00:00:00 --- log: started haskell/01.12.18 00:19:20 --- join: julien (~julien@uu212-190-122-70.unknown.uunet.be) joined #haskell 00:19:32 hi 00:34:28 --- join: shapr (~user@195.156.199.185) joined #haskell 00:34:32 g'mornin 00:39:35 new haskell scripts ? 00:40:21 not me 00:40:34 though I did some reading yesterday 00:40:40 and the Haskell Tutorial is picking up steam 00:40:47 I should probably start a mailing list somewhere 00:41:06 julien: did you write anything new recently? 00:41:13 nope 00:41:20 A lot of reading 00:41:42 I found a good document in a french university about compilation 00:42:19 The teacher uses Haskell to explain his course 00:42:46 sounds good 00:42:52 tried to do any code? 00:44:27 nothing greater than some "Hello world"'s :-) 00:44:43 that's a good way to start 00:45:53 I was reading about network libraries of ghc, and bindings with other libraies 00:46:23 I will try to program a little server this WE 00:46:38 if i have time enough ... 00:48:47 I'd like to change my Mandelbrot to save the fractal I've graphed, and then later increase the resolution of the array, or whatever I've saved it as 00:48:54 that would be spiffy 01:32:41 * juhp releases hsclock-0.01.0 01:33:03 see the announcement on haskell@haskell or 01:33:17 http://www.01.246.ne.jp/~juhp/haskell/hsclock/ 02:42:12 cool, I'll check it out! 02:43:13 whoa spiffy, it's gtk 06:37:02 * sjanes71 wakes up with a bad cold. 06:37:24 * shapr has a bad cold and wants to go to sleep 06:37:46 slashdot is catching up with my ideas for the operating system 06:38:07 article about David Gelerenter and killing the files/directories stuff 07:40:12 I for one don't understand what else you would like to use 07:40:55 one could use something like a database maybe, but it wont really make a difference 08:13:20 people give me a weird look when I try to describe it, I haven't figured out how to describe it fully yet. :) 08:15:17 try it on us 08:15:24 maybe we can help you refine the words 08:15:27 ok 08:15:49 across all the major operating systems today 08:16:14 there's a filesystem where there are places for the system and for the user 08:16:51 in most cases, the user accumulates their files on the "desktop" or whereever the "default" location for files go for each application 08:17:51 so eventually what happens is during the course of using the computer the filesystem starts to rot 08:18:09 and the user may uninstall applications or remove directories 08:18:20 maybe not realizing that they had something important... 08:18:39 or more often, just not knowing where the files are without a search 08:19:16 the metadata on files has shrunk and shrunk to the point where it's not really useful for the user anymore 08:19:43 it's just there to tell the operating system which application to use to "open" the file 08:19:55 or who's allowed to open the file, etc. 08:20:15 this metadata isn't extensible at all by the user, or if it is, it's really not easy to do 08:21:16 if you right-click on a file in W2K Pro, there's a tab called "Summary" 08:21:38 which has some fields like "Title", "Subject", "Keywords" 08:22:13 there's no way to add new kinds of metadata 08:22:38 things like... a "Role"-- aka "Why" you're interested in the dat. 08:22:40 data. 08:23:32 so I imagine a computer should have some kind standard way of associatig metadata with data... that combines all the needs 08:23:33 what would you like to add other than Role? 08:24:30 Role, Task, Goal, When, Where, How, Disclosure, Permanance 08:25:25 ... but more importantly, the user has to be able to create new kinds of metadata as they need it... and the operating system should just be able to index it as if it were "core metadata" 08:25:28 have you tried to put together a mockup with a separate file holding metadata? 08:26:09 I have tried to do it with some shell scripts, but I haven't gotten it to a point where I really like it 08:26:27 it could be simulated with files and directories 08:26:57 it just gets messy because filesystems really want to be hierarchical 08:27:04 boxes within boxes 08:27:55 the extra bits could be done with the recursive boxes just fine 08:28:00 for example, streams in ntfs 08:28:09 they're parts of the file that most programs can't see 08:28:13 hmm 08:28:37 the extra bits you've suggested still sound like they're attached to files 08:29:01 they're really attached to streams in the end 08:29:12 streams of data that have a type 08:30:12 in my mind this o/s should also make versioning of data a core service... ala cvs.. you can eventually tell the o/s to "purge" the data... 08:30:36 but otherwise, all applications then inherit the ability to rollback bad changes 08:31:10 instead of having to do it by themselves 08:32:02 By hand I'm simulating it with a directory called "My Roles" 08:32:25 where I've made subdirectories of various roles... "Webmaster", "SysAdmin", "Home Owner".. etc. 08:32:40 so I collect bookmarks and files and documents in those directories 08:33:05 When I want to put on the "Webmaster" hat, I just find that Webmaster directory 08:33:50 but because of the directories, there's no easy way of showing myself a timeline across everything I've bee doing 08:33:52 been 08:34:05 without having to use the search system 08:34:06 you should check out GZigZag 08:34:19 I think that's part of what you're looking for 08:34:27 I think it's at sf.net/projects/gzigzag 08:35:15 killing my modem connection now.. :) 08:35:50 it's close. 08:36:10 play with it some, read the tutorial 08:36:21 I think it'll help refine your idea 08:36:23 all...Ted Nelson 08:36:29 ahh I meant to write 08:38:23 www.scopeware.com is close to what I want, but I think it's really hard to retrofit an existing op. sys to do it 08:39:15 the other thing I'm really down on is overlapping window interfaces 08:39:17 :) 08:39:40 considering how many times I've lost data because some application pops a dialog on me while I'm typing 08:45:58 biggest problem with scopeware is the buzzworditis and marketing 08:48:05 I use emacs 08:48:15 I've never lost data from a dialog 08:48:27 hehe 08:49:26 I've lost something more important than data because of badly timed dialogs: concentration and temper. :) 08:50:29 yah, that's where emacs comes in handy 08:50:55 --- quit: shapr ("hometime") 09:52:32 --- quit: julien ("Client Exiting") 12:35:50 --- quit: kepler (Remote closed the connection) 14:49:08 --- quit: pHa (Ping timeout: 180 seconds) 14:50:40 --- join: kepler (~kepler@adsl-20-90-85.asm.bellsouth.net) joined #haskell 15:01:53 --- join: pHa (sjh@aircomm-1blk-220-105.air.on.ca) joined #haskell 15:17:07 --- join: dblack (~dblack@ool-18be3b59.dyn.optonline.net) joined #haskell 15:17:07 --- join: chad_ (xpqbof@65.15.137.250) joined #haskell 15:18:18 --- part: chad_ left #haskell 15:25:11 * dblack is a haskell pre-newbie 15:25:17 just bought the thompson book 15:39:55 --- join: chad_ (xpqbof@65.15.137.250) joined #haskell 15:47:51 --- part: chad_ left #haskell 17:16:12 --- join: bashu (~bashu@B8d72.pppool.de) joined #haskell 17:56:01 hello 17:56:08 is someone here and wants to talk? 17:56:45 go ahead! 17:57:38 i would like to know if there already exists a compiler for haskell which creates dynamically linked executables 17:57:55 which platform? 17:58:01 any 17:58:26 ghc creates dl execs 17:58:54 but under unix it can't make dl libs currently 17:59:04 they'll get to it eventually 17:59:09 it can under windoz 17:59:43 yes, hopefully sooner rather later ;) 17:59:49 since when does ghc create dl execs? that must be a new feature, no? 18:00:18 i mean, a relatively recent one 18:11:16 no so recent i think, but i don't remember exactly 18:32:42 --- quit: bashu ("[x]chat") 19:08:31 --- quit: dblack (carter.openprojects.net irc.openprojects.net) 19:08:31 --- quit: juhp (carter.openprojects.net irc.openprojects.net) 19:08:31 --- quit: pHa (carter.openprojects.net irc.openprojects.net) 19:08:31 --- quit: kepler (carter.openprojects.net irc.openprojects.net) 19:08:31 --- quit: dennisb (carter.openprojects.net irc.openprojects.net) 19:08:31 --- quit: jlb (carter.openprojects.net irc.openprojects.net) 19:08:31 --- quit: smkl (carter.openprojects.net irc.openprojects.net) 19:11:50 --- join: smkl (~sami@glubimox.yok.utu.fi) joined #haskell 19:11:50 --- join: juhp (~petersen@firebox-ext.jp.redhat.com) joined #haskell 19:11:50 --- join: jlb (~ktk@jeremeydsl-3.mylinuxisp.com) joined #haskell 19:11:50 --- join: dennisb (~dennis@as2-4-3.an.g.bonet.se) joined #haskell 19:11:50 --- join: kepler (~kepler@adsl-20-90-85.asm.bellsouth.net) joined #haskell 19:11:50 --- join: pHa (sjh@aircomm-1blk-220-105.air.on.ca) joined #haskell 19:11:50 --- join: dblack (~dblack@ool-18be3b59.dyn.optonline.net) joined #haskell 19:30:42 --- quit: pHa (carter.openprojects.net irc.openprojects.net) 19:30:42 --- quit: kepler (carter.openprojects.net irc.openprojects.net) 19:30:42 --- quit: jlb (carter.openprojects.net irc.openprojects.net) 19:30:42 --- quit: smkl (carter.openprojects.net irc.openprojects.net) 19:30:42 --- quit: dennisb (carter.openprojects.net irc.openprojects.net) 19:31:24 --- join: pHa (sjh@aircomm-1blk-220-105.air.on.ca) joined #haskell 19:31:24 --- join: kepler (~kepler@adsl-20-90-85.asm.bellsouth.net) joined #haskell 19:31:24 --- join: dennisb (~dennis@as2-4-3.an.g.bonet.se) joined #haskell 19:31:24 --- join: jlb (~ktk@jeremeydsl-3.mylinuxisp.com) joined #haskell 19:31:24 --- join: smkl (~sami@glubimox.yok.utu.fi) joined #haskell 19:31:46 --- quit: dblack (carter.openprojects.net irc.openprojects.net) 19:31:46 --- quit: juhp (carter.openprojects.net irc.openprojects.net) 19:35:33 --- join: dblack (~dblack@ool-18be3b59.dyn.optonline.net) joined #haskell 19:35:33 --- join: juhp (~petersen@firebox-ext.jp.redhat.com) joined #haskell 23:59:59 --- log: ended haskell/01.12.18