2009-09-01.log

--- Day changed Tue Sep 01 2009
-!- wrldpc2 [n=benny@ool-ad03fe34.dyn.optonline.net] has joined #hplusroadmap00:15
-!- fenn [n=fenn@cpe-72-177-121-73.austin.res.rr.com] has quit [Read error: 104 (Connection reset by peer)]00:34
-!- fenn [n=fenn@cpe-72-177-121-73.austin.res.rr.com] has joined #hplusroadmap00:38
-!- mason_l [n=x@202-89-188-136.static.dsl.amnet.net.au] has joined #hplusroadmap00:57
-!- mason-l [n=x@202-89-188-136.static.dsl.amnet.net.au] has quit [Nick collision from services.]00:58
-!- mason_l is now known as mason-l00:58
-!- splicer [n=patrik@h55n1c1o261.bredband.skanova.com] has quit ["Ex-Chat"]02:18
Utopiahregarding previous discussion on a standard research format, I guess the pharma industry won't let that, especially with public experiment data, happen ( watching http://www.onlinedocumentaries4u.com/2009/08/money-talks.html that made me think of that)04:44
Utopiahcan't even get a standard citation format anyway :/04:45
-!- davidnunez [n=davidnun@209-6-203-217.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com] has joined #hplusroadmap04:45
kanzureUtopiah: bibtex is a suitable standard citation format07:19
-!- goonie_ [n=goonie@neuroblastoma.cs.pdx.edu] has joined #hplusroadmap07:48
Utopiahyep and dominating http://en.wikipedia.org/wiki/Comparison_of_reference_management_software#Import_file_formats07:50
-!- goonie [n=goonie@neuroblastoma.cs.pdx.edu] has quit [Read error: 110 (Connection timed out)]08:00
-!- davidnunez [n=davidnun@209-6-203-217.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com] has left #hplusroadmap []08:22
-!- Phreedom [n=quassel@195.216.211.175] has quit [Read error: 54 (Connection reset by peer)]08:35
-!- Phreedom [n=quassel@195.216.211.175] has joined #hplusroadmap08:35
kanzurewhois goonie?09:12
-!- mason-l [n=x@202-89-188-136.static.dsl.amnet.net.au] has quit [Read error: 113 (No route to host)]09:17
-!- mason_l [n=x@202-89-188-136.static.dsl.amnet.net.au] has joined #hplusroadmap09:17
-!- mason_l is now known as mason-l09:27
kanzureshould an instruction object be a list of strings representing each step (for the human readable version)? or should it be a list of objects that can be called to generate those strings?09:42
fenni imagine you'd go through one to get the other09:44
fennsecond one sounds more flexible09:45
kanzureyes but then I get into this fishy issue where I know that, say, one step is to align two objects into some relative position09:47
-!- davidnunez [n=davidnun@209-6-203-217.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com] has joined #hplusroadmap09:47
kanzuredoes this mean I need an Alignment step object?09:47
-!- superkuh [n=hukrepus@unaffiliated/superkuh] has quit [Read error: 113 (No route to host)]09:48
fennOO doesn't work with instructions really09:48
fennyou want an alignment function (method really)09:48
fennrobot.align(stuff)09:48
fennhuman.align(stuff)09:48
kanzureyes but you're not supposed to care if it is a robot or not09:48
kanzureAlign.robot()09:48
kanzureer09:49
fennno09:49
kanzureAlign.tool(robot)09:49
fennno09:49
fennbad programmer09:49
fennno wonder you hate OO09:49
fennobjects should correspond to physical objects09:49
kanzurerobot.align is better09:49
kanzurebut why should "align" be a function?09:50
kanzurewe want a function for every possible step?09:50
fennyes, then it's a simple mapping from protocol to functions to output09:50
fennwhat else would instructions be for?09:50
fennlist of nouns != instructions09:50
-!- kardan| [n=kardan@p54BE393E.dip.t-dialin.net] has joined #hplusroadmap09:53
kanzurehuman.push() and robot.push() might work in some instances, but when you have a weird tool, "push" might be an ambiguous command .. but not ambiguous to the human object09:55
kanzureso that's why this seems kind of confusing to me09:55
kanzuredoesn't seem optimal09:55
fenntry: agent.push(); except StupidityError: human.push()10:02
fennwhy would it be ambiguous?10:03
-!- |kardan| [n=kardan@p54BE2D6E.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)]10:09
kanzurebecause maybe the machine doesn't have a "push" but instead has these weird servo arm mechanisms that don't quite "push" but could be used to push10:13
-!- superkuh [n=hukrepus@unaffiliated/superkuh] has joined #hplusroadmap10:20
kanzureassembly instructions are not quite the same thing as manufacturing instructions I guess10:43
fennthey are10:45
fennyou can split it up any way you want10:45
kanzuremachining some metal doesn't follow the same form as assembling two legos10:45
fennwhat do you mean 'follow the same form'?10:47
fennthey're both procedures with parameters10:47
kanzureso what if they are procedures?10:47
fennmachine this block with this tool, connect this block to this block10:47
kanzurewhat actually matters are the tools used to do stuff10:47
fennprocedure is how you tell the agent what to do10:48
kanzurethere isn't necessarily an agent10:48
fennwhere agent in {human, collection of robots}10:48
kanzureis a kinematic linkage a robot?10:48
fennof course there's an agent10:48
fennok say you have a stamping press10:48
fennsomeone has to set up the press right? even if it just runs automatically10:49
kanzurethat's a one time setup (hopefully)10:49
fennthey also have to feed it metal, and stop it when it's done10:49
kanzureok?10:49
kanzureall of the previous 'research' on assembly instructions only considers things being mated together10:50
fennwell, yeah, it's lame10:50
fennwelding is not generally considered "assembly" i guess10:51
-!- wrldpc2 [n=benny@ool-ad03fe34.dyn.optonline.net] has quit []10:54
fennis gluing ever considered "assembly"?10:54
fennfwiw OCC recognizes these names http://en.wikipedia.org/wiki/X11_color_names10:57
kanzuresure what about gluing school project models?10:57
fennthat's "craft" not assembly10:57
kanzurecraft =~ crap?10:57
fenni didn't say it :)10:57
fennigraph is weird but i can't think of a better way to do stuff so i guess i like it11:00
drazak_kanzure: who the fuck is eugen?11:05
kanzureeugen is my alter german ego11:05
kanzurehttp://eugen.leitl.org/11:05
kanzurehe worked on a mind uploading device back in the late 90s11:05
kanzurecurrently trying to set up a mobile cryogenics lab11:06
kanzurehttp://web.archive.org/web/20080203123957/http://minduploading.org/research.html11:06
kanzurenematode upload project "A collaboration led by Eugene Leitl. Its aim is to achieve the 3D reconstruction and functional emulation of individual specimens of C.elegans from scanned images of the vitrified original. "11:07
kanzuredata: http://web.archive.org/web/20070814044636/minduploading.org/research/data.nematodeupload.html11:07
parolangIf it helps any, cutting is a different operation than assembling, and both are rather abstract operations and are both involved in manufacturing.  You might want to overload the subtraction and addition operators to represent cutting and assembling, respectively.11:22
parolang(Or if you don't like to call gluing parts together "assembly" you can use "combining" instead.)11:25
kanzurewhy aren't there any papers on automatically generated instructions for more than assemblies?11:31
kanzureI mean, more than part mating11:31
kanzurethere's a field of "process planning" but that's usually just trying to figure out what operations to use11:32
kanzurenot necessarily how to represent those operations or anything11:32
kanzureah maybe "construction" should be the operative keyword here11:34
kanzureIIRC, there's a journal called "Automation in Construction"11:35
fennparolang: we've had this discussion many times.. i don't think overloading + and - is worthwhile. they don't contain enough information to do anything useful11:45
parolangAlright.  I thought I was saying something new :)  I think I'm just fond of the kind of operator overloading of boolean logic, and thought this was a natural extension of that.11:47
fennalright why can't i run any OCC GUI stuff now12:27
fennraise "Please set the CSF_GraphicShr environment variable."12:27
fennTypeError: exceptions must be classes or instances, not str12:27
fenni understand the error but i don't know why i never got it before12:27
fennweird.12:29
kanzure['align interface stud along the mating axis defined by interface anti stud', 'push anti stud with 10 N along Vector(0.0, 1.0, 0.0)']12:31
kanzureCSF_GraphicShr just randomly became unset?12:31
kanzureis it set to anything? echo $CSF_GraphicShr12:31
CIA-32skdb: kanzure * r c5f601d /doc/proposals/techniques.py: some really bad OOP for treating techniques as objects and spitting out instructions12:35
fennat least it's in doc/proposals/ so i can just ignore it :)12:39
-!- davidnunez [n=davidnun@209-6-203-217.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com] has quit []12:45
-!- genehacker [i=genehack@wireless-128-62-47-18.public.utexas.edu] has joined #hplusroadmap12:45
kanzureit works doesn't it?12:46
kanzurewhat's so bad about it?12:46
* kanzure questions the existence of the witch of agnesia12:50
-!- genehacker [i=genehack@wireless-128-62-47-18.public.utexas.edu] has quit [Read error: 60 (Operation timed out)]13:04
-!- genehacker [i=genehack@wireless-128-62-168-43.public.utexas.edu] has joined #hplusroadmap13:23
fennhttp://adl.serveftp.org/lab/fenn/pngs/connection_graph.png13:24
drazak_kanzure: so eugen is cool if a little impolite seeming?13:24
drazak_kanzure: I basically just pasted a protocol from current protocols in molecular biology to the mailing list, oh well13:27
kanzurefenn: what does that png show?13:38
kanzureyes eugen is "cool"13:39
kanzureok sorry didn't see the other side of the image13:42
-!- splicer [n=patrik@h55n1c1o261.bredband.skanova.com] has joined #hplusroadmap13:59
-!- genehacker [i=genehack@wireless-128-62-168-43.public.utexas.edu] has quit [Read error: 60 (Operation timed out)]14:01
kanzureafter analyzing genehacker's timeout log, i've concluded that he has absolutely no pattern whatsoever in his class schedule14:16
fenntimeout is usually just bad wifi14:21
fennor is it because he's closing his laptop or something?14:21
kanzurehe's closing his laptop14:23
CIA-32skdb: fenn * r d1c4eb5 /geom/geom.py: mate_connection should have no side effects14:26
CIA-32skdb: fenn * r ddc9e3a / (core/interface.py paths.py): start tracking the connection graph14:26
kanzurewhy are you dumping graphviz instead of yaml?14:31
fennyaml sucks at graphs, you said so yourself14:42
kanzureare we really sure we want packages to save their data in graphviz?14:42
fennno14:42
fenni just havent figured out a good way to do it yet14:42
fennbricks don't have identifiers is part of the problem14:43
kanzuremaybe we should give them better names14:43
fennthe names are fine14:43
kanzurea name isn't an identifier?14:43
fennyou can have multiple instances of each brick14:44
fennto make a graph you have to refer to the instance14:44
kanzureright14:44
fennpython has a hash function, so..14:44
fennuh, it's different for every instance14:44
fennor we could just trust yaml to keep track of its pointers14:44
kanzureI tried dumping graphs in yaml before 14:44
kanzureand it just looked ugly14:44
kanzurebut it worked14:45
kanzurenot like I know anyone who modifies graphs by hand anyway14:45
CIA-32skdb: fenn * r 2aa4a1b / (geom/geom.py paths.py): move coincidence_fixer from paths to geom, only connect compatible interfaces14:46
CIA-32skdb: fenn * r a266c7d / (core/part.py geom/geom.py): go away14:46
fennigraph isn't yaml-friendly at all14:48
fenni guess i could just save all the positions and then detect coincidences :P14:53
-!- davidnunez [n=davidnun@209-6-203-217.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com] has joined #hplusroadmap15:30
kanzureheh the guy who made haas' first product has a wikipedia user account: http://en.wikipedia.org/wiki/User:Kurt_Zierhut15:44
-!- kso512 [n=kso@cpe-66-68-153-104.austin.res.rr.com] has joined #hplusroadmap15:50
kso512hi15:50
kanzurehey kso512 15:51
kso512hiya...  saw you over on #austin, read some of your site, and am interested in this whole "apt-get sandwich" idea...15:55
kso512not sure how i can help, but i have experience with linux & circuit design...  interested in fabbing ideas15:56
CIA-32skdb: fenn * r 33a08db /gui/gui.py: colored interface arrows15:56
CIA-32skdb: fenn * r 4d185f8 /core/interface.py: igraph del_part16:05
CIA-32skdb: fenn * r d0df82d / (geom/geom.py paths.py): try to detect mismatched interfaces. doesnt work yet16:05
kanzurekso512: that's great :)16:08
kanzurefenn just committed some code to that btw16:08
kanzureI'm sitting around on the couch in the automated design lab16:08
fenndunno why i didn't think of it before.. i can sort the xyz coordinates of interfaces, then compare only those in a certain set of slices17:00
xp_prgI think I am going to die, Drew Endy may come to my DIYBIO meeting at the Tech Shop!!!17:01
fenni hope he slaughters you17:02
xp_prgit would be my honor to be slaughtered by him17:03
* xp_prg places his neck on the chopping block for Drew Endy17:04
-!- wrldpc2 [n=benny@ool-ad03fe34.dyn.optonline.net] has joined #hplusroadmap17:04
ybitgrrs i want a full cad file for a car17:37
ybita cad file with all car parts*..17:38
ybithttp://72.22.64.51/cgi-bin/csupload/CSUpload.cgi?database=Cars%203D.db&command=viewupload&id=null maybe...17:39
ybitit is occ 6.3 being used in skdb, si?17:49
fennyes because that's what pythonOCC wants18:13
-!- mason-l [n=x@202-89-188-136.static.dsl.amnet.net.au] has quit [Read error: 110 (Connection timed out)]18:13
-!- mason-l [n=x@202-89-188-136.static.dsl.amnet.net.au] has joined #hplusroadmap19:07
-!- davidnunez [n=davidnun@209-6-203-217.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com] has quit []19:15
-!- davidnunez [n=davidnun@209-6-203-217.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com] has joined #hplusroadmap19:19
fennthat's a lotta smoke http://www.wildfiretoday.com/storage/post-images/Station_Fire_2.jpg http://www.wildfiretoday.com/storage/post-images/Smoke_Map_8-31-2009.jpg19:21
drazak_kanzure: didn't you rattle on one day about TCMS?19:47
-!- genehacker [i=genehack@wireless-128-62-62-60.public.utexas.edu] has joined #hplusroadmap19:51
kanzuredrazak_: yes?19:53
kanzuredrazak_: sorry, I already replied to that email19:54
genehackerdoes anyone know if peptide microarrays are good for anything?19:56
kanzurepeptide affinity?19:57
genehackerwhat's that?19:58
kanzurefiguring out what sticks to peptides19:58
genehackerok19:58
-!- genehacker2 [i=genehack@wireless-128-62-62-60.public.utexas.edu] has joined #hplusroadmap19:59
genehacker2and NPPOC protected peptosynthesis might be easier than DNA synthesis19:59
kanzurewhy do you want to do peptosynthesis?19:59
genehacker2well possibly for synthesizing antibodies for proteins we want20:01
genehacker2possibly for testing if something is present in a sample20:01
genehacker2also for lulz20:01
genehacker2it would allow us to test if that maskless lithography thing works20:02
kanzurehow about a microscope20:03
genehacker2that's what I want to test20:04
genehacker2http://www.evitamins.com/product_grid2.asp?catID=4920:06
genehacker2wonder if these are pure enough or cheap enough to consider using...20:06
-!- genehacker [i=genehack@wireless-128-62-62-60.public.utexas.edu] has quit [Read error: 110 (Connection timed out)]20:17
genehacker2hmmm....20:32
genehacker2going to mars is expensive because of the cost of getting a heavy spaceship with rad shielding into orbit20:33
katsmeow-afkuse a damned comet :-/20:33
katsmeow-afkit will supply you with water on the way20:33
genehacker2yeah20:34
katsmeow-afkuse a hunk of the moon, there's enough mass there to spare some20:34
genehacker2or use a ram accelerator, whatever it's called to shoot the shielding into orbit20:35
genehacker2after all ramaccelerators aren't really suitable for humans20:35
katsmeow-afkthe olde Orion project, but use neutron bombs20:35
genehacker2hardly practical for shooting sats anyway20:36
kanzureHow about you build the shielding in orbit?20:36
kanzurewhat the fuck is wrong with you people?20:36
genehacker2but water or some other form of rad shielding20:36
katsmeow-afkdidn't i say that?20:36
genehacker2from what?20:36
kanzurekatsmeow-afk: yes but you're not a person20:36
katsmeow-afk[20:28] <katsmeow-afk> use a damned comet :-/20:36
katsmeow-afk[20:29] <katsmeow-afk> it will supply you with water on the way20:36
kanzureyes but nobody listens to you20:36
katsmeow-afk[20:29] <katsmeow-afk> use a hunk of the moon, there's enough mass there to spare some20:36
katsmeow-afkmy appologies20:37
kanzure:)20:37
kanzureno worries. nobody listens to me either.20:37
genehacker2the ram accelerator thing is one of the cheapest options20:37
kanzurecheap doesn't matter IMHO. you get a lot of benefit for being able to do manufacturing in space.20:37
katsmeow-afki don't think i would20:37
genehacker2taking out hunks of moon requires building an accelerator on the moon20:37
kanzurealso why would you pay for the moon?20:37
kanzureno, you wouldn't have to build an accelerator20:38
kanzureyou don't need an accelerator to dig on the earth do you?20:38
genehacker2yeah you don't20:38
genehacker2you want to assemble ships in orbit20:38
katsmeow-afkthe moon is a much smaller gravity well20:38
genehacker2from hunks of moon20:38
genehacker2you need to get those hunks of moon in orbit somehow20:39
katsmeow-afkno, you wanna assemble the rad shield from hunks of moon20:39
katsmeow-afknot the ship itself20:39
genehacker2that requires going to the moon20:39
genehacker2a prohibitively expensive option20:39
genehacker2though I wonder20:39
genehacker2if there is enough junk up there in orbit to make a simple radiation shield20:39
genehacker2or at least something semi useful20:39
katsmeow-afkthe current space station will be junk in 10 yrs, use it20:40
genehacker2good idea20:40
katsmeow-afki gotta afk some more, this level of afk isn't enough, bbl20:41
genehacker2hell why not put the shuttle fuel tank into orbit20:41
genehacker2it wouldn't take much20:41
genehacker2I forget the exact figure20:42
kanzureI need more torrents20:42
genehacker2but if one get's it going just a bit faster at the drop off point we could put one in orbit20:42
kanzureany suggestions?20:42
genehacker2you lucky bastard20:43
genehacker2I can't torrent20:43
genehacker2but I can suggest some good stuff that might be worth torrenting20:43
genehacker2what do you want to torrent?20:43
kanzureI just need some trackers20:43
genehacker2would you like a legal torrent to a good videogame?20:44
genehacker2damn20:46
genehacker2shuttle tank weighs 26,500 kg and is made of mostly a lightweight aluminium lithium alloy20:47
genehacker2http://spacestudiesinstitute.wordpress.com/space-art/ssi-sample-slides/20:49
genehacker2here it is20:49
genehacker2http://www.astronautix.com/craft/stsation.htm20:51
genehacker2paper here20:51
-!- superkuh [n=hukrepus@unaffiliated/superkuh] has quit [Read error: 110 (Connection timed out)]21:12
ybithttp://tangible.media.mit.edu/projects/ :: Tangible Bits is our vision of Human Computer Interaction (HCI) which guides our research in the Tangible Media Group.21:34
kanzuretell smari :)21:44
kanzureironically fablab is technically a part of the media lab21:45
kanzureso he should know about that (or at least hypothetically know about that)21:45
kanzurehttp://heybryan.org/~bbishop/docs/rss/torrents/21:46
kanzurethere's some porn torrent trackers in there, fyi21:46
kanzureI parsed a list of torrent trackers and their rss feeds, so there's bound to be all sorts of stuff in there21:47
kanzurehaven't looked at it all yet21:47
-!- drazak_ [n=drazak@drazak.net] has quit [Read error: 110 (Connection timed out)]22:30
kanzuredoes anyone remember how to rip mms streams?22:48
genehacker2mms?22:48
kanzuremms://coe-jaguar-video.engr.utexas.edu/mediasitevideo50/13e474ea-d54e-43df-a4ab-568978846c97.wmv?playbackTicket=5a1d99a2a6a544cd939aee7bd16fed9022:48
kanzureyes22:48
kanzureaha22:48
kanzuremplayer mms://url -dumpstream -dumpfile foo.wmv 22:48
genehacker2why are you dumping to .wmv?22:49
kanzuredoesn't matter22:49
kanzureanyway, that's how you download class videos22:49
genehacker2for a class you're taking?22:52
genehacker2heh22:52
genehacker2at least you don't have a DRM'd textbook22:52
kanzureso the trick seems to be:22:53
kanzuregrep for the link that has FileServer in it22:53
kanzurego to that page22:53
kanzuregrep for the link that has "mms://" in it22:53
kanzurethen fuck the utexas server22:53
genehacker2btw if you can write me a program that prints 2 pages at a time into a pdf and then prints the next page I'll give you a cookie22:54
genehacker2from the DRM'd book22:54
kanzureI don't understand what you want22:54
kanzureprints 2 pages?22:54
kanzureso, prints into a console?22:54
genehacker2the book only lets me print 2 at a time22:54
genehacker2annoying for a human but not for a program22:55
-!- superkuh [n=hukrepus@unaffiliated/superkuh] has joined #hplusroadmap22:57
-!- genehacker2 [i=genehack@wireless-128-62-62-60.public.utexas.edu] has quit [Read error: 60 (Operation timed out)]23:34
-!- genehacker [i=genehack@wireless-128-62-62-60.public.utexas.edu] has joined #hplusroadmap23:35
genehackerinteresting23:37
genehackerthe color camer in my laptop is 640 X48023:37
genehackerand the image quality isn't crap23:38
genehackeryet it's small enough to consider putting into ar glasses23:39

Generated by irclog2html.py 2.15.0.dev0 by Marius Gedminas - find it at mg.pov.lt!