| --- Day changed Sun Sep 14 2008 | ||
| kanzure | No. | 00:02 |
|---|---|---|
| kanzure | How many can there be? 100? 200? | 00:02 |
| kanzure | http://www.mit.edu/people/cdemello/es.html | 00:02 |
| kanzure | You're in luck. :) | 00:02 |
| ybit | i found this http://en.wikipedia.org/wiki/List_of_universities_in_Spain earlier but mit's list might be better | 00:24 |
| ybit | dunno yet | 00:25 |
| bkero | http://digg.com/comedy/How_to_protect_your_car_in_a_hurricane_PIC | 00:29 |
| fenn | certain death | 00:39 |
| ybit | heh | 00:39 |
| kanzure | http://www.pythoncad.org/ | 00:55 |
| kanzure | bleh | 00:56 |
| kanzure | yet another one man team | 00:56 |
| fenn | dead-ish project too | 00:59 |
| fenn | the code was awful, imho | 00:59 |
| fenn | wildcat cad looks interesting, though the dev seems to be sort of obsessed with premature optimization | 01:02 |
| fenn | http://code.google.com/p/wildcat-cad/ | 01:02 |
| kanzure | same problem as browsers | 01:04 |
| kanzure | everything thrown into a single package | 01:04 |
| kanzure | needs to be a little catalysis | 01:04 |
| kanzure | small, independent scripts to make up the functions | 01:04 |
| kanzure | then let somebody else bother with fancy gui bullshit | 01:05 |
| fenn | eh, i guess | 01:05 |
| fenn | brlcad is using that strategy and still there's nothing useful after 3-30 years | 01:05 |
| kanzure | wtf? | 01:05 |
| fenn | i mean brlcad is great if you're doing nuclear tank warfare simulations | 01:05 |
| * kanzure looks | 01:05 | |
| kanzure | it's a simulator? | 01:06 |
| fenn | it's a solid modeler | 01:06 |
| fenn | but the GUI is atrocious | 01:06 |
| fenn | anyway, brlcad is based on the unix philosophy of a million little commands | 01:06 |
| kanzure | does it not perform well or something? | 01:06 |
| fenn | uh, it's just hard to learn, and it doesn't really do "design" | 01:07 |
| kanzure | 'BRL-CAD is a collection of more than 400 tools, utilities, and applications comprising more than a million lines of source code.' | 01:07 |
| kanzure | million lines of code for what? :/ | 01:07 |
| fenn | n-dimensional ray tracer, numerical utilities.. | 01:07 |
| kanzure | 'photon mapping support for realistic images' hahah :) | 01:07 |
| fenn | they do stuff like calculate the x-ray flux to various parts of a person when a nuclear bomb goes off at such-n-such distance | 01:08 |
| fenn | said person being inside a tank or a battleship or whatever | 01:10 |
| fenn | so they have to make models of the tank and all its materials | 01:11 |
| fenn | but they dont actually design the tanks with brlcad | 01:11 |
| fenn | it's been open source for like 3 years, but under development for 30-odd years | 01:12 |
| kanzure | so why would you say they're using that strategy and not getting anything useful if you know they aren't aiming for usable stuff? | 01:12 |
| fenn | "let somebody else bother with the fancy GUI bullshit" doesnt mean anybody will actually bother with it | 01:13 |
| kanzure | wrt brlcad or in general | 01:14 |
| kanzure | because I'm good with cli gcc | 01:14 |
| kanzure | ide is a nice bonus | 01:14 |
| fenn | errr... all i gotta say is command line cad doesn't work so well | 01:16 |
| kanzure | why? consider this argument: | 01:16 |
| kanzure | you're going to be using parts or manipulations to designs that the program has in the db anyway | 01:17 |
| kanzure | so you either know it or you don't, no? | 01:17 |
| fenn | say that again? | 01:17 |
| kanzure | you know how I was talking about repairing a user's requirements when they start typing and clicking in their interface? | 01:19 |
| kanzure | the reason it has to be 'repaired' is because we're translating from la la land in their personal semantic forest over to what the db knows about | 01:19 |
| kanzure | so if you know the items in the db, how they interconnect and so on, or how to generate a flat surface with something attached to it etc., what's the problem with using the cli like that ? | 01:20 |
| kanzure | fenn: did I kill you? :( | 01:32 |
| fenn | eh sorry i got distracted and forgot | 01:33 |
| fenn | i'm just saying it's hard to specify "click here" with a command line | 01:34 |
| fenn | like, this particular edge/face/vertex | 01:34 |
| kanzure | agreed | 01:34 |
| kanzure | but is 'click here' the Right way to do it anyway? | 01:35 |
| fenn | and nobody likes naming every single primitive, so it gets nasty when you have a complex shape and have to figure out which cylinder you are looking at | 01:35 |
| kanzure | I think it would be ok to sacrifice visuals for automatically generated structures, with validation and so on, unless you're specifically designing a final product skin or something, and for that, by all means go do some blender or something | 01:36 |
| fenn | most engineers learn how things work by playing around with them, holding in their hands and moving things back and forth | 01:39 |
| fenn | doing that with a command line requires a whole extra layer of abstract thought | 01:39 |
| fenn | it's "counter-intuitive" | 01:40 |
| fenn | or maybe just not intuitive | 01:40 |
| kanzure | hm, is counterintuitions offensive intuitions or just not intuitions? | 01:40 |
| kanzure | *are | 01:40 |
| fenn | anyway, there's nothing wrong with specifying the constraints via cli, but you _need_ a GUI to see wtf you have in the first place | 01:40 |
| kanzure | no argument there | 01:40 |
| fenn | otherwise it's just a bunch of boxes with lines between them | 01:40 |
| kanzure | what? | 01:41 |
| kanzure | wouldn't the gui show that sometimes too? | 01:41 |
| kanzure | or how would a cli show boxes and lines? I guess if you use ncurses | 01:41 |
| fenn | you know the automated design lab thing | 01:41 |
| kanzure | yes | 01:41 |
| fenn | that's not a GUI | 01:41 |
| fenn | it's boxes with lines between them | 01:41 |
| fenn | wah | 01:41 |
| kanzure | the uml-like graph stuff? sure | 01:41 |
| fenn | labview is also boxes with lines between them, but you can grab the boxes and lines and shake them around and make it do stuff | 01:42 |
| kanzure | still haven't found an open source library for doing stuff like that - even though graphviz should have figured it out a long time | 01:42 |
| kanzure | reasoning about designs on both a geometric, measurement-restricted, and more importantly material basis in the cli is definitely counter/nonintuitive. | 01:43 |
| fenn | graphviz is for diagrams for your peer-reviewed journal article, submitted in LaTeX format | 01:43 |
| kanzure | in fact, reasoning about materials is completely nonintuitive | 01:43 |
| kanzure | except crystallography, which if you look at it is kind of basic "uh, well, you have a force here, so what the hell do you think a huge force is going to do to this material?" | 01:44 |
| fenn | uh, i think you're way off base | 01:44 |
| fenn | nobody goes and looks up the coefficient of blahdyblah every time they pick a material for a part | 01:44 |
| fenn | they have experience with these materials and build up an intuition of how they perform | 01:44 |
| fenn | if it's a critical assembly in some instrument, then yeah you will worry about exact numbers and how the material behaves at extremes | 01:45 |
| fenn | but for most cases you use what you're familiar with | 01:45 |
| fenn | or else there is lots of trial and error, assisted with plenty of FEA | 01:46 |
| kanzure | methinks that trial and error occurs in either cases .. | 01:46 |
| fenn | sure | 01:46 |
| kanzure | or maybe that's why metals are so popular | 01:46 |
| kanzure | not only are they somewhat abundant, but if you reason with them "well, it's strong" then you're good | 01:46 |
| kanzure | and your bases are generally covered | 01:46 |
| kanzure | ok | 01:46 |
| fenn | it's very easy to just go with 6061 aluminum for everything | 01:46 |
| kanzure | but I'm not happy about trusting so much in user intuitions | 01:47 |
| kanzure | wasp nest / open source incrementalism versus -- oh crap, | 01:47 |
| kanzure | what was that wolfram quote earlier today? | 01:47 |
| kanzure | what would wolfram say about the iterative development of a wasp's nest? | 01:47 |
| kanzure | particularly the huge one that was linked to in here a few weeks ago | 01:47 |
| fenn | uh, beehives were one of the few natural systems that satisfied a simple system of constraints? | 01:47 |
| fenn | what with the regular hexagons and all | 01:48 |
| kanzure | cli reasoning about design would only make sense if you're doing more than just geometry/shapes because otherwise just use a 3D GUI playhouse | 01:50 |
| kanzure | unless you don't care about the final shape of course | 01:50 |
| kanzure | which is fine too, but the final shape stuff is only half the design problem | 01:50 |
| kanzure | (the other half is functionality of the components and so on) | 01:50 |
| fenn | and how to make it | 01:51 |
| kanzure | another paper constrained the user's choices for designs to only those things the program knew the cnc router could cut out :) | 01:51 |
| fenn | i liked the nist paper where they specified only the important bits, usually how the pieces related to each other | 01:51 |
| kanzure | so there was a direct genotype-to-phenotype constraining the rules of design choices | 01:51 |
| kanzure | I don't remember that one | 01:52 |
| fenn | then it was up to the program to figure out how to actually machine each part | 01:52 |
| kanzure | hm | 01:52 |
| kanzure | well, there's also issues like the clearance between parts | 01:52 |
| kanzure | dolorean door opening up hitting a ceiling | 01:52 |
| kanzure | erm, that's an easy algorithmic problem | 01:52 |
| kanzure | forget it | 01:52 |
| kanzure | I wonder if "how to make it" is what the designer should be programming .. not the "it" (of cli geom/3D add-sphere-to .. origin stuff), which is just templates that are customized by artists / end-users | 01:53 |
| fenn | nist paper i was talking about: "automated redesign for improving manufacturability" in particular http://www.cs.umd.edu/~nau/vm2/ | 01:57 |
| kanzure | http://www.lpgli.com/ | 02:00 |
| kanzure | ah right | 02:00 |
| fenn | everyone knows that the power of HHO will destroy the imperialist oil baron hegemony! | 02:00 |
| ybit | so, q-cad is that bad? | 02:00 |
| kanzure | hm, I need to get the lab hooked on the Regli/nist work | 02:00 |
| fenn | qcad isnt too bad, as long as you dont try to use blocks | 02:01 |
| fenn | you start to like the context-sensitive buttons after a while | 02:01 |
| ybit | some idiot recently put what should be a spreadsheet in a .dwf file | 02:04 |
| kanzure | manufacturability validation sounds ok, but "how to make it" is a harder problem .. isn't that why you're using the kb/db on the computer anyway? so that you can already use information on how to build separate parts, and then just use rules for combining the stuff together to make the final construction proc | 02:04 |
| ybit | it was for my brother's workplace, i was tasked with figuring out the contents, yay | 02:04 |
| ybit | i'll try keeping dumb stories to myself next time :) | 02:06 |
| kanzure | heh :) | 02:06 |
| kanzure | daily wtf | 02:06 |
| fenn | so it was a drawing of a spreadsheet? | 02:08 |
| kanzure | heh | 02:08 |
| ybit | fenn: indeed | 02:11 |
| kanzure | "how to make it" is just as hard a problem as the nonintuitive material cli stuff | 02:14 |
| kanzure | unless you have fabuntu / Fab Lab version infinity sitting in your basement | 02:14 |
| kanzure | ping? | 02:15 |
| kanzure | nevermind | 02:15 |
| fenn | version 2.7182818 | 02:39 |
| kanzure | hm? | 02:40 |
| fenn | subsequent releases of LaTeX tend towards "e" | 02:41 |
| kanzure | what good is e? :) | 02:41 |
| fenn | what good is infinity? | 02:41 |
| fenn | what good is a newborn baby? | 02:42 |
| kanzure | none! kill it! | 02:42 |
| kanzure | oops, didn't say that | 02:42 |
| kanzure | yandex is rather slow in opera | 02:42 |
| kanzure | I think it might be doing some nasty js | 02:42 |
| kanzure | http://www.youtube.com/watch?v=YOQve3OoiG4 | 03:51 |
| kanzure | best reason ever to not buy their magazine | 03:51 |
| kanzure | wtf, Bill Gates + Jerry Seinfeld commercial? | 04:11 |
| kanzure | most confusing commercial ever | 04:11 |
| kanzure | 'amoeba with a blog' | 04:11 |
| kanzure | we can do this | 04:11 |
| kanzure | quick, somebody get me the genome of an amoeba | 04:11 |
| kanzure | nsh? | 04:12 |
| kanzure | ftp://ftp.ncbi.nlm.nih.gov/gene/ somewhere | 04:13 |
| kanzure | ftp://ftp.ncbi.nlm.nih.gov/genomes/ | 04:14 |
| kanzure | what's amoeba, protozoa or something? | 04:14 |
| kanzure | Naegleria_fowler => "brain-eating amoeba" | 04:17 |
| ybit | http://www.physorg.com/news140527552.html | 04:47 |
| kanzure | oh god | 04:54 |
| kanzure | LHC is on Windows? | 04:54 |
| kanzure | Somebody's trying to prove to me the existence of goals in the brain via http://changingminds.org/explanations/brain/urge_system.htm | 05:20 |
| kanzure | Does anybody else see the problem with this page? | 05:21 |
| kanzure | ""The thalamus in the limbic system ('leopard brain') converts the physical need into an urge within the cortex. | 05:21 |
| kanzure | holy fuck :( | 05:22 |
| fenn | rawr | 05:22 |
| kanzure | seriously though | 05:22 |
| fenn | i think i feel a goal coming on | 05:22 |
| kanzure | ? | 05:22 |
| fenn | goals exist in the mind, not the brain | 05:23 |
| kanzure | guess I've lost my mind | 05:23 |
| fenn | probably | 05:24 |
| fenn | buying land anywhere is cheaper than doing anything in outer space | 05:25 |
| kanzure | on that note I'm currently listening to http://www.last.fm/music/Helloween/Gambling+With+The+Devil/As+Long+As+I+Fall (new album) | 05:25 |
| fenn | for some reason i'm obsessing about wearable computers | 05:26 |
| kanzure | fenn: we're talking about land with enough resources to sustain you | 05:26 |
| kanzure | not some stupid chunk of land without anything usable | 05:26 |
| fenn | what, a few acres at most | 05:27 |
| fenn | in the philippines every citizen is allotted 1.2 acres (hectares?) and it's supposed to be able to sustain them if they farm it well | 05:27 |
| fenn | i dont know if this is realistic but it's on the right order of magnitude | 05:28 |
| kanzure | how does that get you the surface area you need for all of those photons you'd be needing? | 05:28 |
| kanzure | and how does that give you enough material for all of the machinery and so on? | 05:28 |
| fenn | aroo? i'm talking about land on earth | 05:29 |
| kanzure | maybe it does for the machinery - but last time I checked a lot of manufacturing equipment is *huge* | 05:29 |
| kanzure | bah - farming - where are your heavy metals? | 05:30 |
| fenn | the squishy wet kind | 05:30 |
| fenn | you dont need heavy metals to 'sustain you' | 05:30 |
| kanzure | wait, | 05:30 |
| kanzure | are you or are you not talking about the openmanufacturing email I sent? | 05:30 |
| fenn | yes | 05:30 |
| kanzure | so without the manufacturing part, you should just go over to some silly be-green-or-else mailing list I guess | 05:31 |
| fenn | well, he raised a valid point and you sorta brushed him off with 'i dont believe in money' | 05:31 |
| kanzure | http://www.mattmahoney.net/rsi.pdf matt's attempt at a mathematical version of recursive self-improvement .. it's probably wrong, matt tends to be wrong a lot | 05:33 |
| kanzure | let me go check it | 05:33 |
| kanzure | material dependency issues | 05:33 |
| kanzure | what's wrong with that | 05:33 |
| kanzure | it's quite true that many people are screwed | 05:33 |
| kanzure | look, you were born into a system where you don't have any way to make sure you actually have enough resources on the planet to keep you clicking | 05:34 |
| kanzure | of course, thanks to the vast volumes and so on, there is, but still | 05:34 |
| kanzure | that's no guarantee etc. | 05:34 |
| kanzure | I know I'm screwing myself over since what I propose when I speak like that is /hard/, but nobody said things were easy etc. etc. | 05:40 |
| kanzure | re: rsi.pdf, wolfram's lack of faith in iteration to solve problems is interesting | 05:43 |
| fenn | er, whoops i didnt really pay attention to that.. seemed like your usual singularitarian propaganda disguised as science | 05:44 |
| kanzure | i think it is | 05:44 |
| fenn | it doesn't make sense anyway | 05:47 |
| fenn | 'this program counts, therefore it has a goal. therefore recursive self improvement is slow!' | 05:47 |
| kanzure | "represents the goal of finding large prime numbers" | 05:48 |
| kanzure | WTF | 05:48 |
| kanzure | I still don't understand 'goals' | 05:48 |
| kanzure | and I was in a goal cult, I should know goals | 05:48 |
| fenn | goals are just a motivational structure | 05:49 |
| fenn | when you reach them you get to check something off your list and give yourself a cookie | 05:49 |
| fenn | the cookie is not the goal but it provides the positive feedback | 05:49 |
| kanzure | goal is some hypothalamus glutamate imbalance between neurons 15 and 32 ? | 05:49 |
| fenn | the actual goal is something like a unit test | 05:50 |
| kanzure | it's kind of like playing a game with yourself methinks | 05:50 |
| kanzure | but that's just your ability to play games | 05:50 |
| fenn | the goal can even be evaluated externally (make my boyfriend happy, get good grades) | 05:50 |
| kanzure | sometime in 2004 I was going to write an rpg for a side-monitor that would give me points for checking stuff off of http://heybryan.org/todo.html | 05:51 |
| kanzure | (of course, it was todo.hnb at that time, and I only did an html dump in 2005 or something) | 05:51 |
| kanzure | lost the file in 2006 because I was stupid and wasn't doing daily backups | 05:51 |
| fenn | lol rlrpg | 05:51 |
| kanzure | yep | 05:51 |
| fenn | harness the grinding instinct | 05:51 |
| kanzure | whatever happened to that 'effortless economy' eh? | 05:51 |
| kanzure | how could one moment of reality require any more effort than any other? /me is confused | 05:52 |
| fenn | i believe it was stolen by Fanuc, Inc of Japan | 05:52 |
| fenn | well, when you have to do something you dont want to do, and decide to do it anyway, it takes effort | 05:53 |
| kanzure | http://en.wikipedia.org/wiki/Utility_function#Utility_functions | 05:54 |
| kanzure | gah, anything that describes itself starting with microeconomics .. | 05:54 |
| fenn | "consumption of various goods and services" is a really shitty description of utility | 05:55 |
| fenn | basically utility is a quantifiable unit of happiness | 05:56 |
| kanzure | how the hell is it quantifiable | 05:57 |
| kanzure | deep stimulation electrodes or something? | 05:57 |
| fenn | well, quantifying it is the hard part, and dont go off on some stupid wireheading tangent | 05:57 |
| fenn | wireheading != happiness | 05:58 |
| fenn | i dont care how many times the rat pushed the button | 05:58 |
| kanzure | you have a better idea for physically quantifying this crap? | 05:58 |
| kanzure | humans tend to claim it's pleasurable, which is semantic gatewayage or something | 05:58 |
| kanzure | I also don't care about the button btw | 05:58 |
| fenn | what context did utility come up in anyway? | 05:59 |
| kanzure | rsi.pdf | 05:59 |
| kanzure | mattmahoney is from the agi mailing list, full of annoying ideas I must admit | 06:00 |
| kanzure | in this context 'utility functions' are mentioned side-by-side with ai all the time | 06:00 |
| fenn | ah, well in this case you can consider 'utility' to be "having sucessfully satisfied a goal" | 06:01 |
| fenn | unless your AGI has emotions or some crap | 06:01 |
| fenn | wait, actually its more like the derivative of that | 06:03 |
| fenn | bah, semantics | 06:03 |
| kanzure | wtf? | 06:03 |
| kanzure | this is pissing me off | 06:03 |
| kanzure | I think this is one of the few areas in the realm of ideas that has a strong ability to really get me angry | 06:04 |
| fenn | ok, here's where i goofed: supergoal is like your set of values, never to be questioned, it just "is" | 06:05 |
| fenn | goals are tests of whether your situation satisfies the supergoal | 06:05 |
| fenn | subgoals are steps towards goals | 06:05 |
| fenn | so, utility is a measure of how useful something is in achieving your goal | 06:06 |
| fenn | a tool you never thought about before can land in your lap, it has high utility but it was never a goal because you didnt know about its existence beforehand | 06:07 |
| kanzure | rolleyes | 06:07 |
| fenn | it's an ontology, whaddya want | 06:08 |
| fenn | this looks pretty good for a wearable: | 06:11 |
| fenn | http://www.embeddedarm.com/products/board-detail.php?product=TS-7800 | 06:11 |
| kanzure | http://speedmodeling.org/smcfiles/kramer3d_niBCIcontrol.pdf bkero's two output mouse bci stuff | 06:12 |
| kanzure | boots linux in 0.69s | 06:12 |
| kanzure | hehe | 06:12 |
| fenn | 4W power draw | 06:12 |
| fenn | with FPGA!! | 06:13 |
| kanzure | hrm | 06:13 |
| kanzure | mini experiment board? | 06:13 |
| fenn | and it should come with debian installed (if its anything like the other boards) | 06:13 |
| fenn | i dont get why they show two mouse cursors? why not show a robot arm? | 06:15 |
| kanzure | maybe this will help everyone out, | 06:16 |
| kanzure | i'll go propose to the agi freaks that they construct a minimal goal system | 06:16 |
| kanzure | just as there's the minimal cell project | 06:17 |
| kanzure | minimal system that is known to have a 'goal' | 06:17 |
| kanzure | whatever the hell it is | 06:17 |
| fenn | heh i hope it's not the counting program | 06:18 |
| kanzure | 'I was an experiment on the part of Nature, a gamble within the unknown, perhaps for a new purpose, perhaps for nothing, and my only task was to allow this game on the path of primeval depths to take its course, to feel its will within me and make it wholly mine. That or nothing!' - Emil Sinclair, Holocaust Century Eschatologist | 06:19 |
| * kanzure needs to find a quote from the wild | 06:20 | |
| kanzure | aha | 06:25 |
| kanzure | http://heybryan.org/wildness.html | 06:25 |
| kanzure | or http://heybryan.org/quotes.html#wildness | 06:26 |
| kanzure | I felt like that mostly because I see two possibilities here when it comes to RSI: | 06:30 |
| kanzure | (1) the rsi.pdf mentions the problem that it's somewhat uncomputable and not guaranteed, so you could try to get this 'guarantee' through some other obscure mechanism and become a cult of sorts (oops), or | 06:30 |
| kanzure | (2) you could opt to not puritanically enforce 'goalism' (whatever that is), but in the process of this lose such 'guarantees' | 06:31 |
| kanzure | for this (#2), many would call you crazy gambling with so much | 06:31 |
| kanzure | but I'm becoming increasingly confident of it being a good path | 06:31 |
| kanzure | nsh might agree | 06:31 |
| kanzure | poke | 06:31 |
| fenn | if the AI is so smart it should be able to figure out what we wanted anyway | 06:32 |
| kanzure | what? | 06:32 |
| kanzure | what we wanted? | 06:32 |
| fenn | "god, all this time i've been tiling the solar system with smiley faces, and they really just wanted to be happy, how stupid of me!" | 06:33 |
| kanzure | paper clips, not smiley faces | 06:33 |
| kanzure | but anyway | 06:33 |
| fenn | smiley faces is better | 06:33 |
| kanzure | paper clips is an actual reference to eliezer yudkowsky | 06:33 |
| kanzure | or something | 06:33 |
| kanzure | oh god. | 06:33 |
| fenn | yeah he alternates between paperclips and smiley faces | 06:35 |
| fenn | i wonder if "unbrickable design" means they'll give me another one if i break it | 06:36 |
| kanzure | only if you brick it | 06:36 |
| kanzure | or get bricked yourself, as you're using it as a wearable | 06:36 |
| fenn | hmm, you mean it wont surround me with an invisible aura of unbrickableness? | 06:37 |
| kanzure | you know, that's worth an email to their tech support | 06:38 |
| kanzure | are there any meta-map APIs so that nobody has to spend all of their time with just Google Maps? | 07:42 |
| kanzure | this was kind of the point of the AJAX libraries that popped up IIRC | 07:42 |
| kanzure | interesting | 07:52 |
| kanzure | irc channel on mindat.org #mindat | 07:52 |
| kanzure | port 9610 | 07:52 |
| kanzure | 'Because this site contains the copyright material of other members, you agree not to try to extract information from this site for any other use than viewing of this website material. You are not allowed to archive significant portions of this site and you are not allowed to copy data from this site for use in any other product or website.' | 07:56 |
| kanzure | stupid databases | 07:56 |
| nsh | they can eat balls | 07:57 |
| nsh | accessible == free, de facto | 07:57 |
| nsh | don't put something on the web if you want to control it | 07:58 |
| kanzure | http://www.mindat.org/min-124.html | 08:01 |
| kanzure | me is wondering about extracting the information from their Google Maps widget thingy | 08:01 |
| kanzure | http://www.mindat.org/cform/cform_client.js | 08:02 |
| kanzure | yay lack of inline documentation | 08:02 |
| kanzure | jack pot - http://www.mindat.org/gmjs-mineral.php?id=124 | 08:06 |
| kanzure | max is 223 | 08:07 |
| kanzure | Okay. I have the full data set for the location of minerals now. | 08:13 |
| nsh | what you up to? | 08:13 |
| kanzure | http://heybryan.org/~bbishop/docs/mindat_locations.zip | 08:14 |
| fenn | all ur mineralz are belong to us | 08:14 |
| kanzure | forking everyone since they won't do the Internet thing right | 08:15 |
| fenn | but .. but.. you agree! you agree not o try to extract information~!!!! | 08:15 |
| kanzure | it was kinda cute actually | 08:16 |
| kanzure | when I wgetted the php scripts failed with some bad coding error | 08:16 |
| kanzure | but user agent change fixed it | 08:16 |
| kanzure | no rate limiting on the server, so the guy isn't all that serious | 08:16 |
| fenn | i bet you're the first person to actually spider a mineral database | 08:16 |
| kanzure | I was also thinking of making up some silly phylogenetics if people wanted to argue about the origin of minerals and materials and such | 08:18 |
| biopunk | god made them | 08:20 |
| kanzure | You know, technically, Google Maps has a Mars and Luna version .. no reason we can't start making up (perhaps imaginary for starters?) mineral deposits just to make a point (hey look, we can use teh codez!) | 08:20 |
| kanzure | bigger challenge is extracting the metadata from each page | 08:21 |
| kanzure | http://www.mindat.org/min-124.html | 08:22 |
| biopunk | uri geller used to charge mineral companies for running a hand over a map and telling them where they shuoldn't drill | 08:22 |
| biopunk | i love that hoax | 08:23 |
| kanzure | I think it might be a wiki, but I don't see the "EDIT" button that the "ADD/EDIT" button's resulting page tells me of.. | 08:23 |
| nsh | kanzure, have you checked Journal Citation Reports | 08:23 |
| kanzure | Hm? | 08:23 |
| nsh | for impact factor data | 08:23 |
| kanzure | For impact factor? yeah | 08:23 |
| kanzure | that's all in nasty pdfs IIRC | 08:23 |
| kanzure | worth double checking eventually | 08:24 |
| nsh | bah :-/ | 08:24 |
| kanzure | "I mean neptune is like a huge CVD chamber" | 08:24 |
| kanzure | http://www.perl.org.il/pipermail/perl/2003-March/001342.html <- WTF moron alert | 08:49 |
| kanzure | ' Why 'Bad'? Again, if the file is 100 MB in size and I have 1 GB RAM, why is this a problem?' | 08:49 |
| kanzure | bad bad bad bad | 08:49 |
| kanzure | http://www.ph.utexas.edu/~help/faq/linux.html <-- maybe I have a user group nearby (don't bother clicking) | 08:51 |
| kanzure | what are music blogs called? | 08:54 |
| kanzure | mblogs? I know they have vblogs or something, but that's not what I need .. or I guess it could be, hrm. | 08:54 |
| nsh | no idea, sorry | 08:54 |
| willPow3r | mulogs | 08:57 |
| willPow3r | actually no | 08:57 |
| willPow3r | clogs | 08:57 |
| kanzure | maddox said it first | 08:58 |
| kanzure | re: clogs | 08:58 |
| kanzure | Hrm, what encoding format is used for musical notes in MIDI? | 08:59 |
| kanzure | http://interglacial.com/~sburke/midi-perl/ | 08:59 |
| kanzure | Burke again? he seems prolific sorta | 08:59 |
| nsh | http://www.ted.com/index.php/talks/brewster_kahle_builds_a_free_digital_library.html | 09:21 |
| nsh | ^^ gets it | 09:21 |
| kanzure | the lib? | 09:24 |
| kanzure | 'Fetched 29 MB in 39s" Bwahah. This can't get much better. Also, since when did kernel.org get bandwidth? | 09:33 |
| * nsh smiles | 09:33 | |
| bkero | kanzure__: <3 | 10:09 |
| bkero | perfect, but I'd need to run windows | 10:13 |
| willPow3r | is <3 supposed to be a heart or saggy breasts? | 10:16 |
| willPow3r | i never could figure that one out. | 10:17 |
| willPow3r | http://kde-look.org/content/preview.php?preview=1&id=76662&file1=76662-1.jpg&file2=&file3=&name=Feed+Me+A+Fresh+Linux+Distro | 12:57 |
| willPow3r | is there some way to use the guitar hero controller as a midi controller in reason? | 13:10 |
| willPow3r | http://slapyak.wordpress.com/guitar-hero-midi-controller/ | 13:27 |
| willPow3r | yep. sure can | 13:27 |
| kanzure | todo | 15:05 |
| kanzure | - make a bacteria sing on a blog (half way done) | 15:05 |
| kanzure | - mindat rippage | 15:06 |
| kanzure | unfortunately the bacteria song dataset was getting way too large and I had to killall -9 a few things | 15:06 |
| kanzure | 'Actually, I remember reading something about scientists finding a list structure in the brain of a bird singing a song (a moving pointer to the next item in a list sort of thing). But whatever.' | 15:06 |
| kanzure | Some guy on the AGI mailing list is still trying to prove to me the existence of 'goals' in the brain through 'motivation theory'. | 15:06 |
| kanzure | 'When you imagine a goal-state, the relationship is represented in the brain somehow (in the neurons of course). And when evidence of the actualization of that goal-state comes in through the senses, the brains sends an opiate reward, which might make the person want to do whatever that was again in the correct context.' | 15:07 |
| kanzure | yawn | 15:07 |
| kanzure | 'Many things like this are known. And people don't need to understand such at the individual-neuron level to model what happens.' | 15:07 |
| kanzure | I don't even know how to respond to this .. | 15:07 |
| willPow3r | you respond with a 9mm round to the head | 15:34 |
| kanzure | My grandmother is an ancient English/economics major who periodically sends me books that she has read that she really wants me to read too. | 17:23 |
| kanzure | Today I received Vernor Vinge's latest "Rainbow's End". | 17:23 |
| kanzure | grandma's kickin' it up a notch ;-) | 17:24 |
| kanzure | http://vinay.howtolivewiki.com/blog/global/free-guptastan-583 | 17:25 |
| kanzure | yet another nation | 17:25 |
| kanzure | hurray | 17:25 |
| kanzure | http://armthevictims.org | 17:27 |
| kanzure | we must all worship the church of the connection reset by peer! | 17:28 |
| * kanzure found a copy of Engineering Design: a materials and processing approach, Dieter | 17:30 | |
| kanzure | weird, the book cites stuff like '"Committee on Computer Aided Manufacturing"' which doesn't appear on the internet | 18:35 |
| kanzure | so much for computer-aided | 18:35 |
| kanzure | http://www.princeton.edu/~ota/ns20/topic_f.html Office of Technology Assessment, Congress, archived publications | 18:42 |
| kanzure | Does anyone know the origin of the chair swing comic strip? The one illustrating how much everyone sucks. | 19:04 |
| kanzure | Hah. The book is going through an example of the design of a ladle/hook for molten steel processing facilities. It worries me that this is all that the design is based on .. while it's good for a rough idea, something considered the backbone of many industries being so flaky on this aspect? wtf? | 19:22 |
| kanzure | " In your professional career you will have the opportunity to create dozens of original designs" | 19:23 |
| kanzure | hahah | 19:23 |
| kanzure | just imagine the double digit possibilities! | 19:25 |
| kanzure | Hm, wget doesn't like "#" in the URL string, even when escaped. | 19:45 |
| kanzure | What a retarded dataset. The ASM needs some computer scientists to bitchslap them. | 20:01 |
| kanzure | btw, incomplete datasets at the moment - asm, sciencedirect, mindat | 20:02 |
| kanzure | I don't even know how to /fake/ turning this handbook dataset into xml | 20:10 |
| -!- h2i is now known as ybit | 21:24 | |
| kanzure | http://www.ilpi.com/msds/faq/partc.html 'The Defense Environmental Network & Information eXchange (DENIX) has draft DTD's (data type definitions) for MSDS's and they welcome your insight and input on the proposed standard. However, we have not noted any changes in this site (and therefore progress on the DTD) for several years. We have posted a request to their forum seeking further information.' | 22:29 |
| kanzure | http://matml.org/ materials markup language | 22:36 |
| kanzure | http://matdl.org/repository/index.php | 22:40 |
| kanzure | http://matdl.org/repository/view/matdl:967 'building an international materials network' | 22:41 |
| kanzure | unfortunately there's no data in the matdl repo | 22:41 |
| kanzure | matdl.org repo | 22:41 |
| kanzure | oh wait, nevermind | 22:42 |
| kanzure | http://matdl.org/repository/view/matdl:860 <-- this isn't very helpful .. doesn't actually have any information about the item itself | 22:43 |
| kanzure | http://www.matml.org/downloads/MatML3e1.xml <-- However, this does have some relevant tidbits. | 22:44 |
| kanzure | so just like everything else it lacks inertia | 22:48 |
| kanzure | http://matforge.org/ | 22:48 |
| kanzure | http://matdl.org/matdlwiki/index.php/Main_Page | 22:48 |
| kanzure | https://www.cmcrossroads.com/content/view/7768/96/ Making an XML bill of materials in GNU Make | 22:53 |
| kanzure | http://www.openly.com/efirst/ what are they fussing about here? | 22:56 |
| kanzure | http://www.ceramics.nist.gov/srd/scd/scdquery.htm structural ceramics db | 23:02 |
| kanzure | hrm: SpectroML, UnitsML | 23:05 |
| kanzure | http://unitsml.nist.gov/ | 23:06 |
| kanzure | that last one would be useful if they actually got any work done | 23:06 |
| kanzure | but I don't see anything on the site that suggests they are out of their 2006 planning stages | 23:06 |
| kanzure | http://www.matweb.com/ | 23:13 |
| kanzure | http://www.matweb.com/search/DataSheet.aspx?MatGUID=9cfde6787ca84eda998ccd0be9745c8d&ckck=1 | 23:14 |
| kanzure | supposedly this site uses matml but I'm not seeing it | 23:14 |
| kanzure | they offer a link to excel spreadsheet formats, wahoo? | 23:14 |
| kanzure | The NIST guys were gong to convert SCD (structural ceramics db) and HTS (high-temperature superconducting materials db) into the MatML form to show that it's useful, but oddly enough even they knew that "It is likely that this challenge will have to be addressed by someone outside of NIST" (workshop.pdf) | 23:19 |
| kanzure | R. G. Munro, NIST Ceramics Division => "NIST guys" | 23:20 |
| kanzure | probably not just a simple conversion issue | 23:20 |
| kanzure | http://www.ceramics.nist.gov/srd/scd/Z00133.htm <-- what a terrible format | 23:23 |
Generated by irclog2html.py 2.15.0.dev0 by Marius Gedminas - find it at mg.pov.lt!
