| 00:08.16 | ``Erik | heh |
| 00:08.38 | ``Erik | up's the new machine again |
| 00:45.56 | *** part/#brlcad jdoliner (n=jdoliner@c-68-51-76-57.hsd1.il.comcast.net) | |
| 01:24.40 | *** join/#brlcad Ralith (n=ralith@216.162.199.202) | |
| 01:31.29 | *** join/#brlcad Ralith (n=ralith@216.162.199.202) | |
| 02:12.57 | *** join/#brlcad kanzure (i=bryan@66.112.232.233) | |
| 02:16.14 | starseeker | ah, there we go |
| 02:16.23 | starseeker | builds |
| 02:17.14 | brlcad | starseeker: in case you're interested, there is a gsoc mailing list if i didn't mention it before |
| 02:17.21 | brlcad | one specifically for mentors, the other for everyone |
| 02:17.45 | brlcad | you don't have to join, as I relay the important action-required info, but if you're interested |
| 02:19.08 | starseeker | brlcad: Ah, ok - thanks :-) |
| 02:19.13 | starseeker | pulls up gsoc page |
| 02:19.58 | brlcad | the one with everyone can get absurdly noisy as there are about 7000 members |
| 02:20.09 | brlcad | and even the mentor one from time to time (1000 or so members) |
| 02:20.45 | brlcad | but the mentor one is particularly insightful at times, as it's a buffet of open source devs talking about interesting matters |
| 02:21.11 | brlcad | wishes he could easily excavate the dirt between his two basements |
| 02:21.44 | starseeker | hire an army of moles ;-) |
| 02:22.56 | brlcad | I thought about that actually |
| 02:23.02 | brlcad | well not moles, but kids |
| 02:23.26 | brlcad | minimum wage manual labor ftw |
| 02:23.40 | kanzure | gah, trying to get gmsh compile with OCC support, it turns out OCC src is distributed with config.h in the wrong place |
| 02:25.16 | kanzure | brlcad: where is that mailing list? is this it? http://groups.google.com/group/google-summer-of-code-mentors-list |
| 02:26.16 | brlcad | yes |
| 02:27.11 | starseeker | puts in request to be added |
| 02:27.56 | kanzure | don't know how insightful it can be about open source development if even the archives are subscribers-only |
| 02:28.11 | brlcad | it's private with good reason |
| 02:28.30 | brlcad | mentors that are added are verified so that discussions can be private if they're about specific orgs or people |
| 02:30.18 | kanzure | I see. |
| 02:30.32 | kanzure | whaa |
| 02:30.33 | kanzure | http://geuz.org/pipermail/gmsh/2008/003110.html |
| 02:30.34 | brlcad | kanzure: mostly matters of how organizations are run, how they participate with their communities, some legal matter discussions, suggestions on how to improve a community, etc |
| 02:30.36 | kanzure | isn't that me? |
| 02:30.43 | kanzure | this is so disappointing |
| 02:30.54 | brlcad | it's not exactly relevant to the open source community at large either, specific to orgs and mentorship for the most part |
| 02:31.15 | brlcad | with a lot of prominent voices chiming in on what works and doesn't work for them |
| 02:32.05 | brlcad | kanzure: disappointing that you were trying again last year around this time? :) |
| 02:32.17 | kanzure | yes :( |
| 02:32.23 | brlcad | hehe |
| 02:32.53 | starseeker | hah - Battle for Wesnoth is a gsoc project |
| 02:33.11 | kanzure | I think what happens is that I do just enough work to exhaust myself on something only to wait just long enough to forget everything |
| 02:33.25 | kanzure | and then I discover some beautiful post a year later and it seems to have exactly nearly what I'm looking for |
| 02:33.29 | kanzure | and then I find out that it was me. |
| 02:34.05 | brlcad | starseeker: yep, their second year |
| 02:34.07 | brlcad | I pressed on them to apply last year, helped give advice on putting their org application together |
| 02:34.17 | starseeker | awesome :-) |
| 02:34.29 | brlcad | worked out nicely, they're a great group of folks |
| 02:34.32 | *** join/#brlcad Ralith (n=ralith@216.162.199.202) | |
| 02:37.40 | kanzure | there was a Slashdot comment about this recently, something like "isn't it odd that when you search the slashdot archives you sometimes find that one insightful comment from a discussion from a few years ago and when you go to look who wrote it .. it was you?" |
| 02:37.40 | starseeker | hmm - panotools has a "3D extension" idea |
| 02:37.44 | kanzure | but I've lost it. |
| 02:37.45 | starseeker | wonder if they know how hard that is |
| 02:37.57 | starseeker | kanzure: yeah, saw that :-) |
| 02:38.18 | kanzure | <-- karma whore. |
| 03:02.12 | brlcad | ~kanzure++ |
| 03:30.43 | starseeker | grrrrrrr |
| 03:30.53 | starseeker | has had it with the auto_path annoyance and starts tracking |
| 03:41.24 | kanzure | huh, anyone know who pacman87 is? |
| 03:42.52 | *** join/#brlcad pacman87 (i=500@resnet-46-102.dorm.utexas.edu) | |
| 03:42.57 | kanzure | hey pacman87 |
| 03:43.16 | brlcad | kanzure, of course |
| 03:43.28 | pacman87 | kanzure: hi |
| 03:43.32 | kanzure | he's probably living a block away from me |
| 03:43.45 | pacman87 | who, me? |
| 03:43.50 | kanzure | nods |
| 03:43.51 | kanzure | jester? |
| 03:43.56 | pacman87 | blanton |
| 03:44.00 | kanzure | castilian |
| 03:44.10 | pacman87 | ah |
| 03:44.24 | pacman87 | you're ME freshman? |
| 03:44.28 | kanzure | yes |
| 03:44.51 | pacman87 | what classes are you taking? |
| 03:44.58 | kanzure | nothing interesting |
| 03:45.11 | pacman87 | yeah, that was my first sem, too |
| 03:45.30 | pacman87 | then i took 18 hours second sem |
| 03:45.44 | kanzure | I did that backwards, but anyway |
| 03:45.53 | kanzure | didn't know that there was anyone working with brlcad on campus here |
| 03:46.06 | pacman87 | i was GSoC last year |
| 03:46.17 | brlcad | and a fantastic gsocer at that |
| 03:46.20 | pacman87 | :D |
| 03:46.22 | kanzure | hm |
| 03:46.24 | brlcad | awesome new primitives |
| 03:46.25 | kanzure | we should probably meet up some time. |
| 03:46.31 | pacman87 | added the hyp and started the rev |
| 03:46.39 | brlcad | almost finished rev |
| 03:46.43 | kanzure | maybe I'll show you the austin fab lab that I'm helping to throw together |
| 03:48.05 | starseeker | brlcad: Ah HA |
| 03:48.30 | starseeker | something, somewhere on my system, is setting ITCL_LIBRARY and ITK_LIBRARY |
| 03:48.36 | brlcad | "take on me" is a great ah-ha song |
| 03:48.49 | starseeker | looks like libtclcad is treating that as an explicit override |
| 03:48.56 | starseeker | heh |
| 03:48.57 | brlcad | that's a sure-fire way to screw up autopathing |
| 03:49.09 | pacman87 | anyone want to help me debug an "undefined reference" problem in a .so for something unrelated to brlcad? |
| 03:49.15 | brlcad | those are supposed to be explicit overrides |
| 03:49.21 | kanzure | sure, why not |
| 03:49.27 | brlcad | tcl_library, tk_library, itcl_library, itk_library |
| 03:49.30 | starseeker | shouldn't that be disabled if itcl/itk is specified as the internal copy? |
| 03:49.42 | brlcad | nope |
| 03:49.59 | brlcad | it's an environment var override -- the environment is wrong |
| 03:50.11 | kanzure | the environment is never wrong :) |
| 03:50.12 | kanzure | snickers |
| 03:50.14 | brlcad | like setting ld_library_path and expecting it to find the right libs |
| 03:50.22 | kanzure | starseeker: what are you working on? |
| 03:50.44 | pacman87 | i'm trying to compile the qt4 branch of moto4lin, and it uses p2kmoto (libp2kmoto.so) |
| 03:50.49 | starseeker | archer won't start on my system, because it's mixing the internal tcl/tk and system itcl/itk |
| 03:50.55 | brlcad | the entire point of the vars is so that regardless of what it was compiled for, I can at run-time make it use anything |
| 03:51.26 | pacman87 | moto4lin builds libp2kqt.so, which is trying to link to symbols defined in libp2kmoto.so |
| 03:52.39 | brlcad | starseeker: mixing internal tcl/tk with system itcl/itk 'can' work, but depends on the revisions, search paths, and init files it finds |
| 03:53.12 | brlcad | if it finds the 3.2 itcl init files first, it won't load a 3.4 install |
| 03:53.58 | pacman87 | "nm /usr/local/lib/libp2kmoto.so" shows the symbols are defined, but "nm libp2kqt.so" says they're undefined |
| 03:54.11 | pacman87 | and i don't know how to fix it |
| 03:56.53 | brlcad | pacman87: er, if it's a partially resolved lib, just means you have to provide both libs on the linker line (in order) |
| 03:57.31 | pacman87 | and the linker line would be in the Makefile? |
| 03:57.49 | brlcad | depends how they do their build, but yeah usually |
| 03:58.07 | brlcad | can always link by hand if need be too |
| 03:59.19 | pacman87 | how would i do that? |
| 04:05.06 | brlcad | well paste a log of it failing |
| 04:05.08 | starseeker | well, that's closer |
| 04:05.18 | brlcad | that should show you the compile/linker line |
| 04:05.21 | starseeker | looks like maybe iwidgets is messing up again... |
| 04:05.37 | brlcad | really? haven't seen iwidgets be a problem in years |
| 04:05.46 | brlcad | there is an iwidgets override too iirc |
| 04:05.58 | pacman87 | brlcad: i think i fixed it |
| 04:06.13 | pacman87 | i uncommented two lines in the CMakeLists.txt file relating to p2kmoto |
| 04:06.39 | brlcad | :)_ |
| 04:06.58 | pacman87 | ouch, segfault |
| 04:07.00 | starseeker | hang on, here's the log: http://pastebin.bzflag.bz/d3e0053b5 |
| 04:10.52 | starseeker | well, setting IWIDGETS_LIBRARY didn't fix it |
| 04:11.35 | brlcad | starseeker: looks like you're getting a system iwidgets |
| 04:11.45 | brlcad | bob has a modification on exactly that line it's failing on |
| 04:11.52 | starseeker | yeah, not surprsied |
| 04:12.03 | starseeker | is looking to figure out how to point it to local version |
| 04:12.06 | pacman87 | it probably doesn't help that moto4lin hasn't been touched in 14 months |
| 04:12.36 | starseeker | gentoo must be cluttering up the environment with a bunch of explicit setting of flags or something |
| 04:12.43 | starseeker | s/flags/paths |
| 04:12.45 | brlcad | starseeker: it's tcl, you can update the system iwidgets file directly too |
| 04:12.52 | brlcad | see if that at least fixes it |
| 04:12.54 | starseeker | true |
| 04:13.08 | brlcad | diff your system panedwindow.itk to the one in src/other/iwidgets |
| 04:13.21 | brlcad | er, src/other/iwidgets/generic/panedwindow.itk |
| 04:13.37 | brlcad | should be at least three lines, two commented out with "Bob Modification" |
| 04:14.12 | brlcad | looks like it's a bug in iwidgets that he fixed |
| 04:14.28 | starseeker | yep, see it |
| 04:15.33 | brlcad | it uses _ret without sanity checking it first, the for loop might be empty if there are no active panes |
| 04:15.46 | starseeker | yeah, that did it |
| 04:15.46 | brlcad | resulting in the error trace you saw |
| 04:16.51 | brlcad | bets ArcherCore.tcl could just wrap that line in a try/catch and it'd also prevent the failure |
| 04:17.19 | brlcad | that'd be good to try, cause if it worked, then it'd work with vanially iwidgets |
| 04:17.39 | starseeker | I think I saw three separate modifications to that file though |
| 04:17.48 | starseeker | any chance of upstream incorporating it? |
| 04:19.22 | brlcad | try that |
| 04:19.36 | CIA-40 | BRL-CAD: 03brlcad * r34104 10/brlcad/trunk/src/tclscripts/archer/ArcherCore.tcl: try catching the error in case we're using a system iwidgets that doesn't have bob's patch for referencing the unitialized _ret var in panedwindow |
| 04:19.36 | brlcad | i'm sure upstream would |
| 04:19.45 | brlcad | that is the that I was referring to, btw |
| 04:20.17 | starseeker | er, sept 2002 last iwidgets release - nevermind |
| 04:20.29 | brlcad | it's considered "done" iirc |
| 04:20.36 | starseeker | winces |
| 04:20.43 | starseeker | we're the only ones who hit that bug? |
| 04:21.50 | starseeker | ah |
| 04:22.05 | starseeker | so if it fails it can be safely ignored? |
| 04:22.14 | starseeker | that's handy |
| 04:23.49 | brlcad | if it fails, there are no active panes yet, so nothing needs to happen |
| 04:24.03 | brlcad | "should" work.. but give it a try |
| 04:24.15 | brlcad | might be more places it needs catched |
| 04:27.14 | brlcad | waits to hear the result... |
| 04:33.20 | starseeker | heh - sorry, had stuff to take care of - one sec... |
| 04:34.59 | starseeker | yep, looks like that got it |
| 04:35.03 | starseeker | nice :-) |
| 04:37.28 | brlcad | the iwidgets dev still responds to patches, so would probably be worthwhile to clean up and upload a diff |
| 04:38.28 | starseeker | rebuilds with ogl enabled to properly test archer |
| 04:40.23 | starseeker | brlcad: IIRC, we disabled ogl because it was crashing on some platforms? |
| 04:40.38 | starseeker | plus the slow updating with rt? |
| 04:40.55 | brlcad | more "bad behavior" than crashing |
| 04:41.02 | starseeker | ah |
| 04:42.16 | starseeker | well, I've never debugged ogl but there's always a first time... |
| 04:45.02 | brlcad | slow remote fb updates |
| 04:45.20 | starseeker | I thought that was for everything? |
| 04:45.31 | starseeker | recalls poking at libpkg but not seeing anything conclusive |
| 04:45.40 | starseeker | was that ogl specific? |
| 04:46.21 | brlcad | I don't recall, but it was at least remotefb->oglfb with specific sizes being much worse than others |
| 04:46.54 | brlcad | and iirc, wasn't in the libpkg layer, think that came up clean |
| 04:47.13 | starseeker | seemed to be |
| 04:49.45 | brlcad | was directly in the fbserv or sending side |
| 04:50.04 | brlcad | not crossing the wire |
| 05:43.13 | *** join/#brlcad ``Erik_ (n=erik@c-76-111-12-116.hsd1.md.comcast.net) | |
| 05:44.25 | Ralith | ogl was crashing on freebsd w/ nvidia drivers, at least for a while |
| 06:36.44 | *** join/#brlcad Ralith (n=ralith@216.162.199.202) | |
| 07:15.58 | *** join/#brlcad _sushi_ (n=_sushi_@84-72-93-63.dclient.hispeed.ch) | |
| 07:24.46 | *** join/#brlcad d_rossberg (n=rossberg@bz.bzflag.bz) | |
| 07:33.18 | Ralith | brlcad: is the content portion of a GSoC application publicly visible? |
| 07:34.27 | brlcad | Ralith: only if the student(s) post their application to the wiki or elsewhere beforehand |
| 07:34.40 | brlcad | the content becomes visible after students are selected |
| 07:35.24 | Ralith | huh. The tooltips suggest that only the abstract is ever published like that. |
| 07:47.49 | brlcad | ah, that may be |
| 07:48.18 | brlcad | yeah, was just abstrcat in previous years |
| 07:48.31 | brlcad | there's a small window to tweak those before they go live usually too |
| 07:53.55 | Ralith | so the content *isn't* ever public? |
| 07:54.03 | Ralith | at least in the past? |
| 07:54.42 | Ralith | de-TCLification application submitted. |
| 07:58.17 | Ralith | brb |
| 07:59.03 | *** join/#brlcad Ralith (n=ralith@216.162.199.202) | |
| 08:02.27 | Ralith | looking forward to any/all feedback. |
| 08:04.45 | madant | Ralith: I think google's concern is with privacy issues and all :D |
| 08:05.32 | Ralith | madant: well, I don't think I have a reasonable expectation of privacy in the legal sense for that sort of thing. |
| 08:06.20 | madant | :) |
| 08:06.58 | madant | apparently some people are :D like when we started an introduction thread in the group and some people had a problem with that too :D |
| 08:07.12 | Ralith | O.o |
| 08:07.19 | Ralith | people are weird. |
| 08:07.31 | Ralith | not that a soc app is the worst of things a future employer might find upon googling. |
| 08:07.42 | madant | hahaha :D |
| 08:08.21 | Ralith | "But I don't want the internet to know I'm ga^H^Han amateur engineer!" |
| 08:08.25 | madant | I think the best approach as brlcad mentioned is to post the application at appspot and then reflect the same on the wiki and even better on the mailing list.. :) i will be doing that today :) |
| 08:08.41 | Ralith | what's this now? |
| 08:09.04 | Ralith | applicants are expected to mirror their applications on the mailing list and the wiki? |
| 08:09.25 | madant | not as a compulsion or anything :) |
| 08:10.26 | madant | the only technical necessity is the appspot application :) |
| 08:11.08 | Ralith | well, sure, but you're not *technically* expected to make contributions to establish your ability to read and write useful code. |
| 08:11.33 | Ralith | I just didn't see any mention of an expectation to repost your application anywhere on the wiki instructions |
| 08:11.51 | Ralith | brlcad: can you clarify? |
| 08:13.58 | brlcad | Ralith: great (regarding the detcl), should have feedback sometime tomorrow hopefully |
| 08:14.19 | brlcad | or you can keep pinging any of the mentors here too ;) |
| 08:14.23 | madant | Ralith: did you read the "Applications, wiki, and early submission" mail from brlcad :) .. ah he is here :D |
| 08:14.52 | Ralith | brlcad: heh |
| 08:15.07 | Ralith | madant: I only just recently got on the ML, and I don't usually keep a close eye on my email; let me load up thunderbird. |
| 08:15.31 | Ralith | is very much an IRC person. |
| 08:16.10 | madant | http://www.nabble.com/Applications%2C-wiki%2C-and-early-submission-td22731057.html |
| 08:16.11 | brlcad | and that's right -- there's no "requirement" to cross-post to the wiki -- just some folks like to do that to share their application with others or to get discussion/feedback before it goes into the socghop interface |
| 08:16.15 | Ralith | oh, crap, forgot to confirm my subscription. |
| 08:16.25 | Ralith | madant: thanks |
| 08:16.36 | Ralith | brlcad: ah, so no point to once it's on google? |
| 08:17.08 | Ralith | thought the wiki bit was odd because those concerned would be able to see it on google already, and that would explain it. |
| 08:17.41 | brlcad | Ralith: no -- the same point could still hold, sharing the details of your application with the community at large |
| 08:18.02 | Ralith | hm, that reminds me, I should probably grab a wiki account. |
| 08:18.12 | Ralith | ooh, recaptcha! |
| 08:18.14 | Ralith | kudos! |
| 08:18.34 | madant | :) i love it too :D |
| 08:19.07 | Ralith | oh, looks like I already registered and forgot. |
| 08:19.09 | brlcad | there's no real "secret" to an application imho, some orgs require them to be hashed out publicly before they go into the app interface even |
| 08:20.53 | brlcad | example, http://my.bzflag.org/w/User:IneQuation.pl |
| 08:21.28 | brlcad | or http://my.bzflag.org/wiki/index.php?title=User:Nightstrike&oldid=5617 |
| 08:22.08 | brlcad | or http://my.bzflag.org/w/User:Will07c5 .. where we could quickly tell the student that his priorities were almot all wrong originally and woefully lacking in detail |
| 08:28.41 | CIA-40 | BRL-CAD: 03Ralith 07http://brlcad.org * r1314 10/wiki/User:Ralith: Page created w/ GSoC applications |
| 08:28.53 | Ralith | CIA is neat. |
| 08:29.15 | Ralith | are the bots open source? |
| 08:29.22 | Ralith | i.e. would it be easy to host one independent of the service? |
| 08:29.58 | brlcad | yeah, the whole system can be retrieved and run independently |
| 08:30.17 | brlcad | it's a pretty complex system though |
| 08:30.45 | Ralith | neat! |
| 08:30.56 | Ralith | it's the ability to do so that matters, really |
| 08:31.07 | brlcad | the bots themselves are pretty simple -- they just say what they're told to say |
| 08:31.14 | Ralith | although I can't help but think that it might not be hard to make a lightweight version that just attached to a single repo and/or set of RSS feeds |
| 08:31.46 | brlcad | if that's all you want, there are probably easier means (like libIRC or an existing bot like supybot or blootbot) |
| 08:32.03 | Ralith | I thought that was the major selling point? |
| 08:33.05 | brlcad | selling point of what? |
| 08:33.54 | brlcad | cia's focus is more around commit data, which isn't caputureed well with rss fields and suffers other drawbacks |
| 08:34.24 | brlcad | there are bots (supy for example) that will poll an rss feed and announce changes |
| 08:34.35 | brlcad | tend to be rather flakey |
| 08:35.38 | Ralith | I mean, I thought the main point of CIA was to offer a commit announce bot. |
| 08:36.02 | Ralith | you could implement a bot by more reliable means than RSS feeds, e.g. post-commit hooks; I just mentioned RSS feeds cuz they're nice and generic. |
| 08:36.03 | brlcad | yeah, one of the main original points |
| 08:36.32 | brlcad | cia uses / prefers post-commit hooks |
| 08:36.52 | brlcad | or it'll take e-mail notifications |
| 08:37.03 | brlcad | or it'll take xml rpc updates |
| 08:37.12 | brlcad | or it can poll for updates |
| 08:38.35 | Ralith | right |
| 08:38.42 | Ralith | but it seems a little odd to use a centralized service for that |
| 08:38.48 | Ralith | rather than instantiate something lightweight for each project |
| 08:39.19 | brlcad | the complexity in the system is that it's geared with performance and scalability in mind, also keeps track of daily commit histories, automatically presents the data through a load-balanced web site, and a little bit more |
| 08:39.52 | brlcad | mostly because it's so easy to set up and someone else maintains it |
| 08:40.42 | brlcad | you just drop in a commit notifier, and that's it .. everything else including running bots, the web site, the project listing on the site, all happening automatically |
| 08:41.47 | brlcad | kinda like how ibot is in about 100 channels .. even though there are dozens of irc factoid bots out there that projects could/do use by themselves |
| 08:42.27 | brlcad | someone else manages and maintains it, and I don't have to care |
| 08:44.11 | archivist | I run a multi channel bot, can be fun |
| 08:44.55 | Ralith | good point. |
| 08:45.10 | Ralith | that makes a lot of sense |
| 11:09.34 | *** join/#brlcad madant_ (n=d@117.196.143.31) | |
| 11:10.22 | d-lo | *readreadread* |
| 11:10.36 | d-lo | mernin all |
| 11:10.52 | madant_ | mernin d-lo :) |
| 11:27.52 | *** join/#brlcad madant_ (n=d@117.196.135.15) | |
| 12:16.12 | starseeker | hmm. built with enable-profiling, but don't see any a.out files from rt |
| 12:22.50 | *** join/#brlcad poolio (n=poolio@bz.bzflag.bz) | |
| 12:28.45 | ``Erik | hm, what is the bots webpage url? |
| 12:29.08 | ``Erik | d-lo: you missed the big jg crash in e79! |
| 12:31.58 | d-lo | I know. |
| 12:32.21 | d-lo | Care level is pretty low. Good weather (minus the tornado :/) this weekend. hard to stay indoors :) |
| 12:33.11 | ``Erik | it was sloppy and slow, there were plenty of long breaks of activity :) |
| 12:33.40 | ``Erik | breaks from activity, rather |
| 12:34.04 | ``Erik | yesterdays hail almost makes me happy my car is in the shop and not infront of the house heh |
| 12:35.06 | *** join/#brlcad madant (n=d@117.196.137.242) | |
| 12:35.56 | d-lo | heh, thats kinda funny. We didn't get hail up at our place. just a nice view of the 'nado to the NW of us and a spit of rain. |
| 13:10.29 | *** join/#brlcad madant_ (n=d@117.196.138.162) | |
| 13:21.41 | starseeker | uh... tornado? |
| 13:21.47 | starseeker | pics? |
| 13:24.28 | d-lo | Heh, didn't snap any. I looked out the window to see some nasty looking storm clouds. Flipped on the TV only to see News channel 8 basicly saying "OMGOMGOMGOMGTORNADO". Looked out the window again and saw what looked like a 'funnel', but mostly obscured by neighboring houses. |
| 13:24.41 | starseeker | ah :-) |
| 13:24.53 | d-lo | Power went out abou then, and I called the ball and got the family into the basement. |
| 13:25.05 | starseeker | indeed |
| 13:25.46 | d-lo | Nothing bad happened, infact, the rain was quite pathetic. no cool air-raid sirens from the local VFDs... no awesome 2" hail... just about 5 minutes of torrential downpour. |
| 13:25.51 | d-lo | then... all over. |
| 13:26.00 | d-lo | su was out 30 mins later :/ |
| 13:26.04 | d-lo | su= sun |
| 13:26.24 | starseeker | phew |
| 13:27.20 | d-lo | yeah, I guess I should be more thankful than disappointed :) |
| 13:28.19 | starseeker | unless you REALLY don't like your current roof/home siding |
| 13:31.39 | d-lo | I just wanted to see a car get thrown like a toy tonka truck.... its on my Bucket list. |
| 13:38.28 | starseeker | ah |
| 13:38.43 | starseeker | well, you could take up storm chasing as a hobby ;-) |
| 13:38.51 | starseeker | doubt the insurance guys would go for it though |
| 13:41.00 | d-lo | Well, if i ever become independantly wealthy, a widower and without children.... then storm chasing might be fun :) |
| 14:31.12 | starseeker | realizes he needs to bring set theory/Venn diagram logic into the point question |
| 14:31.39 | starseeker | wonder if ASCII has set notation characters |
| 14:46.48 | *** join/#brlcad mafm_ (n=mafm@223.Red-83-49-86.dynamicIP.rima-tde.net) | |
| 14:53.31 | mafm_ | hi |
| 15:10.00 | *** join/#brlcad cad73 (n=803f1020@bz.bzflag.bz) | |
| 15:10.33 | *** join/#brlcad madant (n=d@117.196.142.56) | |
| 15:26.28 | *** join/#brlcad ziactn (n=7934900c@bz.bzflag.bz) | |
| 15:33.56 | CIA-40 | BRL-CAD: 03brlcad * r34105 10/brlcad/trunk/src/external/Cubit/ (Makefile.am g-sat.cpp g-sat.cxx): Rename g-sat.cxx to g-sat.cpp to be consistent with the other C++ sources in the 'brlcad' module. |
| 15:43.27 | CIA-40 | BRL-CAD: 03brlcad * r34106 10/brlcad/trunk/src/librt/ (bezier_2d_isect.c comb.c cut.c db5_bin.c db5_types.c): apparently don't need to include nmg.h, not using nmg routines/structures. |
| 15:47.11 | madant | brlcad never sleeps :P |
| 15:48.13 | CIA-40 | BRL-CAD: 03brlcad * r34107 10/brlcad/trunk/TODO: reverting a very old decision. separate the nmg routines from librt into their own library. |
| 15:51.15 | CIA-40 | BRL-CAD: 03brlcad * r34108 10/brlcad/trunk/TODO: downgrade expected list of items for next release until later save for a few relatively simple ones |
| 15:51.58 | CIA-40 | BRL-CAD: 03brlcad * r34109 10/brlcad/trunk/TODO: bob made the inside command work on edit primitives (this was on-schedule, just hadn't been moved up for current release) |
| 15:55.26 | *** join/#brlcad elite01 (n=omg@unaffiliated/elite01) | |
| 15:59.59 | CIA-40 | BRL-CAD: 03brlcad * r34110 10/brlcad/trunk/misc/win32-msvc8/ (Makefile.am g_diff/ g_lint/ g_transfer/): start of stubs for the rest of the gtools |
| 16:02.01 | CIA-40 | BRL-CAD: 03brlcad * r34111 10/brlcad/trunk/misc/win32-msvc8/ (7 files in 4 dirs): step closer on gtools stub using g_qa.vcproj as forked base |
| 16:04.34 | CIA-40 | BRL-CAD: 03brlcad * r34112 10/brlcad/trunk/misc/win32-msvc8/ (3 files in 3 dirs): last step, update content to compile the right source files. probably still need minor source file changes to compile on windows. |
| 16:06.59 | CIA-40 | BRL-CAD: 03brlcad * r34113 10/brlcad/trunk/misc/win32-msvc8/brlcad/brlcad.sln: can't imagine the uuid's are right, but this will hook the three new gtools into the windows build -- g_diff, g_lint, and g_transfer. |
| 16:11.43 | CIA-40 | BRL-CAD: 03brlcad * r34114 10/brlcad/trunk/misc/win32-msvc8/ (4 files in 4 dirs): update the deps. only g_qa needs libged, g_transfer needs libpkg and winsock |
| 16:12.46 | CIA-40 | BRL-CAD: 03brlcad * r34115 10/brlcad/trunk/src/gtools/g_diff.c: g_diff doesn't use ged |
| 16:34.02 | CIA-40 | BRL-CAD: 03brlcad * r34116 10/brlcad/trunk/src/gtools/ (g_diff.c g_lint.c g_qa.c g_transfer.c): ws, style consistency cleanup |
| 16:35.58 | CIA-40 | BRL-CAD: 03brlcad * r34117 10/brlcad/trunk/src/gtools/g_diff.c: ugh, libged atrociously still has wdb_obj and friends (e.g., wdb_create_cmd) so DO have to include ged.h here. suckage. |
| 16:36.56 | *** join/#brlcad ibot (i=ibot@rikers.org) | |
| 16:36.56 | *** topic/#brlcad is BRL-CAD Open Source Solid Modeling || http://brlcad.org || http://sf.net/projects/brlcad || Release 7.14.4 posted this weekend (20090313) || GSoC 2009 Begins! | |
| 16:37.31 | CIA-40 | BRL-CAD: 03brlcad * r34118 10/brlcad/trunk/misc/win32-msvc8/g_diff/g_diff.vcproj: g_diff does use libged for the blasted wdb_obj funcs like wdb_create_cmd, wdb_get_tcl, and wdb_init_obj. |
| 16:42.54 | *** join/#brlcad dreeves (n=c752f348@bz.bzflag.bz) | |
| 16:43.38 | CIA-40 | BRL-CAD: 03brlcad * r34119 10/brlcad/trunk/ (TODO include/ged.h): annotate the fixme items that the wdb and view objects in libged still need to be refactored/renamed |
| 16:46.44 | CIA-40 | BRL-CAD: 03brlcad * r34120 10/brlcad/trunk/TODO: might not 'compile', but g_lint is hooked into the windows build now so remove the todo until we know there's a problem. |
| 16:57.22 | brlcad | Ralith: interesting thread on /. about firefox's planned taskbar -- that's very similar to what we were discussing the other day |
| 16:57.26 | brlcad | http://people.mozilla.com/~faaborg/files/labs/taskFoxi1.png |
| 16:58.37 | brlcad | (particularly idea2 if you want to get specific on overlay integration and look n' feel) |
| 16:59.04 | brlcad | except for the silly ellipses around the action/verbs |
| 17:24.17 | brlcad | things are looking really well for those that have already put an application in.. submission counts are quite low this year thusfar |
| 17:24.39 | brlcad | much better chances that previous years .. wonder how much of a "get it in at the last minute" rush there will be |
| 17:24.48 | madant | :) |
| 17:25.09 | madant | i expected a lot more discussion on the irc myself |
| 17:25.26 | brlcad | every year seems to be a different twist |
| 17:25.33 | madant | will submit in an hour :) |
| 17:25.35 | mafm | maybe they extend the period for 1 month this year :P |
| 17:25.44 | brlcad | mafm: unlikely |
| 17:25.48 | madant | mafm lol |
| 17:26.14 | brlcad | i mean, maybe .. but last year the week or so extension didn't really do much but add some last minute desperation submissions that were pretty bad |
| 17:26.33 | brlcad | got the count up, but none that made it |
| 17:26.57 | madant | and besides i wonder if google is really concerned about increasing the total number of applications this year :) |
| 17:27.04 | brlcad | this year, there just hasn't been nearly as much advertising (by google, didn't get the /. headline, etc) |
| 17:27.17 | brlcad | yeah, I'm not worried |
| 17:27.27 | brlcad | I mean, we're only asking for 3 or 4 slots |
| 17:27.31 | mafm | Debian also talks about less and less applicants |
| 17:27.42 | mafm | anybody taking the GUI one? |
| 17:27.47 | brlcad | so we only need 3 or 4 good applications, and we already have at least two that I'd be happy with :) |
| 17:28.50 | brlcad | mafm: actually, there is a good application in application in to continue what you started |
| 17:29.23 | d-lo | whoa.... deja-vu |
| 17:29.37 | brlcad | s/a good application/an/ :) |
| 17:29.40 | d-lo | needs to know if it was *the same* black cat..... |
| 17:30.01 | d-lo | ;) |
| 17:30.05 | brlcad | not enough samples to qualify it as good or bad yet :) |
| 17:30.44 | madant | can't figure out the word limit for the proposal |
| 17:31.00 | mafm | 7500 last year, IIRC |
| 17:31.07 | mafm | erm, characters |
| 17:31.13 | madant | haha :D |
| 17:31.50 | mafm | brlcad: one of the two that you're happy with? from whom? |
| 17:32.00 | madant | wonders whether writing or reading is tougher |
| 17:32.38 | brlcad | mafm: why applicants of course |
| 17:33.37 | madant | brlcad: :D |
| 17:33.48 | mafm | I mean the nickname, so I'll talk to him/her if I see $1 |
| 17:34.15 | brlcad | no point in calling people out, hang around and you'd find out ;) |
| 17:34.53 | brlcad | i know what you meant, just not going to get into who has submitted what apps unless they reveal it themselves just for sake of discretion until selections are made |
| 17:34.54 | madant | :) besides it is not fair to announce names before the final list is out |
| 17:35.35 | mafm | it was just curiosity, no list is definitive until all applications are in, anyway |
| 17:35.42 | brlcad | also what's cool about putting your proposals on the wiki -- so other students can see what they're up against and can improve each other's work or submit for other areas |
| 17:36.39 | mafm | I didn't found the new ones in the wiki, where are they linked? |
| 17:36.44 | brlcad | thinks everyone should submit for (exactly) two ideas regardless just so desirable individuals aren't left out because of a competitive submission topic |
| 17:37.06 | mafm | (that's why I came to the channel, to ask directly :) ) |
| 17:37.09 | brlcad | like the web geometry database last year.. geez. There was more than a half dozen submissions for it |
| 17:37.25 | brlcad | you can check the recent changes |
| 17:37.25 | madant | and none got selected :O |
| 17:37.30 | brlcad | yep |
| 17:37.34 | brlcad | actually one did |
| 17:37.47 | brlcad | but she was lost to a coin toss with another org that conflicted |
| 17:38.09 | brlcad | so one of you four last year got lucky :) |
| 17:38.17 | madant | ah nice :) |
| 17:39.23 | brlcad | and yes, there was an actual coin toss by leslie since we both wanted her with pretty equal justifications and the student wanted to work with both equally |
| 17:39.40 | mafm | I had already done that, but only saw something new for libpc (aside from older devels) in the last few days |
| 17:39.45 | mafm | gonna dig further |
| 17:39.50 | madant | brlcad: wow talk about coins and destiny :P |
| 17:40.24 | brlcad | mafm: does it matter? shouldn't you be working on your proposals? :) |
| 17:40.32 | brlcad | or do you intend to mentor? |
| 17:40.35 | brlcad | or just watch? :) |
| 17:40.47 | brlcad | (all three are options, of course) |
| 17:41.23 | brlcad | madant: yeah, i'm pretty happy with last year's results though |
| 17:41.43 | brlcad | wished more progress could be made, of course, but that can almost always be said ;) |
| 17:41.47 | madant | mafm: http://brlcad.org/wiki/User:Ralith read this ? |
| 17:42.37 | mafm | madant: reading it right now |
| 17:42.56 | madant | brlcad: :D lets hope this year is much more productive.. I feel like we have a bunch of very interesting people.. i mean the ones who came on the channel at least |
| 17:42.57 | mafm | brlcad: I don't think that I'm eligible, and besides that I don't think that I would apply this year |
| 17:43.11 | mafm | just checking the proposals for curiosity |
| 17:43.34 | mafm | and to maybe help people in that part, even if not officially a mentor |
| 17:43.47 | mafm | but if it's ralith, he's already an insider :) |
| 17:43.52 | brlcad | mafm: btw, I believe I am about ready to make an executive decision regarding the gui |
| 17:44.07 | brlcad | mafm: ah right, understand |
| 17:44.12 | brlcad | then you should mentor! |
| 17:45.13 | brlcad | ralith IS an insider .. but also just getting started with coding on BRL-CAD, getting new devs up and coding is always good even if they've been in the channel for years :) |
| 17:45.27 | brlcad | mafm: even as a backup mentor, you'll at least get a t-shirt |
| 17:45.30 | madant | Mentor Manuel Montecelo :) |
| 17:45.31 | brlcad | :) |
| 17:46.26 | mafm | well, it depends on the obligations |
| 17:47.22 | mafm | my father died a few days ago and I have tonnes of things to do, delayed and new ones due to this "ocurrence" |
| 17:47.59 | mafm | what's the executive decision about the gui, using another toolkit (reading proposal right now)? |
| 17:49.26 | madant | mafm: really sorry to hear that. I know irc doesnt offer a good channel to convey what one feels very clearly .. nyways.. hope he had a great time |
| 17:49.45 | mafm | thanks |
| 17:49.57 | mafm | he did, most of the time |
| 17:50.28 | madant | today was the death anniversary of my grandmom. went to the cemetery, church etc :) |
| 17:50.33 | mafm | he was kind of a Big Fish kind of person, not so exaggerated, but well |
| 17:51.33 | madant | mafm: hehe :) my dad is a Big Fish kind of person too :D |
| 17:52.16 | mafm | he died of cancer with multiple "branches", I think that the Big Fish character also did |
| 17:52.48 | mafm | so the last months were not very pleasant :( |
| 17:52.56 | madant | hmm. can imagine |
| 17:53.57 | madant | brlcad: should i first post my assesment of the current status of libpc plus my plan first to the group / wiki and then submit to appspot like after 2 days probably ? |
| 17:54.11 | mafm | but well, he did not suffer so much as it was expected etc, so at least it's something |
| 17:54.42 | madant | mafm: when was this ? |
| 17:54.59 | mafm | died? less than 2 weeks ago |
| 17:55.31 | madant | ok .. what are ur plans now that school is over ? |
| 17:55.42 | mafm | madant: you were from india, right? when you talk about church, which one? |
| 17:55.55 | madant | well I am a roman catholic :) |
| 17:56.16 | madant | has ancestors who were baptised by St. Thomas around 2000 years ago apparently :D |
| 17:56.41 | madant | is definitely not very religious though :P |
| 17:56.58 | mafm | are you from Goa or some region with more european culture, or doesn't have anything to do? |
| 17:57.16 | madant | mafm: nope 100 % Indian :P |
| 17:57.42 | mafm | I'm not officially out of school, the final project to get the degree was to be presented in 20th of march, but it couldn't be... so I still have to finish some bits and present it |
| 17:58.49 | mafm | that's why I don't know exactly whether I would be eligible for gsoc :) |
| 17:59.29 | madant | ah.. u could ask Leslie you know.. at the channel ? |
| 18:00.10 | mafm | IIRC Goa even had candidates for the Pope "elections", and never went there but several friends did |
| 18:00.30 | mafm | well, I don't know whether I'd like to apply to something |
| 18:00.54 | mafm | my first option would be to continue the GUI project, but well :D |
| 18:01.39 | mafm | and after many years without holidays, it doesn't look a very attractive prospect (even if gsoc is) |
| 18:01.49 | mafm | summer holidays, that is |
| 18:02.35 | madant | :) |
| 18:03.37 | mafm | btw, I was supposed to be roman catholic too, but I only believed as a child |
| 18:03.44 | brlcad | mafm: really sorry to hear about your father, I can only imagine the suffering and pain but hope you're well |
| 18:04.34 | brlcad | madant: sure, that'd work |
| 18:04.37 | mafm | besides, nobody expects the spanish inquisition, with "almost-fanatical devotion to the pope" chief weapon, etc :P |
| 18:04.59 | madant | mafm: me too :) I was even an altar boy. |
| 18:05.20 | mafm | brlcad: thanks too :) |
| 18:05.49 | madant | goes to church on sundays so that his mom doesn't feel "All is lost" :D |
| 18:06.42 | mafm | I'm more or less OK, the worse part was the initial shock, and to pretend that everything was alright (so neither him or other people in the family knew the truth) |
| 18:07.50 | mafm | I stopped going to the church at around 15 I think, and now that my father died the priest wanted me to repent and come back to the shepherd :P |
| 18:08.05 | madant | :) |
| 18:08.25 | mafm | brlcad: so the dramatic change was about Qt? |
| 18:08.41 | brlcad | hm? |
| 18:08.43 | brlcad | dramatic change? |
| 18:10.10 | mafm | brlcad: well, whatever the executive decision would be :) |
| 18:11.31 | hippieindamakin8 | hey ppl |
| 18:11.39 | mafm | hi hippieindamakin8 |
| 18:11.57 | hippieindamakin8 | hey mafm ! so are u mentoring this year ? |
| 18:12.46 | mafm | dunno yet |
| 18:13.38 | hippieindamakin8 | madant, you are from goa or kerala ? |
| 18:13.56 | mafm | I wasn't involved in development after gsoc except for a bit towards the end of last year, IIRC |
| 18:15.24 | madant | hippieindamakin8: kerala :) |
| 18:16.23 | hippieindamakin8 | madant, i just googled you and ur current location is thrissur apparently :) |
| 18:19.59 | madant | wow :) |
| 18:20.10 | madant | well it is around 40 kilometer off |
| 18:20.33 | madant | fears google |
| 18:20.38 | hippieindamakin8 | madant, cant expect more from google maps :). |
| 18:20.56 | hippieindamakin8 | madant, that must be an awesome picturesque country side then |
| 18:21.26 | d-lo | if you do a traceroute to www.google.com, you will see that the primary DNS server is www.sky.net ;) |
| 18:22.44 | mafm | madant: you're in for another edition with a new proposal for libpc? |
| 18:23.05 | madant | yep :) |
| 18:23.25 | madant | god's own country :) |
| 18:23.36 | madant | or countryside as the case maybe :P |
| 18:24.30 | mafm | well, good luck then :) |
| 18:24.41 | madant | mafm: yes, would love to make some progress :) |
| 18:25.38 | madant | mafm: thank you :) , i do need a lot of luck to stay on track :) |
| 18:28.59 | mafm | madant: what are you doing at school, degree, master...? |
| 18:29.55 | madant | mafm: masters |
| 18:30.17 | madant | mafm: you planning on a masters after ur thesis ? |
| 18:30.18 | hippieindamakin8 | madant, where ? |
| 18:31.40 | starseeker | is starting to wonder if he'll need the constraint system to resolve this. erk |
| 18:31.41 | madant | hippieindamakin8: i have two options :) 1 is IISc , 2 IIMB |
| 18:32.04 | madant | starseeker: :) resolve what ? |
| 18:32.05 | mafm | madant: not sure yet, I don't like much the opportunities in the univs around, and I cannot easily move elsewhere in the next months |
| 18:32.40 | hippieindamakin8 | madant, in architechture ? or smthing else like cs ? |
| 18:32.56 | hippieindamakin8 | i.e at IISC |
| 18:33.03 | madant | IISc : Sustainability IIMB : MBA :) |
| 18:33.14 | starseeker | madant: looking at comparing two points with error bounds, deciding if they are similar or not. In isolation that's pretty simple, but if there are three or more points "in range" look out |
| 18:33.26 | starseeker | er s/similar/the same/ |
| 18:33.58 | madant | starseeker: did u see the discussion on checking the collinearity of three points using the determinant ? |
| 18:34.36 | hippieindamakin8 | madant, wat is this discussion abt ? |
| 18:34.50 | madant | wonders whether that would be any use here though |
| 18:35.21 | madant | it was an excerpt from Beautiful Code book |
| 18:35.30 | starseeker | madant: that might help when it comes time to look at implementation techniques, but first I need a decision methodology |
| 18:36.19 | madant | starseeker: can you point me to the problem ? as in any doc /code ? i remember seeing u and sean talking about it but didn't pay attention |
| 18:36.32 | starseeker | i.e. given 3,4,...,n points with overlapping error bounds, how do I decide which ones should be regarded as being the same point and which should be treated as different? |
| 18:36.51 | starseeker | there are a few notes in doc/TODO.BREP |
| 18:37.37 | CIA-40 | BRL-CAD: 03117.196.142.56 07http://brlcad.org * r1315 10/wiki/User:Homovulgaris: Proposal Draft Added |
| 18:38.05 | madant | starseeker: error bounds? |
| 18:38.53 | starseeker | BRL-CAD has a tolerance on dimensions - anything smaller than that tolerance can be ignored when, for example, doing overlap checks |
| 18:39.24 | starseeker | so any point has a value (x,y,z) |
| 18:39.51 | starseeker | x has a tolerance or error bound of delta_x |
| 18:39.56 | starseeker | same for y and z |
| 18:40.16 | starseeker | x, y, and z themselves are presumably stored as floating point numbers |
| 18:41.27 | madant | ok.. so different points would be points which are further apart than sqrt(delta_x^2+delta_y^2+delta_z^2) |
| 18:41.29 | starseeker | so given two points (x1,y1,z1) and (x2,y2,z2), if their delta ranges overlap they might need to be treated as being the "same" point |
| 18:42.14 | starseeker | madant: maybe, but I'm not sure if the error bound is spherical or cubic in x,y,z space |
| 18:42.36 | hippieindamakin8 | starseeker, it is cubic in space |
| 18:42.47 | starseeker | that's been my assumption |
| 18:42.55 | hippieindamakin8 | starseeker, and it depends on the extent of overlap isnt it ? |
| 18:43.25 | starseeker | hippieindamakin8: it might. these are the issues I'm trying to work through now |
| 18:43.35 | madant | hmm.. in effec the check is whether (x1-delta_x1,x1+delta_x1) and (x2-delta_x2,x2+delta_x2) intervals overlap right ? |
| 18:43.41 | starseeker | right |
| 18:43.43 | madant | and similarly for other coordinates |
| 18:44.11 | hippieindamakin8 | starseeker, cubic if error is the worst case |
| 18:44.29 | starseeker | but if one point overlaps with two other points and those two other points don't overlap, it makes no sense to treat both of those points as being the same as the original point |
| 18:44.37 | hippieindamakin8 | *if error is taken in the worst case |
| 18:44.52 | madant | hmm.. so if there are n coordinates you the worst case is checking all the C(n,2) combinations right ? |
| 18:44.59 | starseeker | A may equal B, and A may equal C, but if B != C then A may NOT equal B AND C |
| 18:45.37 | hippieindamakin8 | madant, rt but split into regions |
| 18:45.40 | starseeker | madant: I'm not sure - hopefully some sort of decision metric can be found, but it might be that to correctly decide things such a worst case is unavoidable |
| 18:45.43 | hippieindamakin8 | *the space |
| 18:46.38 | madant | starseeker: interesting problem ? did you find any literature on this ? |
| 18:46.58 | starseeker | madant: Not yet. I've been looking, but it may be I don't know the correct keywords |
| 18:46.58 | madant | s/interesting problem ? / interesting problem ! :) |
| 18:47.50 | starseeker | it gets even more complex if you allow different points to have different deltas |
| 18:48.09 | pacman87 | starseeker: have you looked at clustering algorithms? |
| 18:48.46 | madant | so in the above scenario what do u want the result to be ? A,B and C all "mutually" different |
| 18:49.25 | starseeker | madant: that's one of the questions - what the "correct" answer should be |
| 18:49.32 | pacman87 | if two points are the same, do you treat that location as the mean of the two points? |
| 18:49.53 | starseeker | pacman87: no decision made about how to handle it yet :-) |
| 18:50.09 | starseeker | pacman87: clustering? no I hadn't |
| 18:50.37 | madant | starseeker: what is the final purpose of categorizing the points ? i mean is the result to be used in some other process ? |
| 18:50.37 | hippieindamakin8 | starseeker, why cant there be an algo which in its worst case is O(n^2), where you check if each point lies in the error bound a point and go checking for each point |
| 18:51.02 | starseeker | hippieindamakin8: That's probably what will happen, or something like it |
| 18:51.09 | hippieindamakin8 | and this can be worked upon using randomized algos |
| 18:51.19 | pacman87 | http://en.wikipedia.org/wiki/Cluster_analysis |
| 18:51.26 | hippieindamakin8 | or planar subdivision and parallel analysis |
| 18:51.53 | starseeker | madant: When we evaluate breps with a ray, we may hit edge cases where the question of point equality will come up. If so, we want to have a consistent procedure implemented that we can call |
| 18:52.55 | starseeker | pacman87: Hmm, yeah, looks like I was thinking about clustering algorithms without knowing what they were called :) |
| 18:54.20 | hippieindamakin8 | pacman87, yup :P (i worked on k-means when working on my industrial intern) |
| 18:54.24 | pacman87 | one of my programming assignments covered it, but in the context of data mining |
| 18:54.27 | *** join/#brlcad boerrebj_ (n=boerrebj@ti0036a380-1283.bb.online.no) | |
| 18:55.03 | pacman87 | off to class, back later |
| 18:59.29 | hippieindamakin8 | starseeker, similar to wat i said.. kmeans takes the worst possible time of O(n^2) and the expected value is O(n) which isnt bad |
| 18:59.35 | hippieindamakin8 | and it is easier to implement |
| 19:01.36 | *** join/#brlcad boerrebj__ (n=boerrebj@ti0036a380-1776.bb.online.no) | |
| 19:01.37 | madant | still can't figure out how k-means algorithm solves starseeker's dilemma of what the result should be |
| 19:02.13 | hippieindamakin8 | again ponders over the problem. |
| 19:22.15 | *** join/#brlcad ChanServ (ChanServ@services.) | |
| 19:22.15 | *** join/#brlcad samrose (n=samrose@c-24-11-214-181.hsd1.mi.comcast.net) | |
| 19:22.15 | *** join/#brlcad boerrebj__ (n=boerrebj@ti0036a380-1776.bb.online.no) [NETSPLIT VICTIM] | |
| 19:22.15 | *** join/#brlcad boerrebj_ (n=boerrebj@ti0036a380-1283.bb.online.no) [NETSPLIT VICTIM] | |
| 19:22.15 | *** join/#brlcad mafm (n=mafm@223.Red-83-49-86.dynamicIP.rima-tde.net) [NETSPLIT VICTIM] | |
| 19:22.15 | *** join/#brlcad poolio (n=poolio@bz.bzflag.bz) [NETSPLIT VICTIM] | |
| 19:22.15 | *** join/#brlcad Ralith (n=ralith@216.162.199.202) [NETSPLIT VICTIM] | |
| 19:22.15 | *** join/#brlcad ``Erik (n=erik@c-76-111-12-116.hsd1.md.comcast.net) [NETSPLIT VICTIM] | |
| 19:22.15 | *** join/#brlcad MinuteElectron (n=MinuteEl@unaffiliated/minuteelectron) [NETSPLIT VICTIM] | |
| 19:22.15 | *** join/#brlcad hippieindamakin8 (n=hippiein@202.3.77.38) [NETSPLIT VICTIM] | |
| 19:22.46 | *** join/#brlcad d-lo (n=claymore@bz.bzflag.bz) [NETSPLIT VICTIM] | |
| 19:22.46 | *** join/#brlcad brlcad (n=sean@bz.bzflag.bz) [NETSPLIT VICTIM] | |
| 19:22.46 | *** join/#brlcad Axman6 (n=Axman6@pdpc/supporter/student/Axman6) [NETSPLIT VICTIM] | |
| 19:22.46 | *** join/#brlcad b0ef (n=b0ef@062016142244.customer.alfanett.no) [NETSPLIT VICTIM] | |
| 19:22.46 | *** join/#brlcad PrezKennedy (i=Matthew@whitecalf.net) [NETSPLIT VICTIM] | |
| 19:22.46 | *** join/#brlcad Maloeran (n=maloeran@glvortex.net) [NETSPLIT VICTIM] | |
| 19:22.46 | *** join/#brlcad dtidrow (n=Don@c-68-62-76-34.hsd1.mi.comcast.net) [NETSPLIT VICTIM] | |
| 19:22.46 | *** join/#brlcad starseeker (n=starseek@bz.bzflag.bz) [NETSPLIT VICTIM] | |
| 19:22.46 | *** join/#brlcad yukonbob (i=1000@s142-179-54-198.bc.hsia.telus.net) [NETSPLIT VICTIM] | |
| 19:22.46 | *** join/#brlcad archivist (n=archivis@host81-149-119-172.in-addr.btopenworld.com) [NETSPLIT VICTIM] | |
| 19:22.46 | *** join/#brlcad bjorkintosh (n=bjork@ip72-204-40-138.fv.ks.cox.net) [NETSPLIT VICTIM] | |
| 19:22.46 | *** join/#brlcad alex_joni (n=juve@emc/board-of-directors/alexjoni) [NETSPLIT VICTIM] | |
| 19:22.46 | *** join/#brlcad CIA-40 (n=CIA@208.69.182.149.simpli.biz) [NETSPLIT VICTIM] | |
| 19:22.46 | *** join/#brlcad piksi (i=piksi@pi-xi.net) [NETSPLIT VICTIM] | |
| 19:22.46 | *** join/#brlcad SWPadnos (n=Me@emc/developer/SWPadnos) [NETSPLIT VICTIM] | |
| 19:22.46 | *** mode/#brlcad [+o ChanServ] by irc.freenode.net | |
| 19:25.12 | hippieindamakin8 | starseeker, if we find out the intersections what next ? if i have found out the combination of cubes in the space which overlap to some extent or completely (provided they all have the same error ), there shall be situations where A intersects B, B intersects C and A doesnt intersect C. |
| 19:25.12 | starseeker | exactly |
| 19:25.39 | hippieindamakin8 | estimates that the above step takes approximately O(nlogn) |
| 19:25.39 | starseeker | probably the thing to do is to find the two closest points within the identified subset, treat those as being the same, and follow the cascade of consequences as far as other decisions |
| 19:25.39 | starseeker | if A is closer to B than it is to C, set A=B and A!=C |
| 19:25.39 | starseeker | but "closer" might not be enough |
| 19:25.39 | madant | starseeker: as pacman87 asked in case two points are found to be the same do u take their median to be the new point ? |
| 19:25.39 | hippieindamakin8 | starseeker, the better thing would be divide the bounding box into a mesh of cubes and then work on it ? |
| 19:25.39 | starseeker | the metric I am considering is relative volume of overlap between A and B's delta box vs. the same volume calculated for A and C |
| 19:25.39 | starseeker | madant: I don't know |
| 19:25.39 | starseeker | hippieindamakin8: what do you mean? |
| 19:25.40 | hippieindamakin8 | as in in each cube all the points are approximated to a single point which is either the centroid of those points or center of the cube |
| 19:25.40 | hippieindamakin8 | and the size of this smaller cube shall be the error bound |
| 19:26.04 | starseeker | hippieindamakin8: you're considering the subcase of many points all within the same collection of bounding boxes? |
| 19:26.41 | hippieindamakin8 | i meant one bounding box (bounding all the space into consideration) |
| 19:26.58 | starseeker | needs to make an organized illustration of the various cases for n=3 points to think about |
| 19:27.35 | starseeker | hippieindamakin8: That might result in more point consolidation that we want |
| 19:28.43 | hippieindamakin8 | starseeker, yeah true but the size of the smaller boxes can be increased to decrease the extent of consolidation obtained |
| 19:29.59 | starseeker | hmm. I'll have to think some more |
| 19:30.13 | hippieindamakin8 | tries to look into the 'approximation algos for geometry ' literature if he can find smthing which can help |
| 20:16.13 | *** join/#brlcad _sushi_ (n=_sushi_@77-58-245-202.dclient.hispeed.ch) | |
| 20:34.49 | CIA-40 | BRL-CAD: 03r_weiss * r34121 10/brlcad/trunk/src/rt/ (main.c opt.c viewarea.c): updates to rtarea adding center computations |
| 20:46.08 | *** join/#brlcad BigAToo (n=BigAToo@pool-96-230-124-64.sbndin.btas.verizon.net) | |
| 21:46.26 | brlcad | alright weiss, progress! |
| 22:00.34 | madant | brlcad: weiss new developer ? |
| 22:01.46 | Ralith | hey mafm! |
| 22:01.57 | Ralith | I see you've seen my proposal. |
| 22:01.59 | mafm | hi Ralith |
| 22:01.59 | Ralith | any comments? |
| 22:02.02 | madant | ah the gui heroes meet :D |
| 22:02.09 | mafm | gui heroes |
| 22:02.13 | Ralith | hehe |
| 22:02.16 | Ralith | I haven't earned that yet. |
| 22:02.16 | mafm | ... :P |
| 22:02.19 | mafm | 1 sec |
| 22:15.00 | *** join/#brlcad kanzure (n=bryan@66.112.232.153) | |
| 22:20.51 | brlcad | mafm: though you had good reason (wrt, "I wasn't involved in development after gsoc except for a bit towards the end of last year"), personal reasons |
| 22:21.03 | brlcad | hopefully you'll be knee deep in code here again soon ;) |
| 22:21.26 | brlcad | someone needs to help weiss with his patch |
| 22:21.42 | brlcad | that needs to be reverted/fixed |
| 22:21.46 | starseeker | somebody rang? ;-) |
| 22:21.51 | starseeker | what's busted? |
| 22:21.58 | starseeker | I haven't tried building yet |
| 22:28.15 | starseeker | hmm - builds OK, doesn't seem to crash... |
| 22:28.40 | brlcad | it's not a compile problem |
| 22:28.47 | brlcad | the patch has some problems |
| 22:28.51 | brlcad | I sent a note to the list |
| 22:29.21 | brlcad | iirc, I *think* he's on the list -- if he's not, he should be |
| 22:30.02 | brlcad | madant: yeah, very green |
| 22:31.00 | starseeker | ah, yes I see it |
| 22:33.42 | brlcad | starseeker: and it's kinda important -- I was going to tag the source release later tonight |
| 22:33.49 | brlcad | most of the changes shouldn't take long at all |
| 22:34.18 | starseeker | revert til after the tag, or fix? that's user visible if left in |
| 22:34.19 | brlcad | and that'd be a good one to make the release so it can be communicated with s2 golks |
| 22:34.22 | starseeker | k |
| 22:34.35 | brlcad | yeah, I wouldn't leave it in for release |
| 22:34.41 | brlcad | I'd revert if he can't fix it today |
| 22:34.57 | brlcad | in fact I'll just revert it right now |
| 22:35.03 | starseeker | he's gone, so it'd be up to him seeing it on the list tonight |
| 22:35.07 | starseeker | k |
| 22:35.36 | Ralith | brlcad: interesting concepts on that firefox image |
| 22:35.51 | Ralith | doesn't look quite as powerful as what you had in mind, but I certainly see the similarities |
| 22:36.06 | brlcad | it's not as powerful just because all they have to work with is a web browser |
| 22:36.14 | brlcad | and they don't have inherint commands already |
| 22:36.23 | brlcad | we already have hundreds to work with |
| 22:36.31 | mafm | brlcad: so what's that executive decision? |
| 22:37.01 | mafm | Ralith: probably you worked with the code more than me lately |
| 22:37.03 | brlcad | mafm: that we need a better gender ratio in this channel |
| 22:37.12 | brlcad | it's a g'damn sausage fest in here |
| 22:37.16 | brlcad | j/k ;) |
| 22:37.21 | mafm | hot chicks? that's fine for me :P |
| 22:37.43 | mafm | Ralith: anyway, if I can be of any help let me know, and I'll try to keep an eye |
| 22:37.53 | Ralith | mafm: I've been toying with the build system more than the actual code, really |
| 22:37.59 | Ralith | oh, that reminds me |
| 22:38.05 | Ralith | mafm: why did you require Ogre 1.7.0? |
| 22:38.21 | Ralith | it seems to work fine on 1.6.1 |
| 22:38.31 | mafm | 1.7.0 as opposed to stable releases? |
| 22:38.38 | mafm | IIRC there was something like 1.4 stable by that time |
| 22:38.55 | brlcad | mafm: there's a couple things I still intend to check on, but basically that we should just run with making Qt work |
| 22:38.56 | mafm | and they didn't have some functionalities that RBGui needed by then |
| 22:39.07 | mafm | so I used trunk, or something like that |
| 22:39.15 | brlcad | only on the thin client front, though, not the entire application back-end or other tools for now |
| 22:39.42 | mafm | I see |
| 22:40.01 | brlcad | still need to verify a couple things with regards to a few widgets |
| 22:40.04 | mafm | I never used Qt, but other than using a precompiler, I heard that it's not very difficult |
| 22:40.09 | brlcad | and how it integrates with ogre |
| 22:40.25 | brlcad | the precompiler is the one big hassle that leaves a bad taste |
| 22:40.30 | mafm | last time that I checked RBGui (january I think) it was in the same state |
| 22:40.37 | brlcad | should see if there are any facilities to avoid using it |
| 22:40.47 | mafm | so, well, it's nice but... :) |
| 22:40.48 | brlcad | yeah, I'm not surprised |
| 22:40.56 | brlcad | rbgui being dead wasn't a problem :) |
| 22:41.35 | brlcad | it's more just where we'd end up after it's fully customized, simple widgets with limited behavior |
| 22:41.56 | Ralith | brlcad: I get the impression that the precompiler contributes a lot to the ease of development with Qt |
| 22:42.06 | Ralith | what's so unappetising about it? |
| 22:42.23 | brlcad | I'd see us wasting a lot of cycles making a hundred dumb things work (like we do with Tk now) like getting select/copy/paste working on a text widget |
| 22:43.06 | brlcad | Ralith: I suppose I just haven't drank the coolaid yet -- I like my sources to compile with a preprocessor and a compiler |
| 22:43.09 | mafm | I had thought about getting rid of Mocha for one, I think that RBGui uses only a few simple classes from it... but if it's not going to be used, one thing less to worry abount :) |
| 22:43.31 | Ralith | brlcad: I certainly understand the sentiment, but I wouldn't write it off immediately. |
| 22:44.03 | Ralith | mafm: yeah, before I realized how apt Qt might be, I saw that as one of the first things I'd try to work on |
| 22:44.11 | Ralith | much of what mocha provides is even already in BRL-CAD somewhere. |
| 22:44.23 | brlcad | Ralith: I'm not writing it off, I just don't like things that lock you in to more than an API |
| 22:44.54 | brlcad | similar to the openthreads declarations, pragmas, and other tool-specific language overlays that some projects run with |
| 22:46.10 | starseeker | thinks he remembers the QT precompiler being fairly essential for hiding a LOT of really nasty details, but it's been a while |
| 22:46.50 | starseeker | last time I was looking at it was when someone was trying to integrate lisp CFFI with QT, iirc |
| 22:47.09 | Ralith | given the degree to which Qt is an entire application framework, rather than just a widget toolkit, I'd imagine general dependency on it might be hard to avoid if we want to take advantage of all that it offers. |
| 22:47.43 | CIA-40 | BRL-CAD: 03brlcad * r34122 10/brlcad/trunk/src/rt/ (main.c opt.c viewarea.c): revert r34121 due to a variety of issues given I'm about to tag a source release. most pertinent to the release is the addition of an rt option (-y) that doesn't apply to any of the other raytrace applications. |
| 22:49.07 | mafm | Ralith: what's your primary OS? FreeBSD? |
| 22:49.15 | Ralith | mafm: linux, at the moment. |
| 22:49.22 | Ralith | freebsd laptop died and I had to get a new system up quick |
| 22:50.31 | mafm | I see |
| 22:51.03 | mafm | RBGui was easy to build for me (with the patches), but that was a while ago :D |
| 22:51.09 | Ralith | patches? |
| 22:51.36 | mafm | yep, src/other includes original RBGui with further patches |
| 22:51.43 | mafm | and the same for Mocha |
| 22:51.55 | mafm | I think that OIS and OGRE did not need them |
| 22:52.00 | *** join/#brlcad hippieindamakin8 (n=hippiein@202.3.77.38) | |
| 22:52.25 | mafm | they were basic patches implementing little stuff, cleaning headers, etc |
| 22:52.52 | Ralith | oh, I'd forgotten about that. |
| 22:52.59 | Ralith | that explains that. |
| 22:54.13 | mafm | +float PosixPlatformManager::getDoubleClickTime( ) |
| 22:54.15 | mafm | +{ |
| 22:54.16 | mafm | + // ... |
| 22:54.18 | mafm | <PROTECTED> |
| 22:54.19 | mafm | :) |
| 22:54.38 | mafm | missing things like that, it didn't even compile -- I guess that nobody had tested it in linux before releasing |
| 22:55.18 | mafm | they were small things like that IIRC, anyway, it must be in SVN logs shortly after importing |
| 22:56.52 | Ralith | yeah, I had to add that in too |
| 22:59.13 | ``Erik | nice |
| 23:01.14 | mafm | I think that I filled some of those with real implementations, but well :) |
| 23:01.30 | mafm | as I said, it must be in SVN |
| 23:01.50 | mafm | does Qt render inside the opengl context, or around it? |
| 23:01.56 | Ralith | it can do either |
| 23:02.04 | Ralith | I plan to do the former |
| 23:02.15 | mafm | nice |
| 23:02.36 | mafm | have you checked some example application or so? |
| 23:02.53 | Ralith | it's all linkified in my official gsoc application |
| 23:02.58 | Ralith | lemme dig that up |
| 23:03.25 | Ralith | http://doc.trolltech.com/4.5/demos-boxes.html |
| 23:03.28 | mafm | oh, I don't have access to it, only to the wiki |
| 23:05.51 | mafm | goody |
| 23:05.58 | madant | Ralith: wow @ the boxes O.o |
| 23:06.03 | Ralith | ? |
| 23:06.12 | mafm | seems like a good replacement of RBGui for this task |
| 23:06.14 | mafm | :) |
| 23:06.17 | Ralith | that's my thought. |
| 23:08.25 | CIA-40 | BRL-CAD: 03Sean 07http://brlcad.org * r1316 10/wiki/Google_Summer_of_Code: update apply url to socghop site |
| 23:08.29 | mafm | btw Ralith, what/where are you studying? |
| 23:09.25 | mafm | for some reason I thought that you were already a teacher or something like that, when you told me last year that you were interested in the project |
| 23:09.49 | Ralith | people always seem to get that kind of impression from me |
| 23:09.53 | Ralith | probably 'cuz I'm such a grammar nazi |
| 23:10.35 | Ralith | actually, I only just got accepted to a college; last year I didn't even qualify for SoC, or I would have applied then, too. |
| 23:11.20 | brlcad | mafm: if I'm undertanding everything I've read to date, especially since 4.0, they've moved towards rendering everything through a generalized backend that gives them an opengl context to work with |
| 23:11.34 | brlcad | allowing the easily zoomable interfaces, simple vector scaling, accelerated rendering |
| 23:11.58 | brlcad | but even if they didn't, yeah -- our entire window should be opengl, that's still the goal |
| 23:12.08 | brlcad | one big context with widgets in it |
| 23:12.09 | madant | this qt looks nothing like my "memory" of Qt :P |
| 23:12.20 | Ralith | hehe |
| 23:12.39 | brlcad | madant: kde gives it a bad name ;) |
| 23:13.28 | madant | is a core gnome-er .. well now xmonader :) |
| 23:13.54 | mafm | vector scaling... one of the things that you wanted last year already :) |
| 23:13.57 | madant | i always feel kde looks childish :P |
| 23:14.32 | Ralith | there we go |
| 23:14.34 | Ralith | application linkified |
| 23:14.35 | CIA-40 | BRL-CAD: 03Ralith 07http://brlcad.org * r1317 10/wiki/User:Ralith: Added links from GUI SoC application |
| 23:19.12 | madant | Ralith: how's reprap ;) ? |
| 23:19.47 | Ralith | madant: in general or in particular? |
| 23:19.55 | madant | in particular ;) |
| 23:20.29 | Ralith | my machine's nearly finished |
| 23:20.41 | madant | awesome :) |
| 23:20.42 | Ralith | just need to wire things up and build the extruder barrel |
| 23:21.03 | madant | hah.. then onto a few kids ;) |
| 23:21.41 | mafm | I read about reprap somewhere in the last few days, can't remember where |
| 23:22.45 | mafm | Ralith: maybe the fact that you didn't apply for soc gave me the impression of being a post-student :D |
| 23:22.59 | Ralith | could be |
| 23:29.21 | brlcad | several of the guys I work with are technically students and a decade or two older than I .. you can be a student at any age / stage of life ;) |
| 23:32.33 | Ralith | not much is as rewarding as learning about cool stuff, after all. |
| 23:38.41 | madant | will be back after a 5km run : 5 am here :D |
| 23:44.42 | brlcad | madant: awesome |
| 23:44.53 | brlcad | see you in 20 minutes ;) |
| 23:52.07 | mafm | I used to go to exams with people of 70 around me |
| 23:53.32 | brlcad | my professors for some courses were students with me in other courses at the university |
| 23:54.21 | Ralith | neat! |
| 23:55.21 | brlcad | I knew I was over my head when my calc III professor was in the same course with me on (graduate level) game theory .. that was three weeks of utter hell |
| 23:55.53 | brlcad | I would have had to drop half my workload just to keep up |
| 23:57.41 | *** join/#brlcad andrecastelo (n=andrecas@189.71.13.123) | |
| 23:59.29 | brlcad | howdy andrecastelo |
