [sterling.openprojects.net] -:- Netjoined: sterling.openprojects.net verne.openprojects.net -:- Plundis [plundis@130.238.23.252] has joined #tunes 12:40am -:- NetSplit: wang.openprojects.net split from sterling.openprojects.net [12:53am] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [wang.openprojects.net] -:- Netjoined: wang.openprojects.net sterling.openprojects.net -:- Plundis [plundis@130.238.23.252] has joined #tunes -:- NetSplit: merril.openprojects.net split from irc.linux.com [12:57am] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [merril.openprojects.net] -:- Netjoined: merril.openprojects.net irc.linux.com -:- smoke [smoke@15dyn71.delft.casema.net] has joined #tunes -:- SignOff thomas: #TUNES (Ping timeout for thomas[193.217.63.152]) -:- NetSplit: irc.linux.com split from adams.openprojects.net [01:22am] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [irc.linux.com] -:- Netjoined: irc.linux.com adams.openprojects.net -:- hcf [nef@me-portland-us1011.javanet.com] has joined #tunes -:- abi [nef@bespin.dhs.org] has joined #tunes -:- smkl [sami@glubimox.yok.utu.fi] has joined #tunes -:- Plundis [plundis@130.238.23.252] has joined #tunes -:- NetSplit: adams.openprojects.net split from devlin.openprojects.net [01:24am] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [adams.openprojects.net] -:- Netjoined: adams.openprojects.net devlin.openprojects.net -:- smoke [smoke@15dyn71.delft.casema.net] has joined #tunes -:- NetSplit: lackey.openprojects.net split from devlin.openprojects.net [01:39am] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [lackey.openprojects.net] -:- Netjoined: lackey.openprojects.net devlin.openprojects.net -:- hcf [nef@me-portland-us1011.javanet.com] has joined #tunes -:- abi [nef@bespin.dhs.org] has joined #tunes -:- smkl [sami@glubimox.yok.utu.fi] has joined #tunes -:- Plundis [plundis@130.238.23.252] has joined #tunes -:- smoke [smoke@15dyn71.delft.casema.net] has joined #tunes -:- thomas [thomas@193.217.63.152] has joined #tunes -:- SignOff hcf: #TUNES (Leaving) -:- NetSplit: adams.openprojects.net split from forward.openprojects.net [02:24am] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [adams.openprojects.net] -:- Netjoined: adams.openprojects.net forward.openprojects.net -:- smoke [smoke@15dyn71.delft.casema.net] has joined #tunes -:- thomas [thomas@193.217.63.152] has joined #tunes -:- SignOff thomas: #TUNES (Ping timeout for thomas[193.217.63.152]) -:- thomas [thomas@193.217.63.152] has joined #tunes -:- smoke` [Xes@130.161.32.9] has joined #tunes -:- smoke` [Xes@130.161.32.9] has left #tunes [] -:- Fare [rideaufr@esmeralda.enst.fr] has joined #Tunes -:- Fufie [stig@tunnel-44-15.vpn.uib.no] has joined #tunes -:- SignOff smoke: #TUNES (Ping timeout for smoke[15dyn71.delft.casema.net]) -:- rares [rares@wtrb-sh5-port77.snet.net] has joined #tunes -:- smoke [smoke@16dyn235.delft.casema.net] has joined #tunes -:- smokie [tw026024@zaalf12.twi.tudelft.nl] has joined #tunes -:- NetSplit: adams.openprojects.net split from merril.openprojects.net [07:51am] -:- BitchX+Deb1an: Press Ctrl-F to see who left Ctrl-E to change to [adams.openprojects.net] -:- SignOff rares: #TUNES (KVirc 1.0.0 Millennium BETA 3) -:- SignOff smokie: #TUNES (BitchX-75+Deb1an -- just do it.) -:- Kaufmann [newbie@dial809.infolink.com.br] has joined #tunes aw, eihrul isn't here I wanted to mock him what has he done? he said it would take a long time for me to write an interpreter for the language we (me and him) were discussing yesterday and I've already written 30% of it since last night granted, most of the functionality will just be provided by the underlying Scheme system... at least until I start using the MOP to modify _that_ too. hello? hello is a multi-threaded operating system written in Standard ML and can be found at http://www.ics.hawaii.edu/~esb/prof/proj/hello/ egads 08:30am hehe (upside-down-A) character string S, (flipped-E) operating system O such that name(O) = S (Kaufmann's Conjecture) :) sorry.. had to pick up a phone that was making loud ringing noises starting off with Scheme or a Lisp is a quick way to get waay ahead in language design you should get it fixed. My phone has never been known to make loud ringing noises; I'm perfectly satisfied with it Fufie, you bet :) but what is the alternative, really? Coding in the support for the primitive data types in what, (I shudder at the thought) C? no, thanks incremental development starting from a reflective Lisp system is the way to go. BBL -:- SignOff Kaufmann: #TUNES (kathyanne gets kick'd/ as leaves fall on still waters/ she is here no more.) 08:40am -:- smoke is now known as avoozl When I want to do something mindless to relax, I reinstall Windows 95. -- Jean-Louis Gassee Gakuk! -:- SignOff smkl: #TUNES (Ping timeout for smkl[glubimox.yok.utu.fi]) what was that Alan Kay quote about smalltalk making all programmers users or something? -:- smkl [sami@glubimox.yok.utu.fi] has joined #tunes -:- thomas [thomas@193.217.63.152] has joined #tunes for our finnish friends: what was that Alan Kay quote about smalltalk making all programmers users or something? 09:50am -:- avoozl is now known as smoke -:- rares [rares@wtrb-sh2-port140.snet.net] has joined #tunes KVirc 1.0.0 'Millennium' BETA 3 by Szymon Stefanek umm looks like you can do taxes over the net from Linux :) 11:30am -:- SignOff rares: #TUNES (Ping timeout for rares[wtrb-sh2-port140.snet.net]) -:- mibin [mibin@an1-846.tiscalinet.it] has joined #tunes -:- rares [rares@wtrb-sh4-port147.snet.net] has joined #tunes -:- mgx [mgx@vd-g404.student.utc.sk] has joined #tunes * Fare/#tunes alien -i ocaml and hevea RPMs on bespin -:- SignOff mgx: #TUNES (mgx) zsh: segmentation fault ocaml ouch 01:10pm maybe I ought to have used the "RH 6.1" rpm instead of the "RH 6.0" one? bespin has old libc ... grumble * Fare/#tunes gets the .tar.gz, then isn't there a 2.04 deb now? ld.so doesn't complain smkl: in potato. But bespin is slink I wouldn't want to dist-upgrade a machine in .wa.us when I'm in .fr It must be possible, but doing it without either experience or a fallback tactic would be preposterous or disasterous * Fare/#tunes plays brightside.wav once again * Fare/#tunes just wanted to run hevea on bespin * Fare/#tunes is cleaning WhyNewOS and converting it to LaTeX * Fare/#tunes is also looking for nice quotations to put below each chapter's title (what's the name for these, already?) 01:20pm RT-Linux methods patented on /. do they license for use in any DFSG-free software? 01:30pm yes but they make restriction to BSD and HURD which is really stupid since Debian BSD and Debian Hurd are coming out evil 01:40pm * Fare/#tunes make ocamlopt.opt * smkl/#tunes installs his prompt generator hehe prompt generator? it generates $PS1 02:00pm with a fortune cookie in each, from a special cookie jar? no, that would be precmd() bespin doesn't have fortune ?! yes it does -:- SignOff rares: #TUNES (Read error to rares[wtrb-sh4-port147.snet.net]: No route to host) even before I did an apt-get install maybe you're using the wrong path? hmm it's just not in my path -:- rares [rares@wtrb-sh3-port80.snet.net] has joined #tunes /usr/games/fortune indeed * Fare/#tunes doesn't understand why "games" are in /usr/games, except for all those X arcade stuff in /usr/X11R6/bin UN*X filesystems suck, anyway 02:10pm yeah we need /usr/X11R6/games go tell that to the FSSTND guys (whatever the name is today) maybe #debian people will care -:- SignOff mibin: #TUNES (Ping timeout for mibin[an1-846.tiscalinet.it]) I think we need a meta file format with meta paths we need get rid of the file concept file equals collection of bytres sorry can't do it it would cause the universe to iumplode who cares about collections of bytes? I want objects. and in what form shall we store those oifbjects it's still a collection of bytes *WE* won't store them. The system will. in what form and it could use trits instead of bits, or 20-bit words, for all I care dude were talking about breaking rules as solid as 2+2=4 nope, win2k would have caused the destruction of universe already if that was possible with current technology It could use quantum logic, or dimension 2.5 fractal data storage, instead of linear sequences actually they're electronic signals but asnyway I don't give a fucking damn, as long as it works you can't control analog as easily or manipulate it as easily in a meaningful way as digital it could use the strong force instead of electric signals, or armies of monkeys, for all I care well the designer needs to give a damn NO the IMPLEMENTER does it could be chemical agents instead of electrical signals the DESIGNER needs to be somewhat down to earth the user can do almost anything they like it could be fluctuations of The Force the DESIGNER needs to solve the problem of the USER in a way _affordably feasible_ by the IMPLEMENTER and how would you record those fluctuations he doesn't need make things STRAIGHTFORWARD to the latter 02:20pm files are the cheap solution to the implementer -- quite justifiable 20 years ago. Completely obsolete nowadays, yet historically entrenched the IMPLEMENTER is living within the limits of the laws of the universe which are much more conservati ve than say everyday experiences Next you're going to tell me the Universe is obsolete you don't get it. I do get it have you read the glossary entry on orthogonal persistence? things are built from blocks have you ever used an HP28? or maybe even a wordprocessor? yes I have used word processors do you look at the sequences of bytes in a word processor? no but the computer program has to in order to put it on the screen btw, disks are usually divided in cylinders, "sides" and sectors, which is not sequential they're still read sequentially once they're mapped and searched and memory is organized as a dynamic tree of cache, not as a sequential array and data on disk is scattered among several blocks in a non-sequential way and entries in a hash-table are not sequential and the video gun scans across the screen not randoml;y _some_ things are intrinsically sequential _some_ things are intrinsically hierarchical _some_ things are intrinsically FOO mapping everything to that FOO concept is stupid but everything is a sexp ! no, everything has a sexp _representation_ here's the problem the storage technologies and the processing technologies are serial no matter how parallel we try to make them and every time we add a parallel layer we need to add a serial layer (cause and effect you've heard of it not?) to hook it to the real world don't confuse the _thing_ and its _representation_ meta-confusion is the root of OO evil. we you can only store a representation silly rares: there's not just _one_ representation rares: there are many. The system knows best which to use. 02:30pm rares: ever heard of distributed shared memories? Not very sequential. here's the problem sequential systems exist because the world we live in is a world of cause and effect I could equally say that the world we live in is intrinsically parallel sure and my shoes are scattered all around the room but they're only meaning full if they're paired or that there is no cause and effect, just history if we can't conbtrol then how can we possibly use it? or that sequential time is a local concept that relativity ensures cannot be globally shared rares: we can control things w/o their being sequential the arrow of time still flows forward and/or we need not control them down to such details that global sequencing be needed rares: the point is about local vs global, not about one-wayness independent things can happen in dependent things must happen in sequence insert parallel someplace there when you have two computers talking to each other, what "intrinsic sequencing" do you see among the memory constituted by their distributed system? of course it is you can't read distributed memory without mapping regions together which make sense rubbish. take a byte in the first computer and another in the other, which is intrinsically at lowest address? that's not the point memory isn't intrinsically sequentially addressed. tyou still need the next important meaningful byte no, it could be decimal stuff instead of bytes or it could use basis three, or 2^20 addressing could be not flat (a URL is a non-fixed-size address for data that isn't fixed-size either) next you'll telme the fact that high energy physoicists keep inventing more particles to satisfy their odd ball theories like Greeks inventing Gods doesn't seem odd there has existed decimal computers (ENIAC, for instance). And there exists 20-bit computers (F21, for instance) there needs to be a consistent system otherwise yiou only get special purpose machines 02:40pm no. there needs be a consistent way to interchange data the way data is actually stored doesn't matter surely, data sent over the network, or saved on a tape, must be marshalled into a stream no one denies it it does if you want one machine capable of handling all of it without needing to buy a decimal drive for some stuff and a binary drive for other stuff and boy the cycles wasted on conversion that doesn't mean the internal format during normal operations be anything remotely similar you don't expect RMI-marshalled streams to be the internal format of Java implementations. well let me put it to you asnotherway would you rather have a video memory grab to be your internal wordprocessor format nmaking it impossible to change video memory grab would be inefficient and video-size dependent the problem is most peoplright oops meant right but having the wordprocessor format be internally lists of objects in a persistent store seems right to me btw, the (what's its name? cedar?) system at Xerox did exactly that and framemaker (or what's that LISP-based word processor, already?) does that, too and framemaker saves to alien technology filesystems? or to files exactly. the filesystem is an alien system with which to interchange using byte streams. it's not the native format rares: read J. Liedtke's article on 15 years of experience with Eumel maybe you'll get a clue may be here's my point any squeak expert around? it seems to me that squeak images store source internally as structures, not text, don't they? 02:50pm although of course you can "export" the source as files on the filesystem, if you like -:- eihrul [lee@usr5-ppp89.lvdi.net] has joined #tunes i like files, collections of files, and collections of collections of files, because I don't need a proprietary patent license to access them file and collections of file make everything trivial in terms of concept you can argue all you want about triviality in volume but in concept it's directly accessibile I'm a designer, a coder, and a user and I sdon't need some CEO from some company telling m telling me they've invented something I can't build from building blocks rares: who talks about proprietary patents? I'm talking about free software that's not what I mean rares: indeed proprietary software has been a TREMENDOUS brake to technological progress. I've argued that in my article mpfas http://fare.tunes.org/articles/ll99/ * eihrul/#tunes is sick of moronic valentine's day of fools advertisements... why are you looking at them i'm not... but they pollute the radio you try to listen to some metallica... and they're selling roses! hmm tunes mlist has produced a book about combinators smkl: anything wrong with that? nope maybe iepos should sort it out and make a real e-book Fare: you're aware that'd it'd probably take as long as Tunes itself has taken to do that? :) this is funny rares: btw, the Windows NT filesystem layout is proprietary. here I am arguing general purpose and he's arguing against and his paper agues for general purpose against me eben 03:00pm even even uh? okay here's my view on all this (nice paper btw): which paper? which paper is that? abi: shut up, you bitch! mpfas oh, thanks did you just read it??? hey be nice to abi no the whole but I get it see what you're going for has one problem until you design a machine that can boot off a meta object (which is possible) you'll always have a dirty system dirty as in the way Windows runs on top of DOS the system will *always* be dirty. It's just a matter of being less and less dirty. Much less than current systems. 03:10pm actaully for the sake of cleaning up the monstrous redundancies in 3-D programming I have half a mind to create phase space RAM -:- pyro [tcn@cci-209150250032.clarityconnect.net] has joined #tunes hey gusy phase space RAM ? Finite State Machine RAM pyro! gusy is not here abi: gusy? rumour has it gusy is not here guys he mant pyro: what do you mant? heh.. what's up? nothing arguing about dead space and objects don't let me interrupt :) oh no the more the funnier Fare, anything that can deflate gcc's complexity is good for me since I'm meant to write a meta-GC, I'd rather be talking about space and dead objects pyro: retro doing things other than crashing yet? eihrul: the keyboard driver works now I need to get the interpreter running pyro: what was the catch? anything worth checking out of cvs for? it's not in cvs :) eih: retro3 is not on cvs :( doh retro3? retro3 is, like, not on cvs :( you're rewriting it again? 03:20pm yeah, in forth. had to be done. too much of retro2 was in asm, eh? * Fare/#Tunes objects retro not being on CVS 'objects to'? well.. NASM kinda sucks that too asm sucks in some ways yes, in others no right now the only thing I see holding us back is the general spectre of illiteracy Forth/asm is better it's a matter of deciding where you're going to force ppl to work rares: what kind of illiteracy do you have in mind? the fact that ppl generally don't like to learn new things rares: do you like to learn new things? pyro: CVS is a tool to share code. pyro: why not share it? like tunes could make writing a cross compiler extremely easy I love to learn new things why do you think I argue all day :) I don't know.. Fare: maybe I'll just upload my code to tunes.org and your nightly script can add it to cvs :) tyes scratch your left ear w/ your right hand Fare: yes, so why doesn't core use it? :) pyro: sure NickPerv is spamming me pyro: but what about *submitting* changes to you? pyro: and why not use cvs commit -m "no comment" as a way to upload? arrest this man.. he talks like a fridge.. he's like a detuned radio Fare: I don't have a permanent net connection... after a night of hacking I don't feel like logging on and uploading if I had a permanent connection you all could barge right in :) this is what you get that's how we do it at work (on the LAN, not the internet) Radiohead rox I have given all I can.. it's not enough 03:30pm rares: indeed.. think I have at least a dozen cds, singles and stuff pyro: it's not like you must update _every night_ pyro: you can hack a lot, and defer the cvs commit until much later commiting often is a Good Thing though pyro: alternatively, you could have a cron job connect you and disconnect automatically I've done that in the past just a matter of programming your /etc/ppp/ip-up* I turn my computer off when I'm done with it so there :) use anacron, then: it connects when it can :) ftp work fine so.. * Fare/#Tunes now has a permanently online server cvs works fine, too Fare, I have one quesstion how you suppose we might describe say a plane in all of its incarnations without needing several textbooks to be stored in memory metaprogramming can get seriously hairy and btw, rsync is more seamless than ftp rsync is ok ftp is the one thing I can count on rares: of course it will get hairy. But still simpler than doing things by hand it's not about being simple. It's about being _as simple as possible_, and no less. I hope th no ;less is included otherwise this discussion could very well become Tovalds vs Tannenbaum and you already know where I stand and the fate of both :) 03:40pm what happened to tannenbaum? Tanenbaum you mean I guess pyro: he is still a professor and still teaches OS courses, still uses Minix.. and Torvalds... still at transmeta? * eihrul/#tunes shudders. a tanenbaum OS course would conceivably be frightening if he holds grudges yes if his books are any indication... I think a lot of people criticises tanenbaum based on the mail-discussion.. and little else * eihrul/#tunes has read both of his OS books. and that's what i base my criticism on :) and compared to other OS books they aren't too bad i think they criticise him because of trhe success of Linux but then again who was he to think a student couldn't whip a teacher's arse unlike most professors he has actually implemented something Fufie: besides sharing 50% of their content... rares: umm.. tanenbaum never wanted to make Linux rares: tanenbaum wanted to and probably still wants to use Minix for teaching.. not making it the greatest OS in the world is it simple enough? Linux can not be used for teaching not necessarily clean enough, however... Minix almost makes Linux look clean... Minix is pretty simple and pretty well documented Minix however suffers a lot from being implemented on too small compuuters.. heh seen Downix you'd prolly be better to use Xinu for teaching OSes or Mach... if you really hate your students Mach isn't hehe Fiasco-L4 isn't half bad, either... Minix vs Linux flamefest summary: minix is teaching tool, while linux is a learning tool. (shameless self-quote) I think people should read tanenbaum's books I think ppl should code :) and respect that people are allowed to be wrong.. as tanenbaum was Fufie: or 1.5 of them... eihrul: maybe.. :) he wasn't necessarily anymore wrong than Linus is and was... both microkernels and monolithics are dead... :) 03:50pm well, not dead, but reaching the end of their usefulnesses... not about that.. about his views on copyights, the net, and such stuff copyrights oh, ok :) ideas do NOT deserve respect, especially wrong ideas. only people deserve respect. Fufie: what are his views? oh by I think I' pyro: check the mail-discussion or flamefest.. and see the foreword on the Minix 2nd edition book gah X needs to go on a fucking diet Fufie: so we should NOT respect that people be wrong; we shal l only respect the people that are wrong. fufmann: was the foreword posterior to the flamefest? pyro: he was very wrong about GPL and about the "bazaar" model fare: the foreword in the 2nd ed is after the flamefest and when linux was gaining in popularity.. around 96 I think fare: no, people are not wrong -:- AlonzoTG [Alonzo@216-164-132-228.s482.tnt2.lnhva.md.dialup.rcn.com] has joined #tunes crap no Borders near me GPL vs. PD... I prefer bugroff to PD maybe I also prefer bugroff to GPL eh? * Fufie/#tunes prefers the world of Winnie the Pooh ah yes the Tao of Pooh the only difference between GPL and PD is, GPL binds people to keep derivative works open-source.. right? that and distribution itself GPL says: you should have the source the same way you can pop the hood on your car (source code must be available on readily available media if it's distributed in the public...) umm.. GPL predates open-source open-source isn't the same as liberated code linux sucks no, complete freedom predates them all. "free software," to be politically correct exactly I can pop the hood and change it Fare: but complete freedom is also the freedom to deny others freedom... ATG: yeah, but it still sucks less than most which is not necessarily an ideal freedom... complete freedom is but a platonic ideal (which is a lot, granted) eihrul: bullshit eh, not bullshit... eihrul you have the right to deny people freedom in the GPL you only have to give source to those to whom you give binaries * Fare/#Tunes is reading a Great book on freedom: Hayek's "Law, Legislation, and Liberty" rares: i'm not talking about the derned GPL... 04:00pm perhaps the greatest book of all times on the subject Fare: explain, if otherwise... ei: explain what? how does complete freedom not give one the freedom to deny it to others? D0S is the most widely used, fastest, smallest, most reliable, easiest OS that ever existed. rofl * rares/#tunes dies laughing eih: because freedom is based on law DOS IS STILL THE MOST WIDELY USED OS IN EXISTANCE!!! it's small but it's still a piece of shit! it is based on the respect of property why do I still use it? fare: bullshit Fare: the word *complete* excludes the idea of law, in my opinion of _natural_ property. Yes it's used to hold up windows like a mother and her cub fare: freedom is a much earlier concept than private property and written law eih: Hayek has quite nice a chapter about that Fufmann: Hayek has quite a nice chapter about that, too -:- water [water@tnt-9-220.tscnet.net] has joined #tunes ATG: {} IS STILL THE MOST WIDELY USED OS IN EXISTANCE!!! and private property is theft put in system Gah we is fsckd hey all welcome to the looney bin... hey water smaller, faster, more reliable, easier than DOS. loons here? join the freedom flamefest loonie tunes i've been working on groking rewrite stuff water: any profound solutions to syntax? priv property is areaction to theft and endagerment Fufmann: I invite you to read the book by Hayek; even the first chapters are worth just by their cybernetic account of society some interesting stuff on integration of higher-order logic with higher-order fp eih: nothing profound yet however, as FAH reminds us, economists practiced cybernetics long before the word was coined by physicists; even before it was popularized by biologists. dude it take 2 ppl to creat a culture Elitism makes me wanna hurl as if you need a phd to grok the world water: doesn't squeak internally store method source as structure rather than text? water: know much linear algebra? atg: yes GTH fare: i don't know for sure Is there anything you don't know???? = =\ rares: you're prompting me a hayek quote: The [classical] liberal, of course, does not deny that there are some superior people - he is not an egalitarian - but he denies that anyone has authority to decide who these superior people are. -- F. A. Hayek, "Why I Am Not a Conservative" atg: yes, plenty of things liberals are usually elitists water: what's the color of the socks I'm wearing? don't you know? heh nope it's the conservatives who are the litists wow. quite a lot of nonsense floating about today they're the ones who need to be in control Fufmann: don't you confuse liberals with "liberals" they're the ones who buy into hype and sell people hype and elitism with "elitism", either water: #tunes has a delicate signal/noise karmic balance... I have yet to meet a loser who is a liberal LOL ei eih: btw, what did we agree on for a quoting mechanism? anyway.. (crawl :to 'bed) eih: or did we even reacha conclusion? goodnight 04:10pm water: we merely noted it, but did not elaborate later it's like tunes if it's simple at the core it's indescribanbly complex at the surface if it's simple at the surface it's in describably complicated at the core if it has one =\ eih: what do you think of the self idea (mirrors as primitive functions)? mirrors? mirrors are slow as well rares: but what if a simple surface is used as the simple core of another system? abi forget mirrors water: I forgot mirrors eihrul I have yet to see one good luck water: they're fine, for now hm would allow addition with minimal fuss... btw, eih, when do you want to consider the lisp code releasable, requirements-wise? eih: ok that's a good point man it's like the puritans they just had to cut the whole wilderness geezus when slate is actually a concrete specification (and most of that specification has been put into it) alright 100 lines of lisp code is not much to release I'm out of here have fun see ya eih: but it demonstrates concepts at least thx Farecool discussion water: though... building off the object being it's own meta-object, mirrors are almost not needed :) I will leave you with one idea make your objects tunbelable tunnelable even hm i'll have to think on that one also, assignment to data slots is another interesting issue... heh.. I finally found a good Linux! any thoughts/concerns? well, if it were to be done at all, you'd need to do it as in Self... if something requuuires a redundant but necessary step in the global but unnecessary in the local creat temporary bridges between those elements pyro: CLIeNUX? for each foo slot, a setFoo slot/method would be required... Fare: yeah :) It's small, comes with Lynx and Forth set up.. perfect ;) well see yall -:- rares [rares@wtrb-sh3-port80.snet.net] has left #tunes [Segmentation] water: and setSetFoo, and setSetSetFoo... and set(Set)+Foo... sorry, was afk hm 04:20pm is that a joke? (setSetFoo...) pyro: didn't Rick H contact you about contributing to retro? again CVS would be a nice idea for collaboratoin... water: why should it be? :) fare: this rewrite book has on average 7 lambda symbols per page water: sure billy would hate it :) you and billy are mismatched as for impedance eih: those things belong in the bmo water: the idea of setting slots seems inelegant, now :) I happen to like billy eih: and they form a regress eih: but we're not setting slots Fare: he wants to make a bootable version of his Forth variant eih: you can look on it as a msg which returns the new object value and updates all references pyro: good idea He could use my bootloader & drivers, and an XCOM or ELF loader to load his interpreter (which is in C) ouch, bad idea Water: when will you release your sentient AI? atg: uhh it's not scheduled yet, how's that for an answer? looser. underachiever! you just like knowing a bunch of shit to make me feel bad. =\ eih: still there? what do you think? the right answer was "a few days after you die" 04:30pm atg: when he's ready to terminate you water: well, it merely seems an extension to the original concept... eih: right but it avoids the data metaphor, doesn't it? 'data metaphor'? clarify i thought you had misgivings about stored state not quite so what's the problem? with that, nothing going afk for a few, b(r)b(l) so why setSetFoo? doh well, if setFoo is a slot containing a method to set slot foo, then it to is subject to being 'settable' but why by the object itself? oh, you'd have it part of the mo? not sure well, that's what i've been almost thinking the mo governs the existence/application of the accessor or reifying slots as slot objects... and then applying values to them to coerce them to 'hold' another value... but if the accessor is a slot in the object, then it has state which can be assigned something's wrong about this argument we're assuming something incoherent 04:40pm for instance, if the accessor is an implicit slot (its presence/application determined by the bmo dynamically), then it's not part of the object's state but it's behavioral meta-state i'm gonna go play w/ clienux :) -:- SignOff pyro: #TUNES (.) water: well, may i go now, i have a head-ache? :) so accessors themselves are slots dynamically generated by the bmo heh ok well, yes, that seems the cleanest way to handle them thus far cleaner, i should say -:- SignOff eihrul: #TUNES (sleep) hm the object's behavioral meta-state would implementationally need to be stored with the object for simplicity's sake * AlonzoTG/#tunes is working on the amazing gnu Fifth programming language! =) you know there's a Fith, right? 04:50pm no fuck you 05:00pm -:- eihrul [lee@usr5-ppp89.lvdi.net] has joined #tunes wb re oh yes, i said something that probably bears repeating: the object's behavioral meta-state would implementationally need to be stored with the object for simplicity's sake or rather, modifying the mo should just simply be immutable and cause explicit cloning well, very easy to have the behavior if an object mo's itself :) bleh merely an observation (though it excludes the idea of migrating between meta-spaces somewhat, or atleast complicates it) yes, that bothers me 05:20pm darn it, rewrite is making me shift my ideas a bit to where? well, i have this idea of behavioral spec via equations it might make the math system rather clean in fact, extremely clean well, wasn't that what kauf suggested? i forget, what did he suggest exactly? about specifying state and behavior of object as relations instead hm i'd like to have both top-down and bottom-up, so to speak so you get categories and objects with slots the more the merrier (though not to the point of non-orthogonality) yeah all these ideas interacting really bug me 05:30pm well, it would be simple to implement re-write systems on top of slate maybe i should take a little break from this stuff :P and do what? * eihrul/#tunes is one of those unfortunate people who have nothing better to do. i don't know, maybe start dating again but girlfriends are not orthogonal, consistent, nor extensible exactly you might as well program in C! :P =((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((( -:- smoke [smoke@16dyn235.delft.casema.net] has left #tunes [] * AlonzoTG/#tunes signs up for cryofreze so that he can come back when bigger brains are available.... rmm, don't your cells literally rip things up? turning into weapons of microscopic destruction? i'd prefer my parasynthetic nervous system operation to be automated by implants 05:40pm suicide is the simpler option not only that, but death is consistent and orthogonal... though not necessarily extensible * water/#tunes tries to describe his persona for an online match-making service in seattle how the hell should i describe myself? nothing i do really describes who i am -:- present [xdef@209-6-184-130.c3-0.wth-ubr1.sbo-wth.ma.cable.rcn.com] has joined #tunes welcome to the present :) If albert einstein was a moron, you are a super genius... hehe water: rmm, describe the slate language, in full eih: to whom? :) to the slate page, would be nice perhaps to a LyX or your post-script editor of choice... what's missing from it? perhaps the details of the rewriting system... oh yeah, the ones that don't exist yet? :) You are to Alan Turing as Nikola Tesla was to Edison. 05:50pm * eihrul/#tunes wonders if too much DOS has had a perverse effect on AlonzoTG's brain. well, mind, i should say -:- SignOff water: #TUNES (Ping timeout for water[tnt-9-220.tscnet.net]) 06:00pm i wonder if there are any examples out there of what a program written in the TUNES environment would look like? -:- Sancio [jsancio@BUSW044-0B01-101.BU.EDU] has joined #tunes om DOS is the absolut king of all OSes... NO OTHER OS RUNS ON OVER 90 MILLION COMPUTERS!!! Today you can go into any office anywhere and find a DOS based computer. present: well, a program... doesn't mean it's any good Name one OS that is easier to use. one just one echo... what is echo? eihrul: is 'program' not a good term to use? a 'program' is a 'program' is a 'program' I would just like to know if there are any practical-type examples of 'toons in action' meta-programming systems, reflective languages lots of various things that implement various componenents of tunes etc, etc 06:10pm im interested i guess in the HLL that would embody some of those concepts i actually campe upon tunes while reading about MS's 'intentional programming' well, it depends on whether you consider tunes to be the hll language itself... s/hll language/hll i dont really grok the concepts well enough to be able to make that distinction.. from what i understand, one of the components is a new type of programming language which would incorporate some ideas from formal methods, etc.. i dont know what im talking about tho -:- Sancio [jsancio@BUSW044-0B01-101.BU.EDU] has left #tunes [] To Do on this page: Show how the real program would look like on our HLL. heh 06:20pm -:- air [brand@p0wer.qzx.com] has joined #tunes -:- water [water@tnt-9-97.tscnet.net] has joined #tunes wow im in the irc log on tunes.rog org rather heh 06:50pm -:- nietzsche [nobody@216-164-138-111.s111.tnt5.lnhva.md.dialup.rcn.com] has joined #tunes cool! hello, nietzshe :) hello water, like the nick? yep wel,, the namesake, rather well. thanks. whats this channel about? i'd check the webpage, but unfortunatly i'm involved in a large download, and I have a slow connection. mp3 warez abi tunes? tunes is, like, a free reflective computing system at http://www.tunes.org or for programming languages what the internet is to networks or like flypaper for hallucinogenic programmers or discombobulated or obnubilated or AOTA i.e. vaporware right now hmm. sounds interesting. anything up there right now, like a simple FAQ, or just conceptual stuff? there's an faq on the site thanks, i'll check it out some time 07:00pm grrr, damn ie4 i need to get that external modem so i don't have to use 'doze any more what should it be external? err, why? well i have this us robotics pci pnp modem which won't play with beos or linux is it a soft winmodem? i don't think so i think it's the pci aspect u need to turn of pnp, which you can probably do with a jumper on the board yes i tried that already or maybe a config program -:- ult [noone@user-37kbasq.dialup.mindspring.com] has joined #Tunes still didnt work in linux? thats odd... *cackle* God: Nietzsche is dead. Nietzsche: God is dead. does it have jumpers for vvv Linux sucks because it makes it hard for you to configure PCI devices. Nietzsche is pietzsche, but Sartre is smarter! present: configurable bios ive never had any problem with that ` `21A+ v oops uh i was wiping corn flakes off my keyboard *cackle* how does microsoft's intentional programming ideas relate to toons HLL? if anyone knows... it doesn't. Microsoft doesn't know anything about intensional programming. their version of it just allows middle-level managers to pretend that they understand code what is intentional programming? =\ it claims to have some facilities for specifying constraints on input/output for functions which can be proven formally 07:10pm reading thru their stuff, im still fuzzy on what it does tho yes, which takes time it's just a complicated high-level overlay for languages like c++ do you think it boils down to a bunch of annotations on c code? oh mostly one think i saw was that implementations of 'actions' would be chosen based on the 'domain' think=thing lag sux. is that a fancy kind of #include or #define or is 'domain' more fundamental? =\ * water/#tunes doesn't much care 3.11 sucks almost as much as linux.... lag kills. It just lost controll of netscape again... linux sucks almost as much as ATG's mom... now if I close all netscape windows, the netscape engine will never leave memory and I can't restore the system... Kind of like how Linux will loose controll of the console to doom or not be able to restart X when it quits.... Netscape will do that on any platform. eeek but on a slightly better platform it would still be killable... Like Linux? One thing that fscking OWNZ about linux: So easy to kill processes. is one of tunes' HLL goals to have the language be formally provable and somehow avoid the time problem of proving semantics? present: No. heh so where do formal methods come into play * ult/#Tunes cackles -:- SignOff AlonzoTG: #TUNES (Have Nice Day :)) * ult/#Tunes is bored, as you can tell ;) ult's no tunes spokesman the idea is to allow formal specs *easily* well there are formal specs and there are formal specs..... one spec would govern every single operation, another would impost constraints n'stuff impost=impose are the ideas similar to LARCH ? i oopsies. i'm not familiar with larch larch is a formal specification language -:- SignOff nietzsche: #TUNES ([x]chat) abi, LARCH? LARCH is, like, a formal specification language Gah. using the language, you can model the semantics of different languages (e.g.: c, c++, modula ) 07:20pm what makes it different from Z or denotational semantics? they made a distinction between 'operational' and 'denotational' semantics im kind of fuzzy on this.. so what kind of theory is larch based on? logic? first-order predicate logic ok from larch, came the 'lclint' program. which does some very basic kinds of semantics checking on C programs fscking laaaaaaaag shut up ult no one cares like easy to find memory leaks, functions which break basic kinds of 'contracts' etc.. bite me water present: i prefer hll's like smalltalk or lisp or maude Hey,wow,alternet is messed up. How odd. present: How would larch deal with varied higher-order languages? ult, i don't think it even tries to approach that Didn't think so =P that's not good maude? i've never heard of that do u haave a url? abi maude maude is probably a reflective rewriting logic language at http://maude.csl.sri.com/ or http://www.csl.sri.com/~duran/ it's good for specifying all sorts of programming ideas cool concurrency, gc, higher-order fp, ... 07:30pm although the syntax is terrible heh Damn it! Damn the internet to HELL! ive played with some model checkers (SPIN and SMV), but not really with the logic stuff * ult/#Tunes notes that alternet ought to hire some competant admins and cable-layers. alternet sux who gives a rat's a**? Heh. reliable-access.net is fscking up... so are you again, I repeat, bite me. are there any mechanically verifiable languages that are easy to use and that have similar power to C? sigh present: Visual Basic heh vb barely has types this is pointless, what with ult the ass being here -:- flippy [rdelfin@200.34.111.111] has joined #tunes i'll bbl tonight -:- water [water@tnt-9-97.tscnet.net] has left #tunes [] cya.. Hello everyone =) present: Exactly...if you hit your head on the keyboard the code would run -- your checker would consist of return(TRUE); -:- SignOff flippy: #TUNES (Leaving) well supposedly u cant crash vb Well, vb is a safe language to a point java is safer You *can* crash stuff, but not with pure vb code. Java is safe because you can't import native libraries and such easily. VB is like Lisp: It's as safe as the underlying implementation. That's not so good in VB's case. yep.. vb has some weird semantics too i was horrified to find that if (x) ... and if not(x) ... could both be false *cackle* if x is 'null' that should really be an exception i think 07:40pm Well, that just doesn't make sense. Silly not operator. i like python now, tho i havent done much work w/it its extremely readable python is kinda like VB only with a bit more sense put into it. yah bleh im gonna do some work later -:- SignOff present: #TUNES (Read error to present[209-6-184-130.c3-0.wth-ubr1.sbo-wth.ma.cable.rcn.com]: EOF from client) -:- rares [rares@wtrb-sh1-port49.snet.net] has joined #tunes okay if I see another editor that looks like every other damned editor i'll die you'd think ppl; forgot what legos are 07:50pm -:- SignOff eihrul: #TUNES (Ping timeout for eihrul[usr5-ppp89.lvdi.net]) -:- hcf [nef@me-portland-us739.javanet.com] has joined #tunes -:- SignOff ult: #TUNES (Read error to ult[user-37kbasq.dialup.mindspring.com]: Connection reset by peer) -:- ult [noone@user-37kbai2.dialup.mindspring.com] has joined #Tunes -:- eihrul [lee@usr5-ppp113.lvdi.net] has joined #tunes eihrul: how do i get rid of the following warning? main.c:39: warning: passing arg 3 of `pthread_create' from incompatible pointer type main.c:39: warning: passing arg 3 of `pthread_create' from incompatible pointer type pthread_create(&ai_thread, NULL, processAI, NULL); void processAI(); void * processAI (void *); -:- hcf has changed the topic on channel #tunes to: TUNES, Free Reflective Computing System || slate so i have to tell the function to use an arg even tho i dont need it? yes... because if you tried to cast it... it'd be technically bad-style hmm ok it works now 08:50pm standards committeees oinderful ain't they -:- SignOff rares: #TUNES (KVirc 1.0.0 Millennium BETA 3) 09:00pm -:- water [water@tnt-9-15.tscnet.net] has joined #tunes hey, look, it's water. 09:30pm -:- water [water@tnt-9-15.tscnet.net] has left #tunes [] ult: wanna do this for me? Wot? finish my openspace server :) not particularly ;) its easy just write an function that finds any object within range of the current object whats the best way to do that? * ult/#Tunes was about to go to bed actually... depends on the object how? you might just do a regexp based on object names linked lists btrees arrays hash tables uhh i dont think u understand somehow network the objects together each object represents a ship *cackle* and i when another ship gets close enough then i need to know oh 2d or 3d? 2d ((x0-x1)^2+(y0-y1)^2)^(1/2) for all objects simplest way if your map is sector based you just have to do it for objects in adjecant sectors (try and cull some of the ships out so you don't do 500000 operations) don't bother taking the square root eihrul: ya Well, you can just square the maximum distance, sure/ do your distance calculations with distance squared if using pythagoras 09:40pm how many light years can a warp capable ship scan? hmm well how fast is warp? warp 1 is speed of light Hmm. At speed of light scanning is rather ... irrellevant I would think scan distance limitations don't make sense since we can get -extremely- accurate visual data of -extremly- distant objects from space warp 2 is 10 times that wapr 9 is 1516 times Um. So your planning on a pretty massive universe? Is openspace turn-based? it will be 256 light years acroos and u can connect servers to make it bigger Well: If a openspace turn is a real-time half hour it is not turn based the scan distance per turn should just be how far light travels in a half hour and using a time scale of 1 it will take real time air, does it have any form of timing mechanism? yes u can speed up game play for faster games or let it play out for years then just detect everything in (c * (seconds since scan started)) meters range. uhh (If you want scanning to be a stationary action) based on sensor technology each ship can see so far wot kind of sensors though? Thing is, in space...there's not exactly a lot out there to block your signal. so basically you can see -reaaaaaallllllly- far -:- rares [rares@wtrb-sh1-port57.snet.net] has joined #tunes in space time is the limiting factor...things really far away will have moved by the time you scan them -- perhaps you should do that nah -:- SignOff hcf: #TUNES (Leaving) i will have planets move yes but they're faraway therefore they've moved very little in field of view rares, quite true. I love Emergence so putting any kind of temporal/physical limitation on scanning is unrealistic -- but since your making a game(This is a game, I assume), you really ought to pick an arbitrary distance of say, 32 light years...;) 09:50pm there is a max sensor limit and u cant see beyond it it makes it simpler you could always detect anything in the system period but you get varying levels of information based on the distance hmm genetic cross between irc and muds hmmm maybe at 128 light years you can tell whether its a planet or a ship at 64 you can tell whose planet or ship it is at 32 you can tell what kind of planet or ship, at 16 you can tell global climate of planet or general condition of ship (perhaps scale that down a bit, I dunno.) an psycho thriller muds? my sectors are 16 light years, u could 1/4 the map using 128 I never have the patience to study races well perhaps scale that down by a factor or 8 or 16 an = any depends on how many players will be on at any given time well i hope it will attract thousands if its like 2 or 3, fairly good scanners are ok, but if its 400, you must give the ships really crappy scanners. it would be rather boring with less than 50 since it takes 60 days to cross the map at maximum warp how about every so many hops = a light year tho that screw chinese plauers their internet is so twisted everyy few miles is a hop rares? rares is addicted to politics -:- Mister_BogoMips [somewhere@usr26.tamcotec.com] has joined #tunes hops = routing hops rares: what does that have to do with the game? air perhaps you need a smaller map? course that makes no sense now that I think a bit ult: massive multiplayer * ult/#Tunes nods is it real-time? yes so you plan on having thousands of players logged on -simultaneously-? yes and no -:- Mister_BogoMips [somewhere@usr26.tamcotec.com] has left #tunes [] the game plays by itself * ult/#Tunes nods the users log in to see the action and modify the ai of their ships ok air you need a distributed mud some sort of irc mud hybrid could be cool modify the ai? like a Crobots type deal? crobots? yes...you write C routines that make your robot chase other robots other robots whose ai is composed of C routines written by other people you see whose ai is best. and hope they don't mate well u setup triggers to make yer ships behave the way u want them. and u can have groups of ships interact... sounds interesting tho is it web-based? its kinda like starcraft, but u program the ai ok so graphic-intensive? the game has clients that display whatever info u want u can have a simple 2d map or an opengl view * ult/#Tunes nods sounds cool tho. 10:00pm Are all ships the same? depending on what the client programmer does the X11 client will support opengl Why not have different 'guilds' the players can join to get different starting ships and ship upgrades lar1 is writing a win32 client, dont know if he will do opengl mercenaries that can buy upgrades to their ships, militaries that get better ships based on performance but can not upgrade their ships with $$ each player is given the same set of ships to begin the game and u can research thousands of technologies and design yer own ships ok and technologies can be added at any time during play oh, so it is more like starcraft, you control an empire of ships? yes but u can customize yer empire and make them smarter individually? yes, u can build yer own ships with varying capabilities and looks and u can even sell ships if i have 2000 battlecruisers, can I give 500 one AI, 1499 another AI, and 1 a third AI (My command ship ;) yes cool. * ult/#Tunes is very interested. What language is the AI programmed in? logo NOOOOOOOOOO! cept the windows client will have visual logo :) gah cuz windows users arent the brightest maybe i'll have to make a similar mud-like thing based on Forth for the AI... logo is a good choice cuz it makes that turtle move around just fine stupid turtle... uhh, forth and logo are the same, just backwards and logo is like lisp but easier you like logo, don't you? yes, so k. i like forth too maybe you could give options * ult/#Tunes shrugs it doesn't matter much as long as the language itself is powerful. i wrote a logo for my hp48 which uses forth oh well. I need to get to bed. cya later. -:- SignOff ult: #TUNES (snooze) -:- SignOff rares: #TUNES (KVirc 1.0.0 Millennium BETA 3) 10:10pm -:- SignOff air: #TUNES (http://www.qzx.com/ :: sleep) -:- SignOff eihrul: #TUNES (Leaving) -:- _ruiner_ [DIY@ppp019.wi.centurytel.net] has joined #tunes -:- SignOff _ruiner_: #TUNES (destroy what destroys you) -:- present [xdef@209-6-184-130.c3-0.wth-ubr1.sbo-wth.ma.cable.rcn.com] has joined #tunes -:- topic unset by present on #tunes -:- present has changed the topic on channel #tunes to: TUNES, Free Reflective Computing System || slate shit -:- present has changed the topic on channel #tunes to: TUNES, Free Reflective Computing System || slate abi random abi slate slate is probably a unifying/hybridizing of self/beta/lisp at http://www.tunes.org/~water/slate-home.html 11:30pm [msg(TUNES)] newlog 2000.0211 IRC log ended Fri Feb 11 00:00:01 2000