| 00:32.45 | brlcad | giving we're probably only going to accept no more than 4 students, maybe as low as 2, it's a good idea to impress ;) |
| 02:04.43 | *** join/#brlcad me_ (n=me@c-69-141-247-21.hsd1.nj.comcast.net) | |
| 02:32.21 | *** join/#brlcad Twingy (n=justin@74.92.144.217) | |
| 03:19.28 | *** join/#brlcad minusinsk (n=jishi@83.234.35.158) | |
| 04:17.26 | brlcad | mmm, annotations and parametrics will be really cool |
| 04:17.37 | brlcad | even believes that guy might actually be able to implement it |
| 04:22.26 | yukonbob | waves in -- feels it's been interesting chatter in here last couple days |
| 04:41.05 | *** join/#brlcad Twingy (n=justin@74.92.144.217) | |
| 04:49.47 | *** join/#brlcad cad75 (n=dde88e7e@bz.bzflag.bz) | |
| 05:19.31 | hippieindamakin8 | so how many applications have u recieved so far ? |
| 05:53.06 | *** join/#brlcad Z80-Boy (n=clock@217-162-109-63.dclient.hispeed.ch) | |
| 06:00.39 | brlcad | hippieindamakin8: worry about yours :) |
| 06:02.31 | brlcad | I don't know if you participated last year, but your chances are considerably higher the earlier you get it in .. and chances tend to be much higher with orgs that take on fewer students (like us!) if the applications are focused |
| 06:07.33 | Z80-Boy | brlcad: I may need to use BRL-CAD to model a monitor, wall holder, a cabinet and a room corner because I want to make a custom wall holder for a monitor |
| 06:08.12 | Z80-Boy | brlcad: also, althought it might sound funny in the company of gadgets like the Hubble Telescope, Ronja has been published on a IEEE conference. |
| 06:09.15 | brlcad | cool, which one? |
| 06:09.32 | Z80-Boy | TENCON 2006 |
| 06:09.40 | Z80-Boy | Do you want to know the ISBN? :] |
| 06:10.00 | Z80-Boy | finds it funny that a design like Ronja has an ISBN |
| 06:10.24 | brlcad | in india? |
| 06:10.28 | brlcad | ooh, 2006 |
| 06:10.33 | Z80-Boy | Asia |
| 06:10.37 | Z80-Boy | I don't know where exactly |
| 06:10.59 | brlcad | hong kong |
| 06:11.19 | brlcad | http://www.ieee.org.hk/tencon2006/ |
| 06:11.44 | brlcad | so did you publish something, or one of your users? |
| 06:12.46 | Z80-Boy | One of the users did some modification - replace the LED with a laser diode - and wrote an article about it. Because I designed virtually the whole device, he put me there as a co-author |
| 06:12.51 | Z80-Boy | http://ronja.twibright.com/web.php |
| 06:12.58 | Z80-Boy | there's a link to abstract and full text |
| 06:13.14 | Z80-Boy | He put me there as "Karel Kulhavy, Twibright Laboratories, Prague, Czech Republic" ;-) |
| 06:16.36 | Z80-Boy | Someone else wrote a thesis about underwater Ronja: "the future tactical ocean environment will be increasingly complicated. In addition to traditional communication links there will be a proliferation of unmanned vehicles..." |
| 06:17.19 | Z80-Boy | Apparently the idea is to mount Ronja on some kind of combat submarines? |
| 06:17.34 | Z80-Boy | They should make sure they are big enough not to sink under the weight of Ronja :] |
| 06:18.26 | brlcad | light-based transmission would not be my first thought for underwater comms |
| 06:19.58 | Z80-Boy | you could communicate with torpedoes |
| 06:20.06 | Z80-Boy | send the "we don't like you" message/ |
| 06:20.36 | Z80-Boy | submarines use like 20kHz to connect with the land, don't they? |
| 06:47.00 | brlcad | dunnos |
| 06:53.05 | cosurgi | brlcad: good morning, I see that you are compiling :-) |
| 06:53.14 | brlcad | :) |
| 06:53.17 | cosurgi | looks through backlog |
| 06:53.26 | brlcad | and now going to bed :) |
| 06:53.32 | cosurgi | (it's 7:50 here) |
| 06:53.54 | Z80-Boy | is listening to 7 years and 50 days from groove coverage |
| 06:54.10 | brlcad | -5 here |
| 06:59.31 | cosurgi | heh :) (still reading - not easy with 2y daughter sitting at keyboard ;) |
| 06:59.40 | cosurgi | is compilation succesfull? |
| 08:03.17 | *** join/#brlcad clock_ (n=clock@zux221-122-143.adsl.green.ch) | |
| 09:04.18 | *** join/#brlcad jdoliner (n=jdoliner@cpe-24-59-109-153.twcny.res.rr.com) | |
| 09:20.35 | *** join/#brlcad Elperion (n=Bary@p548761A2.dip.t-dialin.net) | |
| 09:42.23 | *** join/#brlcad F-fisher (n=29dd105b@bz.bzflag.bz) | |
| 09:43.35 | F-fisher | hello |
| 09:43.47 | F-fisher | is there smeone here ?? |
| 09:44.26 | F-fisher | no, i'm alone |
| 09:58.38 | *** part/#brlcad jdoliner (n=jdoliner@cpe-24-59-109-153.twcny.res.rr.com) | |
| 10:06.48 | *** join/#brlcad nmh_2Grajw (n=nmh@gw.nomh.org) | |
| 10:11.58 | nmh_2Grajw | hi, I was hoping someone could point out how to get the brlcad install process to provide libblt2.4.so (in addition to libblt2.4.a). I am installing from source. |
| 10:29.46 | cosurgi | nmh_2Grajw: debian? |
| 10:30.02 | nmh_2Grajw | consurgi: macos 10.4 |
| 10:30.13 | cosurgi | I tried this: |
| 10:31.17 | cosurgi | svn checkout https://brlcad.svn.sourceforge.net/svnroot/brlcad/brlcad/trunk brlcad ; cd brlcad ; sh autogen.sh ; ./configure --enable-all --enable-optimized ; make LDFLAGS="-L/home/janek/20-Programowanie/10-cpp/51-Brlcad/brlcad/src/other/tcl/unix -ltcl8.5 -L/home/janek/20-Programowanie/10-cpp/51-Brlcad/brlcad/src/other/tk/unix -ltk8.5" |
| 10:31.48 | cosurgi | but it was on debian. Then I stumbled on some debian-related error, and brlcad is now looking into it. |
| 10:32.13 | cosurgi | but it's night in US now, so he is sleeping ;) |
| 10:33.11 | nmh_2Grajw | consurgi: well, this seems better than what I have right now. Would you mind telling me what error you hit? |
| 10:34.41 | cosurgi | sth libtool related: libtool: link: cannot find the library `../../src/libfft/libfft.la' or unhandled argument `../../src/libfft/libfft.la', /home/janek/20-Programowanie/10-cpp/51-Brlcad/brlcad/src/conv/.libs/lt-asc2g: error while loading shared libraries: librt.so.19: cannot open shared object file: No such file or directory |
| 10:36.35 | nmh_2Grajw | consurgi: I will just hope to not see that myself. Thanks for the suggestion |
| 10:37.18 | cosurgi | he said that libtool on debian is broken :) |
| 10:38.01 | cosurgi | now I must do some work, but later I will either install libtool from source, or hope that brlcad will fins a fix for this :) |
| 10:38.27 | nmh_2Grajw | consurgi: have fun with your work |
| 10:40.06 | cosurgi | heh, thanks :> |
| 11:20.23 | starseeker | Z80-Boy: The Ronja screenshot would be a good one to add to the Gallery: http://images.twibright.com/images/ronja/promotion/screenshots/brl-cad.png |
| 11:21.01 | clock_ | starseeker: yes |
| 11:22.57 | clock_ | cosurgi: and I cannot start mged now - it hangs :) |
| 11:39.33 | starseeker | clock_: What's the url for checking out from the ronja src tree? |
| 11:39.47 | clock_ | starseeker: not possible at the moment |
| 11:39.53 | starseeker | Ah |
| 11:39.55 | clock_ | Is there a way how to display SVN on HTTP? |
| 11:47.46 | *** join/#brlcad cad20 (n=63fd349a@bz.bzflag.bz) | |
| 12:39.48 | ``Erik | lock: http://en.wikipedia.org/wiki/Communication_with_submarines |
| 12:39.53 | ``Erik | s/^/c/ |
| 12:40.28 | ``Erik | /t lurk, damnit. |
| 13:11.57 | clock_ | oh yes |
| 13:12.13 | clock_ | pumping tons of AC into the earth so that submarines can catch a tiny fraction of that |
| 13:12.57 | clock_ | Sime ELF communication with submarines: if the submarines stop seeing the 50Hz and 60Hz signals, they know the end of civilization has just passed. |
| 13:13.01 | clock_ | Sime -> Simple |
| 13:46.39 | brlcad | added screenshot |
| 13:47.40 | clock_ | brlcad: my screenshot? |
| 13:47.53 | brlcad | yes |
| 13:47.59 | clock_ | jumps around |
| 13:49.24 | brlcad | http://brlcad.org/gallery/s/screenshots/ |
| 13:52.01 | clock_ | cassini, interesting |
| 13:52.41 | clock_ | did they actually use BRL-CAD on Hubble and Cassinin development? |
| 13:52.50 | brlcad | hubble, yes |
| 13:52.52 | brlcad | cassini, no |
| 13:53.06 | clock_ | Who did Hubble? ESA? NASA? |
| 13:53.19 | brlcad | hubble involved lots of people |
| 13:53.34 | brlcad | stsi, nasa |
| 13:53.38 | clock_ | what did they use BRL-CAD for? |
| 13:54.03 | clock_ | Cassini is European isn't it? |
| 13:54.28 | brlcad | visualizations, optics |
| 13:55.03 | clock_ | I guess they surely had some good pseudoreason why waste money on expensive commercial modeller, when USA have developed a free one from their state budget |
| 13:55.05 | brlcad | http://en.wikipedia.org/wiki/Cassini-Huygens says it was joint |
| 13:55.27 | brlcad | oh, they still spend on the expensive |
| 13:55.31 | clock_ | Designed in a THC epiphany? |
| 13:55.43 | clock_ | ("it was joint") |
| 13:56.18 | clock_ | So they actually raytraced the optics in the Hubble with BRL-CAD? |
| 13:56.56 | brlcad | we had nothing to do with that mirror ;) |
| 13:57.10 | clock_ | they used the wrong material code? |
| 13:57.17 | clock_ | assumed air instead of vacuum? |
| 13:57.36 | brlcad | no |
| 13:57.53 | brlcad | the first mirror in hubble had a flaw, out of focus due to manufacturing flaw |
| 13:57.59 | clock_ | oh for a mirror it doesn't actually matter |
| 13:58.19 | clock_ | they didn't try to focus a desk lamp on a wall first before sending it into the sky? |
| 13:58.31 | clock_ | that's how I asses lens quality |
| 13:59.21 | brlcad | "they" involved thousands of people, dozens of major companies, several governments |
| 13:59.34 | clock_ | oh companies and government, they always screw things up |
| 13:59.43 | clock_ | if it were made by enthusiasts in garage, it would surely work fine |
| 14:00.45 | brlcad | I'd already added a ronja rendering to the renderings section: http://brlcad.org/gallery/s/renderings/ |
| 14:02.31 | clock_ | jumps around |
| 14:02.38 | clock_ | the hubble looks a bit like an oversize Ronja |
| 14:02.45 | clock_ | I am sure it would have a good range |
| 14:03.07 | clock_ | One night I |
| 14:03.22 | clock_ | ll climb up the sky with a ladder and unscrew secretly their mirror and put it into my Ronja |
| 14:04.26 | clock_ | We should beam the Ronja instructions into the deep space |
| 14:04.45 | clock_ | then all aliens will build it to download pr0n and we can easily detect them as planets shining in red |
| 14:06.56 | clock_ | brlcad: can BRL-CAD simulate also Fresnel lens if I use the equation to produce lots of truncated general cones? |
| 14:18.39 | brlcad | no reason why not, but wouldn't use tgcs |
| 14:19.03 | clock_ | which ones then to make those grooves? |
| 14:19.22 | ``Erik | <-- flips css off |
| 14:32.58 | *** join/#brlcad cad66 (n=6161d635@bz.bzflag.bz) | |
| 14:33.14 | brlcad | hello cad66 |
| 14:33.19 | brlcad | bye bye cad66 |
| 14:34.45 | brlcad | clock_: a fresnel lens is just a bunch of layered paraloids with sections cut away |
| 14:35.12 | clock_ | brlcad: should I model it like that? |
| 14:35.17 | brlcad | so i'd think you'd want to use layers of ELLs or EPAs |
| 14:35.47 | clock_ | brlcad: do you think they use paraboloids when they manufacture? Or cones? |
| 14:36.00 | clock_ | An ideal Fresnel should have paraboloids of course |
| 14:36.10 | brlcad | I think they just mill it out of a single sheet of glass |
| 14:37.43 | brlcad | getting the prism cuts is the hard part |
| 14:38.08 | brlcad | at least for a lens |
| 14:51.15 | cosurgi | I asked debian dev about licese problems, summarizing: brlcad licenses BSD+LGPL+public_comain indeed conflicts with GPL used in QT. |
| 14:51.26 | clock_ | AFAIK they make a precision mold and they mold it from PMMA. |
| 14:51.50 | clock_ | cosurgi: do they really? |
| 14:51.51 | cosurgi | brlcad: but, it is OK to communicate through a socket between a GPLed GUI frontend and brlcad. |
| 14:54.17 | cosurgi | his reasoning is that GUI frontend is just a client code and it can talk with anything which uses a certain socket communication API. It doesn't matter that it talks with brlcad. Also it's not a problem that the API is defined by brlcad, because BSD license is less restrictive. |
| 14:54.31 | cosurgi | if it was other way round: brlcad using QT - it's a conflict. |
| 14:54.51 | cosurgi | But the frontend GUI (eg. QT) is using brlcad - no conflict |
| 14:55.11 | cosurgi | so we have here a conflict only in one direction. |
| 14:55.46 | cosurgi | which is good for us :) |
| 14:56.59 | cosurgi | So we would have licence conflict if such GUI frontend used direct library calls. But if we talk through socket we are safe. |
| 14:57.15 | cosurgi | I was relieved to hear that :) |
| 14:58.39 | cosurgi | anyway, I'm on my regular work now. I'll get back to compiling brlcad in the evening. |
| 15:03.52 | brlcad | cosurgi: they dont' conflict in the sense that it would just require that the front-end be gpl :) |
| 15:04.11 | brlcad | ah, which you later say |
| 15:04.18 | brlcad | should read all before commenting |
| 15:05.01 | brlcad | fwiw, the sources are almost all LGPL, not BSD -- the BSD portions are the build system, documentation, and data files |
| 15:06.00 | brlcad | ahh, and it looks like laptop lid has closed, or ip changed |
| 15:06.52 | brlcad | I was able to reproduce the failure as expected with no problem (and was just now going to dig into a workaround) :) |
| 15:13.11 | cosurgi | brlcad: yeah my IP changed. sorry for that |
| 15:13.28 | cosurgi | brlcad: did you use screen? do you have my current IP ? |
| 15:13.50 | cosurgi | 77.253.129.137 |
| 15:14.21 | cosurgi | brlcad: I have no influence on my ISP behaviour. That's why I'm using dyndns |
| 15:15.30 | cosurgi | brlcad: if you find a workaround I will be grateful :) |
| 15:18.40 | cosurgi | absurd.kozicki.pl or absurd.homelinux.net should work for you. |
| 15:22.56 | brlcad | ah, cool |
| 15:23.23 | brlcad | i'm in |
| 15:24.15 | cosurgi | great. but I must leave now for aprox 1h. Get kid from kindergarten :-)~ |
| 15:24.35 | cosurgi | see you later :) |
| 15:26.01 | brlcad | cya! |
| 15:27.24 | brlcad | ahh, running screen inside of screen is .. challenging |
| 15:58.24 | ``Erik | nah it ain't |
| 15:58.29 | ``Erik | I do it all the time :D |
| 15:59.39 | ``Erik | http://pastebin.bzflag.bz/d790904dd <-- makes it trivial |
| 16:02.00 | brlcad | how do you handle things like ctrl-a a, ctrl-a e |
| 16:02.24 | brlcad | ctrl-a d even |
| 16:03.42 | ``Erik | ^a a d |
| 16:10.53 | *** join/#brlcad docelic (n=docelic@78.134.200.134) | |
| 16:40.01 | cosurgi | brlcad: ok, I'm back |
| 16:40.59 | cosurgi | if there's anything I can help with, shoot. |
| 16:42.08 | brlcad | cosurgi: no problems, I'm just (slowly) debugging libtool while context-switching with 4 other things |
| 16:42.14 | brlcad | making progress little by little though :) |
| 16:42.23 | cosurgi | sure, I'm very grateful |
| 16:43.12 | cosurgi | making my mind on how great a CAD GUI inside yade would be :) Of course first written as a separate app, but easily embeddable inside yade. |
| 16:43.37 | brlcad | heh |
| 16:49.42 | *** join/#brlcad rcampos (n=rafaelca@micro85.comp.ufscar.br) | |
| 16:52.28 | brlcad | hello rcampos |
| 16:53.19 | rcampos | Hi there. |
| 17:05.26 | *** join/#brlcad rcampos (n=rafaelca@200.18.97.86) | |
| 17:05.36 | rcampos | hi there. |
| 17:06.25 | brlcad | heh, hello again |
| 17:06.56 | brlcad | rcampos: did you have a question, or just stopping by to check out the hot babes? :) |
| 17:07.27 | alex_joni | brlcad: forgot rendered in the description :) |
| 17:08.04 | rcampos | I do have a question... just ran into some tech problems |
| 17:08.09 | rcampos | oops... again... |
| 17:08.24 | rcampos | very sorry! i'll try reconnecting |
| 17:09.40 | brlcad | alex_joni: what description? |
| 17:11.54 | alex_joni | hot babes.. |
| 17:15.38 | *** join/#brlcad rcampos (n=rafaelca@200.18.97.102) | |
| 17:16.32 | rcampos | here i am again. everything should be fine this time. |
| 17:17.32 | cosurgi | brlcad: do you have dimiensioning lines in brlcad? |
| 17:18.28 | brlcad | nope |
| 17:19.16 | brlcad | nothing fundamentally limiting there, though -- mostly just the user interface |
| 17:20.07 | cosurgi | brlcad: what's the difficulty in adding them, say.. on scale 0..10 ? |
| 17:20.10 | *** join/#brlcad micges (n=michu@public-gprs39842.centertel.pl) | |
| 17:20.27 | brlcad | it really depends how and where they are added |
| 17:20.46 | *** part/#brlcad micges (n=michu@public-gprs39842.centertel.pl) | |
| 17:21.23 | brlcad | also referring to dimensions on a 2D or 3D view (or both)? |
| 17:21.28 | cosurgi | on which side they should be added: brlcad, or GUI frontend (unlikely for me?) |
| 17:22.57 | brlcad | well, on the gui if you just want them as a visual means for inputting values .. that's almost entirely just a UI detail to pull out lines, see the dimensions, type in some new values |
| 17:23.08 | brlcad | but then to make them persist, you have to have some sort of annotation support |
| 17:23.24 | brlcad | (hence the gsoc item for annotations) .. and that goes into the backend with the geometry |
| 17:24.01 | brlcad | on the backend, I think adding those would actually be pretty darn easy |
| 17:24.01 | cosurgi | both, I think. A dimension line should have at least: two points and a Local Coordinate System. The dimension line goes along X axis of local coordinate system from point A to point B. It also can have: a third point - the dimension line goes through this point. And other less important attributes - like font, font size, etc. |
| 17:24.27 | brlcad | probably a couple weeks at best to add the support on the backend as it's *really* similar to the way sketches are implemented now |
| 17:25.03 | cosurgi | ok, so good to hear this. |
| 17:25.42 | brlcad | the UI side of things would be more work to do anything more than display them -- add a couple more weeks if you wanted to click and drag on them, type in values |
| 17:26.10 | brlcad | that is really where it's probably better to just plan for it in the new UI |
| 17:26.55 | cosurgi | ok. Don't worry about UI now :-) I'm asking to make sure that there's place for them inside brlcad. |
| 17:27.19 | brlcad | yeah |
| 17:27.31 | brlcad | technically you can do it now through a really horrible hack |
| 17:28.04 | cosurgi | I cannot tell how far I will go (impossible) with frontend. But my motivation is to have an AutoCAD replacement on linux. So my focus is mostly on replicating AutoCAD's interface. |
| 17:28.40 | brlcad | one of the users actually implemented annotations as 3D geometry, completely tcl scripted hack but end result was pretty impressive |
| 17:28.54 | cosurgi | ok :) |
| 17:29.07 | brlcad | it would auto-label all objects being displayed |
| 17:29.32 | cosurgi | Also I don't know how slow I can be due to my time limitations. If anyone springs up in GSoC to do that - it would be great for me :) |
| 17:29.34 | brlcad | the way he did it though is the sort that makes you snort your milk if you're not expecting it :) |
| 17:29.46 | cosurgi | heheh :) |
| 17:30.17 | brlcad | he actually creates the fonts as 3D geometry, positioning them in the scene according to the given view |
| 17:30.20 | *** join/#brlcad louipc (n=louipc@bas8-toronto63-1096669639.dsl.bell.ca) | |
| 17:30.43 | brlcad | so you have all these tiny little polygons spelling out names and labels with lines leading from the labels to the objects |
| 17:30.46 | brlcad | pretty funny |
| 17:30.53 | cosurgi | I see :-) |
| 17:31.16 | brlcad | all done as an mged tcl procedure too ... which is just bizzare :) |
| 17:32.02 | brlcad | no worries about the time, i'm a very patient guy -- some things in open source happen *very* slowly :) |
| 17:33.19 | rcampos | this is a great window for my question. i'm interested in proposing the GI renderer development for the GSoC, but I'm not that confident about the timeframe. |
| 17:33.59 | rcampos | Path tracing is quite tricky, so is a student expected to deliver it finished by the end of the term? |
| 17:36.19 | brlcad | rcampos: so long as the proposal is detailed in what you're proposing and what you see the deliverables as being, we're completely satisfied |
| 17:37.28 | brlcad | whether you can finish what you propose is based on your background and experience -- so you could conceivably be a first year student with minimal programming experience, so the timeline should naturally reflect that in how far they expect to get |
| 17:38.18 | brlcad | the focus is more on attracting new long-term developers, so there's really no problem if you don't "finish", but a really strong desire that you keep working on it and expand your interests into other parts of brl-cad development |
| 17:39.50 | brlcad | so yeah, don't promise the impossible, be conservative .. and show that you're interested in becoming a long-term contributor ;) |
| 17:41.04 | rcampos | I see. I'm very interested in rendering, so working with brl-cad in the long-run, specifically on the librt library and the GI module to be added would be great. |
| 17:42.39 | brlcad | sounds great to me too ;) |
| 17:42.57 | brlcad | I've wanted to implement that myself for years |
| 17:43.16 | rcampos | this might sound naive, but there are options in developing a GI renderer, such as the most adequate algorithm (e.g. metro or radiosity). |
| 17:43.37 | rcampos | Should major choices be made prior to project submission? |
| 17:44.05 | ``Erik | could be part of the proposal, would help focus the timeline |
| 17:44.05 | brlcad | yeah, just to show competency, you need to pick one or the other since they're such different approaches |
| 17:44.42 | brlcad | radiosity usually performs better, but is usually much harder to implement (especially with implicit geometry) |
| 17:44.56 | ``Erik | hides the path tracer already in there O.o O:-) |
| 17:45.14 | brlcad | MLT is basically bidirectional path-tracing, a much simpler algorithm, generates gorgeous picutres, but run-times are really long |
| 17:45.45 | rcampos | Maybe some research on performance tuning the MLT. |
| 17:46.04 | brlcad | ``Erik: if it were easy to use, a librt one wouldn't have made the list ;) |
| 17:46.16 | brlcad | if it didn't require facetization, a librt one wouldn't have made the list |
| 17:46.32 | ``Erik | true, that's an ugly requirement |
| 17:47.00 | ``Erik | but without a serious CSG optimizer... |
| 17:47.22 | brlcad | if someone can get one working with librt, and tie is hooked into librt for bots, then it could be a reasonable balance |
| 17:48.15 | brlcad | it's not like rise was actually fast .. and at the order it took, 2-10x longer really doesn't matter -- you're still going distributed and running for days, so throw 2-10x hardware at it |
| 17:48.27 | ``Erik | but so many models are built in a fashion that "seems right" to a modeler who doesn't know the actual cost when it comes to intersecting and weaving |
| 17:50.54 | rcampos | the CSG optimizer seems to be more important at the moment than the GI renderer. am i wrong? |
| 17:51.09 | brlcad | rcampos: nah, they're both "important" |
| 17:51.14 | ``Erik | they're just different |
| 17:51.25 | ``Erik | make it fast or make it pretty... |
| 17:51.42 | brlcad | you really need to know your stuff to work on the CSG optimizer though -- that's already highly tuned code, and changes would have to be very careful |
| 17:52.02 | brlcad | propose both ;) |
| 17:52.10 | brlcad | just not in the same proposal |
| 17:52.24 | rcampos | hum, i'm not sure. not really familiar with CSG. i am with RT, though. |
| 17:52.28 | brlcad | and only after doing your preferred first with detail |
| 17:52.45 | brlcad | csg is easy to learn |
| 17:52.58 | brlcad | the optimization algorithms for csg can be tricky though |
| 17:53.07 | ``Erik | it's just the geometric version of boolean logic |
| 17:53.09 | brlcad | and optimized csg *code* can be even trickier |
| 17:53.19 | ``Erik | "just" :D |
| 17:53.28 | ``Erik | like, academic "trivial" |
| 17:54.06 | rcampos | oh, i'm used to that! |
| 17:54.47 | ``Erik | dangit, ucw seems to have been mostly abandoned :/ almost 2 years ago |
| 17:55.32 | rcampos | brlcad: i'm very into RT, but my background supports me better on the GI front. only one well-done proposal. what do you suggest? |
| 17:56.01 | cosurgi | brlcad: is it possible to do .dxf export ? |
| 17:56.16 | brlcad | cosurgi: g-dxf |
| 17:56.37 | cosurgi | oh, so it works right now? great! :) |
| 17:56.41 | brlcad | that will polygonalize the model since dxf doesn't support implicits |
| 17:57.11 | brlcad | rcampos: i'm not sure what you're asking |
| 17:57.48 | cosurgi | yeah, I know. I asked because that would allow people to work in brlcad GUI frontend, then sell their drawings to people who use autocad. |
| 17:57.54 | brlcad | rcampos: only propose the csg optimization task if you think you could do something there -- all the items on the task list are rather important and would be great to have .. |
| 17:58.14 | rcampos | brlcad: I really like the CSG optimizer idea, but I'm afraid it might be overwhelming. |
| 17:59.06 | brlcad | I could give you some references to research papers for it probably in a day or two to give you a better feel, or better yet just search for CSG optimization papers |
| 17:59.24 | brlcad | otherwise, focus on the GI submission first |
| 17:59.46 | brlcad | you can always stub in an application for CSG and work on it or decide through next week |
| 18:00.02 | brlcad | there's a week and a half to review and expand upon the submissions after the 31st |
| 18:00.17 | brlcad | but you have to submit *something* by the 31st or you're cut out |
| 18:00.42 | brlcad | mentors will start reviewing and refining next week with the candidates |
| 18:00.57 | rcampos | brlcad: i'll have a rather detailed proposal on GI by tomorrow evening - should I submit it and alter if needed, or put it up for comments beforehand? |
| 18:01.15 | brlcad | rcampos: either way works just fine |
| 18:01.40 | brlcad | if you want, you can share it to the devel mailing list to get more eyes on it beforehand |
| 18:01.56 | rcampos | great. i'll definitely do that :) |
| 18:03.24 | brlcad | sooner would definitely be better than later, if last year is any indication it's just going to get busier and busier until the deadline |
| 18:04.41 | rcampos | yes, and that's not good. i'll have it on the mailing list by tomorrow - it'll give me enough time to review. |
| 18:04.59 | brlcad | sounds like a plan |
| 18:05.15 | brlcad | welcome to toss it up into a page on the wiki if you like |
| 18:06.16 | rcampos | another great idea. is doing both redundant?! maybe having a link on the email to the list should suffice, then. |
| 18:08.16 | brlcad | nah, it's not redundant |
| 18:08.27 | brlcad | some folks won't click a link, some will -- its up to you |
| 18:08.48 | rcampos | noted! |
| 18:09.36 | rcampos | thanks a lot for the help so far :) |
| 18:09.41 | ``Erik | but if you attach a .doc or .ppt, we'll take turns smackin' ya :D |
| 18:10.57 | rcampos | hehe, i would'nt do that, rest assured! |
| 18:22.37 | cosurgi | whoopps, my backup box just got oops |
| 18:23.56 | cosurgi | brlcad: did you get a message on screen, too? :) |
| 18:28.55 | brlcad | cosurgi: yup |
| 18:31.47 | cosurgi | It's 7y old. I figured I still can use it for something. I've put into it raid5 3*500GB=1000GB and makes backups twice per day from cron with rsnapshot. I oopses when CPU overheats, so I'm uninstalling distcc from it, now :) |
| 18:32.08 | cosurgi | s/I oopses/It oopses/ |
| 18:35.39 | ``Erik | hrm, I think my primary server at home is 9 years old |
| 18:36.06 | ``Erik | need to buy a fan for the 8 yr old one O.o |
| 18:37.40 | cosurgi | heh, I gave my box which would deserve such title, to my friend. (the debian dev I was talking about earlier). It's his primary server now and is is 9y old. Still kicks butt :) |
| 18:38.27 | cosurgi | dual PIII, 600MHz |
| 18:39.15 | ``Erik | I had a dual p133 with scsi long ago, screamed for compiling (sucked for raytracing) |
| 18:41.34 | cosurgi | :^) |
| 18:43.14 | ``Erik | make a nice coffee table, too |
| 18:43.41 | yukonbob | waves in |
| 18:44.26 | ``Erik | looked like http://members.jcom.home.ne.jp/labyrin/maniki/maniiki01/dsc0079.jpg but far more abused |
| 18:49.17 | MinuteElectron | my home server is 98! |
| 18:57.38 | ``Erik | from '98, or running 98? O.o :> *duck* *run* |
| 18:58.40 | yukonbob | or from 1910? (a difference engine?) |
| 18:58.55 | brlcad | has fun showing the tcl guys why their regex imlementation sucks |
| 18:59.17 | ``Erik | heh |
| 18:59.17 | ``Erik | uhm |
| 18:59.24 | yukonbob | regex was just reworked w/ some help from postres, apparently |
| 18:59.26 | brlcad | interesting little test pattern.. ((a{%d}){%d}){%d} ... |
| 18:59.37 | yukonbob | *postgres |
| 18:59.45 | ``Erik | I saw a report where tcl's regex smoked perls a while back |
| 19:00.08 | brlcad | with that little pattern, you end up blowing the stack quickly the way tcl implements it |
| 19:00.16 | brlcad | or running out of memory |
| 19:00.23 | yukonbob | I wish there was an emitcon for my face for that comment, ``Erik |
| 19:00.25 | ``Erik | (by pushing perl into a degredate case where it basically had to backtrack every single time) |
| 19:00.57 | brlcad | yukonbob: try this in your tclsh: set i 40 |
| 19:00.59 | brlcad | then: regexp [subst ((a{$i}){$i}){$i}] x |
| 19:01.02 | yukonbob | ``Erik: was this a single weird case, or a common pattern |
| 19:01.10 | brlcad | it should dies somewhere between 20 and 40 |
| 19:01.33 | brlcad | I thought that was pretty odd .. so I wrote a little test program |
| 19:01.33 | yukonbob | brlcad: what version of tcl are you running? |
| 19:01.42 | brlcad | 8.5 latest |
| 19:01.45 | yukonbob | nods |
| 19:01.57 | ``Erik | it was a common case pushed to the extreme |
| 19:02.00 | brlcad | like I said, this was debugging with the tcl guys just a lil while ago |
| 19:02.05 | ``Erik | something like a 4k character pattern |
| 19:02.13 | yukonbob | ``Erik: interesting -- I honestly wouldn't have expected that... |
| 19:02.27 | ``Erik | 'cept perl got too slow to wait for at 10 characters or something |
| 19:02.33 | brlcad | quite a drastic difference -- gnu's is an utter pig (takes minutes just to compile) but seems to not run out of memory here until somewhere 200+ -- bsd seems to just hit a hard-coded uchar limit but is nearly instant even at 255 -- tcl's dies around 36 here |
| 19:02.59 | brlcad | you can see the diff between bsd's and gnu's right off the bat if you compile http://brlcad.org/tmp/regtest.c against both libs |
| 19:03.02 | yukonbob | tcl regex == Henry Spencer |
| 19:03.25 | brlcad | spencer did the old bsd one too |
| 19:03.43 | brlcad | so maybe it's just been broken since :) |
| 19:03.48 | brlcad | s/broken/optimized/ |
| 19:04.06 | ``Erik | http://swtch.com/~rsc/regexp/regexp1.html |
| 19:05.21 | ``Erik | of course, you always have the jwz quote to go along with that... |
| 19:05.36 | brlcad | this must be a different sample pattern for ((a{%d}){%d}){%d} |
| 19:05.46 | brlcad | because tcl and gnu both are epic fail |
| 19:05.55 | ``Erik | Some people, when confronted with a problem, think "I know, I'll use regular expressions." Now they have two problems. --- jwz comp.lang.emacs |
| 19:06.09 | brlcad | tcl at least dies gracefully, gnu's starts to take down the machine and seems to inf loop |
| 19:06.29 | brlcad | bsd's doesn't seem to have any problem |
| 19:06.43 | ``Erik | mebbe try an old tcl (like 8.0?) to see if tcl screwed up their fugly cases to a slight boost in trivial cases? |
| 19:06.55 | ``Erik | up until 256 characters? :D |
| 19:07.06 | brlcad | interesting that you can see a distinct difference between the bsd on mac os x and the bsd we ship (about 1sec to compile 256 with osx, about .1 sec with old bsd) |
| 19:07.16 | brlcad | it's not characters |
| 19:07.18 | yukonbob | core-dumped on your test, br |
| 19:07.22 | yukonbob | *brlcad |
| 19:07.55 | brlcad | yukonbob: tclsh did? |
| 19:07.55 | yukonbob | yup |
| 19:07.55 | yukonbob | 8.4.16 |
| 19:07.55 | brlcad | heh, well at least 8.5 is better in that regard |
| 19:08.12 | brlcad | it should say "couldn't compile regular expression pattern: nfa has too many states" |
| 19:08.22 | ``Erik | um, which "old bsd"? there're a couple fbsd 8 boxes floating around |
| 19:08.34 | brlcad | src/other/libregex |
| 19:08.44 | ``Erik | tcl and bsd shouldn't be using nfa's |
| 19:09.10 | ``Erik | should be a big fugly dfa with a list of state pointers |
| 19:09.20 | ``Erik | iirc |
| 19:09.39 | ``Erik | tcl, anyways... I'm guessing about bsd |
| 19:10.09 | brlcad | like he said, 8.5's regex was modified |
| 19:10.13 | ``Erik | does the mac one want to work in unicode? or do you see a lot of system activity (microkernel abuse) |
| 19:11.02 | brlcad | wth are you talking about? :) .. you've not compiled that regtest.c /me thinks |
| 19:11.19 | ``Erik | no? |
| 19:11.24 | brlcad | it's nearly instant on bsd (both old and new) to compile the expression |
| 19:11.47 | brlcad | we're talking about just compiling the expression, running it on nothing |
| 19:12.08 | brlcad | the expression is very "brief", but expands nasty state-wise |
| 19:12.18 | yukonbob | my 8.4 dies SIGSEGV SIG_DFL |
| 19:12.35 | brlcad | e.g. try running this: egrep '((a{255}){255}){255}' |
| 19:12.46 | ``Erik | are you talking about the bsd impl on a mac, or the bsd impl on bsd? |
| 19:12.49 | brlcad | that will use gnu's regex lib .. should run out of memory |
| 19:13.05 | brlcad | they're not much different iirc |
| 19:13.17 | brlcad | but was referring to on a mac and our old bsd impl |
| 19:13.39 | ``Erik | ah |
| 19:13.51 | ``Erik | thought you were mixing os's, as well |
| 19:14.28 | *** join/#brlcad jdoliner (n=jdoliner@cpe-24-59-109-153.twcny.res.rr.com) | |
| 19:14.29 | ``Erik | huh, that stopped at 4 gigs for me |
| 19:14.46 | ``Erik | still burning cpu, but not eating any more memory |
| 19:14.51 | brlcad | there, just ran it on .bz too .. slow, but it completes to 255 with no problem before hitting the uchar limit |
| 19:16.17 | brlcad | i'm believe gnu's just get stuck in an inf loop, or it's just got some N^3'd time going on -- it's been compiling here for a half hour at 200 |
| 19:17.07 | brlcad | to run the test, just gcc regtest.c && ./a.out 200 |
| 19:17.45 | jdoliner | is anyone here a possible mentor of a project on CSG evaluation of BREP objects. |
| 19:17.53 | brlcad | hugs jdoliner |
| 19:18.05 | ``Erik | :o |
| 19:18.32 | brlcad | jdoliner: heh, yes -- there are several mentors here (and on the mailing list) |
| 19:18.46 | *** join/#brlcad rafaelcampos (n=rafaelca@200.18.97.175) | |
| 19:18.52 | jdoliner | great |
| 19:19.56 | brlcad | we do more communal mentoring, so just speak up -- you might have to hit up both irc and ml to get a response sometimes, but the mentor's role is mainly to make sure you're on track and for evaluating you |
| 19:20.10 | brlcad | jdoliner: what's your background? |
| 19:20.47 | jdoliner | Well I'm presently a mathematics major at The University of Chicago |
| 19:21.38 | jdoliner | I've also had formal education in both Scheme and C at Uchicago |
| 19:22.17 | jdoliner | And I spent the summer working at Fermi Lab where I had to do some coding in Fortran on a simulation project |
| 19:22.24 | yukonbob | reads backtrace on #tcl |
| 19:23.55 | jdoliner | Also I have experience in computer art so I've worked with both B-Rep and CSG before |
| 19:24.32 | brlcad | on a coding level or at an application level? |
| 19:25.13 | ``Erik | looks around |
| 19:25.31 | ``Erik | you can't escape us, brlcad, you're becoming surrounded by coders who like sexy curvy languages |
| 19:26.13 | brlcad | i was just going to say that I'll try not to hold the fortran and scheme against him :) |
| 19:26.40 | jdoliner | Largely on application level the closest I've gotten with them in coding is with Pov-ray |
| 19:26.41 | ``Erik | each has their benefit... scheme teaches you how to really program, and floortrash teaches you how to put up with bull |
| 19:26.45 | ``Erik | *cough* O:-) |
| 19:26.52 | brlcad | j/k, I actually do like lisp, never really gotten into scheme .. just fails on a practical level for me |
| 19:27.08 | brlcad | much like how I like smalltalk, but not practical :) |
| 19:27.18 | jdoliner | yeah the physicist are so outdated |
| 19:27.30 | ``Erik | I d'no, depends on the impl, unfortunately :( gauche is a nice pragmatic scheme |
| 19:27.50 | jdoliner | I think when the program had originally been written Fortran was new |
| 19:27.54 | ``Erik | hooks to gtk, opengl, various rdbms's, good unix extensions, etc |
| 19:28.01 | brlcad | jdoliner: hm, then how strong is your math background with respect to spline surfaces and manifold geometry? :-) |
| 19:28.10 | brlcad | doing CSG of BREP is a bit tricky |
| 19:28.27 | brlcad | i mean it actually shouldn't be more than probably a few hundred lines of code.. but it's dense stuff to sort out |
| 19:28.47 | ``Erik | classroom scheme would like C without libraries or functions, so a lot of people have a lame view of the language |
| 19:28.52 | ``Erik | would be like |
| 19:30.05 | ``Erik | shouldn've eaten today, is feelin' awful goofy |
| 19:30.15 | brlcad | jdoliner: see if you can do some searches for CSG evaluation of BREP papers -- that really is a hot topic feature, but there's no sugar coating that it's trivial -- you'll need to instill some confidence that you can implement a given algorithm for doing the coding needed |
| 19:30.56 | brlcad | ignores the language banter :P |
| 19:30.56 | jdoliner | K |
| 19:32.35 | ``Erik | brlcad: just a slew of nurbs in a complex csg (no other primitives), and evaluate it to the minimal 'union only' patch set? |
| 19:32.51 | brlcad | yes |
| 19:33.22 | brlcad | basically performing splinesurface-on-splinesurface trimmings per the csg operations |
| 19:33.51 | brlcad | hell, I bet we actually already implement a full-blown algorithm that would work just fine in the nmg sources |
| 19:33.55 | brlcad | but it's in the nmg sources... |
| 19:34.47 | ``Erik | 'would work just fine', what, after throwing it all away and starting over? :D |
| 19:35.07 | brlcad | jdoliner: another very similar but a hell of a lot more straight-forward is to propose implementing/updating all our existing primitives with a describe-me-as-a-brep function |
| 19:35.44 | brlcad | we have about 30 primitives that you'd have to work on, about a third of those are already "done" but using the wrong data structures |
| 19:35.47 | ``Erik | bot, dsp, and metaball might be challenging for that one |
| 19:36.11 | brlcad | point-sample, turn into patches |
| 19:36.32 | brlcad | bot and dsp should be trivial |
| 19:36.46 | brlcad | each bot face is a brep face |
| 19:36.53 | brlcad | it already *is* in brep form |
| 19:37.32 | ``Erik | but what it's supposed to be smoothed? like a tesselated sphere.... 1k nurbs when it should be 4? |
| 19:38.01 | brlcad | the meatball would be a little more tricky |
| 19:38.04 | ``Erik | <-- was thinking dsp and bot would need curve fitting |
| 19:38.16 | brlcad | no, bot is what it is -- those are the faces |
| 19:38.52 | brlcad | dsp would probably be five faces for the sides and bottom, and then at least one if not many sub-pieces for the surface |
| 19:39.17 | brlcad | dsp has two modes, one is flat-faced -- those would be direct to N flat-faces brep faces |
| 19:39.41 | brlcad | the smoothed version, though, would be harder -- "should" come across as just one patch with a helluva lot of control points |
| 19:40.19 | ``Erik | heh, yay for funny stuff on mailing lists |
| 19:40.20 | brlcad | undoubtedly have to localize, though, as you can have discontinuities in the dsp (zero'd regions) |
| 19:40.24 | ``Erik | > apache. Well tested, lots of features. |
| 19:40.24 | ``Erik | well, tested... :) |
| 19:49.25 | *** join/#brlcad Z80-Boy (n=clock@77-56-82-60.dclient.hispeed.ch) | |
| 20:06.22 | brlcad | kicks CIA-33 |
| 20:06.23 | CIA-33 | ow |
| 20:09.53 | ``Erik | annoying, the tclConfig.sh installed in the latest fbsd is broken :/ |
| 20:38.04 | *** join/#brlcad jgay (n=jgay@fsf/staff/jgay) | |
| 20:38.46 | jgay | brlcad when you get a chance, tell me what you think about http://endsoftpatents.org |
| 20:46.12 | ``Erik | heh, I originally read that as "ends of patents" O.o |
| 21:32.35 | ``Erik | wanders home, later kids |
| 22:05.14 | brlcad | jgay: in what regard? |
| 22:05.33 | brlcad | it reads like a political rant, but that's because it is |
| 22:05.53 | brlcad | I've seen that site before iirc, pretty sure |
| 22:15.20 | louipc | hehe do you think if brl-cad became a real competitor that some companies out there would try suing? |
| 22:15.22 | jgay | brlcad, cool. It's pretty new. It's supposed to be informative, kind of boring, and simple. |
| 22:17.39 | jgay | brlcad, I didn't know if anybody in this project had a position on software patents. I am helping to write an amicus brief for an important court case (In re Bilski). There are a lot of patents on CAD and related software. You are likely violating dozens if not hundreds of software patents. |
| 22:17.40 | brlcad | louipc: not likely |
| 22:18.35 | brlcad | oh, it's entirely possible, but then if they were to sue, they'd be suing uncle sam .. and uncle sam gets to decide if you're allowed to sue |
| 22:18.52 | louipc | yep that's what I was thinking |
| 22:18.59 | jgay | brlcad - not so. More likely they'd sue a company using the software or selling it inside of a distribution. |
| 22:19.16 | louipc | that might be precedent setting though |
| 22:19.52 | jgay | remember software has both copyright and patents, but they are very different laws with very different requirements |
| 22:20.12 | jgay | under copyright law you are safe, under patent law, uncle sam may be safe, but others might not be. |
| 22:20.51 | brlcad | jgay: as for a position, I'm fairly against software patents but they frankly aren't a day-to-day concern (or an issue that has *ever* really come up with major significance) .. at least to date of course |
| 22:20.51 | jgay | e.g., TrendMicro is being sued for how they are using ClamAV, a free software package that they did not develop. |
| 22:20.55 | louipc | depending on what a judge says |
| 22:21.42 | jgay | brlcad, right, you may not be personally effected, but others in the free software community (and generally programming community) are very much effected on a day to day basis. |
| 22:22.28 | brlcad | the legal concerns on our end tend to be much more heavily weighted towards issues of information security , the license terms, issues with the GPL, classification levels, FOUO data, and disclosure |
| 22:23.16 | jgay | brlcad, of course, I understand that. With software patents, however, you needn't ever know of the patents or technology you infringe upon to be found infringing upon the patents. |
| 22:23.30 | brlcad | i know others are affected, just following the headlines it's clear a lot of the problems brewing |
| 22:24.26 | yukonbob | BRL-CAD to plaintif -- let me introduce my legal team -- m1 bradley, blackhawk, and stryker. |
| 22:26.15 | jgay | so, you guys are safe, you can't infringe upon patents, because the government asks you to develop brl-cad |
| 22:26.39 | jgay | however, the government could be asked to pay damages or costs even when it claims eminent domain over the patents. |
| 22:27.43 | brlcad | do patents have to defend against prior art as well? |
| 22:27.49 | brlcad | (software patents) |
| 22:27.54 | jgay | my question really wasn't about whether or not you guys are personally effected, but whether or not anybody would be interested in signing onto the amicus brief. |
| 22:28.52 | jgay | brlcad, yes. But, there are about 70 software patents granted per day. |
| 22:29.07 | brlcad | nods |
| 22:29.36 | jgay | So, they are relatively safe because of their numbers. To look into prior art for even a fraction of those is a daunting task. |
| 22:30.07 | louipc | there are some pretty ridiculous patents and any judge with half a brain should just throw them out |
| 22:30.35 | jgay | louipc, I recommend reading over endsoftpatents.org -- we have done a good job of explaining a whole lot about software patents. |
| 22:31.12 | jgay | Anyhow, I was trying to an extend an offer that if anybody was interested in taking a stance against software patents, and hopes to see them eliminated, the time would be now. |
| 22:31.29 | brlcad | jgay: so you are involved directly with endsoftpatents.org |
| 22:31.41 | jgay | I didn't mean to flood the channel or force it down y'alls throats :-) |
| 22:32.15 | brlcad | I gathered that you weren't gloria or ben |
| 22:32.16 | jgay | brlcad, yes. I am the campaigns manager for the free software foundation. We helped launch the ESP campaign, which is a coalition effort. |
| 22:32.26 | brlcad | got it |
| 22:32.53 | jgay | It's not limited to free software or user freedom, so it's not an in house thing. But, when it comes to doing my part for the free software community, or campaigning like activities, I step up to the plate. |
| 22:33.22 | brlcad | quite noble of you |
| 22:33.30 | brlcad | that work takes a lot of time and effort |
| 22:33.38 | brlcad | and is rarely really appreciated directly |
| 22:34.01 | brlcad | (except by others that have also stepped up to the plate) |
| 22:34.06 | jgay | thanks -- I enjoy staying in the shadows most of the time :-) |
| 22:35.17 | louipc | you guys are lobbying for a patent reform bill? |
| 22:36.42 | jgay | louipc, I don't think we are lobbying, but we are recommending that congress consider reforming the patent system through legislation. |
| 22:37.02 | louipc | oh |
| 22:37.36 | jgay | louipc, 501(c)3 organizations are allowed to encourage bills to be passed and influence legislation, but they are not allowed to help political campaigns or put more than some small percentage of their budget each year toward efforts that are considered lobbying. |
| 22:38.42 | jgay | I think you might also have to justify such actions as being directly related to your mission. Clearly something that we consider to threaten the creation and adoption of free software directly threatens our mission. |
| 22:38.49 | louipc | ah that second part sucks |
| 22:39.23 | louipc | so you have to make a profit corporation to be allowed to lobby agressively hah |
| 22:39.39 | jgay | no, I think it's 501(c)4 |
| 22:40.12 | jgay | So, you don't get a tax break, but you can qualify for grants. |
| 22:41.25 | brlcad | jgay: has anyone worked on tracking down the various "patent whore" clearing house companies -- or is that akin to chasing down the storm botnet? |
| 22:41.54 | brlcad | to identify the list of companies that abuse the system the worse that file for the intention of extorting |
| 22:42.05 | brlcad | s/worse/worst/ |
| 22:42.10 | jgay | brlcad, it's just not worth it -- I mean, if in the legal community you are known as a troll -- how can it get any worse? |
| 22:42.45 | jgay | So, IBM & Microsoft hold a tremendous number of patents |
| 22:43.13 | jgay | But, some of the worst patent holders are law companies that exist for the sole purpose of suing people for patent infringement. |
| 22:43.28 | brlcad | right, it's the latter I'm referring to |
| 22:43.40 | brlcad | calls them "patent whores" |
| 22:43.43 | jgay | They collect software patents and sue big companies at critical times. In business it's referred to as "the patent tax," that you expect to pay for being big. |
| 22:44.07 | brlcad | because it's cheaper to pay than to fight |
| 22:44.21 | jgay | Right, and because there is a good chance you'll lose. |
| 22:44.33 | jgay | I mean, most everybody is infringing software patents. |
| 22:44.46 | jgay | Our list of people on the site are companies like the green bay packers, MLB.com, etc. |
| 22:45.34 | jgay | You don't have to develop software, you just have to use it to make money, or design a web site to work a certian way, etc. |
| 22:50.08 | louipc | yeah that's pretty dumb |
| 22:55.30 | starseeker | rather doubts the patent mess will get straightened out anytime soon - the money and financial incentive to the legal establishment is lined up behind making it more difficult to do anything legally, not less... |
| 22:56.07 | louipc | damned lawyers |
| 22:57.09 | starseeker | wonders how many examples in human history there are of legal establishments working to make themselves less of a nuisance |
| 22:58.02 | starseeker | Ah, well. |
| 23:00.45 | starseeker | jgay: Out of curiosity, are there websites that attempt to make a concrete case FOR software patents? |
| 23:00.55 | jgay | starseeker -- not that I know of. |
| 23:01.54 | jgay | starseeker I've actually tried to find some pro-software patent arguments that were based on facts ... but the only ones I could find rae based on generalization, like, "Patents are important for innovation." |
| 23:02.18 | jgay | All of the business case studies I've read have not found any link between innovation (even for a single case) and software patents. |
| 23:02.45 | jgay | but, I'm not as well versed as some. So i don't claim to be a definitive source of all things software patents. |
| 23:02.48 | starseeker | always comes back to that "promote the progress of science and useful arts" bit... |
| 23:03.18 | starseeker | If we're going to stop people from doing constructive work, I'd like to be sure we are getting a better return in the end. |
| 23:04.04 | starseeker | wonders if anyone making laws thinks of it in that fashion any longer... |
| 23:04.08 | jgay | starseeker - In re Bilski could be very important in helping to decide the fate of software patents. Remember, software is 1) the only thing that can have both a copyright and a patent, and 2) only started being granted in the mid-1980s. |
| 23:05.37 | jgay | starseeker people don't read bills (readthebill.org), but a lot of us (sunlightfoundation.com, www.leagueoftechvoters.org/, change-congress.org) are doing something to help improve government. |
| 23:06.10 | jgay | *granted patents |
| 23:07.02 | jgay | sorry, I wear multiple hats :-) I should leave the room. The purity of hacking and BRL-CAD is being tainted! :-) |
| 23:07.17 | starseeker | nah, my fault |
| 23:07.41 | jgay | o well, if it ever bothers anybody let me know -- I don't like being "that guy." |
| 23:07.49 | brlcad | jgay: it's good discourse |
| 23:08.31 | brlcad | all for increased awareness |
| 23:08.58 | jgay | cool, well, on to more pressing matters. If I don't get home and make dinner right now my fiance will kill me! |
| 23:09.00 | brlcad | though I'm also a champion of being responsive, not preventive .. the ones responding directly do this should be the ones being attacked (i.e. the big corporations that are being sued) |
| 23:09.27 | jgay | brlcad, yeah, most of the people we have reached out to are companies that have or are being sued. |
| 23:09.48 | jgay | actually, I've spent two days being passed around coprporate switchboards trying to get connected to the well-isolated legal department |
| 23:10.06 | brlcad | that's why I was wondering if having a list of the orgs could help .. formulate some sort of reverse class-action lawsuit |
| 23:10.44 | jgay | brlcad, yeah, as a long term strategy something like that might make sense. But, this Bilski case is actually one of the most important cases we've had re: software patents in a very long time. It could be decisive. |
| 23:11.49 | jgay | So, our amicus brief is important, because the ruling, and the exact wording of it, is very crucial to many parties, and could have a big impact on the state of software patents. |
| 23:12.29 | brlcad | well, hopefully it goes well -- when is it filed? |
| 23:12.41 | jgay | On the ninth |
| 23:12.50 | jgay | hopefully |
| 23:18.54 | jgay | ok, goodnight all! must go make dinner :-) |
| 23:50.15 | *** join/#brlcad Elperion (n=Bary@p54876D23.dip.t-dialin.net) | |
| 23:58.49 | *** join/#brlcad hippieindamakin8 (n=hippiein@203.200.95.130) | |
