--- Day changed Thu Jun 11 2009 | ||
-!- genehacker [n=chatzill@cpe-67-9-157-136.austin.res.rr.com] has joined #hplusroadmap | 02:52 | |
-!- splicer [n=patrik@h189n3c1o261.bredband.skanova.com] has quit [Read error: 60 (Operation timed out)] | 02:59 | |
-!- genehacker_ [n=chatzill@cpe-67-9-157-136.austin.res.rr.com] has joined #hplusroadmap | 03:18 | |
-!- genehacker_ [n=chatzill@cpe-67-9-157-136.austin.res.rr.com] has quit [Read error: 104 (Connection reset by peer)] | 03:25 | |
-!- splicer [n=patrik@h189n3c1o261.bredband.skanova.com] has joined #hplusroadmap | 03:36 | |
-!- genehacker [n=chatzill@cpe-67-9-157-136.austin.res.rr.com] has quit [Read error: 110 (Connection timed out)] | 03:36 | |
-!- elias` [n=c@host217-42-207-213.range217-42.btcentralplus.com] has joined #hplusroadmap | 03:40 | |
-!- elias` [n=c@unaffiliated/elias/x-342423] has quit [Read error: 110 (Connection timed out)] | 05:24 | |
kanzure | 22:16:01 omg/user: ..."Using this as an interesting starting point, the authors ran another series of 201 integrations where | 07:11 |
---|---|---|
kanzure | the semi-major axis of Mars was changed by up to 30 mm. Within 100Myr of starting these integrations, five | 07:11 |
kanzure | of the simulations resulted in Mars being ejected from the solar system." | 07:11 |
kanzure | 22:16:21 omg/user: "All of the remaining 196 cases resulted in various collisions between Earth, Mars, Venus, Mercury, and/or | 07:11 |
kanzure | the Sun." | 07:11 |
-!- elias` [n=c@host217-42-207-213.range217-42.btcentralplus.com] has joined #hplusroadmap | 07:18 | |
kanzure | Harold Stephen Black’s 1927 famous patent that he wrote on a copy of the New | 07:45 |
kanzure | York Times) | 07:45 |
kanzure | http://en.wikibooks.org/wiki/Control_Systems/State-Space_Equations | 07:53 |
kanzure | "I's & C's with integral causality" | 07:58 |
kanzure | oh, sorry | 07:58 |
kanzure | independent energy storage elements: I's & C's with integral causality | 07:59 |
kanzure | dependent energy storage elements: I's & C's with derivative causality | 07:59 |
kanzure | http://www.me.utexas.edu/~bryant/courses/me344/DownloadFiles/LectureNotes/StateEqns.pdf | 08:00 |
kanzure | that was useful. | 08:00 |
-!- elias` [n=c@unaffiliated/elias/x-342423] has quit ["leaving"] | 08:12 | |
kanzure | "causal analysis" | 08:34 |
kanzure | The differential equations describing the dynamics of the system are written in terms of the states of the system. All storage elements (I and C) correspond to stored state variables (P for momentum and Q for displacement, respectively) and equations are written for their time derivatives (i.e. effort and flow). These equations are derived in four steps as described below. | 08:37 |
kanzure | 1. | 08:37 |
kanzure | Observe what the elements (sources, I's, C's, and R's) are giving to the system and write down their equations looking at the causalities and using variables for strong bonds. | 08:37 |
kanzure | 2. | 08:37 |
kanzure | Write down equations for the junctions and the two-port elements for the variables for the strong bonds. | 08:37 |
kanzure | 3. | 08:37 |
kanzure | Replace the variables, which are expressed in terms of states in other equations. Continue sorting and replacement till the right side of the entire set of equations are expressed in terms of states and system parameters only. | 08:37 |
kanzure | 4. | 08:37 |
kanzure | If some equations are still not completely reduced, there is the existence of some kind of a loop (algebraic loop, causal loop or differential causality, as shall be discussed later). Try solving those as a set of linear equations either through substitution or matrix inversion. | 08:37 |
kanzure | And finally, erase all trivial equations other than those for derivatives of state variables and write them in terms of state variables. | 08:38 |
kanzure | http://www.bondgraph.info/about.html#eqn | 08:38 |
-!- ybit [n=ybit@unaffiliated/ybit] has joined #hplusroadmap | 09:26 | |
-!- ybit [n=ybit@unaffiliated/ybit] has quit [Remote closed the connection] | 09:54 | |
kanzure | ok it seems to be working today | 10:09 |
kanzure | does anyone know where I should put my server now that I can't keep it at home? | 10:09 |
kanzure | fenn: http://www.cs.utexas.edu/~suriya/UT-wireless/ | 10:09 |
kanzure | aha. disabling dynamic word wrap helps in kwrite. | 10:33 |
-!- ybit [n=ybit@unaffiliated/ybit] has joined #hplusroadmap | 10:55 | |
-!- any40177394 [n=someone@75-120-1-122.dyn.centurytel.net] has joined #hplusroadmap | 12:42 | |
-!- samrose [n=samrose@adsl-68-73-194-95.dsl.sfldmi.ameritech.net] has joined #hplusroadmap | 12:47 | |
-!- katsmeow-afk [n=someone@75-120-44-144.dyn.centurytel.net] has quit [Read error: 110 (Connection timed out)] | 12:50 | |
-!- any25481156 [n=someone@75-120-36-94.dyn.centurytel.net] has joined #hplusroadmap | 12:52 | |
-!- any25481156 is now known as katsmeow-afk | 12:52 | |
-!- any50187595 [n=someone@75-120-4-74.dyn.centurytel.net] has joined #hplusroadmap | 12:53 | |
-!- any52762959 [n=someone@75-120-44-228.dyn.centurytel.net] has joined #hplusroadmap | 12:56 | |
-!- katsmeow-afk [n=someone@75-120-36-94.dyn.centurytel.net] has quit [Nick collision from services.] | 12:56 | |
-!- any52762959 is now known as katsmeow-afk | 12:56 | |
-!- any46776899 [n=someone@75-120-10-123.dyn.centurytel.net] has joined #hplusroadmap | 12:58 | |
-!- kardan| [n=kardan@p54BE775B.dip.t-dialin.net] has joined #hplusroadmap | 13:01 | |
-!- any38918056 [n=someone@75-121-62-199.dyn.centurytel.net] has joined #hplusroadmap | 13:02 | |
-!- katsmeow-afk [n=someone@75-120-44-228.dyn.centurytel.net] has quit [Nick collision from services.] | 13:02 | |
-!- any38918056 is now known as katsmeow-afk | 13:02 | |
-!- any40177394 [n=someone@75-120-1-122.dyn.centurytel.net] has quit [Read error: 110 (Connection timed out)] | 13:03 | |
-!- any74952047 [n=someone@75-120-25-154.dyn.centurytel.net] has joined #hplusroadmap | 13:05 | |
-!- any07454642 [n=someone@75-120-218-15.dyn.centurytel.net] has joined #hplusroadmap | 13:07 | |
-!- katsmeow-afk [n=someone@75-121-62-199.dyn.centurytel.net] has quit [Nick collision from services.] | 13:07 | |
-!- any07454642 is now known as katsmeow-afk | 13:07 | |
-!- any46309011 [n=someone@75-120-219-74.dyn.centurytel.net] has joined #hplusroadmap | 13:11 | |
-!- any50187595 [n=someone@75-120-4-74.dyn.centurytel.net] has quit [Read error: 110 (Connection timed out)] | 13:14 | |
-!- any46776899 [n=someone@75-120-10-123.dyn.centurytel.net] has quit [Read error: 110 (Connection timed out)] | 13:20 | |
-!- any49026262 [n=someone@99.194.253.130] has joined #hplusroadmap | 13:23 | |
-!- katsmeow-afk [n=someone@75-120-218-15.dyn.centurytel.net] has quit [Nick collision from services.] | 13:23 | |
-!- any49026262 is now known as katsmeow-afk | 13:23 | |
-!- any74952047 [n=someone@75-120-25-154.dyn.centurytel.net] has quit [Read error: 110 (Connection timed out)] | 13:25 | |
-!- cis-action [n=cis-acti@146-115-127-170.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com] has joined #hplusroadmap | 13:39 | |
-!- any46309011 [n=someone@75-120-219-74.dyn.centurytel.net] has quit [Read error: 110 (Connection timed out)] | 13:41 | |
-!- samrose [n=samrose@adsl-68-73-194-95.dsl.sfldmi.ameritech.net] has quit [Connection timed out] | 14:13 | |
-!- elias` [n=c@host217-42-207-213.range217-42.btcentralplus.com] has joined #hplusroadmap | 14:41 | |
kanzure | http://www.bondgraph.info/about.html has broken links but the images are actually there. so don't be fooled. http://adl.serveftp.org/papers/bondgraphs/eqn/www.bondgraph.info/about.html has the fixed version, or something | 16:04 |
-!- kardan| [n=kardan@p54BE775B.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] | 17:05 | |
-!- kardan| [n=kardan@p54BE396E.dip.t-dialin.net] has joined #hplusroadmap | 17:11 | |
-!- elias` [n=c@unaffiliated/elias/x-342423] has quit [Read error: 110 (Connection timed out)] | 17:34 | |
--- Log closed Thu Jun 11 17:42:57 2009 | ||
--- Log opened Thu Jun 11 19:26:48 2009 | ||
-!- kanzure [n=bryan@cpe-72-177-123-170.austin.res.rr.com] has joined #hplusroadmap | 19:26 | |
-!- Irssi: #hplusroadmap: Total of 17 nicks [0 ops, 0 halfops, 0 voices, 17 normal] | 19:26 | |
-!- Irssi: Join to #hplusroadmap was synced in 1 secs | 19:26 | |
kanzure | blargh | 19:26 |
fenn | you didnt miss anything | 19:27 |
kanzure | my soul | 19:27 |
kanzure | todo: check if the morning zipping of GraphSynth 2 works. | 19:29 |
drazak_ | I'm a h+ist that believes in the soul, just a little weird | 19:30 |
kanzure | http://www.xmms.org/files/1.2.x/xmms-1.2.11.tar.bz2 | 19:30 |
kanzure | drazak_: sorry, what? | 19:30 |
kanzure | what do you believe in? my translators are down. | 19:30 |
fenn | drazak_: it's a useful concept that gains much traction among the gentiles | 19:30 |
fenn | see maslow's hierarchy of needs | 19:31 |
drazak_ | seen it | 19:31 |
drazak_ | I should restate, also | 19:31 |
drazak_ | I believe in shit like psionguild.org | 19:31 |
fenn | whatever that is, you probably shouldn't have said so | 19:32 |
drazak_ | probably :P | 19:32 |
fenn | i built several orgone accumulator/amplifiers but they didnt seem to do anything | 19:35 |
fenn | or at least all the weird coincidences turned out really sucky for me personally | 19:36 |
drazak_ | well, I know nothing about that | 19:37 |
fenn | it's basically the same thing, but with toys | 19:37 |
drazak_ | right | 19:38 |
drazak_ | that's called radionics | 19:38 |
drazak_ | in our group | 19:39 |
drazak_ | I know shit for shinola about radionics | 19:39 |
fenn | radionics is more like quantum entanglement stuff | 19:39 |
fenn | the stuff i was doing was supposed to be like holography | 19:39 |
fenn | it did make some interesting ice crystals | 19:39 |
fenn | but that's sort of hard to prove causality | 19:40 |
fenn | anyway i've taken up diax's rake, which says to believe in something because it's right, not because you like it | 19:41 |
fenn | not sure what the equivalent philosophy is called in earth terms | 19:42 |
kanzure | was diax the retarded brother of occam? | 19:47 |
drazak_ | he must have cut himself with his razor, shaving | 19:49 |
fenn | he was gardening outside the temple, watching the numerologists play dice, then one day he just snapped and beat the crap out of them with his rake | 19:50 |
-!- genehacker [n=chatzill@cpe-67-9-157-136.austin.res.rr.com] has joined #hplusroadmap | 19:55 | |
genehacker | holy cow | 19:56 |
genehacker | high altitude wind power could be cheaper than coal or natural gas | 19:56 |
kanzure | hehe, sata might vouch for me to get into any class despite pre-reqs | 19:57 |
genehacker | heh | 19:58 |
genehacker | awesome | 19:58 |
kanzure | what should I take | 19:58 |
genehacker | emailed you | 19:58 |
kanzure | M 340L ? | 19:59 |
genehacker | yes | 19:59 |
genehacker | what classes are you taking? | 19:59 |
kanzure | M408D, CH301, ME205. | 19:59 |
genehacker | ME205? | 19:59 |
kanzure | yeah | 20:00 |
genehacker | dang you haven't even started taking fun classes yet | 20:00 |
kanzure | it's really really boring | 20:00 |
genehacker | chem 301? | 20:00 |
kanzure | yes :( | 20:00 |
genehacker | is that O chem? | 20:00 |
kanzure | no | 20:00 |
kanzure | I don't qualify for o-chem. | 20:00 |
kanzure | much less p-chem | 20:00 |
genehacker | you know what I think this is a conspiracy by the luddites of the world to hold people like us back | 20:01 |
genehacker | luckily I found a backdoor around it | 20:01 |
kanzure | pre-reqs suck. | 20:01 |
kanzure | what is the backdoor? | 20:01 |
genehacker | to take college in highschool | 20:01 |
kanzure | that doesn't work if your parents don't let you | 20:02 |
fenn | http://192.168.1.86/mp3 | 20:02 |
genehacker | yeah and there is only one in texas so | 20:02 |
kanzure | not true at all, genehacker | 20:02 |
genehacker | ok | 20:02 |
kanzure | there are many high schools within universities | 20:02 |
kanzure | and many community colleges | 20:02 |
kanzure | but it doesn't seem that parents are interested in letting their kids attend those classes, for some reason | 20:02 |
genehacker | well anyway you should figure out how to automate your homework | 20:03 |
kanzure | it only works if I know how to do the homework in the first place | 20:03 |
genehacker | btw are there any graph grammars for english? | 20:03 |
kanzure | you mean string grammars? like, regular linguistic grammers | 20:03 |
kanzure | *grammars | 20:03 |
kanzure | ack | 20:03 |
genehacker | yeah | 20:03 |
kanzure | yeah it's defined in the dictionary | 20:03 |
genehacker | I'm wondering if it's possible to make an automatic arguing machine | 20:03 |
kanzure | you mean, me? | 20:04 |
genehacker | arguments are somewhat algorithmic | 20:04 |
genehacker | no | 20:04 |
kanzure | one time I came up with a way to make a bullshit machine | 20:04 |
fenn | graph grammars for english, heh | 20:04 |
kanzure | automatic fallacious argument generator | 20:04 |
drazak_ | kanzure: I did ap chem in hs, I get to start with ochem my first year | 20:04 |
kanzure | drazak_: ap chem conflicted with ap physics in my school. | 20:04 |
genehacker | I'm thinking of a way to pass english class and automatically disprove online kooks | 20:04 |
kanzure | and the pre-req dependency graph thingy made it impossible to take it earlier | 20:04 |
fenn | this is a good one http://192.168.1.86/mp3/Ruth White/Short Circuits/09-solfeggietto.mp3 | 20:05 |
fenn | but the rest of that is painful to listen to | 20:05 |
genehacker | your link is dead fenn | 20:05 |
fenn | local net | 20:05 |
genehacker | oh | 20:05 |
genehacker | where? | 20:05 |
fenn | uh, here to there *points* | 20:06 |
genehacker | also I got a question about automatic design | 20:06 |
kanzure | yeah? | 20:06 |
genehacker | so let's say I'm automatically designing a high altitude wind power system | 20:06 |
kanzure | what is the method of automatic design that you are using? | 20:06 |
genehacker | how do I get the system to take into account factors like storms and ice accumulation on the blades | 20:07 |
kanzure | what system | 20:07 |
genehacker | without a human telling it about it | 20:07 |
genehacker | any system | 20:07 |
kanzure | um. what? | 20:08 |
kanzure | you mean | 20:08 |
kanzure | "how do you make it take something into account that you don't tell it about" | 20:08 |
genehacker | how do you take into account factors that are not readily evident | 20:08 |
kanzure | that's a good question. how do you do that non-automatically anyway? | 20:09 |
genehacker | things like vibration loosing parts | 20:09 |
kanzure | once you know how to do it, you can typically automate it | 20:09 |
genehacker | I don't know | 20:09 |
drazak_ | kanzure: I'm doing ap physics next year | 20:09 |
kanzure | drazak_: get out of high school or else. | 20:10 |
* drazak_ shrugs | 20:10 | |
drazak_ | whatever | 20:10 |
kanzure | genehacker: what I'm currently working on at the moment is a method of extracting state equations for a particular machine design | 20:11 |
kanzure | so if you have a resistor hooked up to a motor, what's the equation? | 20:11 |
kanzure | and if you have a resistor hooked up in parallel to another resistor, what's the equation? (yay kirchoff) | 20:11 |
kanzure | these aren't external environmental variables of course | 20:12 |
kanzure | but I think internal stuff is more important for starters | 20:12 |
drazak_ | wow, my dad might not let me go to the lab in boston, because people in the family have issues | 20:12 |
kanzure | just show up | 20:12 |
drazak_ | right | 20:12 |
drazak_ | go 800mi | 20:12 |
drazak_ | from my parents | 20:12 |
-!- genehacker_ [n=chatzill@cpe-67-9-157-136.austin.res.rr.com] has joined #hplusroadmap | 20:12 | |
drazak_ | with no money | 20:12 |
kanzure | genehacker: what I'm currently working on at the moment is a method of extracting state equations for a particular machine design | 20:13 |
kanzure | so if you have a resistor hooked up to a motor, what's the equation? | 20:13 |
kanzure | and if you have a resistor hooked up in parallel to another resistor, what's the equation? (yay kirchoff) | 20:13 |
kanzure | these aren't external environmental variables of course | 20:13 |
kanzure | but I think internal stuff is more important for starters | 20:13 |
genehacker_ | how do I know this? well I saw a pic with what happened to an airplane that went through a thunderstorm | 20:13 |
drazak_ | 1/R+1/R right? | 20:13 |
genehacker_ | how do we get a machine to do that | 20:13 |
kanzure | genehacker: the search term seems to be "bond graphs" or "linear graphs" | 20:13 |
kanzure | drazak_: Kirchoff's voltage law (KVL) and Kirchoff's current law (KCL) | 20:14 |
kanzure | http://adl.serveftp.org/papers/bondgraphs/ | 20:14 |
drazak_ | right | 20:14 |
genehacker_ | or with the case of parts vibrating off, how do we get machines to know whether they need to do high computational power testing | 20:14 |
kanzure | genehacker: see this: http://adl.serveftp.org/papers/bondgraphs/eqn/www.bondgraph.info/about.html | 20:14 |
genehacker_ | just wondering | 20:14 |
kanzure | genehacker: you mean, "are there certain scenarios where it is definitely known that more advanced models are going to be needed" ? | 20:14 |
kanzure | I think anything involving the maxwell equations might fit under that category | 20:15 |
kanzure | but that's just because I'm bullshitting | 20:15 |
genehacker_ | so how does the machine know that it needs to do see if it needs to do complex FEM simulation to see if vibration from say a tire is going to destroy a part? | 20:17 |
kanzure | fenn: Broadcom Corporation NetXtreme BCM5764M Gigabit Ethernet PCIe (rev 10) | 20:17 |
kanzure | well, there's my ethernet | 20:17 |
kanzure | "Intel Corporation Wireless WiFi Link 5100" | 20:17 |
genehacker_ | will automated design always be semiautomatic because of factors the machine didn't take into account | 20:17 |
kanzure | how would it take into account things that are impossible to take into account | 20:18 |
kanzure | I don't really understand what you're trying to get at | 20:18 |
kanzure | but yeah, there are complex simulations you can do with FEM | 20:18 |
kanzure | so let's take for instance a simple spring-and-mass model or something | 20:18 |
kanzure | and you can model a particular beam as a collection of parts | 20:18 |
kanzure | rather than a single part | 20:18 |
genehacker_ | humans would take it into account, because they've heard about these things | 20:18 |
genehacker_ | err might? | 20:18 |
kanzure | but not all humans because some huamns (like me) don't know about them | 20:19 |
kanzure | *humans | 20:19 |
kanzure | anyway, | 20:19 |
kanzure | the beam model would be an example of a part that could be complexificationized into an FEM model | 20:19 |
kanzure | it's kind of like a "meshing" step | 20:19 |
kanzure | but this doesn't introduce new variables into the equation .. just more of the same :) | 20:19 |
genehacker_ | so how does a machine do it's own real world testing? | 20:19 |
kanzure | and so it creates a higher-order model or something | 20:19 |
kanzure | genehacker: hod lispon had a machine doing real world testing | 20:19 |
kanzure | basically what he did was have a program that would make hypotheses and predictions about the machine itself | 20:20 |
kanzure | because it didn't know how many legs it had or something | 20:20 |
kanzure | so it would try to formulate the best working model for movement or something | 20:20 |
kanzure | so there was a simulation version and a real-world version for a particular machine with many linear actuators | 20:20 |
genehacker_ | so let's say a gas turbine fails | 20:20 |
kanzure | it looked hideous IIRC | 20:20 |
genehacker_ | how does the machine figure out why | 20:20 |
kanzure | by following a procedure | 20:20 |
genehacker_ | the gas turbine has heavily abraded blades | 20:21 |
kanzure | what procedures do engineers use? | 20:21 |
kanzure | those are the procedures that the machine would use. | 20:21 |
genehacker_ | what about things like sounds the turbine makes? | 20:22 |
genehacker_ | patterns of abrasion | 20:22 |
genehacker_ | this sounds like a bad bearing, there's more abrasion here because a part is fragmenting | 20:23 |
fenn | i'm afraid i've been contaminated by "on intelligence" so i can't provide the traditional "hard AI" answer right now | 20:24 |
genehacker_ | will we have to give machines experience packs? | 20:24 |
kanzure | yes but what do you mean by experience packs | 20:25 |
genehacker_ | a whole bunch of data from a whole bunch of different scenarios | 20:26 |
kanzure | we have to encode engineering knowledge into computational tools | 20:26 |
kanzure | so somebody has to learn it once | 20:26 |
kanzure | then they have to write a program | 20:26 |
kanzure | that's generally how automated design works at the moment | 20:26 |
fenn | supposedly (according to kanzure) there's a giant flowchart that the military goes through to figure out what broke when somebody blew up their plane | 20:27 |
genehacker_ | then you run into the problem of having something for everything | 20:27 |
kanzure | why would you have that problem? | 20:27 |
fenn | you mean lack of abstraction? | 20:27 |
kanzure | it should be able to automatically figure out what something to use in the first place | 20:27 |
genehacker_ | I think so | 20:27 |
kanzure | so what's the problem? | 20:27 |
genehacker_ | how do we take the human out | 20:28 |
kanzure | by harvesting their brains (for now) | 20:28 |
fenn | real world testing is not necessarily about failure analysis, it's about making sure the machine functions as intended | 20:28 |
genehacker_ | hmmm... | 20:28 |
fenn | so there's a limited number of things you have to check (the dependencies/requirements of the machine) | 20:28 |
fenn | er, and outputs too | 20:28 |
genehacker_ | do they have any good scans of people's brains yet? | 20:28 |
fenn | no | 20:29 |
genehacker_ | damn | 20:29 |
-!- genehacker [n=chatzill@cpe-67-9-157-136.austin.res.rr.com] has quit [Connection timed out] | 20:29 | |
fenn | i'm still rooting for the nanobots myself | 20:29 |
genehacker_ | how do we make a machine capable of encoding engineering knowledge on it's own | 20:30 |
kanzure | I don't know what that means though | 20:31 |
kanzure | what is "engineering knowledge" and how do you know if you have it or not | 20:31 |
kanzure | that sounds very confusing, genehacker_ | 20:31 |
genehacker_ | or should we just diagonal cutter it out and have people input the knowledge | 20:31 |
genehacker_ | good point | 20:31 |
genehacker_ | how does automatic design work? | 20:32 |
kanzure | it doesn't | 20:32 |
kanzure | it is vaguely defined | 20:32 |
kanzure | at the moment we have some tools for creating different assemblies of different components | 20:32 |
kanzure | but then no way to figure out what the system equations are | 20:32 |
genehacker_ | but not the components themselves? | 20:32 |
kanzure | not anything good | 20:33 |
kanzure | there's some crap, but you can't use it because it's all academic bullshit | 20:33 |
kanzure | and you can't use it because it doesn't capture the right information anyway | 20:33 |
kanzure | although there was some stuff from A-DESIGN (a component library) that was from a catalog | 20:33 |
kanzure | that was actually useful | 20:33 |
genehacker_ | so let me put it another way | 20:33 |
kanzure | unfortunately it doesn't run anymore | 20:33 |
genehacker_ | how do we make a machine that can tell the tone of a paragraph | 20:33 |
kanzure | what is tone? | 20:34 |
genehacker_ | the mood or feeling of a literary work, as the author gives | 20:34 |
kanzure | that didn't help | 20:34 |
genehacker_ | yeah it doesn't | 20:34 |
genehacker_ | the example in english class was the usage of the word peruse | 20:35 |
kanzure | why are you paying for that class | 20:36 |
genehacker_ | peruse is somewhat antiquated so using it makes you sound like you're trying to sound professional but failing at it | 20:37 |
genehacker_ | how do we get machines to know things like that? | 20:37 |
kanzure | according to who though | 20:37 |
kanzure | uh | 20:37 |
kanzure | well it's bullshit so I don't think it counts | 20:37 |
kanzure | it's not really anything that matters or is right .. | 20:37 |
kanzure | it's just some crap that they are making you do | 20:37 |
genehacker_ | yeah it is | 20:38 |
genehacker_ | so how do we get a machine to know if it is sounding like it's joking ? | 20:38 |
kanzure | are you reading questions from your homework? if so, stop it. | 20:39 |
fenn | genehacker_: read "on intelligence" which i can provide the pdf if you like | 20:39 |
fenn | or not | 20:39 |
genehacker_ | my solution is to have the machine look at word usage patterns from websites lots of people go to | 20:39 |
genehacker_ | and copy those | 20:40 |
fenn | ever seen megahal? (hidden markov model chatbot) | 20:41 |
genehacker_ | no | 20:41 |
fenn | its not as cool as you think | 20:41 |
fenn | there is no magic bullet | 20:41 |
genehacker_ | indeed there isn't | 20:41 |
genehacker_ | you'd probably end up with a lot of slang | 20:41 |
genehacker_ | can we get machines to read papers and extract useful information from them is what i'm thinking | 20:42 |
genehacker_ | semantic web stuff | 20:43 |
kanzure | yeah I have some stuff for automatically searching for interesting papers | 20:43 |
kanzure | but it doesn't work yet :) | 20:43 |
kanzure | but it also isn't "analyzing english"- it doesn't have to do that .. | 20:43 |
kanzure | it just needs to go over the citation/reference-topology and look for regularities in "really super interesting mountains of papers" or whatever | 20:43 |
kanzure | (this kind of goes back to my google scholar "search assistant" plans) | 20:44 |
genehacker_ | so do we just a whole lot of people to "wiki up" knowledge for an automatic design engine? | 20:45 |
kanzure | hell no :) | 20:46 |
kanzure | but | 20:46 |
kanzure | look into debian packages | 20:46 |
kanzure | (google it) | 20:46 |
kanzure | colorollary to the anthropic principle: it's weird that there is no magic bullet. | 20:47 |
genehacker_ | well do we have any other approach that works? | 20:50 |
fenn | SCIENCE! | 20:51 |
drazak_ | yes | 20:52 |
genehacker_ | well do you have a framework for SKDB yet? | 20:55 |
kanzure | almost there. I can taste it. | 20:55 |
kanzure | the bond graph and linear graph stuff is like the last on the list | 20:55 |
fenn | not for the next 3 months | 20:55 |
fenn | too busy with ADL crap | 20:55 |
genehacker_ | HOLY COW | 20:56 |
genehacker_ | wolfram alpha knows molecules | 20:56 |
genehacker_ | http://www06.wolframalpha.com/input/?i=pentaerythritol | 20:58 |
genehacker_ | has handy property data | 20:58 |
genehacker_ | perhaps we could use wolfram | 20:59 |
katsmeow-afk | To see full output you need to enable Javascript in your browser | 21:00 |
genehacker_ | what? | 21:01 |
katsmeow-afk | To see full output you need to enable Javascript in your browser | 21:01 |
genehacker_ | I see everything | 21:01 |
katsmeow-afk | ok | 21:02 |
genehacker_ | I have | 21:03 |
genehacker_ | kanzure how do I use the library proxy? | 21:03 |
kanzure | http://ezproxy.lib.utexas.edu/login?url=http://scholar.google.com/ | 21:04 |
katsmeow-afk | wolfram screws up "cow", as do most engines, "placental" is not a term one would use to ask where the cows are at a state fair | 21:04 |
genehacker_ | woohoo | 21:05 |
genehacker_ | I don't care about cows | 21:05 |
katsmeow-afk | wants an eid and password | 21:05 |
katsmeow-afk | i don't care a lot about cows, but i like accuracy in thedata for things i do care about, but know less about | 21:05 |
katsmeow-afk | if it cannot get "cow" right, how do i know it got something else right? | 21:06 |
genehacker_ | here's the deal | 21:07 |
genehacker_ | Wolfram alpha knows information about chemicals | 21:07 |
kanzure | "knows" ? | 21:07 |
kanzure | what are you talking about | 21:07 |
kanzure | it's just a database | 21:07 |
genehacker_ | guess so | 21:08 |
* katsmeow-afk backs outa the channel s l o w l y | 21:08 | |
genehacker_ | but you know all those papers on GNA phosphoramidite synthesis? | 21:08 |
kanzure | what about it | 21:09 |
genehacker_ | well now I can sort of understand them | 21:09 |
kanzure | why not just use chemspider | 21:09 |
kanzure | or wikipedia | 21:09 |
-!- anthonyl [n=duzt@dsl093-216-054.aus1.dsl.speakeasy.net] has quit [Read error: 110 (Connection timed out)] | 21:10 | |
genehacker_ | http://www06.wolframalpha.com/input/?i=4%2C5-dicyanoimidazole | 21:11 |
genehacker_ | alpha uses an awful lot of sources | 21:11 |
genehacker_ | http://en.wikipedia.org/w/index.php?title=Special%3ASearch&search=4%2C5-dicyanoimidazole&go=Go | 21:13 |
genehacker_ | that's why | 21:13 |
kanzure | what about chemspider | 21:13 |
genehacker_ | trying it | 21:14 |
genehacker_ | that's much better than wolfram | 21:19 |
genehacker_ | but it doesn't tell me how to make it | 21:19 |
genehacker_ | and wolfram doesn't either | 21:20 |
genehacker_ | is there anything that does? | 21:22 |
-!- DrTread [n=irchon@32.169.12.50] has joined #hplusroadmap | 21:33 | |
fenn | how about an organic chemist | 21:34 |
DrTread | organic chemists smell bad. :( | 21:34 |
fenn | you could say that about most people | 21:35 |
genehacker_ | heh | 21:35 |
genehacker_ | organic chemists also tend to die of horrible nasty cancers | 21:35 |
DrTread | true. | 21:35 |
DrTread | chemists have the highest suicide rate amongst the sciences. | 21:36 |
genehacker_ | hmmm... | 21:37 |
genehacker_ | now how do I make some O6 benzylguanine? | 21:37 |
DrTread | avoids the cancer problem | 21:37 |
DrTread | find an organic chemist. | 21:38 |
genehacker_ | or use the internet | 21:39 |
genehacker_ | yay google | 21:40 |
DrTread | ok, how do you do it? | 21:41 |
genehacker_ | what? | 21:42 |
genehacker_ | I just type in a chemical name into google with the word synthesis after it | 21:42 |
DrTread | make benzyl guanine | 21:42 |
DrTread | I figures that. sorry, I am on mobil phone. don't want to google. | 21:43 |
kanzure | hello DrTread | 21:44 |
DrTread | hello, kanzure. | 21:45 |
genehacker_ | tornado watch | 21:46 |
genehacker_ | ??? | 21:46 |
DrTread | kanzure, had any interest in building a DIY lazor cutter? | 21:46 |
genehacker_ | ok this benzylguanine stuff is the main component of GNA phosphoramidites | 21:46 |
genehacker_ | YES | 21:46 |
DrTread | ORLY? Ok. let's build. | 21:47 |
fenn | can 100W do thin metal? | 21:47 |
genehacker_ | I believe so fenn | 21:47 |
DrTread | but first, $2000 in parts | 21:47 |
genehacker_ | yes | 21:48 |
DrTread | meh. not really. but how thin & what metal? | 21:48 |
fenn | stainless, say 0.005" | 21:48 |
genehacker_ | ponoko | 21:48 |
genehacker_ | what ponoko has | 21:48 |
DrTread | stainless, no. Al, yes | 21:48 |
fenn | i'd expect Al to be harder to cut | 21:49 |
fenn | higher heat dissipation | 21:49 |
fenn | and more reflective | 21:49 |
DrTread | both of them need a blackener. | 21:49 |
genehacker_ | http://www.ponoko.com/make-and-sell/materials | 21:49 |
genehacker_ | ponoko got rid of metal? | 21:49 |
genehacker_ | DAMN YOU PONOKO! | 21:50 |
DrTread | oh noze! | 21:50 |
genehacker_ | Now I can't make a stirling engine business card | 21:50 |
DrTread | Online Metals or ASAP Source | 21:50 |
DrTread | LOL | 21:51 |
* fenn snickers @ windows crash on news/weather channel | 21:51 | |
genehacker_ | oh noes! | 21:51 |
genehacker_ | tornado must have hit it | 21:51 |
DrTread | the double wall cloud was scary. no tornados here @ 40th & Guadalupe | 21:52 |
genehacker_ | dang | 21:53 |
genehacker_ | that's not too far from where I am | 21:53 |
DrTread | hanging out with friends here. meanwhile, tornado is hitting my house in Bastrop. | 21:55 |
genehacker_ | that sucks | 21:57 |
DrTread | well, I don't know that for sure. | 21:57 |
genehacker_ | hmm... patents on how to make certain chemicals | 21:59 |
genehacker_ | useful | 21:59 |
DrTread | expired patents? | 21:59 |
genehacker_ | don't know | 22:00 |
genehacker_ | but it tells how to synthesize certain precursor chemicals necessary for GNA phosphoramidites | 22:00 |
DrTread | patents don't mix well with open source. :-( | 22:00 |
genehacker_ | it doesn't matter because we aren't selling stuff | 22:01 |
genehacker_ | you know what I want to do right? | 22:01 |
kanzure | it does matter though | 22:01 |
kanzure | because they have the guns | 22:01 |
kanzure | and will shoot you | 22:01 |
genehacker_ | I want to make DNA synthesizer | 22:01 |
DrTread | Hmmmm. ask your lawyers. | 22:01 |
genehacker_ | it's from 1997 | 22:01 |
genehacker_ | well they can't stop bittorrent? | 22:02 |
genehacker_ | so do you know what basic machines are involved in organic chemistry? | 22:02 |
DrTread | true. but chem labs move slower than hax0rz | 22:02 |
genehacker_ | IE distillers and stuff | 22:03 |
DrTread | I do. | 22:03 |
genehacker_ | which is why I want to make a tiny chemistry lab that fits on a desktop | 22:03 |
genehacker_ | you do? | 22:03 |
genehacker_ | mind elaborating? | 22:03 |
DrTread | I do know someone who has that equipment. | 22:03 |
genehacker_ | I don't want to borrow the equipment | 22:03 |
DrTread | are enzymes involved in the synthesis? | 22:04 |
genehacker_ | I want to know what they do | 22:04 |
genehacker_ | unfortunately not | 22:04 |
DrTread | ah. | 22:04 |
genehacker_ | just a whole bunch of crazy stuff about mixing this and that in different solvents and atmospheres | 22:04 |
genehacker_ | or something like that | 22:04 |
DrTread | yes, well, we can discuss that over coffee. I'd | 22:05 |
DrTread | need details of the synthesis. | 22:05 |
genehacker_ | hold up | 22:06 |
genehacker_ | http://www.patentstorm.us/patents/5663338/description.html | 22:06 |
DrTread | hard to do on my iPhone. :) | 22:06 |
genehacker_ | (S)-9-(3-(4,4′-Dimethoxytrityloxy)-2-hydroxypropyl)-O6-benzylguanine | 22:07 |
genehacker_ | (3). Compound 1 (3.10 g, 12.8 mmol) was partially | 22:07 |
genehacker_ | dissolved in anhydrous DMF (25.0 mL) under a nitrogen atmosphere. | 22:07 |
genehacker_ | NaH was added (105 mg, 2.63 mmol, 60% in mineral oil), | 22:07 |
genehacker_ | and the solution was allowed to stir under nitrogen for 1 h. In a | 22:07 |
genehacker_ | separate flask, compound 2 (4.60 g, 12.2 mmol) was dissolved in | 22:07 |
genehacker_ | 26.0 mL of DMF, added to the first solution, and then heated to 90 | 22:08 |
genehacker_ | °C overnight. The next morning, the solution was cooled, all solvent | 22:08 |
genehacker_ | removed, and the resulting oil coevaporated with toluene, redissolved | 22:08 |
genehacker_ | in ethyl acetate, and concentrated again. The product was | 22:08 |
-!- DrTread [n=irchon@32.169.12.50] has quit [Remote closed the connection] | 22:08 | |
genehacker_ | purified via column chromatography starting with 2:1:0.01 hexanes/ | 22:08 |
genehacker_ | acetone/Et3N, then eluting with 3:2:0.01 hexanes/acetone/Et3N to | 22:08 |
genehacker_ | afford compound 3 as a light yellow foam (3.73 g, 47%): | 22:08 |
genehacker_ | (S)-9-(3-(4,4′-Dimethoxytrityloxy)-2-hydroxypropyl)guanine | 22:08 |
genehacker_ | (4). Compound 3 (3.30 g, 5.3 mmol) and Pd/C (1.70 g, 10% | 22:08 |
genehacker_ | on carbon) were suspended in EtOAc (125 mL), and the solution | 22:08 |
genehacker_ | was purged with nitrogen, then hydrogen, and allowed to stir under | 22:08 |
genehacker_ | a hydrogen atmosphere. After 3 h, TLC showed completion of the | 22:08 |
genehacker_ | reaction, and the mixture was filtered through Celite and washed | 22:08 |
genehacker_ | with 5:1 CH2Cl2/MeOH to afford compound 4 as a tan solid (2.73 | 22:08 |
genehacker_ | g, 97%): | 22:08 |
genehacker_ | that's all I'm posting | 22:08 |
genehacker_ | oops | 22:08 |
-!- DrTread [n=irchon@32.169.12.50] has joined #hplusroadmap | 22:08 | |
DrTread | the website killed the app. | 22:09 |
DrTread | sorry. | 22:09 |
genehacker_ | was it all the stuff I posted? | 22:09 |
DrTread | will look @ it later, if you email link | 22:09 |
genehacker_ | are you a chemist of some sort? | 22:09 |
DrTread | I am. inorganic. | 22:10 |
genehacker_ | http://pubs.acs.org/doi/abs/10.1021/jo900365a | 22:10 |
genehacker_ | ah | 22:10 |
DrTread | my company (me) has some modest equipment. | 22:11 |
DrTread | but I know enough organic chem to do many things. | 22:11 |
genehacker_ | then you might not be able to figure out our evil world domination plan that involves a GNA synthesizer capable of synthesizer large genomes | 22:11 |
genehacker_ | *synthesizing | 22:12 |
DrTread | that's a bit beyond my ken, yes. | 22:12 |
-!- ybit [n=ybit@unaffiliated/ybit] has quit [Remote closed the connection] | 22:13 | |
genehacker_ | well It's a bit beyond mine too so... | 22:14 |
DrTread | by large, how many gigabases? | 22:14 |
genehacker_ | 3 megabases | 22:14 |
genehacker_ | in one run | 22:14 |
genehacker_ | or wait was that gigabases? | 22:15 |
DrTread | ah. ok. where will you put them? | 22:15 |
genehacker_ | 31,457,280 bp | 22:15 |
genehacker_ | what do you mean? | 22:16 |
DrTread | in bacteria? mice? people? | 22:16 |
genehacker_ | So I figured out how to make a DNA synthesizer, but not the chemicals for it | 22:16 |
genehacker_ | anything DrTread | 22:17 |
genehacker_ | *anything I want to put them in | 22:17 |
genehacker_ | but most likely bacteria | 22:17 |
genehacker_ | for compiling enzymes and genes | 22:17 |
genehacker_ | instead of buying them, why not download and compile them? | 22:17 |
DrTread | ok. it's pretty much impossible to get the chemicals without a business address. | 22:18 |
DrTread | even simple raw materials. | 22:19 |
genehacker_ | but if I ever figure out how to make it work, I'd like to synthesize mycoplasma laboratorium and boot it up | 22:19 |
genehacker_ | exactly | 22:19 |
genehacker_ | that's exactly the reason I want to make a tiny chemical factory | 22:19 |
DrTread | move to Brazil. it's not as restrictive there. :) | 22:19 |
genehacker_ | I can't | 22:20 |
DrTread | don't speak portugese? | 22:20 |
genehacker_ | no I don't | 22:21 |
DrTread | a guy I met a few years ago is from there, but moved to the US. he had to give up his lab. | 22:21 |
genehacker_ | you see I want a chemical universal constructor | 22:22 |
DrTread | it is much easier fir him to build rail guns here than to do chemistry. | 22:22 |
genehacker_ | wherein we can put in the chemicals we can buy or stuff present and nature and get the stuff we want | 22:22 |
DrTread | there are many ideas for that around. not so many actual machines, however. | 22:23 |
genehacker_ | really? | 22:24 |
DrTread | molecular machines that use air & acetylene as raw materials | 22:24 |
genehacker_ | that's not what I have in mind | 22:25 |
DrTread | you want to create life in your laboratory? | 22:26 |
genehacker_ | yes | 22:28 |
DrTread | meh. life is overrated. :-/ | 22:28 |
genehacker_ | but I'm thinking of having a bunch of miniaturized versions of things like downcells, distilling machines, and what not | 22:28 |
genehacker_ | downs cell requires gravity though it appears though... | 22:29 |
DrTread | centerpetal force is indistinguishable from gravity. | 22:30 |
genehacker_ | true | 22:31 |
DrTread | so if you want to do it in orbit...it's ok | 22:31 |
genehacker_ | but that adds complexity to things | 22:32 |
genehacker_ | if I'm building a tiny downs cell | 22:32 |
genehacker_ | it may not work very well because it's on scale where gravity doesn't play much of a role | 22:33 |
DrTread | ah. microfluiducs. ok. | 22:33 |
genehacker_ | yeah but metallic sodium is fun enough to warrant something large | 22:34 |
genehacker_ | maybe something the size of coke can | 22:35 |
DrTread | so use centrifuges. | 22:35 |
genehacker_ | ok | 22:35 |
DrTread | yes, the more sodium, the better. | 22:35 |
genehacker_ | any way could you name a couple of useful chemical processes for producing basic chemicals? | 22:36 |
DrTread | call Signa/Aldrich? | 22:37 |
genehacker_ | or what are some chemicals one must absolutely have? | 22:37 |
DrTread | water | 22:38 |
genehacker_ | that's abundant | 22:38 |
genehacker_ | no need to synthesize it | 22:38 |
DrTread | beyond that, it depends | 22:38 |
genehacker_ | sulfuric acid? | 22:38 |
genehacker_ | potassium perchlorate? | 22:38 |
genehacker_ | Sodium Hydride? | 22:39 |
DrTread | not so much perchlorate. | 22:39 |
DrTread | sodium hydride? nah | 22:39 |
genehacker_ | ammonia of course | 22:39 |
DrTread | yes. ethanol, acetone, drying agents | 22:40 |
genehacker_ | those are available so those aren't so bad | 22:41 |
DrTread | true. but buying too much acetone can land you on DEA watch lists | 22:41 |
DrTread | the country is insane. | 22:42 |
genehacker_ | sweet | 22:42 |
genehacker_ | bacterial fermentation for producing large quantities of acetone | 22:43 |
DrTread | good. I am all for creating my own economy & industry ouside of the mainstream. | 22:44 |
DrTread | because the mainstream isn't doing anything worthwhile. | 22:44 |
genehacker_ | And I want self-replicating doomsday space robots | 22:44 |
DrTread | ah, star trek. the uss constellation under the command of commodore decker. | 22:46 |
genehacker_ | ??? | 22:46 |
DrTread | the doomsday machine had a hull of neutronium. | 22:47 |
genehacker_ | oh cool | 22:47 |
DrTread | oh, sorry. random STTOS reference. :) | 22:47 |
genehacker_ | hmmm... hydrogen peroxide producing bacteria... | 22:49 |
DrTread | cool! peroxide is usually bad. | 22:50 |
genehacker_ | not to bacteria that produce it and catalase | 22:51 |
genehacker_ | then dump the H2O2 out to kill other bacteria | 22:51 |
DrTread | I figured. there has to be a reason to make the stuff. | 22:52 |
DrTread | it's rather expensive to produce if you're an organism. | 22:53 |
genehacker_ | phosphorous compounds are useful | 22:54 |
katsmeow-afk | i wanna know how to make ammonia from either urine, or atmospheric nitrogen and hydrogen | 22:55 |
DrTread | yep. we couldn't do much without it. | 22:55 |
DrTread | some bacteria are good @ making ammonia out of urine. ever smell a diaper? | 22:57 |
* katsmeow-afk nods | 22:57 | |
DrTread | atmospheric n2 + h2 requires high pressures & a rusty chain as a catalyst. | 22:58 |
katsmeow-afk | why chain? | 22:58 |
DrTread | it fits in the reactor well. | 22:59 |
katsmeow-afk | o | 22:59 |
DrTread | rumor has it that rusty chain was the first catalyst way back when. | 23:00 |
kanzure | well. | 23:00 |
kanzure | http://pygame.org/project/900/ | 23:00 |
kanzure | I'm addicted. | 23:00 |
DrTread | could be a ChemE legend. | 23:00 |
DrTread | oh, kanzure. you should work on more lofty goals. :-) | 23:01 |
kanzure | never underestimate the power of randomly generated maps for 2D top-down space shooter thingies | 23:02 |
kanzure | I was thinking of implementing a way to destroy the walls like in the invaders games | 23:02 |
DrTread | kanzure did you get a reply from travis? | 23:03 |
kanzure | no | 23:04 |
genehacker_ | damn | 23:04 |
genehacker_ | I just had a spider under my microscope but it got away | 23:05 |
kanzure | the map generation algorithm kind of sucks | 23:05 |
genehacker_ | oh well | 23:05 |
kanzure | but I guess it can be improved | 23:05 |
DrTread | I had a paramecium under mine last week | 23:05 |
kanzure | who hasn't. | 23:05 |
DrTread | it died. :( | 23:05 |
genehacker_ | well why don't you improve the algorithm, change the graphics, and sell it for the iphone | 23:06 |
DrTread | I cried. | 23:06 |
genehacker_ | well you could just grow them in cultures | 23:06 |
DrTread | yeah, kanzure. make an iPhone game. make insane $$, then fund our projects. | 23:07 |
kanzure | genehacker_: why are you always about selling stuff on the iphone | 23:07 |
genehacker_ | oh and that mycoplasma laboratorium bacteria? it could be blue | 23:07 |
kanzure | why does everyone assume you can make insane money on the iphone | 23:07 |
DrTread | ttyl. got to go. | 23:07 |
genehacker_ | because a small group of people have made lots of money on the iphone and this small group has been highly publicized | 23:07 |
-!- DrTread [n=irchon@32.169.12.50] has quit [] | 23:08 | |
genehacker_ | well I probably should work on homework | 23:09 |
-!- DrTread [n=irchon@32.169.187.251] has joined #hplusroadmap | 23:15 | |
-!- DrTread [n=irchon@32.169.187.251] has quit [Remote closed the connection] | 23:16 | |
-!- genehacker_ [n=chatzill@cpe-67-9-157-136.austin.res.rr.com] has quit ["ChatZilla 0.9.84 [Firefox 3.0.10/2009042316]"] | 23:20 | |
-!- kardan| [n=kardan@p54BE396E.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] | 23:31 | |
-!- cis-action [n=cis-acti@146-115-127-170.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com] has quit [] | 23:35 | |
kanzure | pymunk is pretty neat too | 23:37 |
kanzure | http://www.pygame.org/project/780/ | 23:37 |
kanzure | pybox2d is nice too | 23:42 |
kanzure | although the theo jansen mechanism example doesn't work because of some poorly specified bit depth field or something | 23:48 |
Generated by irclog2html.py 2.15.0.dev0 by Marius Gedminas - find it at mg.pov.lt!