IRC log for #brlcad on 20091013

00:01.40 *** join/#brlcad Axman6 (n=Axman6@pdpc/supporter/student/Axman6)
00:05.46 Ralith heh
00:37.26 *** join/#brlcad mafm (n=mafm@225.Red-83-45-72.dynamicIP.rima-tde.net)
01:43.28 *** join/#brlcad akafubu (n=akafubu@unaffiliated/akafubu)
02:36.00 *** join/#brlcad akafubu (n=akafubu@unaffiliated/akafubu)
04:04.33 starseeker is fried, but the cedar closet is (essentially) done
04:05.22 starseeker and the gentoo box goes through (yet another) tramatic update, successfully.
04:05.36 starseeker There seem to be a lot of backwards incompatible changes lately
04:05.46 starseeker probably for the best, but eeek
04:45.59 *** join/#brlcad puddingpimp (n=dave@118.93.87.238)
07:40.32 *** join/#brlcad _clock_ (n=_sushi_@80-218-244-105.dclient.hispeed.ch)
07:46.03 Ralith starseeker: this is why I stopped using gentoo.
07:46.31 Ralith "okay, routine system up—OH GOD WHERE'S MY X"
09:27.22 Maloeran Pretty much. With Gentoo, it's best to upgrade only what you strictly need... and hope for the best
11:06.17 d-lo mernin!
11:06.51 archivist with gentoo upgrade to a sane distro asap
11:11.28 ``Erik *yawn*
11:12.22 ``Erik is glad he uses fbsd, updates are neither traumatic nor unsafe O.O
11:13.40 ``Erik there was this one time, in band camp, when I did a massive upgrade on an ancient system and it was a bastard stopgap patch (still is), so several ports failed, but portmanager gives you pre-upgrade pkg's, so it was trivial to undo the mucked up parts of the upgrade... definitely a corner case situation, though :D
11:14.21 ``Erik (I mean, seriously, who in their right mind would still be running 5.2.1? it was so... screwed up that they changed their numbering system just for that one release)
11:15.43 archivist my old debian screwed up on an update, about 5 years ago, its still up though
11:16.58 ``Erik :D stable or testing?
11:17.24 ``Erik I had some interesting experiences with unstable, and stable was just too out of date, I used testing for a lot of my debian boxes back in the day
11:18.10 archivist cant remember but uname responds with 2.4.27-2-686-smp #1 SMP Mon May 16 16:55:31 JST 2005 i686 GNU/Linux
11:18.39 ``Erik should be an /etc/debian file
11:18.43 ``Erik iirc
11:19.00 ``Erik or look at your apt config file :)
11:19.21 archivist 3.1
11:19.56 ``Erik according to this chart, it's "sarge"
11:20.21 ``Erik 6 june 2005, support ended april 2008
11:22.22 archivist these old boxes run and run
11:23.05 archivist it is on its last month or so
11:29.16 archivist this job has ended and I have to move the servers home
12:01.29 Yoshi47 45812:11 mged
12:17.28 *** join/#brlcad brlcad (n=sean@BZ.BZFLAG.BZ)
12:27.59 ``Erik put a bullet in its head, yoshi, it probably put itself in an infinite loop (or went into the "it'll finish.. in a few thousand years" mode)
12:44.43 brlcad blames the ISP
12:51.29 ``Erik eh?
12:51.47 ``Erik (looks like everyone using bz got peered)
12:53.32 _clock_ who has worked on PDP-1?
13:21.17 CIA-33 BRL-CAD: 03bob1961 * r36183 10/brlcad/trunk/src/tclscripts/archer/AttrGroupsDisplayUtility.tcl: Added methods for reading/writing attribute groups and mappings. Added a method to export to png.
13:42.30 brlcad ``Erik: bz rebooted unexpectedly
13:45.28 brlcad _clock_: pdp wasn't interesting until pdp-8, and even then pdp-11 was where it was at
13:48.02 Yoshi47 ``Erik, do i have too? can't i wait till my pc turns off or restarts? lol
13:51.35 brlcad :)
13:51.40 ``Erik pdp7 had some niftiness iirc
13:51.43 Maloeran 45812 minutes? Isn't that like... a month?
13:51.59 ``Erik (wasn't the pdp8 just a souped up 7? or was that the big abi break?)
13:52.03 Yoshi47 yep
13:52.20 ``Erik lisp on a pdp1 was awesome, simh ftw
13:52.29 Maloeran A month. You are being optimistic, I see... :)
13:53.06 Yoshi47 nope just don't have time to work on it anyways
13:53.18 Yoshi47 so just like to see it will finish or not
13:53.53 ``Erik if it doesn't finish in a day or two, it probably won't finish in your lifetime :(
13:53.56 Maloeran I guess it's only using 100% of one CPU core, so it isn't that bad
13:54.02 Yoshi47 yep
13:54.34 Yoshi47 ok well when i need my other core or the power goes out or some idiot comes by and does something then iguess that will be the end
13:54.59 brlcad it's an O(n^3) algorithm .. and that's a pretty big 'n' it's crunching on
13:55.00 Maloeran wonders if there are ways to randomly query the EIP of a running program, to have a clue what it's doing
13:55.20 brlcad so it will probably finish .. just unclear if it's days/weeks/months/years
13:55.53 ``Erik millenia
13:55.58 brlcad (and single-cpu, so it's not exactly burning the midnight oil)
13:57.11 ``Erik let's see, we're in the cenozoic era, what comes next? :D
14:26.13 brlcad the "flying spaghetti monster hath forsaken us" era
14:34.43 ``Erik well
14:34.51 ``Erik that's the biblical name, the age of truth
14:34.55 ``Erik :)
14:49.09 ``Erik hey, uh, starseeker?
14:49.49 ``Erik y'know that old mkVIII or whatever you were putzing with? was survice involved in that?
14:54.48 *** join/#brlcad mafm (n=mafm@225.Red-83-45-72.dynamicIP.rima-tde.net)
15:31.28 CIA-33 BRL-CAD: 03bob1961 * r36184 10/brlcad/trunk/src/tclscripts/archer/Archer.tcl: Tweak the attr method.
15:41.20 *** join/#brlcad starseeker (n=starseek@BZ.BZFLAG.BZ)
15:41.45 starseeker mutter... dead terminal...
15:43.57 CIA-33 BRL-CAD: 03starseeker * r36185 10/brlcad/branches/dmtogl-branch/: Create branch to have a place to experiment with togl
15:57.53 ``Erik heh
15:58.06 ``Erik starseeker, was survice involved in anything related to the mk8?
16:00.19 starseeker ``Erik: I don't believe so
16:00.23 ``Erik huh
16:00.31 starseeker the mk iv yess
16:00.36 ``Erik ahhh
16:01.01 ``Erik one of the old ww1 tanks is on their homepage ticker, in what looks like a point datacloud
16:01.04 ``Erik was curious seeing that
16:01.13 ``Erik (also saw an old rayfarce screenie there)
16:01.36 starseeker oh, the gold one - yes that's a Mark IV
16:01.44 starseeker point cloud
16:06.10 ``Erik ah, 'k
16:06.28 *** join/#brlcad Elrohir (n=kvirc@91.20.224.134)
16:06.31 starseeker or actually, that's probaby the mesh created from the point cloud...
16:06.41 ``Erik yeh, all their stuff tries to build meshe
16:06.43 ``Erik meshes
16:06.51 ``Erik simple local hulling I think
16:07.11 ``Erik fun toys over in the warehouse, though
16:07.28 ``Erik <-- got a tour just after their mini-range went live
16:10.48 ``Erik oh fucking son of a fucking bitch
16:11.01 ``Erik forgot about osX's case idioticity with the fs
16:44.45 CIA-33 BRL-CAD: 03starseeker * r36186 10/brlcad/branches/dmtogl-branch/ (55 files in 6 dirs): First code working toward togl integration into BRL-CAD build.
17:01.40 *** join/#brlcad samrose (n=samrose@c-71-238-70-85.hsd1.mi.comcast.net)
17:18.41 *** join/#brlcad parigaudi (n=quassel@217.91.127.94)
17:27.02 brlcad hello parigaudi
17:28.24 *** join/#brlcad Ralith (n=ralith@d142-058-086-207.wireless.sfu.ca)
17:29.03 brlcad you all shouldn't talk about Ralith like that when he's not here
17:29.07 brlcad oh hi Ralith
17:53.54 brlcad ~16*140
17:53.55 ibot 2240
18:43.29 *** join/#brlcad Ralith (n=ralith@69.90.49.189)
18:47.44 ``Erik cally speaking
18:52.28 Ralith I hear you guys bin talkin bout me behind my back >:|
19:01.34 ``Erik well
19:01.49 ``Erik if you weren't spending all your time off in the distance buried in a sheep...
19:14.25 CIA-33 BRL-CAD: 03starseeker * r36187 10/brlcad/branches/dmtogl-branch/src/other/togl/Makefile.in: Let's see if this makes togl build out of dir...
19:23.47 starseeker ah, good
19:34.33 starseeker brlcad: OK, just to organize my thinking - the goal is for there to be no direct access to ged struct components in the ged code, correct?
19:41.42 starseeker and in this vein, the struct bu_vls ged_result_str currently in struct ged should be replaced with something that supports an API of the form GED_APPEND_RESULT(struct bu_vls) and GED_NEXT_RESULT(struct ged_results)? (probably a bu_list, but that should be hidden as an implementation detail?)
19:41.47 *** join/#brlcad PrezKennedy (i=Matthew@208.43.126.194)
20:13.15 *** join/#brlcad Elrohir (n=kvirc@p5B14E086.dip.t-dialin.net)
20:19.16 *** join/#brlcad mafm (n=mafm@225.Red-83-45-72.dynamicIP.rima-tde.net)
21:03.19 brlcad starseeker: yes, unequivocally
21:03.50 brlcad "struct ged" is a black box, only accessed through api routines/macros
21:04.27 brlcad GED_NEXT_RESULT doesn't make much sense
21:04.34 brlcad you can just keep appending results
21:05.12 brlcad or is that for accessing results?
21:06.50 brlcad would avoid introducing new structures as that defeats much of the goal of making the API as simple as possible
21:08.23 brlcad typedef'd enums are probably okay if there are lists of types/codes, but not containers outside of something simple like a bu_list, or an iterator accessor pattern where you provide a callback that is called per item ala db_walk_tree()
21:57.00 ``Erik struct GED_HERE_BE_DRAGONS
21:59.09 ``Erik (once you truely understand it, I will apogogize to your gf.)
22:07.19 *** join/#brlcad puddingpimp (n=dave@gateway.quickcircuit.co.nz)
22:28.20 starseeker brlcad: yeah, APPEND to add results and NEXT to iterate over the list/array of results
22:31.08 starseeker but those were just made up on the spur of the moment
22:31.59 starseeker doesn't want to dive into rewiring libged and then find out he did it the wrong way...
22:36.08 starseeker is liking bu_lists of bu_vls full pathnames for results, but I must confess a pro/con analysis of that vs. iterator/accessor is probably in order...
22:36.37 starseeker also, where do you envision enums being helpful? (I'm not implying they're not - I'm sure they are - but I'm not following)
23:33.55 Maloeran Gez. So which one of you designed that thing? http://news.bbc.co.uk/2/hi/technology/8302903.stm
23:33.59 Maloeran :)

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.