| 00:06.22 | *** join/#brlcad starseeker (n=CY@ip72-218-16-62.hr.hr.cox.net) |
| 00:06.53 | louipc | wuzzat? |
| 00:07.05 | starseeker | Freebie game using the Tribes2 engine |
| 00:07.11 | louipc | ah |
| 00:07.36 | louipc | sometimes it's necessary |
| 00:07.41 | louipc | but not really for games :P |
| 00:07.46 | starseeker | I should just play bzflag ;-) |
| 00:07.53 | louipc | haha |
| 00:08.02 | louipc | it's fun for awhile |
| 00:08.14 | louipc | maybe just because I'm not a big gamer |
| 00:08.23 | starseeker | Yeah, that's pretty much my take on it |
| 00:08.45 | starseeker | I didn't do enough video games at a young age to acquire the shooting precision |
| 00:08.47 | louipc | I like to play the games to see if I can fix bugs hehe |
| 00:08.59 | louipc | open source stuff |
| 00:10.08 | starseeker | Tribes I was IMHO the high water mark of that type of game - Tribes2 just doesn't have the same feel |
| 00:11.07 | starseeker | Reaper3d actually had a very nice looking engine, but it's been dead for years now |
| 00:11.42 | louipc | I find the problem with a lot of games is that they hog cpu |
| 00:11.47 | louipc | even for dumb 2d games |
| 00:12.12 | starseeker | Indeed |
| 00:12.34 | starseeker | the old school guys from 30 years ago would be aghast at the use of hardware today... |
| 00:13.13 | louipc | yeah |
| 00:13.57 | starseeker | I think if hardware ever does reach some kind of stable performance we will see some major upheaval in programming styles... |
| 00:15.55 | louipc | well we need better innovation in hardware architectures and such |
| 00:16.02 | louipc | no more x86 :P |
| 00:55.58 | brlcad | trainman419: erhm, yeah .. not really any ray-tracer uses opengl accelleration; opengl is general raster image processor, ray-tracing is not a raster rendering |
| 00:56.36 | brlcad | the final rendered image can be displayed into an opengl context, but you're still just displaying an image and don't necessarily need opengl for that |
| 00:57.06 | brlcad | the modeler (mged) could leverage opengl a lot, but mged doesn't require opengl for what it does either |
| 00:57.15 | ``Erik | heh |
| 00:57.19 | trainman419 | ah, it all makes sense now |
| 00:57.40 | ``Erik | erm |
| 00:57.51 | ``Erik | raytracing can use ogl to rasterize... it'd just be, y'know, ass slow |
| 00:58.52 | ``Erik | BAH, wench got drunk at a wedding and now wants to drive :( stupid whore |
| 00:59.31 | ``Erik | :( |
| 01:00.15 | trainman419 | yeah, I'm just used to things like solidworks that show rasteriezed images rather than raytraced images |
| 01:00.44 | ``Erik | BRL-CAD doesn't have a "quick view" functionality, sorry |
| 01:00.46 | Maloeran | Mmhm. You didn't go with her, Erik, despites the free alcohol? :) |
| 01:00.48 | brlcad | they have ray-trace functions too, just not their main graphics display |
| 01:00.58 | brlcad | at least solidworks does |
| 01:00.59 | ``Erik | dislacement issue, mal |
| 01:01.28 | ``Erik | he father is ill and in kc, so that's where she is :/ |
| 01:01.38 | Maloeran | Oh.. |
| 01:01.59 | ``Erik | kansas city, missouri |
| 01:02.08 | brlcad | trainman419: our next modeling interface will intently utilize opengl through and through for the modeling interface |
| 01:02.56 | brlcad | tarzeau: where does the build fail for you with CFLAGS="-O2" LDFLAGS="-Wl,-z,defs" ... that seems to be working for me (on RHEL4) |
| 01:03.00 | trainman419 | awesome! will that be in mged or archer? (or something new?) |
| 01:03.14 | ``Erik | the pasted message was a compiler warning, not an error |
| 01:04.36 | brlcad | trainman419: neither, new interface |
| 01:05.43 | brlcad | he's been working on that for years |
| 01:05.46 | brlcad | little by little |
| 01:05.57 | ``Erik | yeah, but now he has bucks behind him |
| 01:06.00 | brlcad | too draining to only do that, even for him |
| 01:06.07 | brlcad | i mean even before |
| 01:06.14 | brlcad | he started that before you arrived |
| 01:06.18 | brlcad | heck, before I started |
| 01:06.19 | ``Erik | *nod* |
| 01:06.26 | ``Erik | it's the right thing to do |
| 01:06.28 | ``Erik | just not sexy |
| 01:06.34 | brlcad | it is, just tons of grunt |
| 01:07.01 | ``Erik | if he can commit 10% time to it, it's a major win |
| 01:07.09 | brlcad | Bob's not the greatest at refactoring sometimes, a lot of very short term fixes without the refactoring |
| 01:07.21 | brlcad | often detrimental to what he has to deal with down the road |
| 01:07.27 | brlcad | or what others have to do to maintain it.. |
| 01:07.37 | brlcad | the code gets there though |
| 01:07.37 | ``Erik | heh |
| 01:07.56 | ``Erik | I'd hope that if he puts effort there, it'd spin off work |
| 01:08.48 | ``Erik | for laid code, ... windows or tcl :/ |
| 01:09.00 | brlcad | yeah, just having a few more working on development will make a big change |
| 01:09.13 | brlcad | maybe even start to look like what things were like before M3 |
| 01:09.23 | ``Erik | heh |
| 01:09.24 | starseeker | M3? |
| 01:09.27 | ``Erik | :( |
| 01:09.28 | brlcad | still need two more though at minimally |
| 01:09.39 | brlcad | *minimally* |
| 01:09.42 | ``Erik | the project that killed the momentum of BRL-CAD, star |
| 01:09.48 | starseeker | ouch |
| 01:10.45 | ``Erik | the core idea was actually good, but pointy hairs got a hold of it... then gutted "the cad team", then proceeded to chase most of 'em away with micromanagement |
| 01:10.57 | brlcad | rather.. stole motivation, funding, attention, developers, ... |
| 01:11.32 | brlcad | alas, those are more and more turning into just history with the surge of new interest |
| 01:11.57 | brlcad | just took a few years for the turnaround |
| 01:11.58 | starseeker | alas? Are good ideas being lost? |
| 01:12.03 | brlcad | everything happens in waves |
| 01:12.26 | ``Erik | good ideas were had and given... interpretation... ... heh... |
| 01:12.39 | brlcad | "alas", hopefully the brunt of the bad days are behind us now |
| 01:13.12 | brlcad | for our group at least, it's changing |
| 01:13.22 | ``Erik | yeah.... I'm bitter |
| 01:13.23 | ``Erik | *shrug* |
| 01:13.44 | ``Erik | the best aspect I see is that the GPL/LGPL/BSD push worked |
| 01:14.00 | ``Erik | <-- likes to pretend he had a serious roll in that *shrgu* |
| 01:14.10 | starseeker | That's virtually miraculous - I still remember the shock when I saw the announcement |
| 01:14.12 | brlcad | yeah, the diverted attention helped that in a (very) minor regard |
| 01:14.39 | ``Erik | (the part I meant was the automake vs cake side) |
| 01:15.12 | brlcad | yeah, thank jeebus we don't still have to deal with cake hell |
| 01:15.18 | ``Erik | ehehhehe |
| 01:15.28 | ``Erik | I was dumbstruck on seeing tht |
| 01:15.32 | brlcad | for all of the autohell, it wasn't as bad as cakehell |
| 01:15.41 | brlcad | s/wasn't/isn't/ |
| 01:15.53 | ``Erik | and that lee said "it's broken? ok, go fix it"... |
| 01:15.58 | ``Erik | that was a good state |
| 01:16.06 | starseeker | Some of the national labs have open source stuff (Paraview is an example) but I don't recall anything similar coming from the Army side except GRASS |
| 01:16.48 | ``Erik | hum, cool |
| 01:17.03 | brlcad | starseeker: technically grass was just "given away" when it was done, public domained because it was a done/dead project |
| 01:17.06 | ``Erik | I stepped in and was working on the fbsd port and autotools before I even got an account |
| 01:17.18 | starseeker | brlcad: Oh, OK. |
| 01:17.35 | brlcad | only because the community stepped up and a few took ownership (and claimed a copyright after significant changes) did it become open source |
| 01:17.40 | ``Erik | perhaps just "a place to put" the new guy O:-) |
| 01:17.52 | starseeker | I pointed it out to my dad a few times - I don't know if anyone in his group ever took a poke at it or not |
| 01:18.43 | starseeker | It looked/looks impressive, for it's problem domain (not that I'm an expert in that field) |
| 01:19.14 | ``Erik | starseeker: do you use ttcp? ping? tcp/ip? mr muuss was a heavy hitter in those in the name of the army... |
| 01:19.42 | brlcad | as far as I discovered during the legal research, BRL-CAD was the first *active* project in U.S. Army that I could find that went straight to open source licensing |
| 01:19.57 | starseeker | ``Erik: Yes, I'd read about that. I regret arriving too late on the scene to ever get a chance to meet Mike Muss |
| 01:20.06 | ``Erik | me too :( |
| 01:20.27 | starseeker | brlcad: I can believe it. It's a very impressive accomplishment |
| 01:20.28 | brlcad | maybe even first in DoD, but that's not likely .. just nothing really popular/notorious or at least that I was able to find after a lot of searching and contact prodding |
| 01:20.49 | ``Erik | first as in conception date or ? |
| 01:21.14 | brlcad | first to release code legally as OSS |
| 01:21.38 | brlcad | non public-domain |
| 01:21.55 | ``Erik | heh,t hat's artificial, dude |
| 01:21.56 | brlcad | there are plenty of public domain examples that have been picked up |
| 01:22.01 | ``Erik | pd is oss friendly |
| 01:22.55 | ``Erik | also; |
| 01:22.58 | brlcad | not so much that it "died" but that the group that created it no longer was directed to maintain it |
| 01:23.11 | ``Erik | brlcad, *smack* it's MUVES-3, don't sully the name "m3", my car deserves better |
| 01:23.30 | starseeker | brlcad: Ah, that will do it |
| 01:23.59 | ``Erik | many projects have survived the originator... that's the beauty of oss, it CAN survive |
| 01:24.00 | brlcad | yes, PD is OSS, but it's not like it takes any active effort and is almost always done on a "done" code |
| 01:24.35 | brlcad | licenseless vs picking an OSS license |
| 01:24.48 | ``Erik | should I hack in gettext support on ping and do an "army" release? :D |
| 01:25.02 | brlcad | dude, are you totally missing the point? |
| 01:25.24 | brlcad | it's not the same |
| 01:25.41 | starseeker | brlcad: I guess in a sense Maxima falls into that category too - it was originally under DOE control (which is why it still has that rather odd note about US arms export restrictions in the copying file...) |
| 01:25.47 | ``Erik | apparently O.o *shrug* I don't equate gpl and oss :( |
| 01:26.22 | ``Erik | doe has dumped a fair bit of interesting stuff |
| 01:26.23 | brlcad | mabye if I qualified it as an OSS *license* as opposed to just OSS |
| 01:26.58 | ``Erik | unfortunately, 'cougaar' was completely ignored by this certain army group... *cough* |
| 01:27.08 | brlcad | I don't know of another example outside the nat. labs, nasa, and a few other gov't groups that have gone straight to licensed OSS |
| 01:27.33 | ``Erik | isn't pd on the osi list? |
| 01:27.41 | brlcad | PD isn't a license |
| 01:27.54 | brlcad | being PD does make you OSS, but it's not a license |
| 01:27.55 | ``Erik | heh, it's past license :D |
| 01:27.57 | starseeker | PD is simply a renouncing of copyright rights |
| 01:28.42 | ``Erik | starseeker: I'd actaully be a bit scared of that code... heavy assembly (even if well written), classic C style, ... |
| 01:29.05 | ``Erik | lookin' at the 4.3 bsd code was humbling |
| 01:29.26 | starseeker | ``Erik: I suppose that's true - the idea of the debugging they do and the methodology they use is actually what interests me the most |
| 01:29.36 | *** part/#brlcad rpaddock (n=bob_padd@c-24-131-109-50.hsd1.pa.comcast.net) |
| 01:29.50 | ``Erik | I know that faa restrictions of software development is... insane |
| 01:30.15 | starseeker | Assuming this article knew what it was talking about, that is: http://www.fastcompany.com/magazine/06/writestuff.html |
| 01:32.48 | ``Erik | and I'm sure they dont' trust the software 100% :D |
| 01:34.33 | starseeker | ``Erik: Overstated? |
| 01:35.06 | ``Erik | um, cmmi is an interesting notion and probably has value, btu the implementation as it stands is ... |
| 01:35.07 | ``Erik | heh |
| 01:35.39 | ``Erik | amusingly enoughl, gillich said something brilliant about cmmi... "it's a yard stick" |
| 01:35.43 | ``Erik | or something to that nature |
| 01:36.08 | ``Erik | in itself, it offers nothing... it's merely a possible mechanism to measure with |
| 01:36.52 | starseeker | Ah |
| 01:40.24 | starseeker | brlcad: There's a question on the Gentoo bug about AMD64 - for some reason it didn't find itcl/itk if I'm reading that message right, but I don't know why itcl/itk would fail on AMD64? |
| 01:41.02 | ``Erik | tcl/tk/itcl/itk/iwidigts all worsk fine on amd64 fbsd |
| 01:41.14 | starseeker | That's what I thought. Hmm. |
| 01:41.59 | starseeker | He says it works in chrooted 32 bit environment. Odd |
| 01:48.20 | starseeker | Monday morning could be a little rough... |
| 01:49.22 | ``Erik | <- doesn't work on monday :D |
| 01:49.46 | starseeker | Ah - that helps |
| 01:50.28 | brlcad | starseeker: see if they can provide a configure and build log as well as the config.log somewhere |
| 01:50.40 | ``Erik | *grump* stupid wench is drunk as fuck :( |
| 01:51.10 | ``Erik | ~bzpastebin |
| 01:51.19 | ibot | bzpastebin is probably http://pastebin.bzflag.bz a place to put large chunks of text to not flood a channel |
| 01:51.19 | ``Erik | heh |
| 01:51.38 | ``Erik | probably. |
| 01:52.02 | ``Erik | "without their heads, they're powerless!" |
| 01:52.35 | starseeker | brlcad: Request posted. |
| 01:53.27 | brlcad | thanks |
| 01:53.31 | starseeker | np |
| 01:53.36 | ``Erik | I'd totally get on anna faris. |
| 01:53.50 | starseeker | brlcad: Oh, did you see my last question about b-reps? |
| 01:54.34 | brlcad | mm |
| 01:54.34 | brlcad | nope |
| 01:54.53 | starseeker | probably dumb, but if you're bored... |
| 01:55.16 | starseeker | Naive question - the existence of brep at a primitive level - how does that make the core logic of model -> mesh simpler? Would it mean that models would have to be rebuilt with the new primitive or would the conversion of old primitives to b-rep be less expensive? |
| 01:55.38 | brlcad | brep as a primitive provides the base-level representation |
| 01:55.52 | brlcad | you still need a means to load up existing implicits as a brep |
| 01:56.09 | ``Erik | notionally, if we had a fast ->brep conversion method, we know brep->triangles is fast, so we could bypass all the nmg bs |
| 01:56.31 | brlcad | fortunately, that part actually isn't that bad -- you can create an ellipsoid brep if you have the implicit form pretty straightforward |
| 01:56.44 | brlcad | yep, it bypasses the whole mess |
| 01:56.51 | starseeker | Swwweeeettt |
| 01:57.09 | brlcad | the hardest part (and it certainly is non-trivial, but easier than the nmg approach) is CSG evaluation of BREPS |
| 01:57.16 | ``Erik | nmg tries to do some tricky stuff from a poor starting point :( also; it's fucking ugly |
| 01:58.05 | brlcad | csg evaluation of breps involves creating new brep surfaces that trim away various parts of arbitrary surfaces and weaves in other surfaces |
| 01:58.33 | ``Erik | is it still a weave in brep space? |
| 01:58.34 | brlcad | numerically should be a heck of a lot more stable simply because you retain the actual structure instead of something that's been deconstructed already |
| 01:59.10 | starseeker | Nice. IIRC, you said the csg evaluation is the main remaining component? |
| 01:59.25 | ``Erik | the weave notion is brilliant for striaght raytracing, but is bass ackwards for anything else :( |
| 01:59.28 | brlcad | yeah, you still have to attach surfaces from object A to those of object B (assuming they overlap) trimming off the parts on the interior or in negative space |
| 01:59.43 | brlcad | it's not that kind of weaving |
| 01:59.56 | ``Erik | <- thinking boolweave |
| 01:59.56 | brlcad | just my terminology for attaching the two |
| 02:00.03 | brlcad | yeah, nothing at all like boolweave |
| 02:00.33 | brlcad | even current nmg approach has nothing to do with boolweave |
| 02:00.40 | ``Erik | and for straight up "through" raytracing, I doubt I'd have thought of anything that clever |
| 02:01.05 | ``Erik | I'd probably have been stuck on resolving the geometry before shooting it :D |
| 02:01.09 | starseeker | phew - maybe not a dumb question :-) I was afraid I was missing something obvious, but it sounds like a very elegant mathematical workaround |
| 02:01.10 | brlcad | it's approach actually isn't much different either other than the surface patches are an order or two smaller than the average BREP surfaces |
| 02:01.43 | ``Erik | but these days, resolving the geometry is becoming more important |
| 02:02.27 | brlcad | NMGs actually are a BREP.. just most people implicitly mean spline surface boundary representation when they say brep, not n-manifold polygonal boundary representation |
| 02:02.28 | ``Erik | even wireframe, just having the 'cutters' rendered intuitively... |
| 02:02.52 | starseeker | brlcad: Is there a good paper on the mathematics of geometry->brep and brep->mesh conversion mathematics? |
| 02:02.54 | ``Erik | nmg is almost always associated with tesselating |
| 02:02.59 | brlcad | having this done would make it possible to show the interior plot lines too |
| 02:03.54 | brlcad | yeah, technically our nmg lib can handle nurbs too -- that's how stay's nurbs code was written -- but the main and original use is polygonal |
| 02:04.24 | ``Erik | but jay-lo's code doesn't hit nmg? |
| 02:04.27 | brlcad | starseeker: several books, but there's not really a single good paper on it |
| 02:04.36 | brlcad | nope, he avoided nmg entirely |
| 02:04.42 | brlcad | and intentionally |
| 02:04.47 | brlcad | uses opennurbs instead of nmg |
| 02:04.55 | ``Erik | one of these days, he's gonna kick my ass for calling him jay lo |
| 02:04.55 | brlcad | for the representation |
| 02:05.07 | starseeker | brlcad: Books work, if they aren't $$$$$ on Amazon ;-) |
| 02:05.29 | ``Erik | um |
| 02:05.37 | brlcad | starseeker: wait till you're on-site -- there's plenty of books you can get |
| 02:05.44 | brlcad | if you get on-site of course :) |
| 02:05.48 | starseeker | brlcad: cool |
| 02:06.04 | ``Erik | there was that paper that had omfg fast raytracing and ->bot conversion on NURBS, I think? at utah? |
| 02:06.16 | brlcad | yeah, that's a great paper, but not a great into paper |
| 02:06.21 | brlcad | s/into/intro/ |
| 02:06.29 | brlcad | RT06 paper |
| 02:06.30 | ``Erik | heh, no, it assumes base competency |
| 02:07.00 | ``Erik | starseeker: if you're up to the challenge, someone will get you the info. |
| 02:07.44 | ``Erik | on sight or not |
| 02:07.45 | Maloeran | thinks* too |
| 02:07.55 | ``Erik | heh |
| 02:08.06 | ``Erik | write up a paper, submit to ieee rt and siggraph :D |
| 02:08.13 | Maloeran | Sounds boring :) |
| 02:08.16 | brlcad | starseeker: just searching google will pull up a handful of papers |
| 02:08.24 | starseeker | brlcad: OK :-) |
| 02:09.02 | starseeker | ``Erik: This one? Direct and Fast Ray Tracing of NURBS Surfaces - Oliver Abert, Markus Geimer, Stefan Muller |
| 02:09.20 | Maloeran | These curved triangles would require more primitives than nurbs, but are so much cheaper for rays to intersect |
| 02:09.44 | ``Erik | I'm sure they're much less expressive than t-nurbs |
| 02:10.28 | Maloeran | They wouldn't describe nurbs perfectly, but it sure can be a good compromise for performance and accuracy, between nurbs and triangles |
| 02:10.37 | ``Erik | I think so, starseeker |
| 02:10.49 | ``Erik | I think I'm thinking of "Direct and Fast Ray Tracing of NURBS Surfaces |
| 02:10.50 | ``Erik | Oliver Abert, Markus Geimer, Stefan Muller" |
| 02:10.55 | ``Erik | http://www.sci.utah.edu/~wald/RT06/papers/papers_presentations.html |
| 02:11.20 | starseeker | Yep - hmm, no direct link |
| 02:11.39 | ``Erik | I have the book sitting ontop of my winderz spaceheater |
| 02:11.41 | Maloeran | I would scan and send if I had a scanner |
| 02:11.51 | brlcad | e.g. http://www.gup.uni-linz.ac.at/~gk/docs/162CT.pdf |
| 02:11.51 | starseeker | Google to the rescue, I think... |
| 02:11.57 | ``Erik | heh |
| 02:12.06 | brlcad | starseeker: also a pretty good in-depth thesis read: http://cadd.web.cern.ch/cadd/cad_geant_int/thesis/vaitos_main.html |
| 02:12.17 | ``Erik | and mal just sat there coding and gawking at college chicas instead of listening |
| 02:12.21 | ``Erik | :D |
| 02:12.50 | Maloeran | I think you were the one dreaming of college chicas. There was one girl in the entire audience at that conference? :) |
| 02:12.53 | starseeker | Cool - GEANT :-). I wacked my head trying to get that to compile for Gentoo once ;-) |
| 02:12.55 | brlcad | and yeah, that's the rt06 paper |
| 02:13.12 | ``Erik | was a good conf... didn't hurt with the awesomeness like siggraph, but was far better than the javaone pep rally |
| 02:13.35 | ``Erik | there were a few girls in the group, but the eye candy was when we were out of the conf room |
| 02:13.36 | ``Erik | :D |
| 02:13.39 | starseeker | http://www.uni-koblenz.de/~mgm/papers/rt06.pdf seems to be ``Erik's but it's not coughing it up right now - I'll try later |
| 02:13.50 | starseeker | brlcad: Thanks :-) |
| 02:13.54 | ``Erik | (though the girl from unc wasn't hard on the eyes, and smrrt, too) |
| 02:14.03 | brlcad | there actually are a fair bit of good references and implementations for tessellation of breps -- to varying degrees of water tightness |
| 02:14.28 | ``Erik | heh |
| 02:14.29 | ``Erik | um |
| 02:14.38 | ``Erik | during the poster presentation, she had the huge crowd |
| 02:14.44 | Maloeran | Ahah |
| 02:14.55 | ``Erik | something about using gpu's to trace complex structures or something |
| 02:15.09 | ``Erik | uh, she was hardware side iirc |
| 02:15.09 | Maloeran | I think I prefer a girl that can teach me stuff radically different than maths and code |
| 02:15.09 | brlcad | grid of gpus |
| 02:15.34 | ``Erik | brlcad still wanks thinkin' about her :D *duck* |
| 02:16.41 | brlcad | actually don't remember what she even looked like |
| 02:16.46 | brlcad | but do remember that she was there |
| 02:17.34 | ``Erik | heh, I was mroe boggled that something worth seeing came out of unc than that there was a woman presenting |
| 02:17.45 | brlcad | shirley's paper relates a bit to what we've currently got implemented: http://www.cs.utah.edu/~shirley/papers/raynurbs.pdf |
| 02:17.53 | ``Erik | peter shirly? |
| 02:17.59 | brlcad | yeah |
| 02:18.05 | ``Erik | isn't he leeberts advisor? |
| 02:18.27 | brlcad | could be |
| 02:18.48 | ``Erik | peter and ingo were the two people lee was sucking up to |
| 02:18.52 | ``Erik | iirc |
| 02:19.15 | ``Erik | and ingo is a fresh associate, I think |
| 02:19.59 | brlcad | I believe ingo left for intel |
| 02:20.07 | brlcad | started the week of siggraph |
| 02:20.13 | brlcad | that's why he wasn't there |
| 02:20.34 | starseeker | brlcad,``Erik: I've got to call it a night, but I appreciate the references - thanks much! |
| 02:20.41 | brlcad | starseeker: see ya |
| 02:20.56 | ``Erik | huh |
| 02:21.01 | ``Erik | later, starseeker :D |
| 02:21.28 | ``Erik | intel must be droppin' some real coin for a cool job, I mean |
| 02:21.41 | ``Erik | reshetov, now wald? |
| 02:23.53 | Maloeran | Wald writes a lot, but doesn't have much to say |
| 02:24.09 | ``Erik | he's smart and capable, but he tries to dominate discussions, from what I saw |
| 02:24.18 | Maloeran | Yes, he likes to be at the center of things |
| 02:24.46 | ``Erik | it's good that he does what he does, but he aint' the bees knees :D |
| 02:25.16 | louipc | what is good about bees' knees? |
| 02:25.31 | ``Erik | (and, uh, I"d rather have a loud arrogant wald publishing papers than a real clever chicken who just writes proprietary code. *cough*) |
| 02:26.02 | ``Erik | I can read a paper and learn :D |
| 02:26.05 | brlcad | he has gotten stuff done -- simply (and proactively) goes after getting attention for it too |
| 02:26.20 | brlcad | to each their own, getting stuff done is what matters |
| 02:26.30 | brlcad | and if you don't tell anyone about it, then you might as well have not done it |
| 02:26.55 | louipc | indeed you need to ruffle some feathers |
| 02:27.00 | ``Erik | hacking up some brilliant code and then locking the notions down by not putting source or papers out ... |
| 02:27.29 | ``Erik | patents might actually be worse |
| 02:27.40 | ``Erik | :D |
| 02:27.58 | Maloeran | Fine fine, I get the point :) |
| 02:28.10 | ``Erik | anyways, papers are good |
| 02:28.31 | ``Erik | the error smelled of broken system when first pasted :( |
| 02:28.33 | Maloeran | Sure papers are good, I'm just annoyed by the number of papers devoid of any content |
| 02:28.43 | ``Erik | ok... mal... fix that. |
| 02:28.50 | ``Erik | kick out a paper or 3 with content |
| 02:29.21 | ``Erik | most papers value is in tiny incremental steps :/ not too many are brilliant strides of length, these days |
| 02:29.30 | brlcad | ever |
| 02:29.40 | Maloeran | I thought I wrote some 3-4 pages that contained what mattered |
| 02:29.52 | ``Erik | I'm still in awe at the brilliance of CS papers from the 50's and 60's given the state at the time |
| 02:29.53 | brlcad | always the case, not all of them can be seminal papers :) |
| 02:30.27 | brlcad | i was just watching sutherland's sketchpad demo earlier today |
| 02:30.33 | brlcad | seriously impressive for the time |
| 02:30.33 | ``Erik | heh |
| 02:30.39 | ``Erik | make burga buy you real aussie beer. |
| 02:30.46 | ``Erik | or I will fucking kick his ass |
| 02:30.51 | Maloeran | Ahah, fine |
| 02:30.57 | ``Erik | um |
| 02:31.03 | ``Erik | he's in sydney? |
| 02:31.15 | Maloeran | Brisbane, I'll be staying just a day in Sydney before going up there |
| 02:31.16 | ``Erik | I wanna say, uh, perth has something called uh, 'black swan'? |
| 02:31.25 | Maloeran | And what is it? |
| 02:31.27 | brlcad | have a chiko roll while you're there |
| 02:31.28 | ``Erik | that my dad claimed was damn good |
| 02:31.32 | brlcad | i hear they're pretty yummy |
| 02:31.40 | Maloeran | A chicken roll? |
| 02:31.42 | ``Erik | oh, also; hit the beach and romp some of that hot aussie tail |
| 02:31.43 | ``Erik | *cough* |
| 02:32.14 | brlcad | http://en.wikipedia.org/wiki/Chiko_Roll |
| 02:32.26 | Maloeran | I know Erik, I was already instructed by Sophie to do so o.O |
| 02:32.40 | Maloeran | Curious snack |
| 02:33.09 | ``Erik | hum |
| 02:33.10 | brlcad | http://lib.store.yahoo.net/lib/upfromaustralia/chikochick.jpg |
| 02:33.14 | brlcad | http://i10.ebayimg.com/03/i/000/a5/a8/1441_1.JPG |
| 02:33.15 | ``Erik | damnit |
| 02:33.19 | ``Erik | now I want some lumpia :( |
| 02:33.41 | ``Erik | http://en.wikipedia.org/wiki/Lumpia |
| 02:33.58 | brlcad | "was designed to be able to be eaten with one hand whilst drinking a beer with the other" <-- chiko roll |
| 02:34.06 | Maloeran | Inspiring |
| 02:35.20 | ``Erik | the lumpia my mom makes is mostly carrot, cabbage and beef, about the size of a finger, deep fat fried |
| 02:36.09 | Maloeran | I'm mostly looking forward to scuba diving over there |
| 02:36.15 | Maloeran | And rock climbing of course |
| 02:36.15 | ``Erik | heh |
| 02:36.24 | Maloeran | I'll have to find people for that though, Burga doesn't climb |
| 02:36.25 | ``Erik | um, atch out for the sea snakes |
| 02:36.40 | ``Erik | my dad has some stories about running into sea snakes in aussie waters O.o |
| 02:36.52 | ``Erik | highly venemous fuckers |
| 02:37.09 | Maloeran | Are they dangerous even when wearing a wet suit? |
| 02:37.21 | ``Erik | yeah, theyc an be up to 6' long |
| 02:37.48 | ``Erik | barracudas out there, too |
| 02:37.50 | Maloeran | Hum, ouch. I'm reading about these for the first time |
| 02:38.05 | Maloeran | Jellyfishes are usually no big deal with a wet suit |
| 02:38.14 | ``Erik | nah, they're surface stingers |
| 02:38.16 | ``Erik | not biters |
| 02:38.18 | brlcad | Maloeran: give a try at surfing if you're near a coast |
| 02:38.21 | brlcad | some of the best surfing to be had |
| 02:38.35 | ``Erik | teeth go through wetsuits |
| 02:38.40 | CIA-27 | BRL-CAD: 03brlcad 07STABLE * 10brlcad/src/other/tkimg/Makefile.am: gah, merge from head .. need TCLSTUB to be included else there be unresolved symbols on Mac and now also seen on Linux too |
| 02:38.52 | Maloeran | Sure I'll try. I wasn't too good at that when I tried in Mexico |
| 02:39.07 | Maloeran | ( At least in comparison to Sophie who is a snowboard teacher ) |
| 02:39.15 | ``Erik | if ya shake up north a bit, hit okinawa :D I lived there for 3 yrs |
| 02:39.16 | brlcad | meh, still like ten times more likely you're going to get slammed into in your car on the way to the beach than you are to be bit at the beach |
| 02:39.31 | ``Erik | indeed, heh |
| 02:39.41 | ``Erik | don't fuck with an animal, and it won't fuck with you |
| 02:39.43 | ``Erik | :) |
| 02:40.00 | brlcad | unless you can buy her a drink first |
| 02:40.02 | Maloeran | Thanks for this piece of wisdom :) |
| 02:40.09 | ``Erik | hehe |
| 02:40.20 | ``Erik | mmmm, hot aussie tail |
| 02:42.30 | brlcad | that's the second "already fixed on head" issue i've found now .. drat drat |
| 02:43.15 | Maloeran | That's a curious wedding ceremony |
| 02:43.27 | ``Erik | 7.12.0 yo |
| 02:53.36 | brlcad | hm.. setting LDFLAGS does seem to cause it to not pass the X11 flags through, fails at blt |
| 02:55.12 | brlcad | ah, they're actually not listed for blt |
| 02:57.28 | ``Erik | http://www.collegehumor.com/picture:1782347 |
| 03:00.24 | ``Erik | "panburger partner" O.o |
| 03:00.28 | ``Erik | http://www.collegehumor.com/picture:1781996 |
| 03:04.32 | *** join/#brlcad tttttt (n=root@222.190.111.98) |
| 03:04.45 | tttttt | cvvcnbbn |
| 03:05.58 | *** part/#brlcad tttttt (n=root@222.190.111.98) |
| 03:07.04 | ``Erik | ssoommeeoonnee nneeeeddss ttoo ttuurrnn ooffff eecchhoo |
| 03:07.14 | ``Erik | O:-) |
| 03:08.54 | CIA-27 | BRL-CAD: 03brlcad * 10brlcad/src/other/blt/Makefile.am: blt actually makes direct calls to X11, so need to include the goods here else face the wrath of unresolved symbols on some plats (encountered on RHEL) |
| 03:11.05 | brlcad | woot root |
| 03:28.03 | ``Erik | <PROTECTED> |
| 03:28.34 | CIA-27 | BRL-CAD: 03brlcad * 10brlcad/configure.ac: blt *requires* X11 to function, turn it off if x11 isn't available |
| 03:28.40 | CIA-27 | BRL-CAD: 03erikgreenwald * 10brlcad/src/mged/clone.c: I have no clue what a "BUFXIZE" is, but I know what BUFSIZ is... |
| 03:31.06 | bpoole | haha. BUFXIZE? |
| 03:31.53 | ``Erik | that's what was in teh source |
| 03:31.59 | ``Erik | and I rpobably did it :( |
| 03:32.47 | bpoole | gj. |
| 03:33.58 | ``Erik | actually, it was brlcad that did that, heh |
| 03:34.05 | ``Erik | - strncat(suffix, suffix2, BUFSIZ); |
| 03:34.05 | ``Erik | + snprintf(suffix, BUFXIZE, "%s", suffix2); |
| 03:34.35 | ``Erik | bad chris, bad! |
| 03:34.47 | ``Erik | 1.46 -> 1.47 |
| 03:36.24 | brlcad | yep, just a simple typo |
| 03:36.47 | brlcad | not the first, undoubtedly the last |
| 03:37.07 | brlcad | if that's the only one after about 1000 of them, my error rate is prety freaking sweet |
| 03:37.19 | ``Erik | and I felt bad cuz I let a commit go without proper build testing |
| 03:37.38 | brlcad | i even built, for whatever reason it passed the build |
| 03:37.53 | ``Erik | I can't think of why, other than bad timestamping |
| 03:38.01 | brlcad | or some sys header that had it |
| 03:38.03 | ``Erik | there's no cpp switch on that |
| 03:38.17 | ``Erik | BUFXIZE ? O.o |
| 03:38.44 | ``Erik | #include <myfingersaretwistedandunusablebecauseIuseemacs.h> |
| 03:40.09 | brlcad | uh huh, that had nothing to do with emacs |
| 03:40.36 | ``Erik | O:-) |
| 03:40.56 | ``Erik | actually, I lost my emacs session |
| 03:41.17 | ``Erik | power outage dropped the machine, and I was kinda operating in the 'emacs is my os' mode :/ |
| 03:41.46 | bpoole | ``Erik: how's that going for you ? (emacs) |
| 03:41.56 | brlcad | trivial typos are my favorite kind of build errors.. because they're obvious, generally easy to provoke, and easy to fix .. typos, transposed lines, missing terminators, etc |
| 03:42.45 | ``Erik | been busy with other things, ben... got a machine on loan to replace my "stable" machine, so trying to get the ports updated :/ |
| 03:43.58 | brlcad | a good indicator of how involved someone is with a project too -- those involved just fix it and go on, those not as closely involved seem compelled to complain or comment |
| 03:44.26 | brlcad | seen that happen on at least a half-dozen oss projects .. and in-house closed |
| 03:44.57 | brlcad | not a black and white meter, but interesting trend nonetheless |
| 03:45.28 | ``Erik | I fixed it AND whined, and not on paid time, so foff :D |
| 03:45.56 | brlcad | I didn't take that as a whine |
| 03:47.42 | brlcad | talking more the sort that sit on the error without even reading it, throw their hands up and cry "it's broken !! omfg!" |
| 03:48.26 | brlcad | s/more/specifically/ |
| 03:50.29 | ``Erik | we call those "gentoo portage maintainers" ;) *duck* |
| 03:50.33 | ``Erik | O:-) |
| 03:50.36 | brlcad | see it *all* the time on BZ in particular, you would have thought the world had ended |
| 03:51.01 | ``Erik | games are a harsh place to be in general |
| 03:51.14 | brlcad | don't even have to compile-test, there are like two dozen kiddies that will let you know within an hour crying |
| 03:51.54 | ``Erik | <-- generally tries to at least compile to object before committing |
| 03:52.08 | ``Erik | but I've commited assuming it's right before... like the clone MFC ... O:-) |
| 03:52.36 | ``Erik | I've been angling more towards heavy testing though... 'distcheck' on releases, etc |
| 03:52.40 | brlcad | yeah, everyone does it from time to time |
| 03:52.46 | brlcad | i think that's part of the behavior |
| 03:53.08 | ``Erik | my personal projects are generally unbranched... |
| 03:53.12 | CIA-27 | BRL-CAD: 03brlcad 07STABLE * 10brlcad/src/other/blt/Makefile.am: partial merge from head, need x11 |
| 03:53.32 | ``Erik | actually, I have one that's close to another release, but I'm waffling due to license issues |
| 03:53.38 | brlcad | the folks that do that a lot 'generally' tend to be more fluid and not so annoyed/concerned .. those that don't so frequently.. |
| 03:53.54 | ``Erik | the code is under BSD license, I'm trying to decide whether to throw out a version that has gettext support |
| 03:54.00 | brlcad | lee actually tended to do that more and more over the years the less he coded |
| 03:54.17 | ``Erik | internationalization is a good thing, but the license concern.. |
| 03:56.07 | brlcad | I don't like to give anyone shit about it unless they press the point |
| 03:56.10 | brlcad | EVERYONE does it |
| 03:56.12 | ``Erik | if I push on gettext on my personal projects, I might try adding it to brlcad |
| 03:56.13 | brlcad | deal with it |
| 03:56.27 | ``Erik | yeah, everyone does it, but if ya never get grief, you dont' try to get better |
| 03:56.29 | brlcad | gettext is gpl... |
| 03:56.39 | brlcad | that's not true |
| 03:56.40 | ``Erik | when I fuck up, call me on it... I'll try to improve |
| 03:57.02 | ``Erik | <-- more in teh 'harsh abrasive' camp O:-) |
| 03:57.12 | brlcad | i mean there are a few that are pathologically bad at making broken commits.. but they're pretty darn rare |
| 03:57.27 | brlcad | i don't think that helps, myself |
| 03:57.35 | ``Erik | yeah, *shrug* but I'm an asshole |
| 03:57.37 | brlcad | just adds tension |
| 03:58.02 | ``Erik | I haze with a smile, it's about the quality above all |
| 03:58.10 | brlcad | especially when it's really trivial crap that doesn't matter one bit and is really obvious |
| 03:58.44 | brlcad | i mean, optimization of effort -- there are bigger issues to be complaining about and working on than typo commits |
| 03:58.55 | ``Erik | of course, I had some beat-down experiences where doubel-checking before hitting the button seriously matters |
| 03:59.28 | brlcad | yeah, if you're changing core code, or without regard to the impact, understandable |
| 03:59.43 | ``Erik | sysadmin on machines where a typo could mean several millions of dollars lost and standing in a vp's office explaining exactly how ya fucked up and how it'll never happen again... |
| 04:00.09 | ``Erik | so I'm pretty ashamed of that macro issue in the clone.c mfc :( |
| 04:00.17 | brlcad | sysadmin's a diff matter, you don't have the revision control safety net and 'undo' revertability usually |
| 04:00.20 | ``Erik | (yes, it bugs me) |
| 04:00.33 | ``Erik | actually, ya do in a good sa situation |
| 04:00.33 | brlcad | see, that's a meh issue to me other than being so close to release |
| 04:00.43 | brlcad | it would have been caught on the final compilation test |
| 04:00.55 | ``Erik | i wrote a bit of the "move to production" subsystem |
| 04:01.15 | ``Erik | it was all cvs or rcs backed, it all had complete 'undo' capability |
| 04:01.34 | ``Erik | and the machines configurations were all managed in a backed up central repo |
| 04:01.35 | ``Erik | uh |
| 04:01.36 | brlcad | yeah, but that's still pretty rare for most places |
| 04:01.37 | ``Erik | g08 |
| 04:02.04 | ``Erik | replace a machine with new hw? one line and it was generated back to production capability |
| 04:02.37 | ``Erik | from what I've heard from friends, it's pretty common for "decent" organizations. :/ I'm annoyed with our environment |
| 04:02.57 | ``Erik | we get ripped on for being "hobby shop", but the crap that's taken away is very much hobby shop |
| 04:03.08 | ``Erik | and I'm not allowed to go fix it *sigh* |
| 04:04.43 | ``Erik | "boob bones" heehhe |
| 04:04.48 | brlcad | imho (not that you're asking) that's because you either ask for/expect permission (hah) or aren't tactful on showing the benefit or hiding the time spent .. |
| 04:05.21 | ``Erik | yeah... I'm very not tactful and too open :/ |
| 04:05.23 | brlcad | path to frustration |
| 04:05.44 | ``Erik | "shut the fuck up and get the fuck out of my way" worked dandy at fedex, less politics |
| 04:06.02 | ``Erik | and it's the norm for OSS in my experience |
| 04:06.04 | brlcad | better management |
| 04:06.08 | brlcad | for both |
| 04:06.28 | ``Erik | my line manager at fedex... heh |
| 04:06.34 | ``Erik | he actually fell on his sword. |
| 04:06.36 | ``Erik | *cough* |
| 04:07.03 | ``Erik | he sued his manager and won whiel I was leaving... he really omfg fell on his sword |
| 04:07.09 | brlcad | lee has actually gone to bat for me several times over the years |
| 04:07.19 | ``Erik | me too, so has ed |
| 04:07.39 | ``Erik | I don't leverage them enough :) |
| 04:07.58 | brlcad | true dat |
| 04:08.03 | ``Erik | *shrug* |
| 04:09.21 | ``Erik | <-- too honest and up front... folk like dwayne seem to appreciate it, but mgmt doesn't |
| 04:09.39 | brlcad | hm, there is a bsd gettext |
| 04:09.57 | brlcad | http://developer.postgresql.org/~petere/bsd-gettext/ |
| 04:10.03 | ``Erik | at postgres |
| 04:10.04 | ``Erik | yeah |
| 04:10.18 | ``Erik | <- scratches his chin |
| 04:10.25 | brlcad | that would work |
| 04:10.30 | brlcad | the gnu one is teh evil |
| 04:10.45 | ``Erik | one of my apps (gpl) had gnu gettext and I ended up removing it for some reason |
| 04:10.46 | brlcad | license-wise |
| 04:10.51 | ``Erik | um, rand |
| 04:11.04 | ``Erik | I was looking to add it to timestamp, but the license issue... heh |
| 04:11.19 | brlcad | huh? |
| 04:11.41 | ``Erik | http://math.missouristate.edu/~erik/software.php?id=7 |
| 04:11.46 | ``Erik | http://math.missouristate.edu/~erik/software.php?id=95 |
| 04:12.04 | brlcad | ah |
| 04:12.15 | ``Erik | simple little utils :) |
| 04:13.05 | brlcad | what's running your site there? you wrote software.php and have a descriptor file per project or something? |
| 04:13.56 | ``Erik | um, it's backed by postgres |
| 04:14.09 | ``Erik | software.php just talks to the db |
| 04:14.18 | ``Erik | all scratch written, fairly old |
| 04:14.21 | brlcad | k |
| 04:15.01 | ``Erik | went for a semi-/. feel, wrote it just before /. did the registration stuff iirc |
| 04:15.21 | brlcad | looking to (finally) set up something similar for my data archive, but have been *trying* to avoid custom |
| 04:15.55 | brlcad | something minimal-yet-flexible for my various code and non-code projects |
| 04:15.57 | ``Erik | heh, mycrap has been maintenance for a long time, so it doesn't matter to me |
| 04:16.11 | brlcad | couple hundred that I've ever worked on, some private, most public, all already organized hierarchically |
| 04:16.17 | ``Erik | if I had to start from scratch, i'd probably go for a wiki based thing |
| 04:17.08 | brlcad | hmmm.. i suppose mediawiki could handle it with a custom module |
| 04:17.13 | ``Erik | all my crap shows age :( |
| 04:17.31 | ``Erik | 'rand' and 'vp' before image viewers could do what those do |
| 04:17.46 | brlcad | thing is I don't want to add each project individually, hoping to just create a descriptor file throughout my archive that has the access details, and let the module take care of it |
| 04:18.30 | ``Erik | hum, mine has a form to add each project, and af orm to add each OS to each project, and a from for each release, and a form for each news entry... |
| 04:18.57 | brlcad | hrm, more linkage errors on rhel |
| 04:19.27 | ``Erik | hum, looks like my big web move was in jul '99 |
| 04:19.41 | brlcad | yeah, that works for software, but a lot of mine is fairly 'random' .. lots of non-code projects too |
| 04:20.12 | brlcad | some artwork, some literature, photography |
| 04:20.18 | ``Erik | heh, I should rename the 'articles' to 'essays' I think |
| 04:20.55 | ``Erik | and my fbsd/ directory isn't wired to the db or php, it's just cvs'd html |
| 04:20.56 | brlcad | just about any CMS would work, but I so don't want to "populate" a CMS, I want the archive to be "the" data that the site presents |
| 04:21.10 | ``Erik | wait, it does use php, so I can have a + |
| 04:21.27 | ``Erik | mod_lisp looks interesting |
| 04:21.37 | louipc | oh man php is messy |
| 04:21.44 | ``Erik | I need to test if it can persist contained data |
| 04:22.04 | ``Erik | um, like have each connection thread able to access a 'global' structure |
| 04:26.58 | ``Erik | heh, wow |
| 04:27.33 | *** part/#brlcad trainman419 (n=hendrix@75.15.127.134) |
| 04:27.41 | ``Erik | 'gwci' before firefox, ... |
| 04:27.49 | ``Erik | iview... my first and only c&d :D |
| 04:37.10 | ``Erik | ehehhee an entry when I bought my first mac :) |
| 04:54.38 | CIA-27 | BRL-CAD: 03brlcad * 10brlcad/src/libdm/Makefile.am: more undefined references on RHEL, need LIBGL if opengl is enabled |
| 04:55.51 | CIA-27 | BRL-CAD: 03brlcad * 10brlcad/src/librtserver/Makefile.am: more undefined references on RHEL -- need the deps for rt, bn, and bu |
| 05:18.05 | CIA-27 | BRL-CAD: 03brlcad * 10brlcad/configure.ac: more RHELness, multispectral needs symbols from optical |
| 05:21.53 | CIA-27 | BRL-CAD: 03brlcad * 10brlcad/src/adrt/libtie/Makefile.am: uses libm |
| 05:26.30 | CIA-27 | BRL-CAD: 03brlcad * 10brlcad/src/ (8 files in 4 dirs): more flawfinding quellage and bu_exiting |
| 05:26.45 | ``Erik | tequila is evil. |
| 05:28.15 | CIA-27 | BRL-CAD: 03brlcad * 10brlcad/src/mged/tedit.c: quell constness warnings, make the editor string const |
| 05:28.28 | louipc | wowsers |
| 05:28.34 | brlcad | hey ``Erik .. question for ya |
| 05:28.46 | brlcad | probably not the best time if your tequealing |
| 05:28.52 | brlcad | but .. |
| 05:29.38 | brlcad | trying to come up with some way to add -Werror to the build now that we're almost completely warningless on some platforms ..... but not for anything in src/other |
| 05:29.54 | brlcad | any ideas? |
| 05:30.01 | louipc | hacking drunk is a blast |
| 05:30.07 | louipc | well anything drunk is a blast |
| 05:30.10 | brlcad | :) |
| 05:33.58 | ``Erik | heh, uhhhh |
| 05:34.26 | ``Erik | I kinda had the notion of moving $(top_srcdir)/src/other to $(top_srcdir)/contrib |
| 05:34.37 | ``Erik | so abuse can be applied to src/ without crap getting in the mix |
| 05:35.17 | ``Erik | I don't know of any way to undo a cflag without automake bitching |
| 05:35.54 | brlcad | maybe after svn |
| 05:36.07 | ``Erik | *nod* would be ugly on cvs |
| 05:36.09 | brlcad | that's one way I suppose, could then add it to the subtree |
| 05:36.45 | ``Erik | well |
| 05:36.54 | ``Erik | you know my opinion on the src/other dir |
| 05:36.56 | ``Erik | *shrug* |
| 05:37.11 | brlcad | yep |
| 05:37.19 | ``Erik | I was actually a bit upset with how many imports I found in my personal cvs repo :( |
| 05:37.26 | ``Erik | siod, quack 2, ... |
| 05:37.56 | ``Erik | <-- talks it, but doesn't always walk it :( |
| 05:40.27 | ``Erik | now, uh, if you feel like being repetitive, you could always add it to AM_CFLAGS for each dir |
| 05:41.26 | brlcad | yeah, nah, that'd suck |
| 05:41.33 | ``Erik | on personal projects, I like "-W -Wall -Werror -ansi -pedantic", but BRL-CAD doesn't cope well with that rig |
| 05:41.36 | brlcad | the benefit is pretty minimal, just nice housekeeping |
| 05:41.55 | ``Erik | heh |
| 05:42.03 | ``Erik | I almost got one of my project splint clean |
| 05:42.13 | brlcad | yeah, that's pretty standard set |
| 05:42.22 | brlcad | usually use the same |
| 05:42.31 | ``Erik | flawfinder has a couple low hits (getopt not protecting buffers on old platforms, etc) |
| 05:42.54 | brlcad | i wouldn't look at anything below 3, it's got way to many false positives |
| 05:43.08 | ``Erik | um, I think the getopt() one actually lsts as 2 |
| 05:43.09 | brlcad | even most of the 3's |
| 05:43.09 | ``Erik | er |
| 05:43.10 | ``Erik | 3 |
| 05:43.22 | brlcad | there are now about 800 4's left |
| 05:43.35 | brlcad | about 600 are quellable |
| 05:43.46 | brlcad | the rest are false positives |
| 05:43.59 | brlcad | with no reasonable way to shut them up |
| 05:44.00 | ``Erik | I'd imagine that 99% of them are "well handled" to begin with... the kinda thing a new coder could be thrown at |
| 05:44.21 | brlcad | that's where I think coverity will do a better job |
| 05:44.47 | brlcad | yeah, easily |
| 05:45.00 | ``Erik | how's that coming along? those dudes were trying to get something fixed to cope with our weirdness, iirc? |
| 05:45.03 | brlcad | one of a dozen tasks like that |
| 05:45.31 | ``Erik | I want good libbu and libbn documentation :( |
| 05:45.32 | brlcad | yeah, they're just "busy" last I pinged them |
| 05:45.40 | brlcad | bu, bn, and rt |
| 05:45.53 | ``Erik | when I was mucking with vls for clone, I was unhappy at how often I had to go look at code |
| 05:45.58 | brlcad | pkg and wdb for good measure |
| 05:46.22 | ``Erik | and completion felt weak, thus the macros |
| 05:46.33 | brlcad | completion? |
| 05:47.00 | brlcad | you mean funcs missing? |
| 05:47.10 | ``Erik | yeah, I expected strcmp type things |
| 05:47.40 | brlcad | that's just a matter of refactoring -- stuff hasn't really been ever added to any of the libs until/unless it's in use |
| 05:47.49 | ``Erik | total encapsulation type stuff |
| 05:47.52 | brlcad | nobody ever needed/refactored with a strcmp need |
| 05:49.36 | ``Erik | I wonder if vls's shoudl hold some kinda hash value for fast strcmp type needs :/ |
| 05:51.03 | ``Erik | even though there are only 5 calls in the entire source tree, heh |
| 05:51.45 | ``Erik | but there are a fair number of strcmp() with one arg being a vls... |
| 05:53.44 | brlcad | damn alicia keys is hot and a awesome artist |
| 05:55.01 | brlcad | 45 strcmp's that have at least one vls lookup |
| 05:55.02 | ``Erik | in other news; fire is hot, ice is cold, and windows sucks. |
| 05:55.37 | ``Erik | but they're not consistant in which arg is the vls |
| 05:55.43 | ``Erik | easily fixable, but a grind |
| 05:56.33 | brlcad | many of those strcmps could go away too |
| 05:56.58 | brlcad | 1236 calls to strcmp |
| 05:57.35 | brlcad | probably a third should be strncmp or strlcmp |
| 05:57.36 | ``Erik | <-- didn't look at 'em, just noticed when finding strcmp(bu_vls_addr(a),bu_vls_addr(b)) |
| 05:58.02 | ``Erik | heh, I thought you didn't like strlcmp :D too obsd specific or something |
| 05:58.55 | ``Erik | I put in requests to get cygwin and msvc on bobs winderz machine |
| 05:59.32 | ``Erik | if he can make it so his office mate *cough* can build without asking anyone for help, that'd be a good step |
| 06:00.19 | brlcad | strlcmp is a bad example, that's not c89 iirc |
| 06:00.26 | brlcad | but those kinds of functions |
| 06:00.33 | ``Erik | no, not c90, not c99... obsd |
| 06:00.35 | brlcad | to check length at same time |
| 06:00.47 | ``Erik | strncmp() is the best portable afaik |
| 06:00.54 | brlcad | yeah, that one is c89 |
| 06:01.43 | ``Erik | my obsd builds said to use strlcmp() instead, that's the only place I've seen that func come up |
| 06:04.36 | brlcad | yeah, strncmp is a pain to remember that it's length and not size, though less important for cmp |
| 06:04.53 | brlcad | more important for cpy, strncpy vs strlcpy |
| 06:05.08 | ``Erik | heh |
| 06:06.47 | brlcad | cept it allocates, can't use buffers |
| 06:07.05 | ``Erik | most uses of strcpy() follow a malloc |
| 06:07.46 | brlcad | eh |
| 06:08.16 | ``Erik | sorry, they do |
| 06:08.34 | brlcad | I don't believe that, at least it's just as likely that someone copies into a char[] than that they have an alloc'd char* |
| 06:08.43 | ``Erik | go ahead, script up something to grind on sf to look for that pattern |
| 06:08.48 | ``Erik | it's.... common |
| 06:08.52 | brlcad | you're the one making the claim |
| 06:08.57 | ``Erik | heh |
| 06:09.06 | brlcad | i'm not gonna waste my time doing that :) |
| 06:09.11 | ``Erik | I'm also the one who's excessively lazy ;) |
| 06:10.02 | brlcad | besides it doesn't matter -- if people only ever made dups of a string then str[nl]cpy wouldn't have been made and we'd only have strdup :P |
| 06:10.17 | ``Erik | in the code I've laid hands on, I believe that most uses of str*cpy() are into freshly allocated memory (often malloc'd, sometimes stack alloc) |
| 06:10.38 | ``Erik | strcpy predates strdup by quite a bit, though |
| 06:11.07 | brlcad | having gone through nearly 500 instances in brl-cad's code already, it's a fairly even split of buffers and mallocs |
| 06:11.33 | brlcad | entirely unscientific counting of course |
| 06:11.51 | ``Erik | the gnarled hands of an emacs use? |
| 06:11.53 | ``Erik | user? |
| 06:12.25 | brlcad | just recollections of the edits since I've been checking other things while I cleanup (like bombing, !!!!'s, allocs, etc) |
| 06:13.05 | ``Erik | I wonder if user message format should be addressed in hacking |
| 06:13.15 | brlcad | the most annoying is the f'd up inconsistent indentation style on some of the dirs |
| 06:13.16 | ``Erik | consistant message format |
| 06:14.04 | brlcad | think I need to add a style var to the emacs local var footer so that braces aren't jacked on indent |
| 06:14.24 | ``Erik | :) |
| 06:14.40 | brlcad | vim has the same issue, just a matter of setting the style |
| 06:14.57 | ``Erik | probably handled in my overweight .vimrc |
| 06:15.00 | brlcad | I can set it for my emacs, but then that doesn't solve the problem |
| 06:16.19 | brlcad | moss's code is one of the worst offenders |
| 06:16.27 | ``Erik | heh |
| 06:16.41 | brlcad | his code often smells like s2 |
| 06:16.47 | ``Erik | hum, splain it to me, lucy |
| 06:16.53 | ``Erik | rt asks for a ray |
| 06:16.57 | ``Erik | librt shoots it |
| 06:16.58 | brlcad | (not surprising since that's what he worked on) |
| 06:17.07 | ``Erik | liboptical figures out what to do with a hit |
| 06:17.20 | ``Erik | no? where are the actual hand-offs in all that? |
| 06:17.30 | brlcad | that's about right |
| 06:17.52 | ``Erik | suppose I wanted to write a direct hit lighting model |
| 06:17.55 | brlcad | then liboptical sets the value, the rt sends to libfb and/or whatever output channel via the application struct |
| 06:17.58 | ``Erik | -l99 |
| 06:18.10 | ``Erik | how do I stub it in? |
| 06:18.14 | brlcad | data passing is done through the resource and application structures |
| 06:18.18 | brlcad | mostly the app structure |
| 06:18.21 | ``Erik | the photon mapping stuff seems horribly wedged |
| 06:18.42 | brlcad | lighting models are handled in rt |
| 06:18.47 | brlcad | it is horribly wedged |
| 06:18.53 | brlcad | justin's not a great example |
| 06:19.01 | ``Erik | I saw three big siwtches for lighting model in rt/ I think |
| 06:19.03 | brlcad | er, his code |
| 06:19.17 | brlcad | yeah, they're about it |
| 06:19.34 | ``Erik | so there's no map table or anything? it's just hard coded in? |
| 06:20.22 | ``Erik | I should probably look at the uhhhh normal coloring |
| 06:21.01 | ``Erik | seemed there was no clear contract between rt and liboptical... |
| 06:21.17 | ``Erik | <-- kinda wants to write a 'toon' shader :) |
| 06:21.24 | brlcad | that's what the view files do -- handoffs between whatever optic model you're using |
| 06:22.08 | ``Erik | um, are the shadow rays thrown in rt/view.c ? |
| 06:23.02 | brlcad | shadow rays are specific to a phong-type shader, phong actually shoots them |
| 06:23.20 | brlcad | (i.e. in liboptical) |
| 06:24.18 | brlcad | hence the variety of rt_shootray calls in liboptical |
| 06:24.54 | ``Erik | hm |
| 06:25.20 | ``Erik | don't be too terribly surprised if you see a new lighting model in the semi-near future :) |
| 06:25.37 | ``Erik | it seems like one of those things with a good thinking/grunt ratio |
| 06:26.16 | brlcad | fun stuff |
| 06:26.19 | ``Erik | I think I got most of the grunt in clone out of the way, but the thinking involves some major user insight :/ |
| 06:26.50 | ``Erik | classic mirror works only on cardinal axis... I want it to work on arbitrary vectors... |
| 06:26.51 | ``Erik | etc |
| 06:27.09 | brlcad | alas, I haven't coded up a new/clean lighting model yet, so no path of lookups like there are for new shaders and prims |
| 06:27.19 | brlcad | but there pretty straightforward |
| 06:27.32 | ``Erik | um |
| 06:27.34 | ``Erik | speaking of prims |
| 06:27.39 | brlcad | there are a couple really simple ones that do things like visualize the material thickness per ray |
| 06:27.40 | ``Erik | g-xxx.c ... should be fixed |
| 06:28.06 | ``Erik | the normal model looks like a good place to dig in to me |
| 06:28.17 | ``Erik | since I just want first hit shading |
| 06:28.59 | ``Erik | I dunno if I said it, but we need a function to generate a ray set correlating to an arbitrary ae for both ortho and perspective |
| 06:29.02 | brlcad | rtedge probably could have been done as a lighting model |
| 06:29.08 | ``Erik | so g_qa.c can have non-cardinal runs |
| 06:29.14 | brlcad | heh |
| 06:29.20 | brlcad | yeah, I've been saying that for years too |
| 06:29.30 | brlcad | just haven't gotten to that refactoring |
| 06:29.35 | ``Erik | yeah, heh |
| 06:29.41 | brlcad | its' just pushing a chunk of code from worker into librt |
| 06:29.46 | brlcad | "just" |
| 06:29.48 | ``Erik | I keep saying "yup, we need it, uh, maybe after XXX" |
| 06:30.14 | ``Erik | many of the rt's duplicate functionality in not quite the same manners :( |
| 06:30.19 | ``Erik | and I don't like that |
| 06:30.57 | ``Erik | heh, rt should boil down to if(strncmp(*argv, "... |
| 06:30.58 | ``Erik | O:-) |
| 06:31.01 | ``Erik | imho |
| 06:31.25 | brlcad | yeah, and much of it is like that .. just not enough |
| 06:31.49 | brlcad | the entire "rt backend" -- which cake actually made more obvious -- belongs in a lib |
| 06:31.56 | ``Erik | oh, uh, image conversion, I'm tempted to grab libconvert and shove it in src/other/ and redo the entire pix conversion suite to be two func calls |
| 06:32.21 | ``Erik | hey, I was given a cd and told to make it automake |
| 06:32.26 | ``Erik | I had no clue what I was lookin' at :D |
| 06:33.03 | brlcad | not saying it's right or wrong, just autoobfuscated :) |
| 06:33.32 | ``Erik | cake still gives me the willies |
| 06:34.04 | ``Erik | if I'd have known any utility cake providd above and beyond, I would've commented it in the am |
| 06:35.11 | ``Erik | <-- got a few dirty looks when he first started cuz he kept referring to the "hit by a bus" situation as a need to document properly... |
| 06:35.45 | ``Erik | guess it hit too close to home for a few folk :/ |
| 06:36.43 | brlcad | no time to make things simple for the noobs :) |
| 06:36.51 | brlcad | that's valuable coding time |
| 06:37.03 | ``Erik | heh |
| 06:37.20 | ``Erik | should I point out the tragic flaw there? :D |
| 06:37.48 | brlcad | nope |
| 06:39.01 | ``Erik | y'know |
| 06:39.14 | brlcad | yup |
| 06:39.19 | ``Erik | much of libbu's functionality is found in glib |
| 06:39.34 | ``Erik | how heretical would it be to ponder that? |
| 06:39.41 | brlcad | sure is, the nature of a basic library |
| 06:39.51 | brlcad | also much of it is in the stl too |
| 06:40.05 | ``Erik | stl is for c++ weenies, not real coders |
| 06:40.13 | brlcad | as well as the boost libs, glib, and a dozen others ;) |
| 06:40.16 | ``Erik | last I heard, stl wasn't very... standard... |
| 06:40.24 | brlcad | that's old news |
| 06:40.35 | ``Erik | the msvc users were hand deleting their own stl and importing sgi's |
| 06:40.45 | brlcad | it's pretty solid now |
| 06:41.01 | brlcad | anythings past msvc 6 |
| 06:41.10 | brlcad | msvc 6 and before sucked ass |
| 06:41.11 | ``Erik | msvc6 is still used |
| 06:41.15 | ``Erik | heh |
| 06:41.19 | ``Erik | I have a cd with msvc1 |
| 06:41.22 | brlcad | exceptionally rare |
| 06:41.26 | brlcad | that's ancient bustage |
| 06:41.30 | ``Erik | up against the cd with borland 4.51 |
| 06:41.41 | ``Erik | wanna trip into suckage? I can loan 'em... :D |
| 06:41.54 | brlcad | i have vc6 somewhere |
| 06:41.56 | brlcad | it's not worth it |
| 06:42.05 | brlcad | it was outright broken and incompliant in many ways |
| 06:42.21 | ``Erik | I used msvc5 and 6, it convinced me to rely on cygwin/mingw32 |
| 06:42.36 | brlcad | can take care of the issues with some things (C code isn't too horrible, C++ is horrible, though) |
| 06:42.48 | ``Erik | heh |
| 06:42.51 | brlcad | 7 and 8 were vast improvements |
| 06:43.01 | ``Erik | at the time, msvc supported c++ a whole lot better than gcc |
| 06:43.09 | ``Erik | uh, egcs, for the cool kids |
| 06:43.46 | brlcad | yeah, caring about msvc6 is like caring about gcc1/egcs portability |
| 06:43.54 | brlcad | you just don't do it |
| 06:44.17 | brlcad | unless you have somebody funding that waste of time |
| 06:45.00 | ``Erik | :/ I was a big c++ proponent for a while, then I got heavy into leenewx and realized C could do all that stuff without much work... then many years later, I got into smalltalk,and then I "got" oo (I think)... and c++ just ain't there... objc is a lot closer |
| 06:46.21 | brlcad | smalltalk is one of my favorite languages |
| 06:46.26 | ``Erik | amusingly enough, ruby is probably my favorite "oo" language, and I've run into several situations where ruby just doesn't hold up so I switch to thscheme or lithp inthtead |
| 06:46.29 | brlcad | favorite impractical languages |
| 06:46.58 | ``Erik | the crypt game in squeak... w00t :D |
| 06:47.40 | brlcad | c++ best and worst feature is that it doesn't really confine you to any paradigm and actually lets you implement three or four fundamentally different styles |
| 06:48.02 | ``Erik | 'cept, uh, it doens't do a couple that it claims to |
| 06:48.15 | ``Erik | the biggest feature of oo, imho, is multi-dispatch |
| 06:48.23 | brlcad | you can seriously abuse the template layer and get lispness, you can go pure OO, you can go run-time ala objc, functional, etc |
| 06:48.29 | ``Erik | and alan kay has said the same... he has harsh things to say about c++ |
| 06:48.57 | ``Erik | I loaned out the uh, patterns book |
| 06:49.00 | brlcad | that just means the power is in the hands of the coder to know what to do and what NOT to do |
| 06:49.16 | ``Erik | the notion of a design pattern, to me, is an indication of a failure in the language to do its duty |
| 06:49.53 | brlcad | eh, you have design patterns in just about every language |
| 06:49.57 | ``Erik | I have a set of slides somewhere, um, design patterns for ml |
| 06:50.00 | brlcad | it's just a .. pattern of use |
| 06:50.14 | ``Erik | and MOST of the c++ listed ones disappear, because "the language just handles that" |
| 06:51.21 | brlcad | yeah.. ml's a great practical language :P |
| 06:51.30 | brlcad | a clear success :) |
| 06:51.32 | ``Erik | heh |
| 06:51.42 | ``Erik | it's a good language |
| 06:51.46 | ``Erik | it just has no good apps |
| 06:51.49 | brlcad | heh |
| 06:51.53 | brlcad | that's not all it's missing :) |
| 06:52.25 | ``Erik | <-- thtill a lithp weenie, tho won't argue |
| 06:52.28 | brlcad | otherwise it'd have a significant devout following like the python/ruby/scheme/etc camps |
| 06:52.38 | brlcad | it's a great academic language for teaching languages |
| 06:52.45 | ``Erik | I think both breeds have a serious following |
| 06:52.58 | ``Erik | and scheme is kinda the lithp for schools... |
| 06:53.10 | ``Erik | the pathcal of lithpth, even |
| 06:53.36 | ``Erik | still my favorite lisp, but I'm trying to learn common lisp... |
| 06:53.39 | ``Erik | and emacs |
| 06:53.48 | ``Erik | they tell me that's what the grown-ups use |
| 06:54.05 | brlcad | every language has its strengths, weaknesses, and applicability for various purposes |
| 06:54.16 | ``Erik | indeed |
| 06:54.23 | brlcad | one of C++'s best strengths and biggest weaknesses is its flexibility |
| 06:54.27 | ``Erik | only those fluent in a wide variety have anything to say |
| 06:54.49 | brlcad | depends what I'm doing |
| 06:55.02 | brlcad | (for me) |
| 06:55.10 | ``Erik | though I have had on and off urges to use c++ over the last decade or so :/ |
| 06:55.28 | ``Erik | "that'd fit well into c++'s hobbled notion of oo" |
| 06:55.30 | ``Erik | heh |
| 06:55.45 | ``Erik | but at that point, ruby is probably 'nuff |
| 06:55.48 | brlcad | for big architecting, c++ wins for me, good balance of power, performance, and flexibility |
| 06:55.53 | brlcad | (over objc) |
| 06:56.04 | brlcad | for dynamic interactive apps, objc usually wins |
| 06:56.25 | brlcad | to a point, gets nasty as they get bigger |
| 06:56.28 | ``Erik | see, I'd rather start with an obscenely high level language, do bottom up in it, or the skeery bi-directional deisgn of lisp users |
| 06:56.41 | ``Erik | and as the language fails, start re-doing bits in a lower level language |
| 06:57.07 | ``Erik | if I *NEED* to shave that .02% off, switch to C |
| 06:57.51 | brlcad | all of the functional languages fail for me from purely practical standpoints -- admittedly self-defeating reasons -- not useful to use for an app core simply because there aren't many coders for that language and it's a polarizing political/academic debate on the language's utility |
| 06:58.18 | ``Erik | that approach has failed me bigtime, though. I had a top down shooter game that developed a hitch every half second due to naive stop© gc :/ |
| 06:58.37 | brlcad | ew |
| 06:58.56 | ``Erik | should hacked a generational into it |
| 06:59.03 | ``Erik | or at least treadmill |
| 06:59.38 | ``Erik | (amusingly enough... I argued gc's and learned a fair bit from salzman... who went to cmu...) |
| 07:01.08 | ``Erik | hrmph |
| 07:01.21 | ``Erik | I wonder where my big orange and blue "basic for the commodore 64" book went |
| 07:01.31 | ``Erik | it had a chapter on garbage collection, but I ignored it... |
| 07:01.52 | ``Erik | I mean, 64,000 bytes? how can you run out of that??? ... (until I tried to do voxels on the machine) |
| 07:03.12 | *** join/#brlcad IriX64 (n=mariodot@bas2-sudbury98-1128564803.dsl.bell.ca) |
| 07:06.22 | ``Erik | *grouse* |
| 07:07.42 | *** join/#brlcad CIA-27 (i=cia@208.69.182.149) |
| 07:08.16 | ``Erik | this sucks. |
| 07:20.20 | ``Erik | nhttp://www.xkcd.com/ |
| 07:22.10 | ``Erik | http://www.irregularwebcomic.net/ |
| 07:30.21 | IriX64 | i should take slight to that :) |
| 07:30.58 | IriX64 | yea, got cygwin running on vista64. |
| 07:32.53 | IriX64 | singular works, means i can compile brlcad wee |
| 08:04.44 | IriX64 | configure pegs the user cpu meter but kernal time is approx 60 % :) |
| 08:06.50 | tarzeau | brlcad: strange and with -O0 too? |
| 08:07.02 | tarzeau | brlcad: anyways, if you can't reproduce it... i can build debian packages now :) |
| 08:13.03 | *** join/#brlcad MinuteElectron (n=MinuteEl@silentflame/member/pdpc.base.minuteelectron) |
| 08:21.12 | *** join/#brlcad Z80-Boy (n=clock@zux221-122-143.adsl.green.ch) |
| 08:32.42 | *** join/#brlcad IriX64_ (n=mariodot@bas2-sudbury98-1128564803.dsl.bell.ca) |
| 09:12.54 | *** join/#brlcad ibot (i=ibot@rikers.org) |
| 09:12.55 | *** topic/#brlcad is BRL-CAD Open Source Solid Modeling || http://brlcad.org || http://sf.net/projects/brlcad || BRL-CAD is on scan.coverity.com but not completely set up yet, more news TBD || Release 7.10.4 is happening now |
| 09:36.57 | *** join/#brlcad rpaddock (n=bob_padd@c-24-131-109-50.hsd1.oh.comcast.net) |
| 09:43.13 | brlcad | tarzeau: nope, working fine .. not sure how changing the optimization level is supposed to break the build, works fine here |
| 09:43.25 | brlcad | tarzeau: if you have a build log, be interesting to see the error |
| 09:43.35 | brlcad | otherwise, sounds like you have something working on your end |
| 10:08.39 | tarzeau | brlcad: ok thanks |
| 10:09.50 | MinuteElectron | brlcad: busy? |
| 10:15.31 | *** part/#brlcad rpaddock (n=bob_padd@c-24-131-109-50.hsd1.oh.comcast.net) |
| 10:33.52 | *** join/#brlcad elite01 (n=elite01@dslc-082-082-082-010.pools.arcor-ip.net) |
| 12:04.54 | starseeker | Ah hah - now the rt06 download works. |
| 14:09.57 | *** join/#brlcad MinuteElectron (n=MinuteEl@silentflame/member/pdpc.base.minuteelectron) |
| 14:42.51 | Z80-Boy | brlcad: hi |
| 14:47.51 | *** join/#brlcad bpoole (n=bpoole@UNIX34.andrew.cmu.edu) |
| 15:50.04 | *** join/#brlcad MinuteElectron (n=MinuteEl@cpc4-flit1-0-0-cust283.lutn.cable.ntl.com) |
| 16:15.44 | brlcad | MinuteElectron: quite |
| 16:15.49 | MinuteElectron | ok |
| 16:15.58 | MinuteElectron | I won't bother you then, my query is non-essential. |
| 16:16.31 | MinuteElectron | Well, that really depends on your viewpoint - but I'd rather not disturb you and it is nothing integral. |
| 16:16.33 | brlcad | i'll ping you later when I can chat better |
| 16:16.37 | MinuteElectron | ok, sure |
| 16:16.59 | brlcad | tweaked the header some more for the download image |
| 16:17.17 | MinuteElectron | I saw ;) - looking better. |
| 16:17.27 | brlcad | yeh, still not great, but better |
| 16:17.35 | MinuteElectron | True |
| 19:32.20 | *** join/#brlcad Elperion (n=Bary@p54877F62.dip.t-dialin.net) |
| 19:32.58 | *** join/#brlcad elite01 (n=elite01@dslc-082-082-082-010.pools.arcor-ip.net) |
| 19:33.45 | *** join/#brlcad iraytrace (n=iraytrac@trapezewireless23.insecure.utah.edu) |
| 19:34.15 | *** join/#brlcad iraytrace (n=iraytrac@trapezewireless23.insecure.utah.edu) |
| 19:34.58 | *** part/#brlcad iraytrace (n=iraytrac@trapezewireless23.insecure.utah.edu) |
| 20:16.51 | *** join/#brlcad yukonbob (n=yukonbob@CPE001125477e9c-CM0011e6be27b1.cpe.net.cable.rogers.com) |
| 21:40.16 | *** join/#brlcad Z80-Boy (i=clock@77-56-83-7.dclient.hispeed.ch) |
| 21:49.26 | *** join/#brlcad rpaddock (n=bob_padd@c-24-131-109-50.hsd1.oh.comcast.net) |
| 22:33.14 | starseeker | Cool - they FINALLY collect the ebuild messages for multiple ebuilds at the end of the whole install |
| 22:33.32 | ``Erik | huh, tclb2 |
| 22:34.54 | MinuteElectron | brlcad: That thing I was talking about earlier was because I got a bus error from PHP and the guys in ##php said that it was something to do with the binaries. Probably the completely wrong diagnosis but I was worried it might be important. So, yeah. I am going to go to bed now - speak soon I suppose. |
| 22:39.26 | starseeker | This direct raytracing of NURBs looks interesting - I'm curious though, how would such behavior interact in librt with the other types of analysis BRL-CAD is used for? Does the ARL have an internal database of previous analyses it could use to compare the results of the old methods and a NURBS based intersection simulation? |
| 22:45.08 | starseeker | Hmm, hang on - gotta restart to make sure I didn't just nuke my sound drivers... |
| 22:54.59 | ``Erik | heh |
| 23:04.54 | *** join/#brlcad starseeker (n=CY@ip72-218-18-244.hr.hr.cox.net) |
| 23:05.04 | starseeker | Ah, there we go :-) |
| 23:06.07 | ``Erik | heh |
| 23:07.01 | ``Erik | starseeker, we have a good amount of geometry that was modelled in tnurbs using stuff like pro-engineer that was tesselated to convert to our format... we can't share those files, but we can test for ya if you do nurbs work |
| 23:07.15 | ``Erik | right now |
| 23:07.22 | starseeker | hehe :-) |
| 23:07.49 | starseeker | I'm not ready yet to do that level of work - got some homework to do first |
| 23:07.54 | ``Erik | the geometry isn't classified secret or anything, but it's not releasable due to contract constraints |
| 23:08.00 | starseeker | Right |
| 23:08.41 | ``Erik | folk like general dynamics and raytheon dont' embrace the same notions :) |
| 23:08.55 | starseeker | Pity ;-) |
| 23:13.36 | starseeker | ``Erik: From the "public" standpoint, I suppose it would probably be more interesting anyway to model thinks like WWII weapons |
| 23:13.59 | starseeker | Although I doubt enough detailed info survives on most of that to build a model anymore |
| 23:14.07 | ``Erik | um |
| 23:14.19 | ``Erik | it does |
| 23:14.24 | starseeker | Really?? |
| 23:14.28 | ``Erik | the gear still exists, some in private hands |
| 23:14.57 | starseeker | Oh, sure - but that would be quite a project. The blueprints was what I was thinking about... |
| 23:15.02 | ``Erik | wheel your arse out to APG with a caliper and tape measure, got a field full of ww1 and ww2 tanks just sitting there for epeople to gawk at |
| 23:15.10 | ``Erik | I'm sure blueprints exist, but might not be public |
| 23:15.19 | starseeker | point |
| 23:18.06 | ``Erik | um |
| 23:18.11 | ``Erik | buildings are public record |
| 23:18.45 | ``Erik | it was like $20 for a reproduction of a residence like 15 years ago |
| 23:18.56 | starseeker | Hmm. |
| 23:35.21 | starseeker | errrgh. The design documents for the superconducting super collider are online, but they are some sort of funky postscript |
| 23:48.34 | starseeker | Humph. Looks like you can get them but you have to wade through huge bureaucracy to do it. Figures. Same thing with the Saturn V goodies. |
| 23:56.41 | ``Erik | hum, ps is a language intended to spew 2d stuff |
| 23:57.01 | ``Erik | um, we might be able to snarf that up into sketch objects or something |
| 23:57.21 | starseeker | http://www.hep.net/ssc/new/repository.html |
| 23:57.43 | starseeker | Ghostscript can't do much with them out of the box |
| 23:58.08 | starseeker | They've got copyright Apple in there so it might be a capture of some code sent to an Apple printer... |
| 23:58.50 | starseeker | I've dealt with some postscript code intended for printers, and it ain't pretty... |
| 23:59.06 | ``Erik | um |
| 23:59.25 | ``Erik | NeXT used ps as the primary display language, iirc, and jobs may've carried those properties when he went back to apple |
| 23:59.32 | ``Erik | might be display related, not printer related |
| 23:59.33 | ``Erik | :) |
| 23:59.39 | starseeker | Ah :-) |
