| 00:12.54 | *** join/#brlcad Nohla (~Nohla@201.255.215.187) | |
| 00:20.07 | starseeker | eyes bus routes... 5 bucks for a day, vs. $20+ one way in a cab... hrm |
| 00:20.30 | starseeker | anybody know what busses are like in LA these days? |
| 01:08.58 | ``Erik | mmmmm, monster rueban |
| 01:10.10 | ``Erik | given that you'll be reimbursed, the $15 extra should be irrelevant, no? and tha tmuch more peace of mind not having to sit with bums and having ot walk between bus stops and where ya need to get with luggage and stuff, no? |
| 01:10.35 | ``Erik | unless ya'll have someone there who's renting a car or a local friend *shrug* |
| 02:39.10 | *** join/#brlcad CIA-45 (~CIA@208.69.182.149) | |
| 02:47.43 | *** join/#brlcad roberthl (~robert@mediawiki/RobertL) | |
| 02:57.05 | *** join/#brlcad Stattrav (~Stattrav@117.192.131.155) | |
| 03:37.27 | starseeker | ``Erik: doubt I'd be reimbursed for that one :-/ |
| 04:55.13 | *** join/#brlcad R0b0t1 (~Enigma@unaffiliated/r0b0t1) | |
| 08:55.15 | *** join/#brlcad Alexandrus (~nil@p4FE3FB19.dip.t-dialin.net) | |
| 09:38.24 | Alexandrus | http://kccm.dnsalias.org:8080/p2.png |
| 09:38.42 | Alexandrus | the triangle helix is correctly displayed as grid |
| 09:38.56 | Alexandrus | but when i raytrace it, half is missing? |
| 09:39.06 | Alexandrus | what could be the reason for this? |
| 09:43.09 | Alexandrus | (maximum object count?) |
| 09:44.09 | Alexandrus | (it can get even wierder with missing arb6 somewhere in the middle) |
| 09:44.57 | Alexandrus | http://kccm.dnsalias.org:8080/p3.png |
| 09:45.40 | Alexandrus | (same with rtedge) |
| 09:48.06 | Alexandrus | http://kccm.dnsalias.org:8080/p4.png (one detailed winding) |
| 10:12.57 | Alexandrus | for some funny reason it works if i make a region out of it |
| 10:18.14 | *** join/#brlcad Stattrav (~Stattrav@2403:0:500:1:218:deff:fe54:bd88) | |
| 10:26.46 | Alexandrus | http://kccm.dnsalias.org:8080/p5.png (region) |
| 14:09.06 | CIA-45 | BRL-CAD: 03brlcad * r39884 10/brlcad/trunk/regress/repository.sh: look for configure.ac in TOPSRC |
| 14:09.37 | CIA-45 | BRL-CAD: 03brlcad * r39885 10/brlcad/trunk/src/other/libpng/Makefile.am: per-target cppflags aren't kosher yet |
| 14:10.05 | CIA-45 | BRL-CAD: 03brlcad * r39886 10/brlcad/trunk/src/librt/binunif/binunif.c: clarity/cleanup |
| 14:34.20 | ``Erik | hm, "limbo" looks like an interesting game O.o |
| 14:37.40 | ``Erik | heh http://infoworld.com/d/developer-world/google-executive-frustrated-java-c-complexity-375 |
| 15:43.24 | *** join/#brlcad mafm (~mafm@60.Red-80-26-128.dynamicIP.rima-tde.net) | |
| 16:14.51 | *** join/#brlcad Stattrav (~Stattrav@2403:0:500:1:218:deff:fe54:bd88) | |
| 16:36.50 | *** join/#brlcad Nohla (~Nohla@201.255.215.187) | |
| 17:05.40 | *** join/#brlcad willdye (~willdye@fern.dsndata.com) | |
| 17:10.40 | *** join/#brlcad Stattrav (~Stattrav@2403:0:500:1:218:deff:fe54:bd88) | |
| 17:15.06 | *** join/#brlcad Alexandrus (~nil@p4FE3FB19.dip.t-dialin.net) | |
| 17:15.12 | Alexandrus | moin(); |
| 17:24.46 | ``Erik | yargh |
| 17:24.59 | ``Erik | the coil images you posted, were those using ARB8 or BoT? |
| 17:25.05 | Alexandrus | ARB6 |
| 17:25.21 | ``Erik | ah, ok, then I have no guesses at why they wouldn't display |
| 17:25.31 | Alexandrus | i can reproduce it... |
| 17:25.40 | Alexandrus | they display if i order them with r |
| 17:25.41 | ``Erik | (ARB6 is stored internally as ARB8 with 3 points at the same place) |
| 17:25.58 | Alexandrus | ok, i didn't analyze that |
| 17:26.16 | ``Erik | do you have a way to select one of the ARB's that don't display and try raytracing that alone? |
| 17:26.18 | Alexandrus | for the moment, the helix algorithm can accept any "square" form |
| 17:26.29 | Alexandrus | i could give you the script |
| 17:26.40 | Alexandrus | if i pick fewer, they display |
| 17:26.51 | Alexandrus | if i pick more, some are omitted |
| 17:27.25 | ``Erik | hm, would you mind doing a bug report and attaching the script to it, so we have a way of tracking and managing the issue? |
| 17:27.56 | Alexandrus | not a problem |
| 17:28.02 | ``Erik | okie, thanks :) |
| 17:28.39 | Alexandrus | i have to turn back a little |
| 17:30.05 | ``Erik | collapsing the two outside pairs to amke a wedge? |
| 17:30.48 | Alexandrus | no, i have to give you simple primitives again |
| 17:30.55 | Alexandrus | if i put them in a large region, it works |
| 17:31.11 | Alexandrus | now i have to reproduce, if i get it back |
| 17:32.33 | Alexandrus | i got it again |
| 17:32.47 | ``Erik | ok, if not, just state as much as you can... we have a (paid, relatively jr but capable) developer who is between projects right now and looking for something to do, I'm thinking I might try to tric^Wask him if he's interested in looking into the issue :D |
| 17:33.02 | Alexandrus | i have lots for him:P |
| 17:33.18 | Alexandrus | for example why r someregion.r u c$i |
| 17:33.30 | Alexandrus | gets slow with i = 0..20000 |
| 17:33.35 | Alexandrus | (really damn slow) |
| 17:33.49 | ``Erik | he doesn't use irc, so I'm thinking the bug tracker will not only give him a concrete thing to work on, but pull him more into the open source mentality |
| 17:34.12 | Alexandrus | i make a zip with explanation, platform, how to reproduce |
| 17:34.15 | Alexandrus | he will know:) |
| 17:34.20 | Alexandrus | even a pic will be included:) |
| 17:34.32 | ``Erik | heh, probably because tcl is constructing 20000 linked lists to assemble the final command, then parsing it all |
| 17:34.58 | Alexandrus | r is a brl-cad command, isn't it? |
| 17:35.02 | ``Erik | ok, if I remember on monday, I will ask him to see if he'd be willing to |
| 17:35.07 | Alexandrus | and i am giving one extra primitve at a time |
| 17:35.13 | Alexandrus | (not clever..maybe, but still wondering) |
| 17:35.15 | ``Erik | yes, but the list fed to it, the for stuff and the $i, that's tcl |
| 17:35.21 | Alexandrus | (i am very new to tcl...and kind of hat it:P) |
| 17:35.29 | ``Erik | all BRL-CAD does is extend the tcl shell for the mged command promp |
| 17:35.31 | ``Erik | prompt |
| 17:35.42 | ``Erik | r is just a tcl proc |
| 17:35.44 | Alexandrus | $i is a single string |
| 17:35.59 | Alexandrus | usually a name like "c$i" |
| 17:36.24 | ``Erik | hm, so you're running r 20000 times? |
| 17:36.30 | Alexandrus | yes |
| 17:36.42 | ``Erik | ah, hm, every time you run r, it tries to build an optimized tree |
| 17:36.48 | Alexandrus | i didn't know how split a long string into several parameters |
| 17:37.07 | Alexandrus | so it was kind of a temporary solution |
| 17:37.14 | ``Erik | it may be faster to build a big string of "u c$i"'s, then feed it to r once |
| 17:37.15 | Alexandrus | tree is log(n)...isn't it? |
| 17:37.21 | Alexandrus | i tried.. |
| 17:37.30 | ``Erik | search is, for a balanced tree |
| 17:37.31 | Alexandrus | but somehow i don't know how to give that to r |
| 17:37.35 | ``Erik | in the worst case, it's O(n) |
| 17:37.45 | ``Erik | and for optimizing it, I think it's O(nlgn) |
| 17:37.47 | Alexandrus | isn't it balanced? |
| 17:37.54 | ``Erik | optimizing tries to balance it |
| 17:38.01 | Alexandrus | i mean, i have written my own avl-trees once |
| 17:38.09 | Alexandrus | every operaton was log(n) there |
| 17:38.26 | Alexandrus | but maybe your trees have different conditions to fullfill |
| 17:38.30 | ``Erik | hrm, I d'no the r command... I tend to stay low in the C libraries |
| 17:39.04 | ``Erik | I know we have an rb-tree implementation, but I don't think it's used in the comb generation |
| 17:39.12 | Alexandrus | rb? |
| 17:39.27 | Alexandrus | (oh these shortcuts:P) |
| 17:39.28 | ``Erik | I'm guessing the comb stuff is very naive in generating, and then 'optimizes' once it's all built to try to balance it |
| 17:39.31 | ``Erik | red/black |
| 17:39.37 | ``Erik | based off of avl |
| 17:39.42 | Alexandrus | ah, i know this one:) |
| 17:39.51 | Alexandrus | avl trees might be faster here |
| 17:40.00 | ``Erik | and we're not cool enough to have b+ or anything |
| 17:40.03 | Alexandrus | but who knows what it would screw |
| 17:40.17 | Alexandrus | hmm |
| 17:40.34 | ``Erik | and we try to build the trees so they make nice n-ary trees for human interpretation, as well |
| 17:41.08 | ``Erik | I d'no, I can only think of two people who might have a solid clue... one left the team (and is on vacation), the other at siggraph this week |
| 17:41.55 | Alexandrus | hahaha, some heli builder comes along:P |
| 17:42.05 | Alexandrus | and now some guys have to rewrite the core:P |
| 17:42.06 | Alexandrus | hahahaha |
| 17:42.38 | Alexandrus | i think it works pretty well mostly |
| 17:42.58 | ``Erik | oh, we have other things doing stuff like that... at http://brlcad.org/ is a news item for a 'coil builder', which is almost identical to what you're trying to do |
| 17:43.22 | ``Erik | (the spiral thing) |
| 17:43.22 | Alexandrus | i will finish it today |
| 17:43.31 | Alexandrus | yes, but its perl |
| 17:43.40 | Alexandrus | mine can be called with a simple source call |
| 17:43.57 | ``Erik | but the same thing can be written in tcl, in C, in lisp, in ruby, in python, probably in plain shell scripting |
| 17:44.14 | Alexandrus | yes, but as far as i know, i still had to do the helix thing |
| 17:44.21 | ``Erik | even visual basic *cough* |
| 17:44.28 | Alexandrus | its not that i am trying to do something double here |
| 17:44.29 | Alexandrus | i searched |
| 17:44.36 | Alexandrus | makes a cross:P |
| 17:45.09 | ``Erik | in the end, it's creating a set of mged commands to execute to generate the geometry *shrug* |
| 17:45.36 | Alexandrus | tcl commands...the language is a mess |
| 17:45.58 | Alexandrus | i mean, you "can" write programs in it |
| 17:46.03 | ``Erik | <-- doesn't know tcl, has modified some tcl in the past, tries to avoid it, thus the lack of mged knowledge :D |
| 17:46.04 | Alexandrus | but...its weird |
| 17:46.30 | Alexandrus | interesting:P everything depends on tcl in brl-cad |
| 17:46.44 | Alexandrus | i checked manully (added hunderes of tcl85.lib dependencies) |
| 17:46.56 | ``Erik | when ya run 'rt' or 'g-stl', there is no tcl involved :D the libraries require tcl as a dep, but no tcl stuff is used |
| 17:48.00 | Alexandrus | how do you use brlcad? |
| 17:48.38 | ``Erik | um, via C? :D |
| 17:48.52 | ``Erik | I almost never use mged, and when I do, I usually use -c |
| 17:48.52 | Alexandrus | ah, you use it as a library |
| 17:49.09 | Alexandrus | you know, i just do CAD-Work with it |
| 17:49.13 | Alexandrus | which needs flexibility |
| 17:49.15 | ``Erik | well, there're ~400 programs in BRL-CAD, only a handful use the TCL stuff |
| 17:49.21 | Alexandrus | therefore...a script language.. |
| 17:49.44 | Alexandrus | coil and tire for example create a coil.g or tire.g |
| 17:49.48 | Alexandrus | very weird |
| 17:49.57 | Alexandrus | now you have to merge libs manually? |
| 17:50.12 | ``Erik | BRL-CAD used to have it's own approach to that, then tcl came out, there was nothing else like it when it was adopted... now we're struggling to de-tcl stuff ('cept for starseeker, who's doing the opposite) |
| 17:50.14 | Alexandrus | or do you use libged and operate on a database inside the c program? |
| 17:50.25 | ``Erik | I don't generate geometry |
| 17:50.40 | Alexandrus | hehehe, what do you do with it? |
| 17:50.49 | Alexandrus | looks surprised |
| 17:51.04 | ``Erik | the last few things I've done have been writing the metaball primitive, writing the marching cubes converter, and a lot of work in the adrt/libtie/isst interactive raytracing viewer |
| 17:51.42 | Alexandrus | what is the metaball called in brlcad? |
| 17:51.46 | ``Erik | my current interest is finishing the work I started in making rt output directly to png (or tiff, or bmp, or ...) |
| 17:51.47 | Alexandrus | i have a list of primitves here |
| 17:51.49 | ``Erik | metaball |
| 17:51.54 | ``Erik | it may not be in your list |
| 17:52.02 | ``Erik | it was just a couple years ago :) |
| 17:52.06 | ``Erik | do "make mb.s metaball" |
| 17:52.09 | ``Erik | and rt it |
| 17:52.10 | Alexandrus | how is it called today? |
| 17:52.30 | ``Erik | mb.s just being the name I like to use |
| 17:52.39 | ``Erik | some packages call them "blobs" |
| 17:52.49 | Alexandrus | ah, beautifull |
| 17:54.25 | ``Erik | <-- toolmaker, likes to be a toolmaker for toolmakers |
| 17:54.34 | Alexandrus | ähm ok:) |
| 17:54.41 | Alexandrus | i wish for |
| 17:54.56 | Alexandrus | a primitve, which is based on a function given line |
| 17:55.13 | ``Erik | O.o |
| 17:55.15 | Alexandrus | where you are able to give a boolean function for each 2D tangential space |
| 17:55.26 | Alexandrus | which says which regions are filled and which are not:) |
| 17:55.26 | ``Erik | what, like using a rotate or sweep on a bezier spline? |
| 17:55.31 | ``Erik | like in libpc? |
| 17:55.33 | ``Erik | O.o |
| 17:55.40 | Alexandrus | this is there allready? |
| 17:55.51 | ``Erik | kinda sorta, but not ready for end users |
| 17:55.52 | Alexandrus | i am amazed, i made this up when i thought how to do the helix |
| 17:56.14 | Alexandrus | i think its a challange because you have to accept functions |
| 17:56.20 | Alexandrus | and derive them |
| 17:56.31 | Alexandrus | otherwise you can't get the necessary vectors to create a tangential 2D space |
| 17:56.41 | ``Erik | yeah, we haven't looked at that too much as a procdb as we're kinda looking to get NURBS in place so we can do a rotate/sweep on a 2d rep of the thread |
| 17:57.03 | Alexandrus | rep=repition? |
| 17:57.11 | ``Erik | representation |
| 17:57.28 | Alexandrus | ok:) |
| 17:57.49 | Alexandrus | but this wouldn't do a helix |
| 17:58.04 | ``Erik | draw the cut view of one tooth using the 'sketch' primitive, tell it to revolve and sweep, then cap the result, boom, instant bolt |
| 17:58.27 | ``Erik | and with gusseting, rounding, non-flat faces, etc |
| 17:59.02 | Alexandrus | allready running? |
| 17:59.09 | ``Erik | no, we're not quite there yet |
| 17:59.21 | ``Erik | but a lot of effort is going towards doing that |
| 17:59.24 | Alexandrus | hey, i am not so fond of my simple helix solution |
| 17:59.29 | Alexandrus | its working, but its a hack |
| 17:59.50 | ``Erik | we had a GSoC student doing the libpc stuff, two paid people working on getting nurbs into shape (when other emergencies don't show up), ... |
| 18:00.52 | Alexandrus | (and they sure do:P) |
| 18:00.59 | ``Erik | we're trying, honest :D it's been an uphill fight, investing in something as significant as solid NURBS with mgmt looking for the "but how does it help me tomorrow, I don't care about next year" |
| 18:01.51 | ``Erik | we already do a reasonably good job of converting NURBS from rhino3d or STEP to our format and raytracing them (quickly) |
| 18:02.30 | ``Erik | but no editing yet, ... :) |
| 18:03.24 | ``Erik | starseeker has been heavily involved in that stuff... I'm more interested in converting them into triangles and feeding them to OpenGL or libtie for interactive shaded displays/analysis |
| 18:03.34 | CIA-45 | BRL-CAD: 03brlcad * r39887 10/brlcad/trunk/src/util/binfo.c: supposedly removed the binfo tool in r39519 but the source file wasn't actually removed. |
| 18:03.52 | Alexandrus | hmm |
| 18:04.52 | CIA-45 | BRL-CAD: 03brlcad * r39888 10/brlcad/trunk/misc/win32-msvc8/Makefile.am: add shape tool msvc build files to the dist |
| 18:05.14 | Alexandrus | i think opengl to replace the wire-frame modell |
| 18:05.19 | Alexandrus | might be not such a bad idea |
| 18:05.33 | ``Erik | that's one of the goals of the NURBS work |
| 18:05.40 | Alexandrus | one doesn't need perfect rendering every time |
| 18:05.41 | ``Erik | "shaded display" has been on the todo list for a long time |
| 18:06.03 | ``Erik | fast NURBS to triangle conversion will mean dynamic LoD |
| 18:06.21 | CIA-45 | BRL-CAD: 03brlcad * r39889 10/brlcad/trunk/src/other/tktable/Makefile.in: add missing 'mac' dir files to the dist. |
| 18:06.25 | ``Erik | so it'll always be within a pixel or two, and always display fast :) |
| 18:06.39 | Alexandrus | shadows aren't too important |
| 18:06.44 | Alexandrus | normals are easy |
| 18:07.01 | ``Erik | CAD packages generally ignore shadowing, I don't think that's part of our interest |
| 18:07.03 | Alexandrus | (still, possible with shadow maps) |
| 18:07.18 | ``Erik | though with most modern GPU's, stencil shadowing is pretty snappy |
| 18:07.30 | Alexandrus | it has resolution problems though |
| 18:08.10 | Alexandrus | but i think the hardest part is creating meshes... |
| 18:08.18 | CIA-45 | BRL-CAD: 03brlcad * r39890 10/brlcad/trunk/src/other/libpng/Makefile.am: add vstudio files to extra_dist, along with a couple other missing files |
| 18:08.20 | ``Erik | *shrug* there've been experiments with display BRL-CAD geometry using the OGRE engine, stencil shadowing is just a flip of the switch then |
| 18:08.20 | Alexandrus | i wouldn't know how |
| 18:08.52 | ``Erik | that's the part they're looking at now, correct NURBS raytracing, correct primitive->NURBS conversion, and evaluating CSG NURBS into resolved NURBS |
| 18:09.18 | ``Erik | once that's all done, just walk the surface of the resolved nurb to produce tristrips and feed that to ogl or ogre |
| 18:09.22 | Alexandrus | they must do it for stl export... |
| 18:09.43 | ``Erik | we have the NMG tesselator, which USUALLY works, but is really slow |
| 18:09.50 | Alexandrus | hmm...how do you walk a surface of an object which cannot be represented by a nurb |
| 18:09.56 | Alexandrus | (non differentiable) |
| 18:10.07 | ``Erik | and for g-stl, the -8 option enables marching cubes, which is insanely slow and produces incredibly large amounts of triangles when it shouldn't |
| 18:10.08 | Alexandrus | (for example substracting to spheres) |
| 18:10.10 | ``Erik | :D |
| 18:10.32 | ``Erik | the non-differentiable parts would be trims, with an object made of several surface patches |
| 18:11.10 | ``Erik | pairing the trims exactly has been one of the difficulties they're currently looking at closely, iirc |
| 18:11.39 | ``Erik | or was, before the guy doing most of it got one of those emergency requests to write some excel/java related stuff |
| 18:11.53 | Alexandrus | buuuh... |
| 18:11.53 | CIA-45 | BRL-CAD: 03brlcad * r39891 10/brlcad/trunk/src/adrt/Makefile.am: still need isst_tcltk.c added to the dist |
| 18:11.57 | Alexandrus | no comparison |
| 18:12.10 | ``Erik | yeah, but *shrug* gotz ta get paid |
| 18:12.35 | Alexandrus | i heard some here do get paid |
| 18:12.41 | *** join/#brlcad R0b0t1 (~Enigma@unaffiliated/r0b0t1) | |
| 18:17.33 | ``Erik | ayup |
| 19:01.42 | Alexandrus | damn 256k limit is annoying |
| 19:05.35 | Alexandrus | bug is reported |
| 19:34.22 | CIA-45 | BRL-CAD: 03brlcad * r39892 10/brlcad/trunk/misc/win32-msvc8/bolt/tire.vcproj: already copied to bolt |
| 19:37.05 | CIA-45 | BRL-CAD: 03brlcad * r39893 10/brlcad/trunk/src/other/tktable/Makefile.in: there's a README.TXT file in there that needs to be added. |
| 19:38.09 | ``Erik | I think ya'llz fix to tkhtml3 was wrong. I think the tcl script puts the .c and .h files in the same place as the .tcl file, even in an out of dir build. I'll experiment more on monday, but it failed on linux, fbsd and osX for me with clean checkouts |
| 19:42.04 | Alexandrus | @''Erik: may i ask, which area in the USA you are from? |
| 20:05.29 | ``Erik | from or at? I'm currently in the baltimore area, as are the other paid devs |
| 20:05.39 | ``Erik | just north of washington DC |
| 20:05.54 | Alexandrus | from |
| 20:06.02 | Alexandrus | because you use interesting constructions like |
| 20:06.07 | Alexandrus | do'no and ya'llz |
| 20:06.11 | ``Erik | from... seattle... though I've lived in florida and missouri |
| 20:06.13 | Alexandrus | and more i can't recall at the moment |
| 20:06.40 | ``Erik | "d'no" is pretty common, ya'll is from when I lived in missouri I think, not part of my usual speaking patter, but fun to type |
| 20:07.18 | Alexandrus | still, unusual for me |
| 20:07.26 | Alexandrus | and i am chatting in english language a lot |
| 20:07.41 | Alexandrus | (with all my typos) |
| 20:07.53 | ``Erik | most americans who DO use it in colloquial speech will attempt to hide it in talking to people from outside of their regions |
| 20:08.15 | ``Erik | it's kinda associated with being uneducated and 'backwards' |
| 20:08.36 | Alexandrus | we have some people called bavariens |
| 20:08.48 | ``Erik | bavarians, or barbarians? |
| 20:08.48 | Alexandrus | but their language is not understood as backwards |
| 20:08.53 | Alexandrus | no, bavaria |
| 20:08.54 | Alexandrus | Bayern |
| 20:08.56 | ``Erik | <-- notes that his car is from bavaria :D |
| 20:08.58 | Alexandrus | a region in germany |
| 20:09.01 | ``Erik | yes |
| 20:09.09 | Alexandrus | i cannot understand it |
| 20:09.24 | Alexandrus | but it has proud speakers...hahaha |
| 20:09.28 | ``Erik | the swiss might be better suited to understand it O.o |
| 20:09.29 | Alexandrus | they certainly do not hide it |
| 20:09.33 | Alexandrus | no... |
| 20:09.39 | Alexandrus | swiss people talk english to me |
| 20:09.44 | ``Erik | hehehe |
| 20:09.54 | Alexandrus | happened to me in Zurich |
| 20:09.57 | Alexandrus | really weird |
| 20:10.20 | ``Erik | swiss and bavarian are both tautonic languages from roughly the same area, so I'd imagine there're striking similarities... but I'm guessing |
| 20:10.36 | ``Erik | I've never been to europe, my foreign travels were to the far east |
| 20:11.00 | Alexandrus | no simililarities |
| 20:11.06 | ``Erik | huh *shrug* |
| 20:11.11 | Alexandrus | sry |
| 20:11.16 | Alexandrus | but its rather strange here |
| 20:11.18 | ``Erik | was worth a shot ;) |
| 20:11.23 | Alexandrus | nice try.P |
| 20:11.32 | ``Erik | the US has various regionalized languages, like creole |
| 20:11.56 | ``Erik | which is a mix of old french and a little old english, then evolved in it's own direction... |
| 20:12.54 | Alexandrus | you might be baffled listening to this |
| 20:12.57 | ``Erik | I sometimes have trouble understanding southerners and northeasterners here *shrug* |
| 20:13.05 | ``Erik | even though it's all "american" english |
| 20:13.34 | Alexandrus | back in the 50's american movie english was exemplary for good english |
| 20:13.40 | Alexandrus | today...its far more different |
| 20:13.45 | ``Erik | (at least 3 years in japan taught me that if someone doesn't understand you, try rephrasing and speaking slower, not speaking louder) |
| 20:14.07 | ``Erik | yes, I've heard linguists claim that even modern american english is far closer to 1600's british english than modern british english is |
| 20:14.07 | Alexandrus | in my case...try louder:P |
| 20:14.23 | Alexandrus | 1600's...lol |
| 20:14.29 | ``Erik | 1600's/1700's |
| 20:14.37 | ``Erik | when the big colonial migration was |
| 20:14.38 | Alexandrus | you can read Shakespear..haha |
| 20:14.47 | Alexandrus | not so distant from you |
| 20:14.52 | ``Erik | shakespeare doesn't seem to difficult in original form to me |
| 20:15.11 | ``Erik | chaucer is a bit rough though :D |
| 20:15.24 | Alexandrus | have the comedy of errors somewhere here |
| 20:15.35 | ``Erik | I mean, uh, I'm an illiterate american, if it waren't on wrastlin,, ah reck'n it don't matter noen |
| 20:15.39 | ``Erik | none |
| 20:15.45 | Alexandrus | lol |
| 20:15.52 | Alexandrus | waren't...whats that? |
| 20:15.55 | Alexandrus | wraslin? |
| 20:15.57 | Alexandrus | reck'n? |
| 20:16.02 | Alexandrus | neon? |
| 20:16.09 | ``Erik | waren't -> was not |
| 20:16.13 | Alexandrus | weren't |
| 20:16.20 | ``Erik | wraslin' -> americanized wrestling |
| 20:16.28 | ``Erik | reck'n -> reckon |
| 20:16.51 | ``Erik | ah -> I |
| 20:16.59 | ``Erik | don't -> doesn't |
| 20:17.01 | ``Erik | :D |
| 20:17.20 | ``Erik | and, of course, "doesn't matter none", double negative... |
| 20:17.44 | Alexandrus | thanks for the full translation |
| 20:17.59 | ``Erik | scary, no? |
| 20:18.15 | Alexandrus | you are not speaking japanese yet |
| 20:18.20 | Alexandrus | so, I survive |
| 20:18.25 | ``Erik | nihongo oh hanasemasuka? |
| 20:18.38 | Alexandrus | :-: |
| 20:18.39 | ``Erik | o'genki desu ka |
| 20:18.54 | Alexandrus | you don't have the chars? |
| 20:18.55 | Alexandrus | no utf8? |
| 20:19.05 | ``Erik | that is utf8... no utf16/32 |
| 20:19.20 | ``Erik | well, probably could do it, but it's been a long time since I've done hiragana, katakana and kanji |
| 20:19.41 | ``Erik | and I'm using irssi via ssh, so that might not support the full range |
| 20:19.59 | ``Erik | I was a bit impressed that the stack could do an umlaut |
| 20:20.01 | Alexandrus | i guess its only up to the clients if they can display it |
| 20:20.05 | Alexandrus | the IRC protocol doesn't care |
| 20:20.27 | ``Erik | hm, I think rfc1459 might state ascii8 |
| 20:20.44 | ``Erik | was quite a while ago that it was drafted... but modern servers may go above and beyond |
| 20:20.58 | Alexandrus | i wrote an irc client/server once:P |
| 20:21.00 | Alexandrus | sure ignored it |
| 20:21.07 | Alexandrus | i used it for mathematical commucation over irc |
| 20:21.12 | ``Erik | it's a quite simple protocol :) |
| 20:21.14 | Alexandrus | comunication |
| 20:21.19 | Alexandrus | +m |
| 20:21.26 | Alexandrus | it is |
| 20:21.44 | Alexandrus | one could start a telnet and type the commands by hand |
| 20:21.50 | ``Erik | back in the mid 90's, I think most of us knew how to use telnet to use irc |
| 20:22.00 | ``Erik | hehehe, ayup |
| 20:22.03 | Alexandrus | its a bit crap though |
| 20:22.10 | ``Erik | privmsg #brlcad: tada! |
| 20:22.19 | Alexandrus | and then reading all the mess |
| 20:22.21 | Alexandrus | quit |
| 20:22.22 | Alexandrus | :P |
| 20:22.59 | ``Erik | wait, space then colon for a space containing line? |
| 20:23.06 | Alexandrus | oh don't ask me |
| 20:23.07 | ``Erik | like privmsg #brlcad :tada, this is working! |
| 20:23.09 | Alexandrus | i wrote a parser for this once |
| 20:23.12 | ``Erik | so long ago |
| 20:23.13 | Alexandrus | i don't do this manually |
| 20:23.35 | Alexandrus | i don't do assembler any more either*G* |
| 20:23.41 | ``Erik | ooh, I do on occasion |
| 20:24.01 | ``Erik | I'm half poking at getting freebsd working on my arm7 machine, I might have to do some asm and write some drivers |
| 20:24.04 | Alexandrus | i can, but only on microcontrollers |
| 20:24.25 | Alexandrus | drivers, yes, but only a few lines |
| 20:24.34 | Alexandrus | like port access |
| 20:24.42 | Alexandrus | or some special command to use in an inline-procedure |
| 20:24.47 | ``Erik | um, PCI and serial bus drivers, I think |
| 20:24.51 | ``Erik | mebbe some others |
| 20:25.12 | Alexandrus | i wrote a PCI driver in FreePascal once |
| 20:25.16 | ``Erik | arm fell off the list of procs in fbsd's cvs :/ |
| 20:25.16 | Alexandrus | no asm at all |
| 20:25.24 | Alexandrus | hmm, to bad |
| 20:25.32 | ``Erik | so I must put it back. |
| 20:26.09 | ``Erik | other than writing the startup and connection stuff for an i386 os, all my os has just been a couple lines to feed C :/ |
| 20:26.16 | ``Erik | er |
| 20:26.21 | ``Erik | all my asm (recently) has been |
| 20:26.32 | ``Erik | but I did a lot of 6512 asm on the c64c |
| 20:26.33 | Alexandrus | i have written a 200.000 line asm os once.. |
| 20:26.41 | Alexandrus | during my school time:P |
| 20:26.50 | Alexandrus | i know what i reject:P |
| 20:27.09 | ``Erik | mips r2000 was damn sexy |
| 20:27.21 | Alexandrus | still, asm...so damn unportable |
| 20:27.24 | Alexandrus | unreadable |
| 20:27.29 | ``Erik | well |
| 20:27.49 | ``Erik | I'd argue that asm from something like a PPC or r2k or pdp11 was VERy readable |
| 20:27.57 | ``Erik | the x86 stuff is really... really... really really really bad. |
| 20:28.24 | Alexandrus | assembler usually makes out of a simple x:=f(y,z,u,v,w) something long and unreadable |
| 20:28.30 | Alexandrus | if not...its not an assembler anymore |
| 20:28.36 | ``Erik | maloeran is working on a language to make asm less sucky, somewhere between asm and C |
| 20:28.51 | ``Erik | heh |
| 20:29.01 | Alexandrus | i think i prefer moving even higher |
| 20:29.04 | Alexandrus | intention based languages |
| 20:29.24 | ``Erik | only on x86... programs on archs meant to be asm programmed by humans is far far more eradable... indeed, C is sometimes called a portable PDP assembler... |
| 20:29.41 | Alexandrus | (intention:= f:S->R^n for S as set of states of the System) |
| 20:29.42 | ``Erik | <-- has been enamored by CL lately |
| 20:30.05 | Alexandrus | @''Erik: even microcontrollers are programmed in c/c++ today |
| 20:30.15 | Alexandrus | and c is crappy |
| 20:30.20 | ``Erik | I'd rather write code in asm than c++ |
| 20:30.23 | Alexandrus | and even the c coders know that |
| 20:30.34 | Alexandrus | i prefer some high level fpc |
| 20:30.42 | ``Erik | C is awesome... if you're writing an OS for a PDP :D |
| 20:30.53 | Alexandrus | yes, but have a look at BRL-Cad |
| 20:31.10 | Alexandrus | i am sure, lots of work has been done fighting c errors |
| 20:31.13 | Alexandrus | pure c errors |
| 20:31.32 | Alexandrus | and the vls stuff certainly is a fix |
| 20:31.43 | Alexandrus | (one, which even simple pascal had long ago) |
| 20:31.47 | ``Erik | I've made arguments for at least introducing lisp or scheme... you're preaching to the choir, here :D |
| 20:32.07 | Alexandrus | i don't think thats easy to do |
| 20:32.23 | Alexandrus | interfacing between different languages is a challange on its own |
| 20:32.27 | ``Erik | yeah |
| 20:32.40 | ``Erik | I wrote up the 'SWIG-ify BRL-CAD' request, too |
| 20:32.49 | ``Erik | to swig-wrap all the library stuff |
| 20:33.11 | Alexandrus | there are still char x[1024] in there |
| 20:33.11 | ``Erik | even though swig can only implement half of CFFI/UFFI |
| 20:33.14 | Alexandrus | i saw them.. |
| 20:33.17 | ``Erik | yes |
| 20:33.35 | ``Erik | some, we eliminate, some we extend when needed... *shrug* it is what it is |
| 20:33.37 | Alexandrus | dangerous |
| 20:33.48 | Alexandrus | but ok, thats history |
| 20:34.09 | ``Erik | I d'no, char buf[64]; snprintf(buf, 63, "%02x", myval); ... seems safe to me |
| 20:34.27 | ``Erik | where a fixed length buffer is used, there has been a lot of effort to protect it |
| 20:35.06 | Alexandrus | i only do this in especially safeguarded classes in fpc |
| 20:35.15 | ``Erik | fpc? |
| 20:35.18 | Alexandrus | freepascal |
| 20:35.23 | ``Erik | ah |
| 20:35.36 | Alexandrus | i mean, for network layers i use ringbuffers |
| 20:35.57 | Alexandrus | buf[2^20] is more usual here:P |
| 20:35.57 | ``Erik | I kinda like linked page sets |
| 20:36.13 | Alexandrus | are you talking about mmu? |
| 20:36.20 | ``Erik | no, in C |
| 20:36.33 | Alexandrus | don't know what linked page sets are |
| 20:37.07 | ``Erik | struct bufarea { char [BUFSIZ]; struct bufarea *next; int amtused; struct bufarea *root; } |
| 20:37.12 | ``Erik | something along those lines |
| 20:37.23 | Alexandrus | ah, this |
| 20:37.30 | Alexandrus | heheheh, use this too.. |
| 20:37.33 | Alexandrus | but never named it |
| 20:37.48 | ``Erik | a form of memory pooling |
| 20:37.48 | Alexandrus | usually only if i allocate billions of equal blocks in short time |
| 20:38.05 | Alexandrus | but the question is..when do you free a page? |
| 20:38.08 | ``Erik | building a block from a stream, it's useful |
| 20:38.12 | Alexandrus | difficult decision i think |
| 20:38.37 | ``Erik | heh, an unused page is added to the free list, if the free list gets too long, ... |
| 20:38.56 | ``Erik | and once in a while, compact if you're freeing bits inside of pages... |
| 20:38.58 | Alexandrus | i used it in parsers |
| 20:39.06 | ``Erik | eventually, you end up with greenspuns 10th rule |
| 20:39.12 | Alexandrus | :P |
| 20:39.17 | Alexandrus | solves everything, does it:P |
| 20:40.26 | Alexandrus | n't |
| 20:40.33 | ``Erik | notes that in the early 80's, one of his c64 basic books had an entire chapter for garbage collection |
| 20:40.48 | Alexandrus | hehehehe |
| 20:41.32 | Alexandrus | i have wirth's Algorithmen und Datenstrukturen |
| 20:42.02 | Alexandrus | this is my source of inspiration if i am out of ideas |
| 20:45.37 | Alexandrus | must say...i am not sure if garbage collection is ideal |
| 20:45.50 | Alexandrus | its quite expensive at times |
| 20:45.55 | Alexandrus | and often unpredictable |
| 20:52.59 | Alexandrus | i found a solution to pass lots of stuff to r in once step... |
| 20:53.17 | Alexandrus | but really...in tcl there must be a special hack for everything |
| 21:00.20 | ``Erik | generational "treadmill" gc is ... awesome |
| 21:00.55 | ``Erik | also; GC comes in many forms, not just 'general' gc like you see in java, ruby, python, c#, lisp, etc... but refcount like you see in objc, c++, ... |
| 21:01.06 | Alexandrus | i know.. |
| 21:01.22 | Alexandrus | sweep types |
| 21:01.25 | ``Erik | (refcount is trivial to implement in an unsuck performance way, but it has the issue of leaking lost graphs) |
| 21:01.34 | ``Erik | sweep is the most naive and crapiest of the bunch! |
| 21:01.35 | Alexandrus | refcount is quite a challange |
| 21:01.38 | Alexandrus | if you have loops |
| 21:01.54 | ``Erik | yeah, an unreferenced graph just leaks in straight refcount |
| 21:02.21 | ``Erik | mark&sweep is mostly a study in old tech these days, though, modern gc's are a bit ... niftier :D |
| 21:02.33 | ``Erik | as is straight stop© |
| 21:03.03 | Alexandrus | still there are always tradeoffs |
| 21:03.09 | ``Erik | and the generational hack was a bit of sheer brilliance even back then... attached to the new methods (like treadmill), it's quite impressive |
| 21:03.12 | Alexandrus | there is a moment of truth, when the gc runs... |
| 21:03.18 | ``Erik | yes |
| 21:03.38 | Alexandrus | without gc, you have to work properly |
| 21:03.46 | Alexandrus | but no gc effects |
| 21:03.49 | ``Erik | I abandoned a game engine I wrote many many years ago, used a stop© and once I had a nontrivial amount of objects, there was a noticable hitch in the game :( |
| 21:04.15 | Alexandrus | i think gc for script enviroments might work |
| 21:04.23 | Alexandrus | you can call gc before/after loads |
| 21:04.34 | Alexandrus | can' |
| 21:04.38 | Alexandrus | t do much harm there |
| 21:04.41 | ``Erik | BRL-CAD has a lot of garbage collection in it, even in the C parts |
| 21:04.51 | ``Erik | free-lists, p-tables, etc |
| 21:04.56 | Alexandrus | i saw a lot of bu_vls_free... |
| 21:05.00 | ``Erik | greenspuns 10th, yo |
| 21:05.24 | ``Erik | bu_vls_free is a real free, iirc, but in the raytracer where performance matters, you 'free' an object and it goes on a free list instead of actually deallocating |
| 21:05.41 | ``Erik | and the free-list mgmr decides if it's time to actually deallocate a bunch of mem, or just keep going |
| 21:06.29 | Alexandrus | hmm |
| 21:06.32 | ``Erik | that's why we can "allocate" enough to do all the partitions on a shot without ever having to yield quanta for a syscall |
| 21:06.50 | Alexandrus | but free lists are not gc's.. |
| 21:07.05 | ``Erik | sure, they're a scavanging gc |
| 21:07.25 | ``Erik | in fact, they're more of a gc than java's |
| 21:07.33 | Alexandrus | they do not detect themselves, if a memory region is still used, do they? |
| 21:07.35 | ``Erik | since we can actually reduce total memory usage |
| 21:07.49 | ``Erik | java's gc can only add to freelists, it can never reduce the total memory footprint of the jvm |
| 21:07.59 | Alexandrus | hmm...weird |
| 21:08.05 | ``Erik | no, there's explicit release |
| 21:08.09 | ``Erik | but it is collection |
| 21:08.13 | Alexandrus | it should be possible for java to reduce |
| 21:08.24 | ``Erik | you'd think... but they don't |
| 21:08.28 | Alexandrus | i would call this memory pool |
| 21:09.33 | ``Erik | then you'd call most modern gc's a memory pool, since they often tie to the language and add things to a free list when they go out of scope... :D |
| 21:10.15 | Alexandrus | propably |
| 21:10.31 | Alexandrus | for me a gc completly removes any burden on the programmer for mm |
| 21:10.42 | Alexandrus | the gc scans the structure, and decides |
| 21:10.57 | ``Erik | then what do you call boehm-gc? |
| 21:11.15 | ``Erik | it cannot scan structure, yet it's still gc |
| 21:11.40 | ``Erik | (thus it has to be "conservative" gc to avoid removing valid memory) |
| 21:12.14 | ``Erik | I think you're associating all gc with "perfect" gc :D |
| 21:12.44 | ``Erik | even in java, to make SURE something is gc'd, you want to set it to NULL O.o |
| 21:13.16 | Alexandrus | no i am associating with what i read in books about gc's |
| 21:13.19 | Alexandrus | and gc algorithms |
| 21:13.48 | Alexandrus | you have to set it to Null to remove the reference |
| 21:13.51 | Alexandrus | but you do not free it |
| 21:14.01 | Alexandrus | the difference is big |
| 21:14.07 | ``Erik | no, the gc eventually collects it, but you have to make sure the reference is gone |
| 21:14.16 | Alexandrus | especially if you have stuff which is "owned" by several objects |
| 21:14.44 | ``Erik | and in a conservative gc, it's very hard to confirm a reference is gone, so they're "less gc-ey" than the academic "perfect gc" |
| 21:15.09 | ``Erik | herr boehm has some papers out, he's a smart cookie |
| 21:15.14 | Alexandrus | gc's either need a special compiler |
| 21:15.18 | Alexandrus | or inside an interpreter |
| 21:15.20 | ``Erik | might be worth checking out |
| 21:15.27 | ``Erik | boehm-gc is a library for C |
| 21:15.55 | Alexandrus | you know, i don't do c usually:P |
| 21:16.20 | ``Erik | for this situation, I think C and pascal are probably sufficiently similar |
| 21:16.49 | Alexandrus | true, but i don't have this lib |
| 21:16.53 | ``Erik | the two ways of getting memory in C are either static (stack) or dynamic (heap), ... gc only cares about heap |
| 21:17.14 | ``Erik | I think you can translate that to the pascal notion, so his papers should make sense |
| 21:17.28 | Alexandrus | the paper yes |
| 21:17.37 | Alexandrus | if its not behind some cite-wall |
| 21:17.54 | ``Erik | shouldn't be |
| 21:18.00 | ``Erik | http://www.hpl.hp.com/personal/Hans_Boehm/gc/ is a start |
| 21:18.25 | Alexandrus | mark&sweep... |
| 21:18.28 | ``Erik | yes |
| 21:18.38 | ``Erik | so fragmentation is expected |
| 21:18.59 | Alexandrus | hah, moving stuff around in c will not work anywa |
| 21:19.09 | Alexandrus | would need global visible pointers |
| 21:19.33 | ``Erik | http://en.wikipedia.org/wiki/Boehm_garbage_collector mentions a new version being compacting |
| 21:19.51 | ``Erik | precise compacting in .net |
| 21:19.54 | ``Erik | *shrug* |
| 21:20.36 | Alexandrus | conservative...hmm.. |
| 21:20.55 | ``Erik | embedded, I'd still want manual mem mgmt, but gc is not the slow pig it used to be thought |
| 21:21.01 | ``Erik | that's all I'm saying :) |
| 21:21.34 | Alexandrus | memory stuff one can't say easily |
| 21:21.39 | ``Erik | and my notion of embedded is becoming archaic.. I can't consider my 'iphone' an embedded machine |
| 21:21.46 | ``Erik | or my arm7 openrd-client |
| 21:22.00 | ``Erik | I have to call them computers :D |
| 21:22.10 | ``Erik | the PIC16f88's, yeah, those are embedded microcontrollers |
| 21:22.26 | Alexandrus | most Atmels too |
| 21:22.33 | Alexandrus | for motor control etc |
| 21:23.56 | ``Erik | I've never worked with atmel, but I've heard a lot about them.. I went to a "robocon" here and a lot of the people presenting were fond of atmel, they sound a bit beefier than pic, btu not much |
| 21:24.09 | ``Erik | indeed, a lot had switched from pic to atmel |
| 21:24.35 | Alexandrus | atmel's are more compatible |
| 21:24.58 | ``Erik | I think they're more poweful, and a bit more expensive than pic |
| 21:24.59 | Alexandrus | but i heard, pic's are more robust sometimes |
| 21:25.06 | Alexandrus | (temperature) |
| 21:25.12 | ``Erik | heh |
| 21:25.16 | Alexandrus | who cares about cents if its about cents. |
| 21:25.23 | ``Erik | and intel and amd are far more powerful and far less robust ;) |
| 21:25.38 | Alexandrus | there are too many atmels around to say |
| 21:25.43 | ``Erik | depends on how many thousand units you want to make |
| 21:25.46 | Alexandrus | even 4 bit ones |
| 21:26.09 | Alexandrus | even if you build thousands... |
| 21:26.23 | ``Erik | a dedicated die is insanely expensive, fpga's can be prohibitively expensive |
| 21:26.26 | Alexandrus | you are more likely to loose money somewhere else |
| 21:26.38 | Alexandrus | fpga's are insane... |
| 21:26.54 | ``Erik | speaking of, I should get my cyclone2 back from twingy O.o |
| 21:26.55 | Alexandrus | i checked if i could find one for my visual cortex:P |
| 21:27.05 | Alexandrus | but $4000...no... |
| 21:27.13 | ``Erik | you can get a dev board for a few hundred usd |
| 21:27.24 | ``Erik | but then they want to sell you the developer software suite for another grant |
| 21:27.25 | Alexandrus | but its too weak for my purposes |
| 21:27.26 | ``Erik | grand |
| 21:27.34 | ``Erik | yeah |
| 21:27.42 | Alexandrus | i need systolic arrays |
| 21:27.42 | ``Erik | that's why I bought pic chips a few years back |
| 21:27.48 | Alexandrus | flexible, generic systolic arrays |
| 21:27.54 | ``Erik | $20 on a programmer, and I maintained the freebsd port of the programmer software... |
| 21:28.04 | Alexandrus | hmm |
| 21:28.14 | Alexandrus | you can let several mc's work together |
| 21:28.21 | ``Erik | "mc"? |
| 21:28.25 | Alexandrus | microcontrollers |
| 21:28.29 | ``Erik | yes |
| 21:28.32 | ``Erik | they all talk i2c |
| 21:28.44 | ``Erik | indeed, my openrd-client has some i2c crud in it |
| 21:28.45 | Alexandrus | or can |
| 21:28.56 | ``Erik | http://www.globalscaletechnologies.com/t-openrdcdetails.aspx |
| 21:29.05 | ``Erik | that is what I intend to replace my home server with |
| 21:29.32 | Alexandrus | hmm, i think i keep my home server |
| 21:29.35 | ``Erik | 1.2ghz arm7 marvell kirkwood, 512m ram, 4g nand, sd, 7 usb |
| 21:30.06 | ``Erik | my home server is currently a pIII 650mhz with 128m ram, with lots of moving parts and pieces sucking down wattage |
| 21:31.03 | ``Erik | I'd rather have a low power silent machine than a high performance machine... for my home server, that is... for my car, the opposite is true ;) |
| 21:31.04 | Alexandrus | lol |
| 21:31.28 | Alexandrus | mine is in another room |
| 21:31.30 | Alexandrus | and pretty silent |
| 21:32.16 | ``Erik | I keep mine in the study on the first floor... it didn't used to bother me, but it's been exceptionally hot, so the middle and top floors are too warm to be on |
| 21:32.43 | ``Erik | and reducing wattage just seems like a good thing no matter what |
| 21:32.47 | Alexandrus | know, its 23:32 here |
| 21:32.50 | Alexandrus | i love to talk |
| 21:32.56 | Alexandrus | but i am pretty tired |
| 21:32.58 | ``Erik | auf weidersein |
| 21:33.02 | Alexandrus | danke:) |
| 21:33.11 | Alexandrus | sry, but our time zones are a bit too different |
| 21:33.22 | ``Erik | yes, only 5:30 here :) |
| 21:33.24 | Alexandrus | we will speak again:) |
| 21:33.25 | ``Erik | 17:30 |
| 21:33.29 | ``Erik | *wave* |
| 21:33.33 | Alexandrus | bye |
| 22:49.52 | CIA-45 | BRL-CAD: 03brlcad * r39894 10/brlcad/trunk/src/other/tktable/Makefile.in: the README.txt file is already covered by PKG_EXTRA_DIST. double-listing results in permission denied during distcheck due to overwrite attempt. |
| 22:53.54 | brlcad | rx |
| 23:05.33 | ``Erik | aight, the boy, what was I wrong on? O.o |
| 23:11.45 | brlcad | i haven't read backlog |
| 23:11.56 | brlcad | just got reconnected since I left |
| 23:12.14 | brlcad | ponied up the fee |
| 23:12.54 | brlcad | wow, that is some long backlog |
| 23:19.24 | ``Erik | the fee for what, cab? |
| 23:20.24 | ``Erik | I"d rather drop $40 rt for a cab than deal with busses, myselc |
| 23:21.22 | brlcad | fee for internet at the hotel |
| 23:21.41 | ``Erik | apparently severing the power cord is not condusive to running the electric trimmer. :/ |
| 23:21.44 | brlcad | yeah, totally wrt the cab (and it's a $49 flat fare) |
| 23:22.06 | ``Erik | starseeker was struggling, I recommended the cap, he was still struggling when he left |
| 23:23.28 | starseeker | growls... can't they get a decent signal strength??? |
| 23:23.31 | ``Erik | so it's 4:23 there? getting ready to walk across the highway and find some good food? |
| 23:24.00 | brlcad | I'm kind of in a little mecca of interesting places here, at LA Live |
| 23:24.09 | brlcad | it's a new complex they built next to the staples center |
| 23:24.22 | ``Erik | <-- is STILL annoyed that brlcad had to translate "carne quesadilla" from the menu for him |
| 23:24.40 | starseeker | can't believe he got a successful distcheck on Linux with all those problems... |
| 23:24.59 | ``Erik | I said car-nay kay-suh-DEE-yuh, not like I said "kay-zer-dill-er |
| 23:25.00 | ``Erik | " |
| 23:25.21 | starseeker | brlcad: thanks for spotting that tktable README.txt - I saw that on the Mac just before I left |
| 23:25.29 | ``Erik | and the car-nay instead of cadddd-nay couldn't have been THAT tricky |
| 23:26.06 | ``Erik | I really think you are both wrong on the tkhtml3 generated file |
| 23:26.40 | starseeker | ``Erik: it's ending up in build dir for me (are we talking about togl_ws.h or whatever it is?) |
| 23:26.46 | ``Erik | cssprop.c |
| 23:27.03 | ``Erik | I tired three naked machines, bsd linux and osX, they all ended up in srcdir |
| 23:27.12 | ``Erik | if I ran it AGAIN in those, it ended up in blddir |
| 23:27.21 | starseeker | O.o |
| 23:27.48 | starseeker | what the bleep... |
| 23:27.58 | ``Erik | yeah, who the fuck put this fucking tcl shit in |
| 23:28.12 | ``Erik | like I said, I'll do some more testing on monday |
| 23:28.33 | brlcad | starseeker: I can't believe you got a successful distcheck either |
| 23:28.34 | ``Erik | aug28 for fitting |
| 23:28.35 | starseeker | ``Erik: I'm not so much trying to bring more Tcl/Tk in as I am to make what we DO have tolerable |
| 23:28.41 | ``Erik | might go get me a suit while I'm at it |
| 23:28.46 | starseeker | brlcad: both Redhat and gentoo |
| 23:28.47 | ``Erik | last one is too small anymore |
| 23:29.04 | starseeker | ``Erik: cool, thanks! |
| 23:29.25 | starseeker | is probably going to end up with a suit as well, although he has no clue what he'll do with it... |
| 23:29.31 | brlcad | isn't that like testing pork and ham? |
| 23:29.47 | ``Erik | starsucker: for demo, ripped up ac/dc tshirt, right? and bring the ripped up ac/dc tshirt for after the ceremony? |
| 23:30.15 | starseeker | brlcad: heh, pretty much - I wasn't claiming it was "done" but presumably my versions of autotools were different from Redhat's |
| 23:30.25 | ``Erik | 0 build on osX, rhel and fbsd until it goddamn works... |
| 23:30.32 | ``Erik | and if your system is different, eff you :D |
| 23:31.04 | ``Erik | the fbsd one will probably be like gentoo, but more harsh |
| 23:31.05 | starseeker | ``Erik: I knew we weren't done yet (I saw the mac failure, just for starters, and I also knew libpng had some remaining stuff) |
| 23:31.25 | ``Erik | libpng seems to work now, I sat i bobs office and it worked for him, he cussed |
| 23:31.27 | starseeker | I just am astonished the behavior of the tools was so different on a per-platform basis |
| 23:32.14 | starseeker | he cussed that it worked? |
| 23:32.39 | starseeker | why, was he having more fun with something else? |
| 23:32.40 | ``Erik | subconfigging anyone elses config is always ... well, it's like volunteering to shove their grenade up your ass to see how it explodes... it SEEMS like less work at first, but is a shitload more when ya do it, and is a bit more every release |
| 23:32.56 | ``Erik | no, stating it failed and then having it work when I was sitting there got him :) |
| 23:33.11 | starseeker | hehehe |
| 23:33.21 | starseeker | ah yes, that is worth a few choice words |
| 23:33.38 | ``Erik | we got to talk about workshops :) |
| 23:33.40 | brlcad | starseeker: I'd wait until you actively need a suit.. a good suit can set you back $500+ |
| 23:33.58 | brlcad | libpng subconfig isn't done yet, but works for "most" platforms now |
| 23:34.07 | brlcad | you have to help it along on some |
| 23:34.07 | starseeker | brlcad: I get some kind of discount if I get it right around the wedding |
| 23:34.09 | ``Erik | it's good to have one if you don't plan on changing shape, though |
| 23:34.20 | ``Erik | dark grey or brown |
| 23:34.31 | ``Erik | <-- thinking about getting himself a charcoal grey one or so |
| 23:34.50 | ``Erik | mebbe a couple choise applications and I can cinch the db upgrade |
| 23:34.51 | ``Erik | meh |
| 23:34.54 | starseeker | Sarah does like to go to concerts on occasion, other culture... |
| 23:35.09 | brlcad | glances at a few thousand people outside his window |
| 23:35.15 | ``Erik | ok, you say concert and I think you mean smething else than 99.3 % of humans |
| 23:35.36 | starseeker | orchestral performance |
| 23:35.50 | ``Erik | yeahhhhhhhhhh, not what most people think when ya say concert ;0 |
| 23:36.08 | starseeker | is still waking up - no sleep til I got on the plane, and I'm not brlcad :-P |
| 23:37.06 | ``Erik | around nov/dec, I may be dressing up a bit to pull a chuck and bump a level, so'z the suit thing now might be useful *shrug* I d'no |
| 23:37.15 | ``Erik | (2-3, not 3-4, thought) |
| 23:37.41 | starseeker | awaits the string of cardiac arrests with anticipation |
| 23:37.49 | ``Erik | I think I'm acutally the lorest rated person in the office |
| 23:38.18 | ``Erik | starseeker: I used to get chewed out for overdressing. |
| 23:38.40 | ``Erik | I was told to remove the jacket and tie for my interview at fedex |
| 23:38.55 | starseeker | ``Erik: aren't contractors just below pond scum? :-P |
| 23:39.02 | ``Erik | I think the tie gave mgmt type folk at arl willies |
| 23:39.18 | ``Erik | of course not, starseeker |
| 23:39.33 | ``Erik | we woudln't dare violate pc and insult pondscum like that |
| 23:39.34 | ``Erik | :D |
| 23:39.40 | starseeker | hehe |
| 23:40.22 | ``Erik | but I think I'm the only db2 in the building |
| 23:40.31 | starseeker | ah, yeah that sucks |
| 23:41.16 | ``Erik | I was supposed to come in as db3, but I'd worked at places that'd gone under and the 'investigators' didn't bother looking |
| 23:41.33 | starseeker | <snort> figures |
| 23:42.02 | ``Erik | so the initial offer was actaully very insulting, the paystubs bought me maxing db2, but not db3... and lisa and wendy constantly said you haven't been here long enough to apply |
| 23:42.26 | ``Erik | then paul went and changed the rules radically every cycle, and I"d gotten to teh point where I was just figurin' on leavin' |
| 23:42.34 | ``Erik | so ... yeah... here I am. |
| 23:43.01 | starseeker | ah, so now that you're staying it's worth the circus of... what do they call that process... |
| 23:43.28 | ``Erik | (I mean, seriously, the initial offer was about 1/4 of what I was making at the time... I was willing to come over cappeda t 1/2 of it... I must be stupid) |
| 23:44.01 | starseeker | shrugs - you do at least have a steady job in the middle of an economic meltdown |
| 23:44.51 | ``Erik | I got the previous job in the middle of the meltdown... I used to like to pretend that I'm pretty damn good at what I do, now *shrug* fuck it |
| 23:45.04 | brlcad | starseeker: haha, how was that early flight? :) |
| 23:45.15 | starseeker | ``Erik: you know that antiques mall over by Bynum run? - was in there the other day, like half the square area of the whole building was empty |
| 23:45.23 | ``Erik | if it flys before noon, it's an early, yo |
| 23:45.29 | ``Erik | spenciola? |
| 23:45.35 | starseeker | yeah, that one |
| 23:45.39 | starseeker | freaky |
| 23:45.41 | ``Erik | yeah, that used to be a Cmake |
| 23:45.43 | ``Erik | cmart |
| 23:45.47 | ``Erik | as useless as cmake |
| 23:45.57 | ``Erik | that's walking distance from my house, yo |
| 23:46.23 | starseeker | brlcad: yeah, that kinda backfired on me - would have done better to just do something normal |
| 23:46.44 | ``Erik | in fact, driving home, you come out, you hit the stoplight immediately by the cvs... then ya hit hte next stop light, if you turned right on that, you'd be at my house... left, you'd be at daytonas :) |
| 23:46.56 | starseeker | ah, cool |
| 23:47.15 | ``Erik | and I walked to daytonas house with a laptop and a jug of coffee O.o |
| 23:47.20 | ``Erik | it's "on th eway" |
| 23:47.37 | ``Erik | do you use 23 over to 1? |
| 23:47.56 | starseeker | (hoping I didn't get some kind of photo speeding ticket on 695 - some kind of weird flash as cars went by one spot, couldn't tell if my car was a target for one or the car behind me) |
| 23:48.13 | starseeker | ``Erik: 23->bypass->22 |
| 23:48.24 | starseeker | 23->1 if heading into Bel Air |
| 23:48.25 | ``Erik | "bypass" is 1 |
| 23:48.35 | ``Erik | there're 2 1' |
| 23:48.53 | ``Erik | the traffic circle on 23 is a stones throw from my house... literally |
| 23:49.12 | ``Erik | also; 23, ->1, ->prospect mill->22... shave a good mile off |
| 23:49.28 | starseeker | that's a messed up little section of roadway in there - first time I was in Bel Air area looking for something I think I circled around that whole 24/23/1 thing a few times |
| 23:49.32 | ``Erik | when you get back, carpool, I'll show ya the route :) |
| 23:49.59 | starseeker | ``Erik: ah yeah, I've seen people make that turn tried it myself actually |
| 23:50.29 | starseeker | didn't seem to change much time wise <shrug> |
| 23:50.35 | ``Erik | it's a good mile off of the drive, but about the same time |
| 23:50.49 | starseeker | nods - that makes sense |
| 23:50.59 | ``Erik | usually, I end up infront of folk, but sometimes they beat me |
| 23:51.06 | starseeker | I actually end up using the gas station at the other intersection a lot |
| 23:51.09 | starseeker | convenient |
| 23:51.10 | ``Erik | last week, a truck beet me, pissed me off |
| 23:51.14 | starseeker | hehe |
| 23:51.22 | ``Erik | the're two, the wawa and the xtra, those're the two cheapest around |
| 23:51.29 | ``Erik | on 22, past the college |
| 23:51.48 | ``Erik | I use the extra, 93 instead of 91, and no ethanol |
| 23:52.08 | ``Erik | right by la tolteca |
| 23:52.29 | starseeker | ah, right - I've used the wawa sometimes, but for some reason I ususally end up at the one just at the start of my run on 22... fueling up before the long drive I guess... |
| 23:52.50 | starseeker | not BP.. Royal or something? |
| 23:53.14 | starseeker | the wawa I typically hit going home |
| 23:53.31 | ``Erik | at 543 and 22 is a royal |
| 23:53.43 | starseeker | yeah, I think that's the one |
| 23:53.46 | ``Erik | brlcad, ya missed out, we got 'vip' punchcards at la tolteca |
| 23:54.13 | ``Erik | bob swears by it, but every time I've looked, it's been pricier than the two at the college, and they ran out of receipt paper on me |
| 23:55.08 | ``Erik | prospect mill gives me a fun wiggle at the traffic circle, then a nice distance in 6th cutting a mile off, plus a run squeeze ontop 543... |
| 23:55.43 | starseeker | hehe - the immediate thought of several of us was "crap, now there's evidence of how often we eat out..." |
| 23:55.52 | ``Erik | I come pu off of prospect mill slow, flatten down so the tires are in full contact, then open it up in 2nd to boogie up to 60 or 70... good fun :D |
| 23:56.11 | ``Erik | heh |
| 23:56.15 | ``Erik | to whom? |
| 23:56.35 | ``Erik | <steph> la tolteca card? <erik> lawhatnowhuh? |
| 23:56.50 | starseeker | that's how you know who the married/soon to be married folk are - they were the ones worried :-P |
| 23:57.04 | ``Erik | no, no I don't know :D |
| 23:57.36 | ``Erik | that's why I have 'nuff buck banked that I can just go buy a high end dressed out truck if I wanted, or another house O.o |
| 23:58.01 | ``Erik | live po', yo! |
| 23:58.04 | starseeker | heh - in this market, buying another house is less of a challenge than it used to be... |
| 23:58.17 | starseeker | what kinda truck you thinkin? |
| 23:58.20 | brlcad | ``Erik: that usually means you get in the door without a cover charge... :) |
| 23:58.23 | brlcad | you've been paying them? :) |
| 23:58.34 | ``Erik | cover what now? |
| 23:58.42 | brlcad | to get into la tolteca |
| 23:58.55 | ``Erik | I'm thinkin' about buying something like a ram1500 or an upper end dakota |
| 23:59.05 | ``Erik | there's no door fee at la tolteca, fool |
| 23:59.09 | brlcad | you're going to buy north dakota? |
| 23:59.11 | starseeker | no m35? :-P |
| 23:59.37 | ``Erik | no, a dakota, not north dakota... I want something that's actually useful for something |
| 23:59.43 | brlcad | ``Erik: then a vip card is about as useful as tits on a bull |
