--- Day changed Fri Apr 03 2009
00:12 -!- fenn_ is now known as fenn
01:14 -!- any91937210 is now known as useless2
01:28 -!- gene [n=chatzill@wireless-128-62-99-30.public.utexas.edu] has joined #hplusroadmap
01:30 -!- useless2 is now known as katsmeow-afk
01:53 < fenn> http://www.dealextreme.com/details.dx/sku.18431
01:55 < fenn> USB cigarette...
02:17 -!- any86271682 [n=someone@75-120-32-15.dyn.centurytel.net] has joined #hplusroadmap
02:20 -!- wrldpc [n=worldpea@pool-173-48-214-204.bstnma.fios.verizon.net] has joined #hplusroadmap
02:33 -!- katsmeow-afk [n=someone@75-120-31-43.dyn.centurytel.net] has quit [Read error: 110 (Connection timed out)]
03:09 -!- faceface [n=dbolser@bioinformatics.org] has joined #hplusroadmap
03:09 < faceface> hi 
03:10 < faceface> The House Mouse
03:10 < faceface> Atlas of Embryonic Development
03:10 < faceface> (free download) http://genex.hgu.mrc.ac.uk/Atlas/Theiler_book_download.html
03:10 < faceface> emap could be interesting for you 
03:10 < faceface> Computer modeling of stages of mouse embryogenesis to map the expression of different genes.
03:11 < faceface> I attended a lecture of foam, and wanted to use 'surface evolver' as a simulation of embryo genisis
03:12 < faceface> http://www.google.co.uk/search?q=foam+flies+and+functions+surface+evolver&hl=en
03:12 < faceface> Cell adhesion and cortex contractility determine cell patterning in the Drosophilaretina
03:13 < faceface> Because of the resemblance of many epithelial tissues to densely packed soap bubbles, it has been suggested that surface minimization, which drives soap bubble packing, could be governing cell packing as welln
03:13 < faceface> http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2141814
03:13 < faceface> http://genex.hgu.mrc.ac.uk/
03:13 < faceface> http://www.susqu.edu/facstaff/b/brakke/evolver/evolver.html
03:14 < faceface> Sorry if I'm about a year behind you guys (as it usually turns out)
03:47 -!- gene [n=chatzill@wireless-128-62-99-30.public.utexas.edu] has quit [Read error: 110 (Connection timed out)]
04:37 -!- elias` [n=me@62.78.208.74] has joined #hplusroadmap
04:51 -!- gene [n=chatzill@wireless-128-62-99-30.public.utexas.edu] has joined #hplusroadmap
07:43 -!- gene [n=chatzill@wireless-128-62-99-30.public.utexas.edu] has quit ["ChatZilla 0.9.84 [Firefox 3.0.8/2009032609]"]
08:42 < kanzure> huh, that's odd, that's actually something that I might have seen before re: surface minimization of bubbles and cells
08:43 < kanzure> ybit: 
08:43 < kanzure> oops, nevermind. false alarm.
08:53 -!- ybit [n=heath@unaffiliated/ybit] has quit [Remote closed the connection]
08:55 -!- samrose [n=samrose@adsl-99-147-180-206.dsl.lgtpmi.sbcglobal.net] has joined #hplusroadmap
09:21 < fenn> i should have done this a long time ago http://fennetic.net/pub/irc/battery-comparison.png
09:26 < fenn> i cant believe people actually use spreadsheets
09:26 < kanzure> yay for a battery comparison chart
09:26 < fenn> it's just barely workable
09:27 < kanzure> I'm still attempting to heal from the shock of learning that "most engineers just use spreadsheets"
09:28 < fenn> well, what else would they use? there's no good KBE tools
09:28 < kanzure> that's because there's no good knowledge base.
09:28 < faceface> kanzure: usefull?
09:28 < kanzure> ?
09:28 < fenn> true that
09:28 < faceface> foam links / surface evolver
09:28 < fenn> took 2 hours to assemble the battery data :\
09:29 < fenn> ideally every piece of data would have a url source
09:29 < fenn> can't figure out how to do that with a spreadsheet
09:29 < faceface> fenn: start a Semantic MW
09:29 < faceface> media wiki
09:29 -!- anthonyl [n=duzt@dsl093-216-054.aus1.dsl.speakeasy.net] has joined #hplusroadmap
09:29 < fenn> bleah
09:29 < faceface> bah!
09:29 < fenn> last thing i need is wiki bullshit
09:29 < faceface> heheh
09:29 < fenn> seriously, wiki is 90% display/format
09:30 < fenn> this is an associative database
09:30 < faceface> fenn smw is also data
09:30 < faceface> yes
09:30 < fenn> it's focused around building a "page" though
09:30 < kanzure> fenn: do you have the URL sources?
09:30 < fenn> i dont want a page, i want functions and views of data
09:30 < faceface> right, the page name is part of the triple
09:30 < fenn> kanzure: no, not really
09:31 < faceface> fenn: page <=> data, you need to accept that concept when building a DB in SMW, but its not that much of a stretch tbh
09:31 < kanzure> fenn: I think some of that data should be in our skdb metadata format. you can pre-emptively set the CAD to just be a cylinder from pythonOCC :-p
09:31 < fenn> faceface: i have no experience with SMW, only regular mediawiki
09:31 < fenn> so i dont know how it's really done
09:31 < kanzure> (there's an example script on pythonocc for generating cylinders, something like BRepPrim_API.BRepPrimAPI_Cylinder())
09:31 < faceface> fenn: you can start one up on refereata.com
09:31 < fenn> i dont want to start one up
09:31 < faceface> I have prodstat.referata.com and a couple others.
09:31 < fenn> show me something you consider 'good form' if you want to convince me
09:32 < kanzure> god, have I honestly memorized an OCC API call? that thing looks gross
09:32 < faceface> there are some better examples out there than mine.
09:32 < faceface> I don't want to convice you of anything, I just think SMW is going to be big
09:32 < fenn> kanzure: yes of course
09:32 < fenn> the API will change soon anyway to reduce redundundancy
09:32 < kanzure> "sit back and wait" never works
09:32 < kanzure> I don't know if that will actually happen, but I hope it will of course
09:33 < kanzure> how long did you spend assembling battery data?
09:33 < fenn> ^^ 2hr
09:33 < faceface> fenn: here is an example where I 'query' all countries in the triple store... http://prodstat.referata.com/wiki/Concept:Country
09:33 < kanzure> why did you give it to me in a png
09:33 < kanzure> :p
09:33 < fenn> i'm thinking about a few different things
09:33 < fenn> because i dont think everyone on the channel has kspread
09:33 < kanzure> kde spreadsheet?
09:34 < kanzure> what format is the output?
09:34 < fenn> .ods, .csv, gnumeric
09:34 < kanzure> I would be surprised if OpenOffice.org Calc does not import it
09:34 < faceface> here is an example data point http://prodstat.referata.com/wiki/DataPoint:0002:0116:031:1965
09:34 < kanzure> .csv would be nice
09:34 < fenn> i was surprised it didnt export .xls
09:34 < kanzure> .ods is OO.o Calc
09:34 < kanzure> right?
09:34 < faceface> This data point describes the Area Harvested for Potatoes in Afghanistan in the year 1965. 
09:34 < fenn> or at least some kind of plain .xml, not gzipped directory full of files .xml
09:35 < faceface> That text is generated from teh underlying data and associations
09:35 < fenn> ods is open document spreadsheet; there's a separate option for ooo specifically, not sure what the diff is
09:35 < kanzure> you know there's a command to check the diff
09:35 < kanzure> I think it's called `diff`
09:35 < faceface> try xmldiff
09:35 < fenn> hm
09:36 < kanzure> there's an xmldiff? does it do what it says it does?
09:36 < fenn> i don't really care that much as i hate spreadsheets
09:36 < fenn> i was hoping to find a simple way to convert to yaml
09:36 < faceface> I think it does what it can
09:36 < kanzure> fenn: well at least so that I don't have to type up the data if I want to go play around with it
09:36 < kanzure> fenn: oh, take the first row of column names and covert that to variable names in a python class
09:36 < faceface> i.e. send me a text file ;-)
09:36 < kanzure> so you can write a csv2python-class generator: the first row of a .csv file determines the members of the class
09:36 < fenn> kanzure: that's the simple way
09:37 < kanzure> what's wrong with it?
09:37 < kanzure> or, what's better?
09:37 < fenn> i really want to subclass the 18650 battery type and overwrite various values in specific instances
09:37 < kanzure> do you have the dataset describing the 18650 battery types?
09:37 < fenn> only a couple
09:37 < fenn> there are a zillion different batteries out there
09:38 < fenn> that all roughly correspond to the same type of battery compatibility
09:38 < fenn> maybe it's hopeless to try to classify them
09:38 < kanzure> battery compatibility does seem to be slightly more documented than other types of compatibilities
09:38 < fenn> instead do a compatibility checker function that looks at the actual data
09:39 < fenn> well, putting 3.6V 14500 lithium cells in a device for AA's would probably kill it
09:39 < fenn> they are exact same size you see
09:39 < fenn> (thats the point)
09:39 < kanzure> didn't know that. don't have a compatibility checker function.
09:40 < fenn> so, 18650's are much better in terms of Wh/kg Wh/$ and Wh/l, but they dont make battery holders for them
09:40 < fenn> weird eh
09:40 < fenn> i want AA form factor compatibility so i think i'm going to go with 14500 in the initial design
09:41 < fenn> it's probably a terrible mistake though
09:41 < kanzure> what are the AA/AAA battery connections? the ones with springs. does this have a special fit name?
09:41 < fenn> "battery contact"?
09:42 < fenn> |country code=2
09:42 < fenn> |item code=116
09:42 < fenn> |element code=31
09:42 < fenn> |year=1965
09:42 < fenn> |value=15000
09:42 < fenn> |value flag=* 
09:42 < fenn> faceface: are you fucking kidding me?
09:42 < kanzure> heh
09:43 < faceface> fenn: you add the data programatically, or through a form
09:43 < faceface> insert into table values () ()... 
09:43 < fenn> what's the point of using mediawiki then
09:43 < faceface> thats just as ugly 
09:44 < faceface> fenn: it can integrate your data with a remote ontology and publish it on teh semantic web
09:44 < faceface> also its a common platform for communty integration, i.e. using various 'results formats'
09:44 < fenn> how do i edit that datapoint "the right way"?
09:44 < kanzure> so I was putting some thought into an automatic density profile optimization partial differential equation function thingy for arbitrary 2D channel geometries; such that I can solve for a continuous function that outputs channel geometries. I'm thinking that the function that I want is a NURBS curve or Belzier curve that makes up a smooth curve plotted through a set of points. (and then the parallel side of the geometry is either the same geometry, or a straight line- I haven't decided). Then look for a 'target density profile' and do rating for the solutions.
09:44 < faceface> i.e. I can plot my prodstat data using googlecharts, serve it up as an RSS, query it off the semantic web ...
09:45 < kanzure> faceface: it's not 'typed'.
09:45 < faceface> fenn: click edit with form
09:45 < faceface> kanzure: its deeply typed, as in ontologically
09:45 < kanzure> only if you know the ontology
09:45 < faceface> fenn: click "edit with form"
09:45 < fenn> where is that?
09:46 < faceface> kanzure: the ontology can be imported into the wiki, and URI's are linked
09:46 < faceface> (I think... tbh I haven't got into teh ontolgy side yet)
09:46 < faceface> oh your right, the link is missing!
09:46 < faceface> one moment
09:47 < faceface> the link is missing from my site for technical reasons... 
09:47 < faceface> here is what it would look like http://prodstat.referata.com/wiki/Special:EditData/DataPoint/DataPoint:2:515:31:1968
09:47 < faceface> other sites are much better organized (wrt the data) than mine
09:47 < fenn> well that's even worse
09:48 < fenn> i'm expecting something like <country>Afghanistan</country>
09:48 < faceface> kanzure: the typing also happens here http://prodstat.referata.com/wiki/Property:Has_element_code
09:48 < kanzure> say, is there a way to get django to be running in the background constantly so that you don't have to reload imported libraries on each page call?
09:48 < faceface> fenn: sorry to dissapoint ;-)
09:48 < kanzure> (just thinking of pythonOCC+django.)
09:48 < faceface> fenn: check the RDF feed for the semantic web version
09:49 < faceface> the URIs can be configured to be more generic, but I haven't done that yet
09:49 < fenn> which URI?
09:50 < faceface> the URI in the RDF
09:50 < fenn> "http://www.w3.org/2001/XMLSchema#double"?
09:50 < faceface> the real semantic annotation happens here fyi, http://prodstat.referata.com/w/index.php?title=Template:DataPoint&action=edit
09:51 < faceface> fenn: I don't understand the question
09:51 < kanzure>  http://search.cpan.org/search?query=dependency&mode=all hrm. there's an Algorithm::Dependency module
09:53 < faceface> here is a slightly better organized SMW, with an example of analysis over the data... http://discoursedb.org/wiki/Topic_breakdown
09:54 < faceface> also check this if your interested http://discoursedb.org/wiki/Special:BrowseData
09:55 < faceface> but the point is that none of teh stuff I've been pointing you at is 'hard coded' into the wiki, its all user editable. So people can add and edit data, queries, forms etc. etc.
09:55 < kanzure> hm. http://pypi.python.org/pypi/topsort/0.9
09:55 < kanzure> code: http://www.radlogic.com/releases/topsort.py
09:55 < faceface> so once you have data for 50 batteries, other people may come to add more data 
09:55 < fenn> ok, discoursedb is structured a little better
09:56 < faceface> its kind of an exemplar site
09:57 < fenn> it makes my browser sad though :(
09:57 < faceface> too much js? it uses the yahoo library
09:57 < faceface> the 'semantic forms' SMW extension I mean
09:58 < fenn> it's not really clear where the info displayed on the page comes from
09:58 < faceface> also 'semantic drilldown' which tries to build a navigation system that is aware of the structure implicit in the data.
09:58 < faceface> there will be a MW template call somewhere down theres
09:58 < faceface> there
09:58 < faceface> which page?
09:58 < fenn> like on this page http://discoursedb.org/wiki/North_Korean_nuclear_crisis_/_Economic_sanctions_should_be_placed_on_North_Korea
09:59 < fenn> where do all these positions come from?
09:59 < faceface> OK, the text in that page is a 'template call' - kind of mehod call in MW
09:59 < kanzure> fenn, is your git repository on fennetic.net still? and if so where?
09:59 < fenn> kanzure: no, one moment
10:00  * fenn shifts desktops.. too much battery crap :)
10:00 < faceface> for the position method, see http://discoursedb.org/wiki/Template:Position
10:01 < faceface> it uses the magic word 'PAGENAME' so that putting the text {{Position}} in a different page will have a different output
10:02 < faceface> so the position template placed on the North Korean page produces output for North Korea
10:03 < faceface> the "{{#ask:" parts are queries for data
10:07 < faceface> I made the ask queries a bit easier to read... http://discoursedb.org/w/index.php?title=Template:Position&action=edit
10:08 < fenn> i need to do some kind of .htaccess directive to run the .cgi, but i forget how
10:08 < fenn> gitweb.cgi
10:08 < faceface> I think you need to set that globally
10:09 < faceface> won't work in .htaccess iirc
10:09 < fenn> i can't do global config
10:09 < faceface> perhaps the global config has that option enabled in .htaccess... can you read the global config?
10:09 < faceface> (I don't think its enabled by default)
10:09 < kanzure> it works in .htaccess IIRC
10:09 < kanzure> or you have to put it in /var/lib/cgi-bin/ or something silly like that
10:10 < kanzure> ah, /usr/lib/cgi-bin/
10:10 < fenn> i dont want to do that
10:13 < fenn> ahOptions +ExecCGI
10:13 < fenn> AddHandler cgi-script .cgi .pl
10:17 < fenn> ok, formattings kinda weird but gitweb works now
10:19 -!- willPow3r [n=will@66.75.6.181] has joined #hplusroadmap
10:33 -!- willPow3r [n=will@66.75.6.181] has quit [Remote closed the connection]
10:48 -!- willPow3r [n=will@cpe-66-75-6-181.san.res.rr.com] has joined #hplusroadmap
10:50 -!- samrose [n=samrose@adsl-99-147-180-206.dsl.lgtpmi.sbcglobal.net] has quit [Read error: 110 (Connection timed out)]
10:52 < kanzure> http://www.scribd.com/doc/6793556/Assembly-Mates milling machine assembly mates
10:52 < kanzure> http://www.me.metu.edu.tr/courses/me407/documents/Tutorials/SolidWorks_Assembly/Universal_Joint/START_TUTORIAL.htm
10:52 -!- samrose [n=samrose@adsl-99-147-180-206.dsl.lgtpmi.sbcglobal.net] has joined #hplusroadmap
10:55 < kanzure> http://www.theswgeek.com/2008/08/18/design-faster-with-design-library-and-mate-references-pt1/  so here they just use tagging. bleh.
10:55 < kanzure> there used to be part mating in OCC 5.1 or something.
10:55 < kanzure> the class no longer exists in the latest releases
10:58 < kanzure> say, http://designrepository.org/ works
10:58 < kanzure> oh. I've already ripped that dataset. drats.
11:05 -!- willPow3r [n=will@cpe-66-75-6-181.san.res.rr.com] has quit [Remote closed the connection]
11:06 -!- willPow3r [n=will@cpe-66-75-6-181.san.res.rr.com] has joined #hplusroadmap
11:08 < kanzure> hm, I like: http://dimacs.rutgers.edu/Workshops/CompAided/slides/shokoufandeh.ppt
11:09 < kanzure> non-ppt version: http://heybryan.org/~bbishop/docs/Scale-Space%20Representations%20and%20their%20Applications%20to%203D%20Matching%20of%20Solid%20Models.ppt.pdf
11:12 < kanzure> those algorithms can be implemented with OCC
11:16 < fenn> says "thanks to matra datavision/opencascade" on designrepository.org
11:17 < fenn> huh. Thomas Kramer is one of the guys who made EMC
11:23 < kanzure> designrepository.org is just Regli in disguise
11:23 < kanzure> where do you see the "thanks to" and Thomas Kramer?
11:24 < kanzure> http://edge.cs.drexel.edu/repository/old-web/frameset.html
11:25  * kanzure goes to look at a lot of screens for a few hours
11:26 < fenn> ya, there.
11:27 < fenn> have fun 
11:27 < kanzure> I think I will :)
11:28 < fenn> try maximizing a window with google earth
11:28 < kanzure> does Google Earth run under wine?
11:28 < kanzure> hrm, does wine work with x forwarding?
11:28 < fenn> uh, dunno
11:28  * kanzure leaves
11:31 < bkero> kanzure: yes
12:29 -!- Utopiah [n=libre@rps3868.ovh.net] has joined #hplusroadmap
12:35 -!- UtopiahGHML [n=libre@rps3868.ovh.net] has quit [Nick collision from services.]
12:35 -!- Utopiah is now known as UtopiahGHML
13:06 < anthonyl> i thought there was a linux google earth client now
13:11 -!- UtopiahGHML [n=libre@rps3868.ovh.net] has quit ["Changing server"]
13:12 -!- Utopiah [n=libre@rps3868.ovh.net] has joined #hplusroadmap
13:13 -!- Utopiah is now known as UtopiahGHML
13:21 < kanzure> well that was fairly neat
13:22 < kanzure> chromium + opengl + magic-carpet + kde
13:25 -!- xp_prg [n=xp_prg3@99.2.31.217] has joined #hplusroadmap
13:27 -!- willPow3r [n=will@cpe-66-75-6-181.san.res.rr.com] has quit [Read error: 110 (Connection timed out)]
13:35 < UtopiahGHML> (for "log of my life" fans : http://zealog.com/ posted in RWW)
13:48 < fenn> certainly more appropriate than twitter
13:48 -!- any86271682 is now known as katsmeow-afk
13:53 < fenn> could use an API
14:10 < kanzure> so the visualization lab is sort of the type where "you shouldn't reboot any of the machines because nobody knows how to set up the configuration the right way again"
14:11 < kanzure> but otherwise it's great.
14:37 -!- jm [n=jm@p57B9D289.dip.t-dialin.net] has quit [Read error: 60 (Operation timed out)]
14:37 -!- samrose [n=samrose@adsl-99-147-180-206.dsl.lgtpmi.sbcglobal.net] has quit [Remote closed the connection]
14:46 < kanzure> hm, so how am I supposed to get done 8 hours of matrix math in less than 2?
14:46 < kanzure> I guess I can get creative with the machines that I have ssh access to
15:44 -!- jm [n=jm@p57B9D289.dip.t-dialin.net] has joined #hplusroadmap
15:47 -!- jm|earth [n=jm@p57B9C30D.dip.t-dialin.net] has joined #hplusroadmap
15:57 < xp_prg> http://diybio.org/2009/04/02/diybio-san-francisco-glow-in-the-dark-1/
16:02 -!- jm [n=jm@p57B9D289.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)]
17:15 -!- ybit [n=heath@unaffiliated/ybit] has joined #hplusroadmap
17:25 < ybit> hah, what was the 5 second false alarm?
17:26 < kanzure> accidentally had your name in the queue and didn't actually type anything
17:28 < xp_prg> in tcp/ip what is a keep alive packet?
17:28 < xp_prg> is it just some random thing you send every so often?
17:38 < kanzure> http://tldp.org/HOWTO/TCP-Keepalive-HOWTO/overview.html
17:38 < kanzure> it's just the ACK flagged to on
18:03 < kanzure> http://gicl.cs.drexel.edu/wiki/William_C._Regli 
18:04 < kanzure> "
18:04 < kanzure> Regli's academic ancestry can be traced back to Leibniz, via Euler, the Bernoullis and Poisson."
18:04 < kanzure> blah. why can't *my* academic ancestry be traced back to Leibniz? waah.
19:09 -!- Fair [n=Nofaris@adsl-75-42-81-121.dsl.scrm01.sbcglobal.net] has quit [Read error: 104 (Connection reset by peer)]
19:14 -!- gene [n=chatzill@wireless-128-62-99-30.public.utexas.edu] has joined #hplusroadmap
19:16 -!- jm [n=jm@p57B9C30D.dip.t-dialin.net] has joined #hplusroadmap
19:17 < kanzure> fenn: http://packages.python.org/quantities/user/tutorial.html
19:17 -!- jm [n=jm@p57B9C30D.dip.t-dialin.net] has quit [Client Quit]
19:21 < kanzure> another one: http://juanreyero.com/magnitude/
19:35 -!- anthonyl [n=duzt@dsl093-216-054.aus1.dsl.speakeasy.net] has quit [Read error: 110 (Connection timed out)]
20:34 < ybit> at 11:44 http://www.ted.com/index.php/talks/pw_singer_on_robots_of_war.html
20:34 < ybit> what is that?
20:35 < ybit> it looks like something is being screwed into the guy's head
20:39 < ybit> http://www.gizmag.com/go/4004/ :: "The Land Walker: the world's first 340cm bipedal exoskeleton"
20:47 < ybit> wondered how i missed teh bipedal
20:49 < kanzure> is that Sacros?
20:52 < ybit> link 1 or 2?
20:55 < kanzure> um, neither, haven't checked them yet
21:00 -!- xp_prg [n=xp_prg3@99.2.31.217] has quit ["This computer has gone to sleep"]
21:06 < kanzure> http://opencascade.blogspot.com/ nice blog.
21:08 < gene> speaking of exoskeletons has anyone seen lockheedmartin's new exoskeleton?
21:11 < kanzure> no?
21:12 < gene> http://www.lockheedmartin.com/news/press_releases/2009/MFC_022609_LockheedMartinUnveilsExoskeleton.html
21:23 < kanzure> http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=103393 anyone ever play around with Reeb graphs?
21:30 < gene> http://www.pcbmotor.com/
21:30 < gene> check it out
21:30 < gene> flat motoro
21:31 < gene> http://155.69.254.10/users/risc/www/usm-intro.html
21:32 < gene> ultrasonic motor stages that are accurate to 50 nm
21:39 < kanzure> http://www.pascucci.org/topology/contour_tree/
21:40 < kanzure> http://graphics.cs.ucdavis.edu/~sdillard/libtourtre/doc/html/ contour tree library
21:58 < gene> hmmm...
21:59 < gene> how do they make piezoelectric elements?
22:05 < kanzure> they aren't elements. they are crystals.
22:05 < kanzure> also, piezoelectricity is very close to god's gift to man
22:06 < kanzure> fenn: I'm thinking there's more to Regli's work than I originally thought- the Reeb graph stuff is impressive
22:07 < gene> indeed they are
22:07 < gene> link to diy electron microscope that used a piezo buzzer?
22:07 < gene> I'd like to make a nanometer accurate positioning stage
22:08 < gene> that is cheap to make
22:09 < gene> I want damn tiny milling machine
22:11 -!- jm|earth [n=jm@p57B9C30D.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)]
22:14 < kanzure> ".. and is defined as the sum of the geodesic distances g(v, p) from v to the other points {p|p in S}: integral of g(v,p)dS (v is a point in shape S)
22:14 < kanzure> did they fail their latex or do they really mean sum and then mean to show an integral?
22:18 < kanzure> oh, a geodesic distance integral
22:33 -!- gene [n=chatzill@wireless-128-62-99-30.public.utexas.edu] has quit [Read error: 104 (Connection reset by peer)]
22:37 < kanzure> hahah
22:37 < kanzure> blender implements reebgraphs
22:37 < kanzure> yay
22:37 < kanzure> http://www.google.com/codesearch/p?hl=en#bl6RANvgn8I/trunk/source/blender/src/reeb.c&q=%22reeb%20graph%22
22:40 < kanzure> bwahah: https://svn.blender.org/svnroot/bf-blender/branches/harmonic-skeleton/release/scripts/reeb.py
22:40 -!- gene [n=chatzill@wireless-128-62-99-30.public.utexas.edu] has joined #hplusroadmap
22:45 < kanzure> http://blenderocv.googlecode.com/svn/trunk/source/blender/src/reeb.c
22:55 < kanzure> doesn't seem to be the same type of Reeb graph
23:19 < fenn> reeb graph is a way to represent topology?
23:24 < fenn> pq looks halfway decent
23:28 -!- gene [n=chatzill@wireless-128-62-99-30.public.utexas.edu] has quit ["ChatZilla 0.9.84 [Firefox 3.0.8/2009032609]"]
23:56 < kanzure> reeb graph is apparently the way that blender generates skeletons
23:57 < kanzure> regli was using reeb graphs for part matching, similarity search and apparently part mating (I guess symmetry in the reeb graph would indicate a way to connect the two graphs together?)
23:58 < kanzure> gene- that piezo thingy was just a buzzer from a headphone cut into four sections with a knife.