| --- Day changed Sat Oct 04 2008 | ||
| kanzure | fenn: how's the thinking coming with inventory systems or alternatives? | 02:56 |
|---|---|---|
| gene | kanzure | 04:15 |
| gene | are there any diy microprocessors? | 04:16 |
| splicer | Not sure there is a need. There are very cheap microcontrollers that are easy to handle. | 04:45 |
| ybit | i think kde is dead | 04:47 |
| ybit | awesome still works | 04:47 |
| * ybit needs to open the lappy tonight or tomorrow | 04:47 | |
| ybit | and i think i will hold off on grabbing the matweb.zip file until this lappy is fixed | 04:48 |
| * ybit needs to backup soon | 04:49 | |
| ybit | i don't think my baby will last much longer :| | 04:49 |
| splicer | hmm.. I'm gonna do one later today too | 04:51 |
| ybit | splicer: do you use an offsite backup service? | 04:51 |
| splicer | i just copy the db, as sql | 04:52 |
| ybit | to...? | 04:52 |
| splicer | local HD.. occacianally DVD. I don't have the amout of data I suspect you guys have. | 04:53 |
| gene | the question is, is it possible to make microcontrollers with the complexity of the one in say a 1998 pc | 05:00 |
| gene | splicer you there? | 05:01 |
| splicer | here | 05:02 |
| gene | not to make our own microcontrollers | 05:02 |
| gene | but making semiconductor devices is necessary to make something that can self-replicate | 05:03 |
| splicer | you are not talking biology now? | 05:04 |
| gene | yeah | 05:04 |
| gene | one of the main goals of this forum is the design of self-replicating machines | 05:04 |
| gene | and also the biohacking toolkit | 05:04 |
| gene | for the biohacking toolkit a way needs to be found to synthesize sequences of DNA | 05:05 |
| gene | a DNA compiler is needed | 05:05 |
| gene | one that is sufficiently cheap and easy to use | 05:06 |
| splicer | I have kindof avoided the self replicating machins discussion... I guess computers can be made out of very simple standard parts with simple instructionsets and good parallellisation. | 05:07 |
| gene | mechanical computers? | 05:08 |
| splicer | yeah | 05:08 |
| gene | I've been thinking fluidic computers | 05:08 |
| splicer | ah | 05:08 |
| gene | computers that use fluidic logic gates and such | 05:09 |
| splicer | ok | 05:09 |
| gene | they don't require the tolerances necessary for mechanical computers | 05:09 |
| splicer | I'd like to find out how brains work | 05:10 |
| gene | ask kanzure | 05:10 |
| gene | he has a bunch of papers | 05:10 |
| splicer | I know how his brain works | 05:10 |
| gene | and is doing some work with computational neuroscience | 05:10 |
| gene | me too | 05:11 |
| splicer | ok | 05:11 |
| * ybit is on furlough until this issue is resolved. later everyone | 05:12 | |
| gene | so do you have access to any interesting equipment? | 05:14 |
| splicer | me?.. not really | 05:17 |
| splicer | it seems to me that the brain is the holy grail when it comes to biological computing. | 05:18 |
| gene | yeah | 05:18 |
| splicer | I think the blue brain project is probably the state of the art right now... and they don't know much | 05:19 |
| gene | yup | 05:19 |
| gene | they know quite a lot | 05:20 |
| gene | but not enough | 05:20 |
| splicer | it's like the beginning of the beginning | 05:21 |
| gene | that pretty much describes | 05:22 |
| gene | it | 05:22 |
| splicer | For me it's a bit like when a philosopher friend asked me once "Don't you ever think about the meaning of it all" and I thought about it and came to the conclusion that a lot of smart people I knew were thinking about it constantly and If they found out they would probably tell me. | 05:24 |
| gene | yeah | 05:26 |
| gene_ | huh? | 05:47 |
| gene_ | Did I win or something? | 05:48 |
| bkero | you win the game! | 06:04 |
| bkero | (I don't remember typing that) | 06:05 |
| gene_ | what game | 06:05 |
| gene_ | I don't know about the game | 06:06 |
| gene_ | some one told me not to think about it | 06:06 |
| bkero | http://en.wikipedia.org/wiki/The_Game_(mind_game) | 06:06 |
| gene_ | strange | 06:07 |
| gene_ | I can't read that page | 06:07 |
| bkero | lol | 06:07 |
| gene_ | it shows up | 06:07 |
| gene_ | I can't read it | 06:07 |
| bkero | I've lost the game too many times. 06:07 <+Mogget> 3 bkero 1452 0 secs | 06:08 |
| gene_ | I can't read what you just typed | 06:08 |
| gene_ | it is possible to win this game | 06:09 |
| bkero | Yes | 06:09 |
| gene_ | people in the amazon without internet win the game | 06:09 |
| bkero | Hahaha thursdays daily show is great. http://www.hulu.com/watch/37533/the-daily-show-with-jon-stewart-thu-oct-2-2008#s-p1-so-i0 | 06:17 |
| -!- Irssi: #hplusroadmap: Total of 16 nicks [0 ops, 0 halfops, 0 voices, 16 normal] | 06:17 | |
| gene_ | I truly believe brownian motion equations can be applied to talk on the internet | 06:21 |
| bkero | brownian motion equations? | 06:21 |
| gene_ | in other words | 06:24 |
| gene_ | the current topic of a forum can be represented as x,y coordinates | 06:25 |
| bkero | orly | 06:26 |
| gene_ | the forum randomely shifts from the original or current topic | 06:26 |
| gene_ | see right now? | 06:27 |
| gene_ | what was the original topic? | 06:27 |
| splicer | good episode. Love the daily show. | 07:12 |
| kanzure | http://groups.google.com/group/getarticles | 13:09 |
| kanzure | wtf's wrong with my server? | 13:19 |
| kanzure | the screen has switched to some sort of kernel dump | 13:19 |
| kanzure | "0kb pages_scanned:0 all_unreclaimable? no" | 13:19 |
| kanzure | reboot :( | 13:19 |
| kanzure | "/dev/hda1 has gone 225 days without being checked, check forced" | 13:22 |
| kanzure | bwahah, fear my uptime! | 13:22 |
| kanzure | re: diy microprocessors http://opencores.org/ | 13:35 |
| -!- kanzure__ is now known as kanzure_ | 13:37 | |
| -!- Utopiah is now known as UtopiahGHML | 14:10 | |
| UtopiahGHML | anybody watching videos with sound at >1x speed? | 15:06 |
| UtopiahGHML | good way to do that with synchro sound? | 15:06 |
| UtopiahGHML | good way to do that with Flash videos? | 15:06 |
| UtopiahGHML | good way to do all that online integrated to my usual browsing environment? (FF with WinXP) | 15:06 |
| fenn | download flash video and use mplayer | 15:10 |
| fenn | there are firefox add-ons to download just about anything, but i usually use a command line script | 15:10 |
| fenn | and give it the url as an argument | 15:11 |
| UtopiahGHML | like keepvid.com | 15:11 |
| UtopiahGHML | ok so mplayer support that well, cool | 15:11 |
| fenn | every once in a while there is some video i can't play (very rare) | 15:11 |
| UtopiahGHML | I tride it a while ago (>2 years AFAIK) and it wasn't so cool | 15:11 |
| UtopiahGHML | but can't mplayer actually read videos from an URL? | 15:13 |
| fenn | not a yourube url | 15:14 |
| UtopiahGHML | and a translated url from your script? | 15:14 |
| fenn | so, this works: youtube-dl.py -g http://www.youtube.com/watch?v=4N3Nm3vwkpY |xargs mplayer -speed 2 | 15:22 |
| UtopiahGHML | and it's possible to call that within a greasemonkey script I guess, hmmm sounds like the best solution | 15:23 |
| UtopiahGHML | (like doing a little x2 x4 x8 button on the side of the flash player) | 15:23 |
| fenn | can't seek though, you'd have to download first for that | 15:23 |
| UtopiahGHML | that's ok as long as I can pause/play | 15:24 |
| fenn | did i mention i hate flash? | 15:25 |
| UtopiahGHML | what do you hate more, AI people or Flash? | 15:25 |
| fenn | flash | 15:25 |
| fenn | at least i can choose to ignore ai people | 15:25 |
| UtopiahGHML | you can ignore flash with adblock | 15:29 |
| kanzure | argh | 17:15 |
| kanzure | I have a problem | 17:15 |
| kanzure | Shumaker's Process Pipe Drafting, Adventures from the Technology Underground (catapults, pulsejets, rail guns, flamethrowers, tesla coils, air cannons, garage warriors), Unit Processes in Organic Synthesis, Modern Manufacturing Process Engineering | 17:17 |
| kanzure | and my grandmother sent me another Vinge book that she's read .. I'm worried that next she's going to start sending me Sterling stuff or something | 17:19 |
| kanzure | http://tools.google.com/gapminder/ or http://gapminder.org/ for the main site - seems to be a statistics datawarehouse of some sort | 17:27 |
| kanzure | downloadable figures? http://www.gapminder.org/downloads/documentation/ | 17:27 |
| kanzure | http://www.tubecad.com/2004/blog0018.htm OTL amplifier design .. main site is http://www.tubecad.com/index.htm "John Broskie's Guide to Tube Circuit Analysis & Design" | 17:34 |
| gene | So who here knows about Charles Michael Collins | 18:54 |
| gene | he claims to have made a replicator in 1998 | 18:55 |
| gene | http://www.geocities.com/charles_c_22191/_home.html | 18:55 |
| kanzure | 1996. | 18:55 |
| kanzure | http://www.google.com/patents?vid=USPAT5764518 | 18:55 |
| gene | is he crazy? | 18:55 |
| gene | He's claiming patent infringement on reprap! | 18:56 |
| gene | or does his replicator actually replicate | 18:58 |
| gene | I can't figure out how it works | 18:59 |
| kanzure | it's about time I wrote the validator anyway .. I'll do it when I get back into the lab next | 19:02 |
| kanzure | if I write the validator and throw it up for download, gene, would you use it the next time you think you have a replicator? | 19:02 |
| kanzure | this validator checks for minimal requirements of self-replication | 19:02 |
| kanzure | the basic idea of skdb basically. | 19:02 |
| kanzure | it's not in any good form yet because it's just a simple io checker but whatever. | 19:02 |
| gene | so has collins actually made a replicator or is he some internet conspiracy nut? | 19:03 |
| gene | I tend to favor explanation 2 | 19:03 |
| gene | I don't think I have a replicator | 19:03 |
| kanzure | then why are you asking about this Charles Michael Collins fellow? | 19:06 |
| gene | I want to know if he is a nutcase or not | 19:06 |
| gene | if we should consider his ideas | 19:07 |
| gene | or if we should consider them as noise | 19:08 |
| gene | It cerntainly is an interesting approach | 19:09 |
| gene | nothing like the Advanced Automation for Space Missions study | 19:10 |
| kanzure_ | hm, my system locked up | 19:12 |
| gene | plus, it seems like his "alleged" replicator can't replicate fully | 19:12 |
| gene | but this cellular automaton approach | 19:17 |
| gene | might be easier to accomplish | 19:18 |
| gene | just look at the moses replicator | 19:18 |
| gene | it can assemble itself | 19:18 |
| gene | non autonomously | 19:19 |
| gene | and with human intervention | 19:19 |
| gene | but that's still a step up from reprap | 19:19 |
| fenn | yay overly broad patents | 19:23 |
| fenn | 75 claims! and number 1 covers just about every tool on the planet | 19:24 |
| gene | yes indeed | 19:25 |
| fenn | it looks like he's describing a turing machine | 19:25 |
| gene | maybe he could sue all kingdoms of life for infringing on his patent | 19:26 |
| gene | he allegedly built one in 1996 | 19:26 |
| gene | all there is to support this claim is a blurry photograph | 19:27 |
| gene | just like big foot | 19:27 |
| gene | or the loch ness monster | 19:27 |
| gene | but a cellular automaton approach like he and toth-fejel describe | 19:33 |
| gene | simplifies things a lot | 19:34 |
| gene | it makes things vastly simpler for software to do | 19:36 |
| fenn | lol qbasic | 19:36 |
| * fenn is so moving to guptastan | 19:41 | |
| fenn | rofl | 19:57 |
| fenn | The same nefarious accusers of greed are working in tandem with large multinational corporations who in actual greed are seeking to water down patent laws and other intellectual property rights simply to pay less royalties for innovations within big business. This is the real source of your "open source" movement and "globalization" movements and "patent reforms" and "patent harmonization" movements who play the greed card all | 19:57 |
| nsh | "plat the green card all" | 19:58 |
| nsh | (cut) | 19:58 |
| fenn | all the way to their big banks. | 19:59 |
| fenn | and then he goes on to say that nasa is infringing on his patent merely for showing pictures of colorized tiles on the internet | 20:00 |
| * nsh chuckles | 20:02 | |
| fenn | seems to be angry at freitas, bowyer, et al because he can't get any funding | 20:03 |
| fenn | ok cant read any more of that crap | 20:04 |
| kanzure_ | fenn: re: the patent stuff there "the real reason of your 'open source' movement" | 20:06 |
| kanzure_ | that's funny. | 20:06 |
| kanzure_ | in the engineering class that I'm taking, the professor says that patents are bad because they don't offer engineers protection "it's pre-patent that you want, that way the FBI will hunt down and kill anybody that infringes upon your idea" | 20:06 |
| kanzure_ | methinks everyone is confused about patents | 20:06 |
| kanzure_ | or something | 20:06 |
| kanzure_ | what else would explain all of this? | 20:07 |
| fenn | pre-patent? what's that mean? | 20:07 |
| kanzure_ | in the review queue | 20:07 |
| fenn | why does that offer more protection? | 20:07 |
| kanzure_ | "stuff that the uspto is supposed to get to" | 20:07 |
| kanzure_ | dunno, apparently "it just does" | 20:07 |
| fenn | is that like "patent pending"? | 20:07 |
| kanzure_ | ah, perhaps it is | 20:08 |
| fenn | "everyone is confused about patents" probably because we aren't living under the Rule of Law | 20:08 |
| kanzure_ | under the what? | 20:08 |
| kanzure_ | what are you about to be going on about now? | 20:08 |
| fenn | you know, the idea that law is supposed to be applied equally regardless who you are | 20:09 |
| fenn | but anyway the patent system is impossible now due to computational intractability | 20:09 |
| fenn | especially with people going and patenting any fabricational device with a data storage | 20:09 |
| fenn | or some crap like that | 20:10 |
| kanzure_ | the other day I found that fabuntu was just a collection of three or four scripts on top of ubuntu | 20:10 |
| fenn | my toy from last night: http://fennetic.net/pub/camera/DCP_0836.JPG | 20:10 |
| kanzure_ | the three or four scripts being the tie-in between different cad tools and the device drivers or something | 20:10 |
| fenn | this shows the C-axis rotation better: http://fennetic.net/pub/camera/DCP_0837.JPG | 20:11 |
| fenn | yeah i'm rather unimpressed with the software coming out of gershenfeld's army | 20:11 |
| kanzure_ | he has a friggin army | 20:11 |
| kanzure_ | shouldn't there be, uh, more? | 20:11 |
| fenn | he does | 20:11 |
| kanzure_ | it doesn't seem like the code is to really operate and manage a fablab | 20:12 |
| kanzure_ | although I've not seen any *lab or *shop for that matter that is able to keep track of all of its materials, tools, procurement, etc. anyway | 20:12 |
| fenn | it's because they were fighting with commercial CAM software which is stuck in the 70's | 20:12 |
| kanzure_ | how would you do that? | 20:12 |
| kanzure_ | I'd be much too tempted to just bring in boxes of stuff and not categorize it | 20:12 |
| kanzure_ | I know there's all sorts of content management systems | 20:13 |
| fenn | you mean inventory systems? | 20:13 |
| kanzure_ | yes | 20:13 |
| fenn | it's more than just software | 20:13 |
| kanzure_ | stuff for managing shop floor activity as well. inventory, where things are (or are not), etc. | 20:13 |
| fenn | training monkeys is the hard part | 20:13 |
| fenn | that's why techshop is so lame | 20:13 |
| kanzure_ | training? not just getting them? | 20:13 |
| kanzure_ | techshop is also lame because you have to pay :) | 20:13 |
| fenn | getting monkeys is easy, they're running around all over the place | 20:13 |
| kanzure_ | hrm | 20:14 |
| fenn | its the wild monkeys that come in and abuse your inventory | 20:14 |
| * fenn has been cleaning up messes all day | 20:14 | |
| fenn | s/day/week/ | 20:14 |
| kanzure_ | but really, how much time would you spend making sure you properly check out tools or materials in your own personal shop? | 20:14 |
| kanzure_ | it's not like you're going to want to hack together a few pcb's one day and say "oh, geeze, I should really check this out from the inventory" | 20:15 |
| fenn | i dont check out tools, i put them back where they belong when there's too much crap lying everywhere | 20:15 |
| fenn | however, if the only way to get something was through a computerized storage system, i'd be able to see who last used my frobknob | 20:15 |
| kanzure_ | yeah, but what if somebody just took it? | 20:16 |
| kanzure_ | without swipping a barcode or something? | 20:16 |
| fenn | how do they know where it is? | 20:16 |
| kanzure_ | they see it? | 20:16 |
| fenn | no, it's in surplus tub 15-A | 20:16 |
| kanzure_ | so, surplus tub 15-A under some authentication system? | 20:17 |
| kanzure_ | *is under | 20:17 |
| fenn | well, back when i was thinking about this, i was about to move into a room that used to contain a 1 lane swimming pool | 20:17 |
| kanzure_ | erm, well, I guess this should be modularized and up to the implementation of the *shop/*lab of course, but we're still lacking a good system in general | 20:17 |
| kanzure_ | hm? | 20:18 |
| fenn | so all the storage would be under the floor in the swimming pool, accessible only by robot | 20:18 |
| kanzure_ | ah, so a robotic retrieval system would be my style :) | 20:18 |
| fenn | but it seems kinda slow | 20:18 |
| fenn | i mean waiting 30 seconds for a wrench | 20:18 |
| fenn | where's my RFID tag utopia | 20:18 |
| kanzure_ | and then somebody takes the rfid off | 20:19 |
| fenn | well at least we'll know they took the rfid tag off | 20:19 |
| fenn | most people who borrow tools arent actually thinking about stealing them | 20:19 |
| fenn | people like that should be shot on sight | 20:19 |
| kanzure_ | I'm just worrying about things that can go wrong | 20:19 |
| kanzure_ | things go wrong in file systems all the time | 20:19 |
| kanzure_ | (hdd file systems) | 20:19 |
| kanzure_ | even more so in paper filing systems | 20:19 |
| kanzure_ | hell if I know how wrong things can get in an automated tool shed system | 20:20 |
| fenn | worst case scenario is a nuclear explosion | 20:20 |
| kanzure_ | I guess. | 20:21 |
| fenn | unless you can think of worse | 20:21 |
| fenn | grey goo already happened billions of years ago | 20:21 |
| kanzure_ | what about material procurement? | 20:21 |
| kanzure_ | the way that the superbarges do it is via the typical containers that you see being trucked around on the highways | 20:22 |
| kanzure_ | each of these has a specific standard interface for package delivery | 20:22 |
| fenn | shipping containers are awesome | 20:22 |
| kanzure_ | and during their trip they are hooked up to electrical power at some wattage and voltage and such | 20:22 |
| kanzure_ | :) | 20:22 |
| fenn | before that you had ceramic urns and cargo nets | 20:22 |
| kanzure_ | which is nice since this allows systematic material shipping | 20:22 |
| kanzure_ | ceramic urns and cargo nets? I am not familiar with these. | 20:22 |
| fenn | because you were born in the era of shipping containers | 20:22 |
| kanzure_ | well, yeah | 20:23 |
| fenn | what do you mean "material procurement" anyway? | 20:23 |
| fenn | like a hopper full of bauxite? | 20:23 |
| fenn | a hopper full of fpga's? | 20:24 |
| kanzure_ | "fabuntu-fablab-thingy #234-79A, package delivery 63 for bryan .. contains platonium and 32 ounces of liquid amphetamines .. please open docking bay" :p | 20:24 |
| kanzure_ | erm | 20:24 |
| fenn | so the context is "how to integrate stuff into your inventory"? | 20:24 |
| * kanzure_ is honestly more worried about the monkeys at the reception bay for materials .. "hey, look, a box of new materials and hardware! now what?" | 20:24 | |
| kanzure_ | yeah | 20:25 |
| kanzure_ | specifically when it's not so specific | 20:25 |
| kanzure_ | just hire really really good monkeys? | 20:25 |
| fenn | hmm.. automated assimilation procedures | 20:25 |
| fenn | dunno, that's the classic software interoperability problem | 20:25 |
| kanzure_ | or pray that someone has already added the information into skdb for that package or box or something? | 20:25 |
| kanzure_ | obviously you'd bitchslap anybody who makes something and doesn't have a definition of it in the repositories, but when untagged/untracked materials are gathered from The Black, Cruel Dataless Abyss, .. | 20:26 |
| fenn | and the subcontractor problem.. they'll probably make their own entry and ignore the others already in the db | 20:26 |
| fenn | well you cant have something in the inventory with zero data about it | 20:26 |
| fenn | except for rats and fungus n stuff | 20:26 |
| kanzure_ | let's say you have two shops deployed somewhere | 20:27 |
| kanzure_ | send them two identical boxes | 20:27 |
| fenn | you should know the date it was entered at least | 20:27 |
| kanzure_ | of materials that they do not know about | 20:27 |
| kanzure_ | somehow you have to make sure that both can input the data in some nonredundant manner | 20:27 |
| kanzure_ | or something. | 20:27 |
| kanzure_ | maybe it would be better if they were able to acquire the history of the materials and tools and other things they add to their inventory? | 20:28 |
| fenn | black box number d3b07384d113edec49eaa6238ad5ff00 | 20:28 |
| kanzure_ | i.e., "hey, I found this package off of the streets from some guy named Fred here .. he says he got it from the mines" | 20:28 |
| kanzure_ | heh | 20:28 |
| fenn | go ahead, try to figure out what it is | 20:28 |
| kanzure_ | really this is sort of like the bibliographing/sourcing problem | 20:28 |
| kanzure_ | oh, do you want me to go to #md5 on rizon or something? | 20:28 |
| kanzure_ | their md5 cracking system and such. | 20:29 |
| fenn | i'm wondering if this bot can solve a simple problem | 20:29 |
| kanzure_ | (s/cracking/reverse lookup/) | 20:29 |
| kanzure_ | what do you mean? | 20:29 |
| fenn | whether it works at all | 20:29 |
| fenn | right, well, if you get stuck, just remember "google knows all" | 20:33 |
| kanzure_ | what's going on | 20:33 |
| kanzure | foo? | 20:33 |
| fenn | i have a headache from this ashwaganda stuff | 20:33 |
| fenn | yep 'foo' | 20:34 |
| fenn | did the bot get it? | 20:34 |
| kanzure_ | google got it | 20:34 |
| kanzure_ | the bot will get it though, yes | 20:34 |
| kanzure_ | they have all strings up to 7 digits or something | 20:34 |
| kanzure_ | 7 characters | 20:34 |
| fenn | why is it taking so long? | 20:34 |
| kanzure_ | uh? because I'm lazy? | 20:34 |
| fenn | oh | 20:34 |
| fenn | nevermind then | 20:35 |
| kanzure_ | essentially this is the same problem with recycling. | 20:35 |
| fenn | assimilation | 20:36 |
| kanzure_ | hard enough to get people to remember to throw their plastic bottles in the 'plastic bottle bin', and don't even begin to laugh at the people that have to make the systems to separate the types of plastics :/ | 20:36 |
| fenn | you can separate most plastics by density in a water-alcohol gradient | 20:36 |
| kanzure_ | then why does everybody complaing about the recycling-sorting/classification problem? | 20:37 |
| fenn | i think they just hire people to sort it by hand though | 20:37 |
| fenn | or dump it into the ocean | 20:37 |
| fenn | probably the latter | 20:37 |
| fenn | ever played those electronic 20 question games? | 20:44 |
| kanzure_ | maybe. | 20:44 |
| kanzure_ | I've played the one on the web that claims it's ai | 20:44 |
| fenn | i was thinking we could write a categorization expert system, a search tree sorta like that, but you could add new categories/end nodes | 20:44 |
| kanzure_ | and what would the input be? | 20:45 |
| kanzure_ | the super-measures-everything-identfies-anything system? | 20:45 |
| fenn | so when inventory monkey is trying to put an unclassified item in the system it would make him either find the existing entry to add a new one | 20:45 |
| kanzure_ | oh, ask the monkey the questions? | 20:45 |
| fenn | s/to/or/ | 20:45 |
| kanzure_ | so only monkey-see becomes become do? what about monkey-no-see, such as distinctions that are hard for the untrained to see? | 20:46 |
| kanzure_ | erm | 20:46 |
| kanzure_ | imagine the analytical chem lab | 20:46 |
| kanzure_ | they do some fairly complicated analyses on the materials they receive from forensics peoples, for instance | 20:47 |
| fenn | hmm. supplement with automated analytical testing? | 20:47 |
| kanzure_ | and I know they're following protocols, do we want monkeys to follow protocols? | 20:47 |
| fenn | if monkey receives unmarked baggie of white powder, i dunno how to get a good answer out of him otherwise | 20:47 |
| kanzure_ | yeah, automated analytical testing is one idea, but that sounds like an 'inverse reprap machine' :p | 20:47 |
| kanzure_ | i.e., "has all possible gizmos and gadgets to poke and prod at stuff" | 20:47 |
| kanzure_ | *poke* | 20:47 |
| fenn | dammit there has to be something at least approximating a tricorder | 20:47 |
| fenn | gcms is pretty close | 20:48 |
| fenn | raman spectrometer is good for things that have been surveyed already | 20:48 |
| fenn | biological materials, well, no ideas there | 20:48 |
| kanzure_ | is there a way to get an object's hash other than looking at its overall wavefunction? (we can only get wavefunctions for very, very tiny molecules, and most atoms) | 20:48 |
| kanzure_ | yeah .. | 20:48 |
| kanzure_ | bah | 20:48 |
| fenn | PCR i guess | 20:48 |
| kanzure_ | people have a pain classifying organisms as it is | 20:48 |
| kanzure_ | I've heard horror stories of biologists mistaking obvious flowers and plants that nobody catches simply because we're all clueless | 20:49 |
| fenn | you can get an object's hash by shooting an x-ray laser at it :\ | 20:49 |
| kanzure_ | joke? | 20:49 |
| fenn | then looking at the scatter pattern | 20:49 |
| fenn | no joke | 20:49 |
| fenn | think about it | 20:49 |
| kanzure_ | I'm imaging medical x-rays and all I see is that similar sized objects cast similar 'shadows' and 'masks' not allowing the photons to pass | 20:50 |
| fenn | i'm thinking something more like x-ray crystallography | 20:50 |
| fenn | not a point light source, a planar wave | 20:51 |
| fenn | there was some paper that got me thinking about it.. will look | 20:51 |
| kanzure_ | ooh | 20:52 |
| kanzure_ | what about this | 20:52 |
| kanzure_ | debian has free and nonfree | 20:52 |
| fenn | assimilated and nonassimilated? | 20:52 |
| kanzure_ | so anything that you make that has unsourced materials, has to be labeled as such, i.e. "thar be black magic here" | 20:52 |
| kanzure_ | yes | 20:52 |
| fenn | useful and useless? | 20:52 |
| kanzure_ | heh | 20:52 |
| kanzure_ | basically | 20:52 |
| fenn | 'tainted' | 20:52 |
| fenn | it's a start | 20:53 |
| fenn | varying levels of taintedness too | 20:53 |
| bkero | I'm black magick. | 20:53 |
| fenn | like if i know how to make a computer chip, but dont have the infrastructure | 20:53 |
| fenn | or rare isotopes | 20:53 |
| kanzure_ | it'd also be nice to see if people have previously tried nontainted alternatives so that you don't have to repeat stupid stuff | 20:54 |
| kanzure_ | which would obviously be uploaded in the repo | 20:54 |
| kanzure_ | unless they want me to visit them and cut off their legs | 20:54 |
| fenn | and hopefully in the same category | 20:54 |
| fenn | else you'd never find them | 20:54 |
| kanzure_ | well, surely there should be 'history' and 'use cases' and various reports (even if automatically generated) attached to each part and so on | 20:54 |
| kanzure_ | I was thinking that would be the case anyway | 20:54 |
| fenn | http://www.media.mit.edu/sponsors/uspatents.pdf search for hash | 20:56 |
| fenn | whoever invented the 'select text and type to delete everything' interface needs to be shot | 20:57 |
| kanzure_ | everyone needs to be shot at least once | 20:57 |
| kanzure_ | but what would be a better thing to do with selected text that you keypress on? | 20:57 |
| kanzure_ | besides copy :) | 20:57 |
| fenn | its patent 6584214 | 20:58 |
| bkero | I don't need to be shot. | 20:58 |
| kanzure_ | you've compiled gentoo | 20:58 |
| kanzure_ | yes you do | 20:58 |
| kanzure_ | :p | 20:58 |
| bkero | kanzure_: I do it for a living. ;) | 20:58 |
| bkero | and I have damn good reasons for doing so ;) | 20:59 |
| kanzure_ | shot nonfatally, I mean | 20:59 |
| kanzure_ | "Ouch! Quick! Everybody out of the genetic pool!" | 21:00 |
| bkero | I realize that :P | 21:00 |
| bkero | I'm stirring some bleach into the genetic pool. | 21:00 |
| kanzure_ | fenn: that patent looks like something mostly structural | 21:00 |
| fenn | the physical object hash idea came up because i was thinking about how marketing companies would protect their brand image in the age of molecular nanotech | 21:00 |
| fenn | they would publish hash keys of a "genuine" apple nano-pod or whatever | 21:01 |
| kanzure_ | but we already have people doing reverse hashes of strings | 21:01 |
| fenn | no, they arent doing reverse hashes, they're systematically building a db of forward hashes | 21:01 |
| fenn | there's a difference | 21:01 |
| kanzure_ | since any object would be just a sequence of bits instructing the construction of the object, I fail to see how anything but "well, it has a large state space to permutate through" keeps it safe | 21:01 |
| kanzure_ | okay. | 21:01 |
| kanzure_ | I guess you can just hope nobody can do enough hashes to get the right ipod nano or something | 21:02 |
| fenn | well the problem is it takes a really long time to dissect something down to the individual atoms | 21:02 |
| fenn | so only the originator would have an atom-perfect copy | 21:02 |
| fenn | trade secret, not copyright | 21:03 |
| fenn | blah | 21:03 |
| fenn | not like it matters whose name is on your clothing | 21:04 |
| bkero | I don't have names on my clothing. | 21:05 |
| bkero | MD5 has already been broken. | 21:05 |
| kanzure_ | my clothing has no words on it | 21:05 |
| fenn | bkero: howzat? | 21:05 |
| kanzure_ | oh, that's a gershenfeld patent | 21:06 |
| bkero | HD5 and SHA have both been cracked and/or broken. | 21:06 |
| bkero | *MD5 | 21:06 |
| fenn | i consider a hash broken when you can make two different files with arbitrary content that have the same hash - is that what you mean? | 21:07 |
| bkero | http://en.epochtimes.com/news/7-1-11/50336.html | 21:08 |
| fenn | why would they announce that to the world? | 21:09 |
| bkero | Because they're academics | 21:09 |
| bkero | Why wouldn't you announce it to the world? | 21:10 |
| fenn | because you're working for the chinese gov't? | 21:10 |
| fenn | because you can sell it for billions of dollars? | 21:10 |
| bkero | Some people don't want money. | 21:10 |
| kanzure_ | what he's wondering is why somebody who did want money didn't get it first | 21:11 |
| bkero | It's just you and the rest of us fat americans who want to snort cocaine off a $10,000/night hookers ass | 21:11 |
| kanzure_ | $10k/night? | 21:11 |
| fenn | i could do a lot of things with a cool 'bil | 21:11 |
| bkero | There isn't much of an industry for cracking hashes | 21:12 |
| kanzure_ | hah | 21:12 |
| * kanzure_ has had to crack many a hash | 21:12 | |
| kanzure_ | well, not crack | 21:12 |
| kanzure_ | erm. | 21:12 |
| bkero | You didn't have to create algorithms for cracking hashes though. :P | 21:12 |
| fenn | "I hope efforts to protect against password theft will benefit [from this]" | 21:12 |
| kanzure_ | are you sure it was 'from this' ? heh' | 21:13 |
| fenn | but if you didnt announce to the world how to crack the hash there wouldnt be any problem! | 21:13 |
| bkero | fenn: uh, unless someone else stole her research or figured it out for themselves | 21:13 |
| bkero | obscurity != security | 21:13 |
| fenn | is there such a thing as "unbreakable crypto"? | 21:14 |
| fenn | it seems more like the history of science showing how un-special humans are | 21:14 |
| bkero | fenn: Nobody has cracked RSA yet. | 21:14 |
| kanzure_ | cite thermodynamic arguments against the existence of 'security' here | 21:15 |
| fenn | "ah but we are the only animal that sings sonnets with iambic pentameter" | 21:15 |
| bkero | Need I remind you of the superiority of the dolphins? | 21:16 |
| fenn | please do so | 21:16 |
| bkero | Dolphins: Second most intelligent life-form to have inhabited Earth. Most intelligent of Earths indiginous species. Subordinate only to a race of Hyperintelligent pan-dimensional beings know commmonly as mice. Man has always assumed that he is more intelligent than dolphins because he has accomplished so much: the wheel, New York, Wars, and so on, while all the dolphins ever did was muck about in the water having a good time. However, dolphins consider themselves far more intelligent than man for exactly the same reasons. | 21:16 |
| kanzure_ | Cutoff at "However, dolphins consider" | 21:17 |
| bkero | However, dolphins consider themselves far more intelligent than man for exactly the same reasons. | 21:17 |
| fenn | last i saw they could be coerced to jump through hoop for kippers | 21:18 |
| kanzure_ | I'm not sure it's worth pursuing security like that, fenn | 21:18 |
| kanzure_ | maybe backup strategizing, sure | 21:18 |
| bkero | Curiously enough, the dolphins had long known of the impending destruction of the planet Earth and had made many attempts to alert mankind to the danger; but most of their communications were misinterpreted as amusing attempts to punch footballs or whistle for titbits, so they eventually gave up | 21:18 |
| kanzure_ | but relying on RSA, MD5, ugh. | 21:18 |
| fenn | oh, atom-level product verification? | 21:18 |
| fenn | it could be useful | 21:19 |
| bkero | I wonder if anybody is close to cracking RSA public key encryption. | 21:19 |
| kanzure_ | in corporate situations maybe :) | 21:19 |
| kanzure_ | or govt | 21:19 |
| bkero | DSA got owned, SHA1, SHA0, and MD5 got owned. I think SHA256 and Blowfish are still both alright. | 21:19 |
| fenn | i was thinking more like for checking DNA or its artificial equivalent | 21:19 |
| bkero | The government doesn't have much of a fucking clue about encryption. Sure, they employ cryptographers, but they outsource most of their cracking needs. | 21:20 |
| fenn | bkero: how do you know they havent been sold to the russian mafia for $$$ by unscrupulous math student? | 21:20 |
| kanzure_ | "Today we tell Little Johnny that his mitochondrial hash is wrong." | 21:20 |
| bkero | You can't really have a hash with significant entropy. | 21:20 |
| kanzure_ | what sort of entropy are you talking about here | 21:21 |
| kanzure_ | informational entropy? | 21:21 |
| bkero | Genetic mutation | 21:21 |
| fenn | what does that mean anyway | 21:21 |
| kanzure_ | suppose: scan in your genome, you get your information, you hash it | 21:21 |
| kanzure_ | oh, bkero, sure | 21:21 |
| kanzure_ | mitochondria have something like 13 proteins IIRC | 21:21 |
| kanzure_ | 13,000 bp or something | 21:21 |
| kanzure_ | so yes, it does mutate | 21:21 |
| fenn | kanzure_: i was thinking, you just hash the info in place without scanning it at all | 21:21 |
| kanzure_ | I think you were mentioning something like that a while back, yes | 21:22 |
| fenn | doesn't work with DNA because it's floating around in fluid | 21:22 |
| bkero | XORing it is easier than hashing it. | 21:22 |
| fenn | but some crystalling memory storage is different | 21:22 |
| bkero | Plus, don't we have biological XOR gates already? | 21:22 |
| kanzure_ | hrm | 21:22 |
| kanzure_ | do we ? | 21:22 |
| fenn | XOR is a hash algo, just not a very good one | 21:22 |
| kanzure_ | Hey Dan. | 21:23 |
| bkero | Not a very good one for collisions. | 21:23 |
| facefaceface | anyone good with graph stuff? | 21:23 |
| bkero | It's a very good one for things otherwise | 21:23 |
| facefaceface | hey kanzure_ | 21:23 |
| bkero | Which is why it's used in RAID5 | 21:23 |
| kanzure_ | facefaceface: sort of, yes :) | 21:23 |
| kanzure_ | facefaceface: what sort of graph stuff? | 21:23 |
| facefaceface | here is my questiobn... | 21:23 |
| fenn | it's more of a checksum than a hash | 21:23 |
| facefaceface | hi, I'm looking for some algorithmic advice... I have a network (or graph) structure stored in a table as an 'adjacency list', if I represent this as a matrix, I can calculate the degree of each node by squiring the matrix. Is there an equivalent adjacency list operation? | 21:23 |
| bkero | How is it a checksum? | 21:23 |
| facefaceface | squaring | 21:23 |
| kanzure_ | That's odd. | 21:23 |
| kanzure_ | What's your table structure? | 21:24 |
| facefaceface | Currently I'm using a grouping query over the union of two group by queries... | 21:24 |
| facefaceface | Col 1 = node id x, Col 2 = node id y | 21:24 |
| facefaceface | where each row is an edge | 21:24 |
| kanzure_ | yeah, that's what I've been doing recently | 21:24 |
| fenn | bkero: if the source is different, you get a different output, but it's easy to reverse the function | 21:24 |
| kanzure_ | I have something like nine tables all being selected at once | 21:24 |
| kanzure_ | based off of some ID columns and such | 21:24 |
| kanzure_ | (graph-based mechanical engineering designs, you see) | 21:24 |
| fenn | bkero: "reverse" meaning make two different files with the same checksum | 21:25 |
| facefaceface | OK | 21:25 |
| facefaceface | seems like simply squaring the matrix to calculate the degree is a really elegant solution, however, it is not elegant on an adjacency list | 21:25 |
| kanzure_ | facefaceface: So, I think you might have the wrong type of table structure to calculate the degree of each node. | 21:25 |
| facefaceface | the adj -> mat is messy | 21:26 |
| kanzure_ | how does your adjaceny matrix work when there's more than one edge at that vertice? | 21:26 |
| kanzure_ | like three or four incoming vertices? | 21:26 |
| facefaceface | the primary key is x,y | 21:26 |
| facefaceface | a -> b | 21:26 |
| facefaceface | a -> c | 21:26 |
| facefaceface | a -> d | 21:26 |
| facefaceface | etc | 21:26 |
| kanzure_ | each of those 3 is one row in the table? | 21:26 |
| facefaceface | yes | 21:26 |
| kanzure_ | ah, that's better | 21:26 |
| facefaceface | so I group by ( group by x union group by y ) | 21:27 |
| facefaceface | currently | 21:27 |
| kanzure_ | why can't you just count the number of rows that have the first node = 'A', and the result is therefore your total count? | 21:27 |
| kanzure_ | in MySQL this is something like SELECT count(ID) FROM mytable WHERE A='A' | 21:28 |
| kanzure_ | where table 'mytable' has columns ID (unique primary-key, in MySQL lingo), A, and (presumably also) B | 21:28 |
| facefaceface | sure, but I want to get the answer for every node in one query | 21:28 |
| kanzure_ | hm. | 21:28 |
| fenn | what are you trying to do? find the number of isolated graphs? | 21:28 |
| bkero | fenn: XORing is very cheap though :) | 21:29 |
| fenn | bkero: not if it's a chunk of un-scanned frozen xenon | 21:29 |
| facefaceface | fenn, I'm trying to count the number of edges for each node | 21:29 |
| fenn | for i in graph: i->count++ ? | 21:30 |
| kanzure_ | it would be nice to be able to say "SELECT count(ID) AS $blah-count WHERE $blah IS GROUPED FROM COLUMN 'A'" | 21:30 |
| kanzure_ | fenn: but he wants an sql query :/ | 21:30 |
| fenn | fff fuck sql | 21:30 |
| kanzure_ | sql queries usually don't have scripts inside of them | 21:30 |
| kanzure_ | well, there is one option that I can think of, but it kind of sucks | 21:31 |
| facefaceface | Never mind, I have code, its just HUGE.. | 21:31 |
| bkero | fenn: I'm talking about computationally cheap | 21:31 |
| kanzure_ | facefaceface: Oh, well, hold on a sec | 21:32 |
| fenn | bkero: me too.. hence all the discussion about x-ray lasers as an alternative to laboriously scanning each atom | 21:32 |
| kanzure_ | select * from the table GROUP BY 'A' | 21:32 |
| kanzure_ | then for each result that you get, | 21:32 |
| fenn | bkero: the difference is the computation happens "out there in the ether" instead of on a CPU | 21:32 |
| kanzure_ | generate a query dynamically | 21:32 |
| kanzure_ | SELECT count(ID) FROM mytable WHERE A='$that_dynamic_result' | 21:32 |
| kanzure_ | this is something like two for loops, one within the other | 21:32 |
| facefaceface | http://pastebin.com/d5b7d67b8 | 21:32 |
| kanzure_ | it results in a retarded number of queries to the db though | 21:33 |
| kanzure_ | and it's not your single query :( | 21:33 |
| bkero | That's a lot of queries | 21:33 |
| kanzure_ | bkero: From the pastebin or me? | 21:34 |
| bkero | Your looping selects | 21:34 |
| facefaceface | the paste is a bit overly complex... | 21:37 |
| kanzure_ | uh, anyway, | 21:40 |
| facefaceface | simplified paste... http://pastebin.com/d3f7d37a7 | 21:40 |
| kanzure_ | fenn: untainted would be directly drawn from the earth and the asteroids and distributed to spawn off projects or people's usage | 21:41 |
| kanzure_ | so everything might have to be somewhat tainted in the mean time | 21:41 |
| kanzure_ | unless we have like absolute total confirmation :) | 21:41 |
| kanzure_ | which would mean we still need data monkeys anyway | 21:41 |
| fenn | what is COUNT(*) | 21:42 |
| kanzure_ | counts the number of results with "*" column. | 21:43 |
| kanzure_ | presumably all of them should have that column | 21:43 |
| fenn | kanzure_: still requires an earth and asteroids etc | 21:43 |
| fenn | hence varying degrees of taintedness | 21:43 |
| kanzure_ | but if you're doing a multi-table select statement, results from one table that don't match with another wouldn't have that column or something | 21:43 |
| kanzure_ | fenn: what? | 21:43 |
| kanzure_ | fenn: in the automated procurement instances, presumably you have land surveying equipment and such | 21:43 |
| kanzure_ | so that you at least know preliminarily what the incoming materials should be | 21:44 |
| kanzure_ | then you just need to make sure nothing has damaged it on its way from mining operation to fablab-267a | 21:44 |
| fenn | eh sorry i mixed up taintedness dimensions | 21:44 |
| fenn | there's unobtanium and mysterium | 21:45 |
| kanzure_ | and turtles and pythons. | 21:45 |
| fenn | complexity and nonlinear dynamics! | 21:45 |
| kanzure_ | guess I have to live with the monkeys and automated material analysis equipment ideas for now | 21:47 |
| fenn | is there a 'sql to C converter' | 21:47 |
| kanzure_ | django is python->SQL for what it's worth | 21:47 |
| kanzure_ | it converts the data structures into sql tables and the like | 21:47 |
| kanzure_ | there really should be sql->C-datastructs. | 21:48 |
| kanzure_ | it's not hard. | 21:48 |
| fenn | i meant sql query to c code | 21:48 |
| fenn | anyway, its just me being lazy | 21:48 |
| kanzure_ | it's weird how projects get completed. materials coming from out of nowhere, out of the blue sometimes. this is why I wanted to do the demo website showing four little material sourcing companies putting up responsible XML-like files to say what they have, and to do dynamic bidding contracts (or at least saying "yes, we could give 50 Mg of carbon for this project" as a designer sits there making up some system from the parts + materials databases. | 22:09 |
| kanzure_ | (of course, then the patrons would need to have monkeys too) | 22:10 |
| kanzure_ | erm | 22:10 |
| kanzure_ | after 'materials databases.' It was: | 22:10 |
| kanzure_ | But it also occurs to me that it might be interesting to put everything that is 'tainted', in the sense of (1) having materials that aren't characterized, and (2) having materials that aren't available on the system/network, thrown up on a website where patrons can kindly see what everybody needs to get their Awesome Project rolling. These patrons would just have some preferential selection algorithms to see which designs and projects would be inte | 22:10 |
| fenn | you know about 'rational street performer protocol'? | 22:11 |
| kanzure_ | .. and projects would be .. just press the 'go button' | 22:11 |
| kanzure_ | no? | 22:11 |
| kanzure_ | http://logiarithmic.net/pfh/rspp | 22:11 |
| fenn | takes a while to read and then seems like common sense once you get it: http://www.logarithmic.net/pfh/rspp | 22:11 |
| fenn | he says money but it could be just about anything | 22:12 |
| kanzure_ | 'system of conditional pledges' <-- yay | 22:12 |
| kanzure_ | (which is also the "ebay with agents to proxy for you while you're away just in case you want to be in on that awesome trading circle that spontaneously shows up") | 22:14 |
| kanzure_ | *show up | 22:14 |
| fenn | unfortunately equipment isn't infinitely divisible or easily valuated | 22:14 |
| fenn | so we're sorta back to barter | 22:14 |
| kanzure_ | what do you mean? | 22:14 |
| kanzure_ | I was thinking of something like | 22:14 |
| fenn | you can't pledge half an oscilloscope | 22:14 |
| kanzure_ | "here's my super-awesome rocket" | 22:14 |
| kanzure_ | "I need some metals that I don't have, who would like to help? | 22:15 |
| kanzure_ | " | 22:15 |
| kanzure_ | "ooh, I have teh metals. *click*" | 22:15 |
| fenn | hmm real life example | 22:15 |
| kanzure_ | heh, it would be interesting to see people pledging to go open source after they receive materials to do their first implementation | 22:15 |
| fenn | i sent two netier xl1000 thin client computers to marcin jakubowski, on the premise that he would use them to complete his CNC acetylene torch table | 22:16 |
| kanzure_ | yeah? | 22:16 |
| fenn | however he doesn't have stepper drivers etc | 22:16 |
| kanzure_ | is that just a cnc with a torch tooltip? | 22:16 |
| fenn | now i paid good money for these so i dont want them to just go to waste, i'd rather wait until he has pledges for stepper drives | 22:16 |
| kanzure_ | by "just" I mean "just the most awesome thing ever" | 22:16 |
| fenn | yes, its for cutting out compressed earth block press parts | 22:17 |
| kanzure_ | bwahah. | 22:17 |
| fenn | an equivalent plasma cutter would cost too much | 22:17 |
| fenn | (steel is too thick) | 22:17 |
| fenn | (i guess) | 22:17 |
| fenn | or maybe they just wanted to do something semi-original | 22:17 |
| fenn | anyhoo | 22:18 |
| kanzure_ | it's not right for things to fall off the face of the internet just because somebody couldn't get materials to do it | 22:18 |
| kanzure_ | unless it was unobtanium in the sense of impossibilium | 22:19 |
| fenn | i'm not sure about that | 22:19 |
| fenn | a lot of projects fall under the category of "bad art" | 22:19 |
| kanzure_ | the data should still be there. | 22:19 |
| fenn | take mister collins 'self replicator' for example | 22:20 |
| kanzure_ | would you kill a human for being bad art? | 22:20 |
| kanzure_ | even collins' 'replicator' is in the archive | 22:20 |
| kanzure_ | (uh, maybe) | 22:20 |
| kanzure_ | *the internet archive | 22:20 |
| fenn | kanzure_: there's a difference, a human has intrinsic value, art does not | 22:20 |
| fenn | art has value through being appreciated | 22:20 |
| kanzure_ | "Self-creation is the highest art." | 22:20 |
| fenn | says so-and-so | 22:21 |
| kanzure_ | engineering is not an art? | 22:21 |
| kanzure_ | this is a slippery slope | 22:21 |
| fenn | sometimes it's not | 22:21 |
| kanzure_ | not sure I care enough | 22:21 |
| fenn | anyway the point was there's a lot of dumbasses out there who will be accepting pledges of cool stuff so they can do stupid things with them | 22:21 |
| kanzure_ | doesn't this always happen? | 22:22 |
| fenn | i dont think it's unfair to not give them cool stuff | 22:22 |
| kanzure_ | ok ok | 22:22 |
| kanzure_ | let me qualify my statement, I'm not interested in fairness, I'd just rather not have things be deleted | 22:22 |
| kanzure_ | cleaning up my hdd's is like removing limbs, etc. etc. | 22:22 |
| fenn | that's completely different | 22:22 |
| fenn | data compression problem more than anything | 22:22 |
| kanzure_ | matter compression problem maybe? | 22:23 |
| kanzure_ | hlfkajfkl;as | 22:23 |
| fenn | if you can recreate the half finished project from what information is available, what's the problem? | 22:23 |
| kanzure_ | sourcing materials. | 22:23 |
| kanzure_ | which was what I was talking about in the first place :) | 22:23 |
| fenn | you're going in circles now | 22:24 |
| kanzure_ | oh | 22:24 |
| kanzure_ | I'm confused | 22:24 |
| kanzure_ | what am I supposed to be arguing? | 22:24 |
| fenn | how everyone is supposed to get everything | 22:24 |
| kanzure_ | that's a hard one. | 22:24 |
| fenn | it's "post-scarcity economics" | 22:25 |
| fenn | i think the answer is just to design free hardware with no unobtanium components | 22:26 |
| kanzure_ | what's the sliding scale on unobtanium/taintedness or how is that defined? | 22:26 |
| fenn | and reduce the price of materials to below the cost of keeping track of them | 22:26 |
| kanzure_ | it would be interesting to do it discretely for each material | 22:26 |
| kanzure_ | or each part and tool involved etc | 22:26 |
| kanzure_ | so for each person that wants to get that hardware or make it or whatever, | 22:27 |
| kanzure_ | they get their readout of materials they have in their inventory, as well as potential | 22:27 |
| kanzure_ | what did we call it? potential patrons? | 22:27 |
| fenn | potential donations? | 22:27 |
| kanzure_ | something like that | 22:27 |
| fenn | ooh maybe we can slip the word virtual in there somewhere | 22:27 |
| fenn | virtual donation *click* | 22:28 |
| fenn | i feel so... 90's | 22:28 |
| -!- percent_ is now known as jihaaaaaaad | 22:28 | |
| kanzure_ | do we just randomly add 'virtual' to anything or something? | 22:28 |
| fenn | no, only when it's virtual, silly | 22:28 |
| fenn | i havent actually donated my frobknob yet | 22:29 |
| fenn | its dependent on other virtual donations | 22:29 |
| kanzure_ | bloorgh | 22:29 |
| kanzure_ | Consider The Following | 22:29 |
| fenn | when all conditions are met, then they become real bequeathments (cant think of the right word) | 22:29 |
| kanzure_ | given no automatic metadata for materials and parts in a system, | 22:29 |
| kanzure_ | somebody could really really be smart constricting the growth of a community | 22:30 |
| kanzure_ | as long as that community is dependent on that single repository | 22:30 |
| kanzure_ | (I was thinking of this in the sense of modern economics. Without this context it's a "duh" thing.) | 22:30 |
| kanzure_ | okay, so, more intereting would be to make this more of a reality .. | 22:31 |
| fenn | yes most people just buy something from the store instead of going to the trouble of tracking a free one down | 22:31 |
| kanzure_ | how about a way to calculate an estimated amount of time until a project is from 'completely infeasible' to 'feasible' given the available donors and such? | 22:31 |
| fenn | we have this thing 'freecycle' but its such a joke | 22:31 |
| * kanzure_ uses freecycle | 22:31 | |
| fenn | i have all this crap i want to get rid of but the dumpster is so much easier | 22:32 |
| kanzure_ | I picked up a good number of boxes (both real boxes and boxes of boxen) from freecyclers | 22:32 |
| kanzure_ | also this monitor. | 22:32 |
| fenn | would much rather just take it to a warehouse and let people dig through it, instead of laboriously cataloging and then arranging pickups etc | 22:32 |
| kanzure_ | don't get me started on all of the random driving required for pickups | 22:33 |
| kanzure_ | that's just carpooling problem waiting to be solved | 22:33 |
| fenn | its a bad system | 22:33 |
| kanzure_ | shortest path algorithm, I mean. | 22:33 |
| fenn | 'hubbing it' works | 22:33 |
| kanzure_ | huh? | 22:33 |
| fenn | traveling salesman is a bad situation to be in | 22:33 |
| kanzure_ | the openmanufacturing@googlegroups.com guys wanted an open source automated car to pickup materials from locations on streets | 22:33 |
| kanzure_ | heh | 22:33 |
| fenn | if there were a freecycling center it might be worth it | 22:34 |
| fenn | openmanufacturing when? | 22:34 |
| kanzure_ | weeks ago? | 22:34 |
| fenn | it must have been in one of the three threads :P | 22:35 |
| kanzure_ | anyway, like I was saying | 22:35 |
| kanzure_ | it would be interesting to be able to give estimates about completability, feasability, time-to-manufacture, based off of all of the acquisitions that are required and such | 22:35 |
| kanzure_ | this would give people some interesting numbers to make donations to such initiatives | 22:36 |
| kanzure_ | either for individual projects, or upgrading 'fablab-253894' in their local community, whatever | 22:36 |
| fenn | that sounds like a system waiting to be cheated | 22:36 |
| kanzure_ | corporate goons? | 22:36 |
| fenn | "why of course my fusor 9000 is inherently feasible" | 22:36 |
| kanzure_ | wouldn't that be something that skdb validation scripts are checking? | 22:37 |
| kanzure_ | things requiring more research should be labeled as such. | 22:37 |
| fenn | i just need a five megawatt HV power supply, a porsche, and three sexy interns | 22:37 |
| kanzure_ | peer review too | 22:37 |
| * kanzure_ wishes he had three sexy interns. | 22:37 | |
| fenn | peer review = trust network | 22:38 |
| fenn | (badly implemented, historically) | 22:38 |
| kanzure_ | sure. pgp trust circles. | 22:38 |
| kanzure_ | 'trust circles of wankery' :( | 22:38 |
| fenn | right | 22:38 |
| fenn | with monopoly on funding decisions | 22:38 |
| kanzure_ | aww crap | 22:38 |
| kanzure_ | what sucks is that, as it stands, everybody is stuck on the same boat | 22:39 |
| kanzure_ | in the future it'd be easy to say "go find yer own damn molecular cloud in a few hundred thousand years" but not so much the case here. | 22:40 |
| fenn | semi-OT; emc machine preview/simulator can now show arbitrary geometry: http://emergent.unpy.net/files/sandbox/base_stl.png | 22:40 |
| kanzure_ | maybe we can hope that the donators with the resources are smart enough to avoid the fusor 9000s? | 22:41 |
| fenn | that was the idea behind capitalism | 22:42 |
| fenn | i'll let you decide if that was an optimistic or pessimistic statement | 22:42 |
| kanzure_ | oh, that's right, that's why we call it philanthropy .. you can't guarantee that something is going to work. but you can at least hold people accountable (capitalism?). | 22:42 |
| fenn | you distribute the computation | 22:43 |
| kanzure_ | hm? | 22:43 |
| fenn | however it unfairly(?) forces multiple people to perform the same estimation calculation when it could be shared | 22:43 |
| kanzure_ | also: debian and USPTO have their little guardians blocking the gates from bullshit. uspto less so. | 22:44 |
| fenn | "is the fusor 9000 worth investing X dollars in?" could be hashed and the result shared (tagged with the hash key) | 22:44 |
| fenn | kanzure_: one thing i dont like about debian is precisely that, but i dont know how to resolve it | 22:45 |
| kanzure_ | well, splitting it is one way I would think | 22:45 |
| fenn | its the same thing with wikipedia | 22:45 |
| kanzure_ | forking | 22:45 |
| kanzure_ | also user policies maybe .. "don't download xyz" | 22:45 |
| fenn | well, sorta, the reason it is the way it is, is the structure of the system that generates the social situation | 22:45 |
| fenn | (yuck) | 22:45 |
| fenn | so forking won't do anything different without a shitload of refactoring, because you are running the same system | 22:46 |
| kanzure_ | benevolent dictator that just-so-happens-to-be-running-things-well? (or to-have-stumbled-into-a-good-previous-dictatorship) | 22:46 |
| fenn | WP doesnt really have a benevolent dictator, it's too big | 22:47 |
| kanzure_ | I'm not really sure if you could say they have social /structure/ | 22:47 |
| kanzure_ | I know some admins that seem to have very little, if any ties, to the inner community of adminships | 22:48 |
| fenn | hm. i dont really know. would be nice to see some pretty pictures | 22:48 |
| kanzure_ | that's actually doable .. via looking for people talking to each other on the talk pages | 22:48 |
| fenn | also what do you compare it to? | 22:49 |
| fenn | does myspace have a social structure? what does? | 22:49 |
| kanzure_ | do you remember everybody hating somebody on myspace? | 22:49 |
| kanzure_ | the guy who put himself as everybody's friend | 22:49 |
| kanzure_ | that's hardly structure. even though the graph says so. | 22:49 |
| fenn | show me social structure and we might have something worth studying | 22:51 |
| facefaceface | I wanted to implement the 'opinion-ome', a wiki where you place an opinion, then tag it with a category. | 22:52 |
| kanzure_ | facefaceface: http://canonizer.com/ does that | 22:52 |
| kanzure_ | it sucks. | 22:52 |
| kanzure_ | trust me. | 22:52 |
| fenn | Total Information Awareness program was doing in-depth research on strategies to analyze social networks to determine whether or not U.S. citizens were political threats. | 22:52 |
| facefaceface | then people rate your opinion, and other people rate their rating of your opinion | 22:52 |
| kanzure_ | yes, that's exactly what canonizer is | 22:52 |
| fenn | its quite possible there's a "conspiracy" to break up any significantly organized structures | 22:52 |
| facefaceface | and the rating of their opinions weights their weighting of your opinion | 22:52 |
| kanzure_ | so you're talking about the different types/implementations of social structs, yes fenn? | 22:53 |
| kanzure_ | I wonder what the structure equates to though | 22:53 |
| facefaceface | kanzure_, does it use categories? | 22:53 |
| kanzure_ | facefaceface: yep | 22:53 |
| fenn | just localized order created by rules in a system | 22:53 |
| kanzure_ | and you get to place your vote on issues with various people that share your opinion set or whatever | 22:53 |
| kanzure_ | all sorts of bullshit. | 22:53 |
| * fenn is getting very NKS'y | 22:53 | |
| facefaceface | so my opinion in cookery counts for more, because people who are cooks rate my opinions? | 22:53 |
| kanzure_ | fenn: yeah, but what do the rules /imply/ ? like, debian's social contract. great. but there's no 'consequences'. or "this social structure is simply phpbb. k. done." | 22:54 |
| facefaceface | hmm.. http seems down, but I can talk here | 22:54 |
| fenn | opinionrank algorithm? | 22:54 |
| kanzure_ | or "weekly irc meetings" (which was the case) | 22:54 |
| facefaceface | doh | 22:54 |
| facefaceface | fenn, zactly | 22:54 |
| fenn | kanzure_: well, the structure can possibly say something about the rules that created it | 22:55 |
| fenn | you can then infer what sort of rules lead to different social structures | 22:55 |
| kanzure_ | what do we mean by structure | 22:55 |
| fenn | and decide which rules to use based on your preferred structure | 22:56 |
| fenn | i dont know, i'm sorta like a blind man with no elephants | 22:56 |
| facefaceface | kanzure_, I'll have to look at http://canonizer.com/ more closely - it seems to be what I was thinking about. | 22:56 |
| kanzure_ | facefaceface: I have ssh access if you ever want to do some development work or something | 22:56 |
| kanzure_ | fenn: you mean no turtles. | 22:57 |
| fenn | canonizer has ~zero participation, there must be some reason it's unpopular | 22:57 |
| kanzure_ | there is good reason it's unpopular. | 22:57 |
| kanzure_ | if you meet brent, you start to see why. | 22:57 |
| facefaceface | rules like X believes y, Z believes X, so 'we' should believe y to the extent that X does and to the lesser extent that Z does via X | 22:57 |
| kanzure_ | he's going for goody-goody "everybody's opinion matters! stop telling me your opinion and go systemicize it and make friends with others with the same opinion" | 22:57 |
| kanzure_ | erm | 22:57 |
| kanzure_ | it's worse than I'm making it sound | 22:57 |
| facefaceface | fenn, its hard to get people invoved with anythign that has no payoff | 22:57 |
| kanzure_ | it's also just a bad idea | 22:57 |
| kanzure_ | why would this system be centralized anyway, | 22:58 |
| fenn | facefaceface: thank you | 22:58 |
| facefaceface | kanzure_, why? | 22:58 |
| kanzure_ | "Hello, you've reached the Centralized Opinion System." | 22:58 |
| kanzure_ | "Please sit down to begin your prayers." | 22:58 |
| facefaceface | kanzure_, hello welcome to Nature | 22:58 |
| kanzure_ | the journal? | 22:58 |
| facefaceface | submit your paper and hope to hell we treat you fairly... | 22:58 |
| fenn | it has to be centralized in order to calculate the statistics | 22:58 |
| facefaceface | yes | 22:58 |
| fenn | the complete statistics | 22:59 |
| fenn | otherwise you're back to random sampling | 22:59 |
| fenn | or not-so-random sampling | 22:59 |
| kanzure_ | who says it's complete ... | 22:59 |
| facefaceface | yup, like page rank but more complex... I suppose you could fully decentralize the day everybody starts using the semantic web... | 22:59 |
| fenn | its complete with respect to itself | 22:59 |
| facefaceface | ;-) | 22:59 |
| kanzure_ | just like your opinion is complete with respect to itself? | 22:59 |
| kanzure_ | geeze, this is going nowhere | 22:59 |
| kanzure_ | uhm | 22:59 |
| kanzure_ | speaking of going nowhere, I need to go get some foods | 23:00 |
| facefaceface | ok, but you are wrong | 23:00 |
| facefaceface | ;-) | 23:00 |
| fenn | well, if i knew more about statistics i might be able to explain, but it's not worth the trouble | 23:00 |
| kanzure_ | go write your opinion on canonizer so that I'll never hear it | 23:00 |
| kanzure_ | 'good ideas' *compile* | 23:00 |
| kanzure_ | it's a different type of 'it works' | 23:00 |
| facefaceface | my opinion as a scientist is backed up by other scientists who rate my work on the basis of there judgement | 23:00 |
| kanzure_ | yikes | 23:01 |
| fenn | argument from authority codified | 23:01 |
| facefaceface | its true | 23:01 |
| fenn | trust circle wankery | 23:01 |
| kanzure_ | yay trust circles | 23:01 |
| facefaceface | so when I apply for a job, people say - is this guy any good? Well, has he published... | 23:01 |
| fenn | everyone hold hands and dance for the end of the world | 23:01 |
| kanzure_ | we were talking about that with PLOS at BioBarCamp | 23:01 |
| kanzure_ | everybody was booing and hissing at that stuff | 23:01 |
| facefaceface | I didn't say people were not idiots | 23:02 |
| fenn | what's wrong with PLoS? | 23:02 |
| kanzure_ | nothing | 23:02 |
| kanzure_ | but they were talking about things like 'impact factor' | 23:02 |
| fenn | why were they booing it? | 23:02 |
| kanzure_ | and hiring based off of 'impact factor' | 23:02 |
| fenn | ugh, that shit's the problem, not plos | 23:02 |
| kanzure_ | right | 23:02 |
| kanzure_ | "we were talking about that with PLOS" | 23:02 |
| kanzure_ | "hissing at that stuff" | 23:02 |
| fenn | scientists need to organize some beheadings | 23:02 |
| kanzure_ | nowhere did I say PLOS is doing it | 23:02 |
| facefaceface | how do you rate a scientist? | 23:03 |
| kanzure_ | why would you | 23:03 |
| fenn | i dont care how you rate a scientist | 23:03 |
| facefaceface | how do you decide who to employ? | 23:03 |
| kanzure_ | fuck that shit | 23:03 |
| kanzure_ | employment is a broken concept | 23:03 |
| facefaceface | which papers do you read? | 23:03 |
| kanzure_ | all of them | 23:03 |
| * kanzure_ has all of nature :) | 23:03 | |
| facefaceface | how many papers to you commit to memory? | 23:03 |
| kanzure_ | you commit papers to memory? | 23:03 |
| kanzure_ | honest question | 23:04 |
| facefaceface | how many facts about nature do you assume? | 23:04 |
| kanzure_ | yer doing it wrong | 23:04 |
| facefaceface | not the whole thing | 23:04 |
| fenn | scientist submits a proposal, send results to people in the field to see if it's worth keeping.. are opinions unanimous? if not, why not? | 23:04 |
| facefaceface | kanzure_, I'm doing science | 23:04 |
| kanzure_ | uh | 23:04 |
| fenn | s/results/proposal/ | 23:04 |
| fenn | if there's argument about it, there must be some kernel of truth | 23:05 |
| facefaceface | of course one should keep an open mind, but the basis of any research is information about what has come before. | 23:05 |
| facefaceface | that information is filtered by its 'usefulness' in some more or less objective sense | 23:05 |
| kanzure_ | people are stupid, Eliezer has been working on proving this for the past decade or something :p | 23:05 |
| kanzure_ | god I just cited eliezer | 23:05 |
| kanzure_ | I suck. | 23:05 |
| * kanzure_ eats and goes off to cry | 23:05 | |
| * facefaceface pats kanzure_ on the back | 23:05 | |
| fenn | usefulness isn't objective at all, it's entirely subjective | 23:05 |
| facefaceface | fenn, circles of trust? | 23:06 |
| fenn | what bout them? | 23:06 |
| facefaceface | if the fact confirms your observation, then its useful | 23:06 |
| facefaceface | facts build on facts | 23:06 |
| facefaceface | or perhaps I should say theories | 23:06 |
| fenn | md5sum of 'foo' is d3b07384d113edec49eaa6238ad5ff00 | 23:06 |
| fenn | is that useful? | 23:06 |
| fenn | it's a fact | 23:06 |
| facefaceface | which is why I say "I'm doing science" | 23:06 |
| facefaceface | its useful if it confirms a particular theory about what I believe to be the data in foo | 23:07 |
| fenn | wtf are you parroting | 23:07 |
| facefaceface | confirms or refutes | 23:07 |
| facefaceface | parroting? | 23:07 |
| fenn | say i have 9 metric tons of gold, is it useful? | 23:07 |
| facefaceface | its an object, I thought we were talking about philosophy of science | 23:08 |
| fenn | i'm just trying to show why utility depends on the subject | 23:08 |
| facefaceface | OK, in that sense I take your point... in the *final* analysis *everything* is relative. | 23:08 |
| facefaceface | but in the final analysis, everything is doomed - it doesn't stop others trying to live for ever | 23:09 |
| fenn | if humans had infinite computational ability, simple ("elegant") scientific theories would be less useful | 23:09 |
| facefaceface | it doesn't stop me trying to build a model of the world | 23:09 |
| facefaceface | sure | 23:09 |
| facefaceface | but we don't right? | 23:09 |
| facefaceface | so we approximate | 23:09 |
| fenn | (re "everything is doomed" have you heard of tipler's omega point?) | 23:09 |
| facefaceface | fenn, anyway... I forget why I even went down this road... | 23:09 |
| facefaceface | no | 23:09 |
| fenn | well its a good side track | 23:10 |
| facefaceface | the fact that perhaps we could rate science objectively, if we assess it in terms of some existing models - which is a bit like circle of trust. | 23:10 |
| facefaceface | which is how we tend to rate science... | 23:10 |
| facefaceface | but we call it objective even though we know it isn't really. | 23:11 |
| facefaceface | simply because we take our hypothesis-es for laws | 23:11 |
| facefaceface | hypothesi? | 23:11 |
| fenn | i just dislike the current system of parasite publishing companies having all the 'impact factor' | 23:11 |
| fenn | it stinks of system abuse | 23:12 |
| facefaceface | fenn, agreed (about companies) | 23:12 |
| facefaceface | what is parasite publishing? | 23:12 |
| fenn | elsevier, wiley | 23:12 |
| facefaceface | oh | 23:12 |
| facefaceface | yeah, the opinion-ome is an idea to get rid of them | 23:12 |
| facefaceface | the idea is to set it up and try some algorithms. | 23:12 |
| fenn | so you'd use opinion-ome to rate scientists? | 23:13 |
| fenn | this sorta reminds me of whuffie | 23:13 |
| facefaceface | its all very well discussing ideas for scientific reform, but experements like PLoS One and things like opinon-ome are necessary | 23:13 |
| facefaceface | fenn, yea, or specifically papers by scientists, and hence scientists by inference | 23:13 |
| facefaceface | but only within categories, and conditional on what others thought of what they thought. | 23:14 |
| fenn | one problem i see is the whole principle investigator structure | 23:14 |
| fenn | if you rate papers | 23:14 |
| facefaceface | so 'cliques' of belief could emerge (somehow), representing alternative theories in a field, for example. | 23:14 |
| fenn | graduate student does all the hard work and PI gets credit because his name is on all the papers | 23:14 |
| facefaceface | what is that? | 23:14 |
| fenn | so then PI gets ranked as "awesome demi-god scientist" | 23:15 |
| fenn | and student is "piddling pissant" | 23:15 |
| fenn | just through sheer quantity | 23:15 |
| facefaceface | but the students name should be on there right? | 23:15 |
| fenn | yes but they only have one paper | 23:15 |
| facefaceface | but its a good one right? | 23:15 |
| fenn | yep | 23:15 |
| facefaceface | so they are well placed to get a good next position | 23:16 |
| fenn | somehow i messed up this explanation | 23:16 |
| facefaceface | and continue good work and continue building their ... how the hell do you spell corea? | 23:16 |
| facefaceface | career? | 23:16 |
| facefaceface | ok | 23:16 |
| fenn | heheh you must be british | 23:16 |
| facefaceface | what woudl you say? | 23:17 |
| facefaceface | spell? | 23:17 |
| fenn | career | 23:17 |
| fenn | words.. | 23:17 |
| kanzure_ | increasingly coherent over increasing context | 23:18 |
| facefaceface | oh well... time to work on my 'reply to reviewers' | 23:18 |
| fenn | multiple perspectives are useful though | 23:18 |
| fenn | top-down gets boring | 23:18 |
| fenn | er, s/top-down/birds eye view/ | 23:18 |
| fenn | facefaceface: just remember to wear gloves when handling the explosives | 23:19 |
| kanzure_ | extropy covers many of these topics of growth and managing philosophy of science. | 23:20 |
| kanzure_ | by 'extropy' I refer to the mailing list | 23:20 |
| fenn | tell them to work on something coherent then | 23:20 |
| fenn | extropian manifesto | 23:20 |
| kanzure_ | uh | 23:21 |
| fenn | nevermind | 23:21 |
| fenn | context helps resolve ambiguities due to vagueness | 23:22 |
| kanzure_ | kinda like, say, http://maxmore.com/extprn3.htm | 23:22 |
| facefaceface | kanzure_, canonizer isn't quite what I had in mind, but I'll have a look. | 23:22 |
| fenn | 'extropian principles' document is rather vague and context-less | 23:22 |
| kanzure_ | that's true. | 23:22 |
| kanzure_ | there's been something recent in the mailing list that could be useful here | 23:22 |
| kanzure_ | just a sec | 23:22 |
| fenn | US legal system resolves vagueness with "precedent" which rather sucks | 23:23 |
| kanzure_ | http://lists.extropy.org/pipermail/extropy-chat/2008-October/046112.html | 23:23 |
| fenn | it's turtles all the way down | 23:27 |
| kanzure_ | that's in the next message (click "Next message" at the top/bottom) | 23:27 |
| kanzure_ | hah, in damien's link | 23:29 |
| kanzure_ | "there might be a final theory, a theory of the superturtle, if you will .." | 23:29 |
| -!- kanzure_ is now known as superturtle | 23:29 | |
| superturtle | hehe | 23:29 |
| fenn | whatever, it's just the same thing all over again | 23:29 |
| fenn | plato's allegory of the cave | 23:29 |
| * fenn points to the "no philosphy" sign in the topic | 23:29 | |
| fenn | this is why i'm a chauvinist | 23:30 |
| fenn | hm. impossible to find a good definition of chauvinism | 23:32 |
| fenn | its the idea that "of course i'm right, otherwise i wouldnt believe what i believe" | 23:32 |
| superturtle | chauvinist == "together, we hate everything" ? | 23:32 |
| superturtle | ah. | 23:32 |
| fenn | and also explicitly "and so should you" | 23:33 |
| fenn | that superturtle page.. i like the zooming out picture sequence | 23:35 |
| fenn | "the typical size of a social network is constrained to about 150 members due to possible limits in the capacity of the human communication channel." | 23:37 |
| superturtle | bwahah | 23:37 |
| superturtle | oh, I guess I shouldn't be laughing | 23:37 |
| superturtle | I'm *pretty sure* I keep track of more people than that | 23:38 |
| fenn | certainly | 23:38 |
| superturtle | I don't have an official count though | 23:38 |
| fenn | i probably only keep track of 50 people | 23:38 |
| fenn | it's a bell curve thing really | 23:38 |
| superturtle | well, I be stupid and I sacrifice depth-of-keeping-up for breadth. | 23:38 |
| fenn | anyway it's a regularity that we should be able to spot by studying social networks | 23:39 |
| fenn | not sure if it really matters in the long run | 23:40 |
| superturtle | what if there's a way to know when to abandon ship | 23:42 |
| fenn | when to abandon what? | 23:42 |
| superturtle | when the trust networks turn to pure wankery | 23:42 |
| superturtle | a way to abandon the system and reboot | 23:42 |
| * fenn mumbles something about mutation and objectivity | 23:43 | |
| superturtle | you'll have to mumble a bit more | 23:43 |
| fenn | the problem is a bit harder with science because ostensibly science is the way out | 23:43 |
| fenn | "just look dammit" "we are looking!" | 23:44 |
| superturtle | wait, science? | 23:44 |
| superturtle | I thought we were talking about projects, design and stuff | 23:44 |
| superturtle | which science does fall into I guess | 23:44 |
| fenn | you were talking about trust circle wankery | 23:44 |
| superturtle | right | 23:44 |
| fenn | which i associate with tenured professors reviewing my grant proposal | 23:44 |
| superturtle | you were talking about your fusor 9000 example | 23:44 |
| superturtle | a way to tell when the system is being spammed/flooded or just going to hell | 23:45 |
| fenn | fusor 9000 is just taking advantage of a lack of coherence in the trust network | 23:45 |
| fenn | so its sort of the opposite case | 23:45 |
| fenn | if i said "look it runs java" then i'd be taking advantage of trust circle wankery | 23:46 |
| fenn | but if nobody has any idea what it is, and it gets funded, that's bad system design? | 23:46 |
| * fenn looks up what gave rise to the fusor 9000 comment | 23:46 | |
| fenn | right. you were suggesting that the person submitting the proposal also give an estimate as to its feasibility | 23:47 |
| superturtle | nooo | 23:47 |
| superturtle | not the person | 23:47 |
| fenn | then who? | 23:47 |
| superturtle | teh codes | 23:48 |
| fenn | but.. code is stupid | 23:48 |
| fenn | it doesn't know anything | 23:48 |
| fenn | only what it has been told | 23:48 |
| fenn | complete lack of inference and "objectivity" | 23:48 |
| fenn | oh i dunno maybe you can hack together some neural network to recognize "feasible" characteristics, but that could easily be hacked by system-players | 23:49 |
| superturtle | separation of experimental versus "untainted projects" | 23:49 |
| fenn | (humans can also be gamed/hacked) | 23:49 |
| superturtle | "untainted" meaning "sure thing" | 23:49 |
| fenn | no it does not | 23:49 |
| fenn | untainted means we know what is in it | 23:50 |
| fenn | you have to run the experiment to find the results | 23:50 |
| superturtle | I'm also referring to the designs. | 23:50 |
| superturtle | sure | 23:50 |
| fenn | you cant run the simulator to find the results | 23:50 |
| superturtle | I said experimental versus untained | 23:50 |
| superturtle | not "experimental is untainted" | 23:50 |
| fenn | different dimension | 23:50 |
| fenn | its experimental vs not-experimental | 23:50 |
| fenn | whatever that means | 23:50 |
| fenn | untested vs tested | 23:51 |
| superturtle | it still has the same implications though. like "we know it will work" in the experimental/nonexperimental distinction, compare to "well, it might work if you got xyz material" | 23:51 |
| fenn | tested-works vs tested-fail | 23:51 |
| fenn | i dont think experimental/nonexperimental is very clear | 23:52 |
| fenn | maybe flag something as "worked in the simulator" | 23:53 |
| superturtle | you're not about to argue with me that you can't verify a lego design | 23:53 |
| superturtle | I mean, if it's possible to build it with legos, it's going to be built ;-) | 23:53 |
| superturtle | within constraints of the materials of course | 23:53 |
| superturtle | which are in the system supposedly. | 23:53 |
| fenn | hmm. you havent seen much GA output eh? | 23:53 |
| fenn | they'll hack your lego simulator to bloody bits | 23:53 |
| fenn | "i didnt know legos could do that!" "they cant" | 23:54 |
| superturtle | if you allow the operators in the GA to be only possible moves then how would you get impossible things? | 23:54 |
| fenn | shit happens | 23:54 |
| fenn | the difference between theory and practice is much greater in practice | 23:54 |
| fenn | sorry to get all dogmatic on you | 23:55 |
| superturtle | okay. | 23:56 |
| fenn | certainly i would like to see some GA putting legos together | 23:56 |
| superturtle | I'm okay with never being sure that the design is going to work, or that my code might not compile | 23:56 |
| fenn | good | 23:56 |
| superturtle | we just need a system where everything isn't based on the hope that it will work. | 23:56 |
| fenn | just dont try to lump everything together into the same variable | 23:57 |
| fenn | even with supposedly "tested" code, or "replicable experiments" things can go wrong | 23:59 |
| superturtle | sure | 23:59 |
| fenn | there ought to be a science of testing | 23:59 |
| superturtle | and of debugging | 23:59 |
| fenn | same thing really | 23:59 |
Generated by irclog2html.py 2.15.0.dev0 by Marius Gedminas - find it at mg.pov.lt!
