--- Day changed Sun Aug 31 2008
00:07 < kanzure> argh
00:07 < kanzure> somebody pm'd me asking me "okay, I actually have resources and tools ... what the hell are people really supposed to *do* with this damned kit?"
00:08 < pks> which kit?
00:09 < kanzure> http://biohack.sf.net/
00:09 < kanzure> biotech toolkit thingy
00:10 < pks> iirc there's a tonnnn of stuff in there
00:10 < kanzure> yep
00:11 < kanzure> he wants to know how to be enhanced
00:11 < pks> lol...
00:11 < pks> well
00:11 < pks> who doesn't
00:11 < kanzure> the problem is that there's no quantification of enhancement
00:12 < kanzure> and my general disbelief in the existence of intelligence
00:14 < kanzure> hm
00:14 < kanzure> maybe I could consider "intelligence" in the context of "ability to design better thingies" and then take a limiting context, such as skdb and the parts available to the local thinker, and then say that intelligence can be quantified within these constraints
00:14 < kanzure> but freeform intelligence ... eh
00:29 < kanzure> hey, that sounded moderately intelligent
00:29 < kanzure> cool
00:29 < kanzure> (it passed the "come back 10 minutes later" test)
00:56 < ybit> hehe
00:57 < kanzure> "you constructing ways and technologies which are far out of our horizon"
01:18 -!- marainein [n=marainei@220.253-204-179.VIC.netspace.net.au] has quit [Read error: 113 (No route to host)]
02:48 -!- nsh [n=nsh@87-94-146-186.tampere.customers.dnainternet.fi] has joined #hplusroadmap
02:51 -!- nsh- [n=nsh@87-94-146-186.tampere.customers.dnainternet.fi] has quit [Read error: 110 (Connection timed out)]
03:17 -!- pks [n=pk@c-98-217-103-22.hsd1.ma.comcast.net] has quit []
04:40 -!- Phreedom [n=freedom@ip-194-50-167-184.mir.dn.ua] has quit [Read error: 110 (Connection timed out)]
05:11 -!- Phreedom [n=freedom@ip-194-50-167-184.mir.dn.ua] has joined #hplusroadmap
05:28 -!- marainein [n=marainei@220.253-204-179.VIC.netspace.net.au] has joined #hplusroadmap
06:37 -!- Nofaris [n=Nofaris@adsl-75-42-94-197.dsl.scrm01.sbcglobal.net] has joined #hplusroadmap
08:06 -!- jk4930 [n=jk@p57B73DFB.dip.t-dialin.net] has joined #hplusroadmap
08:30 -!- nsh- [n=nsh@87-94-146-186.tampere.customers.dnainternet.fi] has joined #hplusroadmap
08:32 -!- nsh [n=nsh@87-94-146-186.tampere.customers.dnainternet.fi] has quit [Read error: 110 (Connection timed out)]
10:09 -!- jk4930 [n=jk@p57B73DFB.dip.t-dialin.net] has left #hplusroadmap ["Konversation terminated!"]
10:24 -!- ChrisMT [n=lixasd@cpc2-clif5-0-0-cust516.nott.cable.ntl.com] has quit [Read error: 110 (Connection timed out)]
11:18 -!- Nofaris [n=Nofaris@adsl-75-42-94-197.dsl.scrm01.sbcglobal.net] has quit [Read error: 110 (Connection timed out)]
11:21 -!- nsh- is now known as nsh
11:35 < willPow3r> skdb == serial killer database?
11:35 < nsh> sexual kinetics database
11:36 < willPow3r> fucking krispy kreme hoes never get my order right
11:38 < willPow3r> which is further evidence to support my hypothesis that humanity is doomed
11:52 < kanzure> huh
11:52 < kanzure> Bill Nye is back
11:57 < kanzure> 'Nye was also the assistant of Emmett "Doc" Brown in the live-action segments of Back to the Future: The Animated Series (1991–1993). In one episode, Bill corrected Doc Brown on the pronunciation of gigawatt (not jigawatt), only for Doc Brown to fire back with "Who are you, Bill Nye the science guy?". '
11:57  * kanzure totally missed that
11:57 < kanzure> also, that disc is in the cd tray at the moment
13:16 < kanzure> hey, has anyone ever seen a bash and pipelining implementation where programs have to specify the format of the parameters?
13:28 < kanzure> wait, wouldn't a python shell do that?
14:09 -!- Depucelator [n=Depucela@c-24-118-116-62.hsd1.mn.comcast.net] has joined #hplusroadmap
14:32 < kanzure> Hey Depucelator
14:35 < Depucelator> Hey Bryan
14:35 < kanzure> http://shell.appspot.com/ interactive server-side shell for google apps?
14:37 < Depucelator> I've never used the google app engine; it's only for use with python?
14:38 < Depucelator> ugh, I hate python
14:40 < kanzure> why?
14:40 < Depucelator> whitespace sensitivity mostly
14:40 < kanzure> they say you'll learn to love that
14:40  * kanzure hasn't... yet.
14:41 < Depucelator> I was trying to run this python app on linux, and was getting some cryptic error that really made no sense whatsoever
14:41 < Depucelator> and after hours of trying to get it to work, I accidentally converted all the windows linebreaks to unix linebreaks
14:42 < Depucelator> and it ran perfectly
14:43 < Depucelator> That, to me, is insane in the year 2008
14:47 < kanzure> So I was talking with a guy last night who was asking me what to actually do with the biotech toolkit, what "enhancements" to the human body could be made. He claimed to have money and resources, tools and materials. I'm sure he was lying right out of his ass, but that's not the point ..
14:47 < kanzure> It seems hard to convince people that you're not going to be able to guarantee "enhancements", just as you can't guarantee a weight loss program will work.
14:47 < kanzure> well, there's some metabolism that I'm sure you could guarantee if you investigate their biochemistry enough I suppose
14:47 < kanzure> but that's not the point
14:48 < kanzure> The genotype of alterations doesn't necessarily correlate with a phenotype at all ... so that's why you have to focus on the infrastructure to keep playing around with the genotypes, no matter the phenotype. But it's hard to convert this directly into action since you can't say "well, our goal is to get tools and .. do stuff."
14:48 < kanzure> I'm not sure how to show these types of people this ..
14:49 < kanzure> fenn, not even an operating system has a full 'inventory management system' other than people just remembering where the important directories are .. have there been alternative implementations ? I'm trying to think up some analogies, but none are coming to mind.
15:12 < kanzure> http://en.wikipedia.org/wiki/Electronic_Data_Interchange
15:13 < Depucelator> there was a story about EDIFACT on The Daily WTF just the other day
15:13 < Depucelator> it was pretty funny
15:16 < kanzure> http://thedailywtf.com/Articles/Unix-With-QuikBill!.aspx
15:16 < Depucelator> yeah that's the one
15:18 < kanzure> "“As I’m sure you’re aware,” the presenter spoke as he advanced the slide in the presentation, “our primary focus at Enterprise Business Systems is to enable our clients to formulate key objectives through strategic initiatives to develop a comprehensive strategy that will provide the critical foundation for creating a proactive, synergy-driven directive for utilizing technological approaches to achieve cross-departmental…”"
15:18 < kanzure> that's classic :)
15:19 < kanzure> "“Oh,” the trainer responded, “actually, all that API tech stuff is in the back of the binder.” That answer was quite possibly the only worthwhile bit of information Simon had gotten from the entire day. But it was worth it: at least now he knew where to start."
15:19 < kanzure> awesome
15:21 < kanzure> 'On a tour of the offices, Simon got a chance to see how the accounts payable department worked. When a QuikBill client sends a bill over the modem, the server translates the EDIFACT message to a fixed-column formatted file and saved to disk. A printer then picks up the formatted file and prints the invoice on a form. From there, an accounts payable clerk takes the form off the printer and types it in to their voucher system to be paid.'
15:22 < kanzure> How sad.
15:23 < Depucelator> Wooden table phenomenon
15:23 < kanzure> Hm?
15:23 < Depucelator> the site is the source of endless new terms
15:24 < Depucelator> "the wooden table phenomenon" originates from an article about a school website whose pages are digital camera pictures of a printout on a wooden table of the web site pages
15:24 < Depucelator> which is sadly not far from some of the shit they pull at my workplace
15:25 < kanzure> but how else are you going to make a secure website except by taking a photograph of it?
15:25 < kanzure> I mean, if it works for the mafia holding hostages
15:25 < kanzure> then why not here?
15:25 < Depucelator> one time I received the "source codes" for a program that was printed out, photocopied, scanned in and then each page converted to a PDF file and emailed to me
15:25 < kanzure> wtf?
15:26 < Depucelator> ha ha ha for dynamic webpages they should automatically print out the web page and take a digital camera snapshot of it and upload it to the user
15:26 < kanzure> :)
15:27 < Depucelator> the best ever was this guy who backed up his website by saving them individually while browsing through the inbuilt FTP client in internet explorer
15:27 < Depucelator> except when IE saves webpages, it saves the URL they were saved from as a comment in the page
15:27 < Depucelator> and the urls were all ftp://username:password@domain.com/website.htm
15:30 -!- elias`_ [n=me@cs78208074.pp.htv.fi] has joined #hplusroadmap
15:30 -!- elias` [n=me@unaffiliated/elias/x-342423] has quit [Nick collision from services.]
15:30 -!- elias`_ is now known as elias`
15:31 < Depucelator> en puhu suomea
15:31 < Depucelator> anteeksi
15:54 < nsh> sen on tyhma kieli
15:54 < nsh> *se
15:54 < Depucelator> oh okay, well that clears things up
15:55  * nsh chuckles at above ie ftp story
16:07  * kanzure cries at above ie ftp story
16:15 -!- Nofaris [n=Nofaris@adsl-75-42-94-197.dsl.scrm01.sbcglobal.net] has joined #hplusroadmap
16:30 < nsh> ETOOMUCHFAITHINHUMANITY
16:31 < nsh> remember: a cynic is harder to disappoint
16:33 < kanzure> ETOO?
16:34 < kanzure> also, what if we apply the "trade through many traders" ideas about ebay (if ebay was really doing what it's supposed to be doing) to procurement and material acquisiton ? that way, a designer sitting behind the graphs and design studio can see relative likelihood of people pitching in the materials and such that are required to make something Really Happen
16:35 < kanzure> Eventually you'd think I'd come to think that maybe it's /me/ who is wrong after so many other people 'doin it wrong' but evidently this hasn't happened yet.
16:36 < willPow3r> change to deeply ingrained social process does not come quickly
16:36 < kanzure> nope, this doesn't really need that
16:37 < kanzure> I think it would just as easily apply on a scale of 10 people as >10
16:37 < kanzure> *as it would to >10
16:37 < willPow3r> then start a $multi-million company
16:37 < kanzure> What?
16:37 < kanzure> what are you on about?
16:38 < willPow3r> do an implementation of ebay that caters to the needs of corporations
16:38 < kanzure> willPow3r: it's ridiculously easy to code up ebay
16:39 < kanzure> the implementation isn't the hard part
16:39 < willPow3r> then what is? it takes the guesswork out of bidding for materials
16:40 < kanzure> the hardwork is the social network and convincing people that it's useful
16:40 < kanzure> they don't just magically download and install :-)
16:40 < willPow3r> thats why you would need somebody who's good w/ the business end
16:41 < kanzure> besides, I'm sure there's already an (analogous) "ebay-to-AutoCAD" plugin
16:41 < kanzure> eh?
16:41 < kanzure> this /is/ the business though
16:41 < kanzure> sigh
16:42 < willPow3r> nm
16:42 < willPow3r> i'll just steal your idea then
16:42 < kanzure> maybe I'm missing something, but a business guy can't do anything that I can't
16:42 < kanzure> it's not like he's magical
16:43 < kanzure> so however he acquires a network should be the same for me
16:43 < kanzure> no?
16:43 < willPow3r> to throw a sales pitch, all you need is an abstract of the technology and a "why this is good for you"
16:43 < kanzure> people seriously go for this sort of thing ?
16:43 < willPow3r> if it would save a corporation lots of money, manpower, and time
16:43 < willPow3r> they would
16:44 < willPow3r> you would charge them for the ability to use it
16:44 < willPow3r> a nominal fee
16:44 < kanzure> the thing is that this wouldn't save them money at first because nobody is on the 'system' of dumping files on their server stating what they have for search engines to find
16:44 < kanzure> or for them to have their daemons for strategic "deal bidding" (of course, not automatic -- humans have to review the deals or something)
16:44 < willPow3r> that's the tricky part, i suppose
16:44 < kanzure> yeah, but they're charging for "nothing"
16:45 < willPow3r> it would probably take some effort to begin with, but the ends are worth it, imo
16:45 < kanzure> btw, I hear Chinese factories (for instance) are highly competitive to the point of doing thorough background checks on all of their customers just to make sure they're not their competitor factories in disguise trying to figure out pricing
16:45 < kanzure> yeah, the ends are definitely worthwhile of course
16:46 < kanzure> so having an automatic system for "making deals" from the designer's point of view (feasability checking and such) would scare them shitless
16:46 < willPow3r> you could take that guesswork out of the problem for the procurement side
16:46 < willPow3r> analogous to ebay, it would be like buyer's feedback ratings
16:46 < willPow3r> company profiles, etc
16:47 < kanzure> on ebay you still see the pricing no matter what
16:47 < kanzure> that's what they are worried about (IIRC)
16:47 < willPow3r> and it would make a more global system of capitalism
16:47 < willPow3r> ie, free market economy
16:47 < kanzure> I don't know how unstable their businesses must be,
16:48 < kanzure> but if a few cents of difference between prices between competitors makes such a big difference 
16:48 < kanzure> or they're worried that much
16:48 < kanzure> maybe they have other problems to deal with ?
16:48 < kanzure> heh'
16:49 < willPow3r> on a large scale, a few cents difference when purchasing, say, thousands of tons of materials, makes a notable difference
16:49 < kanzure> http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6V20-4B3JVBN-1&_user=108429&_rdoc=1&_fmt=&_orig=search&_sort=d&view=c&_acct=C000059713&_version=1&_urlVersion=0&_userid=108429&md5=821c16c4f14ccf5ebaf5c7dbb304df38 
16:49 < kanzure> 'Enabling information sharing between E-commerce systems for construction material procurement'
16:49 < kanzure> right
16:50 < kanzure> so they have strong incentive not to do this
16:50 < kanzure> but from the procurement side, it's better if you can see prices more quickly, no?
16:50 < willPow3r> its is
16:51 < willPow3r> thats the point
16:51 < kanzure> imagine designing a new thingy, say a car, and adding a new exhaust pipe and immediately seeing the rework on the factory floor that would be needed, as well as new material costs, etc.
16:51 < willPow3r> and it allows the suppliers to be more competitive
16:51 < kanzure> there might be enough pressure if everyone does it
16:51 < kanzure> but
16:51 < kanzure> since there's finite resources, these material suppliers and so on can easily be a monopoly
16:51 < kanzure> no?
16:52 < willPow3r> with the current system, yea
16:52 < kanzure> so they could just ignore the market forces.
16:52 < kanzure> wtv
16:52 < kanzure> wtf
16:52 < kanzure> I just said 'market forces'
16:54 < willPow3r> with a visible auction-style format, there would be no real possibility of monopoly
16:54 < willPow3r> it would even allow small companies to participate
16:54 < willPow3r> and would definitely spur development of more efficient production processes
16:55 < willPow3r> more $$ would be dumped into research to remain competitive
16:55 < willPow3r> among sellers
16:56 < willPow3r> but the beauty of the system is that goods would be bought by a company, processed to make new goods, then resold on the same system
16:56 < willPow3r> but then again, this idea does reek of utopian miasma
16:56 < kanzure> really it's just XML files on a web server or something
16:57 < kanzure> and RSS informing designers of new tools or materials to incorporate into the "CAD alternative" (I'm thinking the graph grammar methodology is a way to do it without "draw a line" to represent edges of the final product, etc.)
16:57 < kanzure> possibly also pingback servers to tell them of potential deals that are forming
16:57 -!- Nade [n=lixasd@cpc2-clif5-0-0-cust516.nott.cable.ntl.com] has joined #hplusroadmap
16:58 < willPow3r> right, the whole system could be modular
16:58 < kanzure> what I like about it is that broke people :-) such as myself :-) wouldn't have to worry about the actual resources since it's a "press here to make it happen" system where others can donate money or resources to the cause
16:58 < kanzure> so it's a finite/staggard system of 'doing stuff' heh
16:59  * kanzure isn't actually all that broke
16:59  * kanzure assumes he is, though
16:59 < willPow3r> it looks good on paper
16:59 < kanzure> what does?
16:59 < willPow3r> global auction-style e-commerce
17:00 < kanzure> it also looks good on the screen .. software's been doing this for ages, except without the commerce aspects really
17:00 < kanzure> I don't know if you would call it that, the auctions aren't all that centralized of course
17:03 < willPow3r> i'm thinking of an ebay analog that connects buyers with sellers
17:03 < kanzure> Phreedom: We were talking the other day about a designer behind this hypothetical system making, say, the designs for the engines of some particular car. This would be the general 'makefile' including the CFD calculations for all sorts of minor adjustments and simulation of the engine. 
17:03 < kanzure> Then, designers could do all of the pretty graphical stuff and "draw a car". So for the "engine engineering" aspect, could you imagine that being done from a "sea of processes" (*not* a sea of "parts") from which the engineer selects from? Does this make sense to you?
17:03 < willPow3r> it would definitely bring about more connections
17:04 < kanzure> no, there's already that, sort of
17:04 < kanzure> I'm not interested in doing another ebay
17:04 < kanzure> ebay is centralized
17:04 < kanzure> the alternative would be more like bit torrent ;-)
17:04 < kanzure> this way, businesses get to keep their servers to themselves
17:04 < kanzure> their information is completely under their control
17:04 < kanzure> could just be a box sitting in a corner of course
17:05 < kanzure> or hosted somewhere else out on the net
17:05 < kanzure> think of the RSS aggregator sites and what they do .. that's been done with some priceshopper-dot-coms before, but it's not as "open" as it could be
17:06 < willPow3r> oh, i see
17:06 < willPow3r> like google shopping
17:06 < kanzure> uhm, well, I'm pretty sure they are automated
17:06 < kanzure> I'd hate to learn that they're being done by hand
17:06  * kanzure goes back to check the daily wtf
17:07 < willPow3r> google product search or amazon even
17:07 < kanzure> that's posisble
17:07 < kanzure> but think more of the torrent search engines I guess
17:08 < kanzure> and trackers.
17:08  * willPow3r wonders if google product search works for raw materials
17:09 < willPow3r> right, a custom system
17:10 < kanzure> semantic search for raw materials is the wrong way to do it
17:10 < kanzure> instead, the information should be packaged into the units that can be installed into the developer's IDE thingy
17:10 < kanzure> or CAD thingy
17:10 < kanzure> so if you see a new company on the net that offers a zip file of their materials or something, go download it, and then you can incorporate that into your designs and you can see some new possible connections between different processes
17:11 < kanzure> like
17:11 < kanzure> "hey, I finally have that silver I needed! hurrah"
17:11 < willPow3r> right
17:11 < willPow3r> but the point is that the sellers of goods will see what the prices of other sellers is
17:12 < willPow3r> and will be able to drop prices/innovate/etc to remain competitive
17:13 < willPow3r> so what i'm proposing is something that will allow corporations to have multiple possible places of procurement, saving them money
17:13 < willPow3r> and, by owning this system, i would make $$$
17:13 < willPow3r> because that's really the whole point of starting a business
17:14 < willPow3r> you could add on the CAD plugins as a selling point
17:16 < kanzure> oh, what I was thinking of was the "bedroom programmer model" where you wait until enough people donate and then you release your product thingy
17:16 < kanzure> so here it would be that except more automated and the suppliers could feel more confident they aren't getting a bad deal off of their materials
17:16 < willPow3r> you mean investors?
17:17 < kanzure> no, this was specifically mentioned on Slashdot and other sites re: doing some sort of reputation stunts by doing stuff for free, building users and having your fanbase pay your way for your Next Great Thing
17:18 < kanzure> oh
17:18 < kanzure> Phreedom: sorry, that was a stupid question
17:18 < kanzure> any "process" or "ecology of processes" can be modeled by graphs
17:18 < kanzure> see the entire field of ecology or thermodynamic work flows etc.
17:18  * kanzure grabs a bite to eat.
17:23 < kanzure> guess that doesn't mean it's easy to design such things though
17:24 < kanzure> for example, let's take http://www.processengr.com/images-css/Simulation_Project_Diagrams/refinery_flare_capacity.gif
17:24 < kanzure> or other things listed on the 'process simulation' page - http://www.processengr.com/simprojects.html#sim_proj_fumeincenrecov
17:25 < kanzure> it's just a giant system of equations, so I suppose if you start with two "big equations" you can always connect them with intermediate steps that the computer should be able to solve
17:25 < kanzure> after all, this is what Mathematica is supposedly good at, no ?
17:25 < kanzure> don't know if mathematica works with units though
17:25 < kanzure> but the point is still there.
17:38 < Phreedom> kanzure_: eh?
17:38 < kanzure> blah?
17:38 < kanzure> you don't remember our conversation?
17:39 < Phreedom> oh
17:39 < Phreedom> sorry didn't see the question
17:39 < Phreedom> reading now :)
17:39 < kanzure> reading's good
17:40 < Phreedom> kanzure_: ok. I really should write some minimal explanation of what I'm up to
17:40 < Phreedom> I'll do it shortly
17:40 < kanzure> is it similar to this?
17:40 < Phreedom> similar
17:41 < kanzure> Don't leave me hanging ..
17:44 < Phreedom> really it's better explained in a coherent short doc than a page of ramblings
17:45 < kanzure> bah
18:59 < willPow3r> http://www.businessweek.com/ebiz/9909/ec0907.htm
19:00 < willPow3r> "Gray predicts that in five years, 40 percent to 50 percent of all textbooks will be electronic." this was predicted in 1999
19:03 < kanzure> hahah
19:03 < kanzure> well
19:03 < kanzure> of all new textbooks ?
19:03 < willPow3r> yea
19:03 < willPow3r> since new editions come out every other year or so
19:04 < kanzure> well, there's online "supplements" that sometimes includes PDF chaptesr
19:04 < kanzure> *chapters
19:04 < willPow3r> ebooks might not have been such a good idea in 1999, but today the idea would be a lot more acceptable
19:04 < willPow3r> especially since laptops now outsell desktops
19:05 < willPow3r> i have all my books in electronic format
19:05 < willPow3r> downloaded illegally, of course
19:05 < willPow3r> but still, i only carry around a 3 lb laptop and a wacom
19:05 < willPow3r> instead of 20+ lbs of books and binders
19:07 < willPow3r> i happen to have a "wizeup" version of my physics textbook
19:07 < willPow3r> pretty neat, but i think its a few versions old
19:11 < kanzure> PDFs suck
19:11 < willPow3r> you can bookmark them
19:11 < kanzure> even though I have 80 GB of them ..
19:12 < willPow3r> is there an OSS alternative to pdf?
19:12 < willPow3r> except for it being proprietary, i like it
19:12 < kanzure> plain text
19:12 < kanzure> ps, maybe
19:12 < willPow3r> perhaps
19:13 < kanzure> the age of the ascii diagram will come once again!
19:13  * kanzure suicides into pearl harbor
19:13 < ybit> pdf is an open standard, what's wrong with that?
19:13 < willPow3r> lol
19:14 < willPow3r> bbs
19:14 < kanzure> pdf renders terribly
19:14 < kanzure> ybit: ever try 50 pdfs open at once?
19:14 < ybit> 50 files of any format would crash this comp :P
19:14 < kanzure> plain text?
19:14 < ybit> okay, maybe not ascii
19:15 < willPow3r> i'd love to see somebody publish scientific papers in plain text
19:15 < ybit> plain-text, whatever
19:15 < ybit> ascii, utf-9, iso-9880 i believe
19:15 < ybit> think that's about all the plain-text formats
19:15 < ybit> commonly used*
19:16 < ybit> utf-8*
19:16 < kanzure> you fail
19:16 < ybit> heh
19:18 < ybit> wow, i sure did
19:19 < ybit> close, but um, ouch
19:19  * kanzure reads up on graph grammars some more http://www.biocheminfo.org/klotho/grammar.html
19:21 < ybit> just for the record, was thinking of iso/iec8859 character encoding
19:23 < kanzure> 'Formalizing refactorings with graph transformations ' hm
19:42 -!- ybit [n=h@unaffiliated/ybit] has quit [Read error: 104 (Connection reset by peer)]
19:43 -!- ybit [n=h@unaffiliated/ybit] has joined #hplusroadmap
19:52 -!- ybit [n=h@unaffiliated/ybit] has quit [Read error: 104 (Connection reset by peer)]
19:52 -!- h2i [n=h@unaffiliated/ybit] has joined #hplusroadmap
19:59 -!- Nofaris [n=Nofaris@adsl-75-42-94-197.dsl.scrm01.sbcglobal.net] has quit [Read error: 110 (Connection timed out)]
20:09 -!- smurf [n=p@h52n3c1o261.bredband.skanova.com] has joined #hplusroadmap
20:09 -!- smurf is now known as biopunk
20:21 < fenn> pdf is easy enough to convert to plain text
20:22 < fenn> i'd love to find something that converts to html and preserves pictures and layout
20:22 < willPow3r> http://sourceforge.net/projects/pdftohtml/
20:22 < fenn> google's "show as html" kinda sucks
20:23 < fenn> doesn't really work, unfortunately
20:34 < willPow3r> acrobat professional doesn't do a bad job
20:34 < willPow3r> saving a pdf as an html 4.01 w/ CSS
20:36 -!- h2i is now known as ybit
21:37 < kanzure> I find it funny that when I go to forward an event to Charlie (newgenome), 
21:37 < kanzure> my little assistant thinger pops up "Charlie Stross"
21:37 < kanzure> heh'
21:40 < kanzure> http://holtz.org/Library/Social%20Science/Futurology/Stages%20in%20Future%20Evolution%20by%20Baxter.htm
21:48 < kanzure> this generally pisses me off: http://holtz.org/Library/Social%20Science/Futurology/The%20Politics%20of%20Transhumanism.htm
21:48 < kanzure> so nevermind.
21:49 -!- Nofaris [n=Nofaris@75.42.94.197] has joined #hplusroadmap
21:54 < kanzure> I still don't know who this atomjack/fusionanomaly character is ..
22:04 < biopunk> There is always some humor in people predicting the future in detail.
22:04 < kanzure> No, Hughes has always been a pain in the ass.
22:05 < kanzure> He's overly criticial of the extropian crowd because he thinks they just want to blow the government up .. I don't think Hughes is a programmer, I doubt he understands.
22:06 < biopunk> It's not my fight
22:07 < kanzure> It's not a fight at all -- just annoying :-)
22:09 < biopunk> they´re all a bunch of wankers
22:11 < kanzure> who?
22:14 -!- willPow3r_ [n=will@cpe-66-27-118-88.san.res.rr.com] has joined #hplusroadmap
22:14 < biopunk> the transhuman crowd kurzweil, bostrom, the imminst crowd... what's the other swedish neuroscience guy... they're a bunch of mental masturbators
22:15 < fenn> hmm fusionanomaly sounds familiar
22:15 < willPow3r_> philosophers == mental masturbators
22:16 < biopunk> not karl popper surely
22:16 < kanzure> fenn: http://fusionanomaly.net/
22:16 < kanzure> "you ar ewhat you cache" and such
22:16 < biopunk> those guys are not philosophers really either.. they're just nerds
22:17 < willPow3r_> they want their predictions of the future to become reality so bad
22:17 < kanzure> uhm?
22:17 < willPow3r_> that they are blinded by science and scientists that say otherwise
22:17 < kanzure> let's not broadly generalize here, there's a good number of them that are actually building shit :-)
22:17 < kanzure> what's wrong with that ?
22:18  * kanzure just lost interest in the 'graph grammars' crowd .. it's just another case of OOP
22:18 < fenn> don't discount something just because it's OOP
22:18 < kanzure> "nodes are objects and vertices are highly convoluted relations between them"
22:18 < kanzure> did you catch the discussion with Phreedom yesterday, fenn?
22:18  * biopunk fails to see any bigger difference between Kurzweil and Deepak Chopra
22:18 < fenn> sorta
22:19 < kanzure> basically the idea is that the core system that you design isn't an object, you throw some equations together to model the damned thing and then others can go and say "ooh, here's a fancy hubcap" or something
22:19 < kanzure> so the equations and such don't really jump out at me as being object-oriented really
22:19 < kanzure> such as the equations describing a diesel engine
22:19 < kanzure> I'm easily wrong here, I was just thinking about this a few moments ago too
22:19 < kanzure> oh wait, because I brought it up
22:19 < fenn> there's more than one way to skin a diesel engine
22:19 < kanzure> mm
22:20 < kanzure> it's not an object though, it's like a trace of different people's contributions in 'diesel engine design' from history and such .. "oh, you're using a Feynman hyperdrive and a two-legged Drexler coater! awesome!" not just "here's a hunk of points"
22:20 < kanzure> anyway, I need to go back and think
22:21 < fenn> better to refer to things by the principles in how they work, rather than some dead guy's name
22:21 < kanzure> that's fine too 
22:22 < willPow3r_> i am very skeptical about people who refer to themselves as "singularitarians"
22:22 < kanzure> willPow3r_: that just means they love Eliezer
22:22 < fenn> willPow3r_: i dont think anyone seriously refers to themselves as singularitarian
22:22 < willPow3r_> believing too firmly in one possible future is rediculous
22:22 < willPow3r_> some do
22:22 < kanzure> fenn: check the SL4 mailing list
22:22 < kanzure> you'd be surprised
22:23 < kanzure> also, the people I was living with up in California
22:23 < willPow3r_> lots of people even consider it a [i]religion[/i]
22:23 < fenn> it seems to me that the term is more often used by skeptical journalists trying to make some kind of sensation
22:23 < kanzure> 'Modeling the Variability with UML for TRIZ Based CAI System' heh .. buzz word alert.. UMl, TRIZ, CAI, uh oh
22:24 < fenn> what's CAI?
22:25 < fenn> hmm.. a buzzword i see :)
22:25 < kanzure> 'computer aided integration'
22:25 < kanzure> or something
22:25 < fenn> innovation
22:26 < kanzure> oh
22:26 < fenn> another cult :)
22:28 < kanzure> 'In object-oriented programming, programmers can think of a program as a collection of interacting objects, while in functional programming a program can be thought of as a sequence of stateless function evaluations. When programming computers or systems with many processors, process-oriented programming allows programmers to think about applications as sets of concurrent processes acting upon logically shared data structures.'
22:28 < kanzure> http://www.informatik.uni-trier.de/~ley/db/conf/prolamat/index.html prolomat = programming languages for manufacturing
22:30 < kanzure> http://www.informatik.uni-trier.de/~ley/db/conf/prolamat/index.html
22:30 < fenn> concurrent process acting on logically shared data structure = race condition clusterfuck
22:31 < kanzure> heh
22:31 < kanzure> it feels intuitively wrong to call a "chair" an object .. 
22:31 < fenn> if a chair isn't an object, then what is?
22:31 < kanzure> you get into monadology where all things are by necessity related to all other objects
22:32 < fenn> but they are!
22:32 < kanzure> right, so how are you going to write a program to represent that
22:33 < kanzure> oh wait, you'd need the universe
22:33 < fenn> this map/territory problem will show up no matter what philosophy you use to represent the universe
22:33 < kanzure> I would say nothing is an 'object' -- the monitor to my left (or my right) is just the result of the process of the manufacturing going on somewhere (that, sadly, I don't know about) and it's just my lack of knowledge about the world's context that precludes me from knowing more of it .. not that it's this "thing".
22:34 < fenn> the monitor performs a function by nature of its design, which is the same as every other monitor produced in that factory
22:34 < kanzure> you're not really trying to represent the universe .. you're just writing some code to translate some information into other usable information or something
22:35 < fenn> you dont care about the exact isotope ratio of the carbon atoms in the liquid crystal, as long as it works
22:36 < kanzure> if you've done too much abstraction, then you'll miss something and get boned
22:36 < fenn> that's the limits of computation
22:36 < kanzure> "the monitor" is just our grammatical conception of it - a shared word between the two of us - and both of us know that there's certain ways that this word is 'grounded' in reality and that's why we use it (I know you'll be thinking of something roughly like the "monitor" before me). It's not "a monitor" .. it's the shared social reference that I presume you have as well due to the greater context of society that I'm unaware of (seriously, where th
22:38 < kanzure> hm
22:38 < fenn> you're thinking of some assemblage of materials whereas i'm thinking of the function of showing a colored dot array
22:38 < kanzure> that's not computation, that's just data 
22:38 < kanzure> (not your last message)
22:39 < fenn> not having enough data to start with is a different problem altogether
22:39 < fenn> i think we call that 'science' or maybe just 'sensing'
22:39 < kanzure> the problem that I see is that you'll just add more and more objects into the system at varying degrees of "data collection"
22:40 < kanzure> it'd be the bioinformatics problem all over again
22:40 < kanzure> these giant databases just sitting there
22:40 < kanzure> oh, chemspider might be a good example
22:41 < fenn> if i had to pick and choose which data to keep, i'd keep the more abstract functional kind, rather than the tedious more or less arbitrary detail
22:41 < kanzure> and pubchem. millions and millions of cheminformatics
22:41 < kanzure> surely you're not going to code in every possible ways to use a chair
22:41 < kanzure> stepping latter, surface to put things, surface on which to put a person, surface on which to put a general object, ..
22:41 < kanzure> you've basically got cyc all over again :p
22:42 < fenn> "chair" is surface on which to put a person
22:42 < kanzure> *ladder
22:42 < fenn> individual structures that happen to be chairs may or may not hold that 10 ton pickup truck you're putting on top
22:43 < fenn> (i guess pickup trucks dont actually get to 10 tons oh well)
22:44 < fenn> i have some wicker chairs that tend to break if you step on the middle of them
22:44 < kanzure> now, here's the alternative -- as designer, you're sitting in front of your graphs that represent the overall system and the various ways that matter/information/energy is flowing through it or whatever, and "apt-get install chair" would install the programs to do the calculations on the dataformats that you have so that you can get "20 minutes of 500 kg wait time" (it breaks in 20 minutes) (okay, terrible example) 
22:44 < kanzure> these packages in the repo have direct correlates to being built and directly contain the programs required to hash out the math and such..
22:44 < kanzure> and of course the typical instructions we've mentioned time and time again
22:44 < fenn> there isnt much math involved in just determining if something's a chair
22:44 < kanzure> ("step 1 ... move arm left")
22:45 < kanzure> not in determining if it's a chair, but
22:45 < kanzure> let's change chair to a petroleum engineering scenario with a pipeline
22:45 < kanzure> where you want to pipe oil, so you use a pipe, right? and you do the calculations and you figure out your design constraints and so on
22:45 < fenn> ok, what if you decide to pump water through your pipe, is it still a petroleum pipeline?
22:46 < kanzure> within these constrains your procurement network :-) discussed earlier today :-) could conceivably fill in those gaps maybe, but the "objects" are -- at most -- math.
22:46 < kanzure> right, I'm not describing it as a petroleum pipe only .. it's a "way to transfer materials" no ? a "solution to the problem" and the problem is supposedly formerly expressed or something ... right ?
22:46 < kanzure> jfldajflakdja
22:46 < kanzure> I'm not about to write turtles all the way down ..
22:46 < fenn> turtles are useful though
22:47 < fenn> with enough turtles you can put water in a petroleum pipeline and see if it will work
22:47 < fenn> or determine the compression strength of a chair
22:48 < kanzure> gah, like hell I'm going to be damned to turtle herding for my life
22:48 < fenn> i think the question isn't "do we want to use turtles" but rather "how many layers of turtles can we reasonably expect to be able to use at once"
22:48 < kanzure> how do you keep the "layers pure" ?
22:48 < fenn> pure?
22:49 < kanzure> well, you can't have a particle physics definition of one thing and then couple it to an object you only have one line describing ..
22:49 < fenn> you just ignore effects that aren't in your current layer of abstraction
22:49 < fenn> if it fucks up, that's life. try again
22:49 < kanzure> what is that defined by
22:49 < kanzure> that 'current layer of abstraction'
22:49 < fenn> 'chair' is a high level layer of abstraction
22:49 < fenn> bending moments in beams of arbitrary cross section is a slightly lower layer
22:50 < kanzure> I know it's higher, I don't know what context you're using to demand that all objects in your design window are of the same abstraction level
22:50 < fenn> each function has its own context
22:50 < fenn> 'sit' function looks around for chairs
22:50 < kanzure> suppose you zoom in on the interace between a gas pump and a hose
22:51 < fenn> if no chairs are found, you can start trying to construct them out of materials at hand
22:51 < fenn> (such as the ground)
22:51 < fenn> sometimes ground isnt suitable surface to put people on
22:51 < kanzure> "sit function looks around for the chairs" hm? so when you try to demand abstraction level sameness across all of the parts of the system you're designing, you have a sit function look for something? what?
22:52 < fenn> look for objects defined as a chair (or inheriting chair class)
22:53 < fenn> it's not same abstraction level across all parts of the system, it's only within a particular interaction
22:54 < kanzure> this doesn't tell me how to make sure you're operating at the same layer of abstraction across all components in the system ... do you mean that when you throw something new into the pot/the-equation (colloquilism), that it starts to throw errors or something when it can't find a chair or 
22:54 < kanzure> whatever specifics it needs to operate seamlessly? that would make sense, I guess, but just expecting it to magically click together - as if somebody has made all possible levels of abstraction of that random object, eh ..
22:54 < kanzure> hm
22:54 < kanzure> now I thought it was those interactions/transformations that mattered in the first place
22:56 < kanzure> at the interfaces.
22:56 < fenn> 'somebody has made all possible level of abstraction of that object' is probably not true, but we can come up with many of them automatically through class inheritance, and with enough detail in the description
22:57 < fenn> this chair is made of 35mm 304 stainless tubing, we can deduce its density, modulus, tensile strength etc
22:57 < kanzure> does this preserve the ability to "apt-get" the design into a physical structure sitting behind me in my fablab/fabuntulab? those descriptions still have to be instructions rather than natural language, so I'm just checking
22:57 < kanzure> trying to see how this makes sense .. or if it doesn't.
22:58 < fenn> no, you won't be able to make any arbitrary structure out of materials on hand
22:59 < kanzure> then what does it let me do, of all things?
22:59 < fenn> especially if there isn't a detailed enough description of it
22:59 < fenn> if you have a poorly documented object, but actually physically possess an instance of it, the poor representation is still useful even if you can't duplicate it
22:59 < fenn> good lord, english sucks sometimes
22:59 < kanzure> yep
23:00 < fenn> you can teach a man to fish..
23:00 < kanzure> fish for turtles?
23:00 < fenn> but i wont spurn a fish just because i dont have a fishing rod
23:01 < fenn> there's this idea that anyone can make anything, because anything that can be made has been made
23:01 < kanzure> right
23:02 < fenn> so, the reason object oriented descriptions of 'stuff' is good, is that you dont have to recalculate all the possible functions from ground up every time you want to use it
23:03 < kanzure> gentoo
23:03 < fenn> fuck gentoo
23:03 < fenn> i know what i want, and i want it now
23:03 < fenn> wah
23:04 < fenn> so, yeah. in a world of infinite computing power, we wouldn't exist
23:04 < fenn> you know about the tipler omega point>?
23:05 < kanzure> yes
23:06 < kanzure> okay, so let's say that those functions have to be calculated at least once
23:06 < kanzure> I agree that it's worth encapsulating, perhaps within the shell of a turtle
23:07 < kanzure> but it still has to exist, no?
23:07 < fenn> what has to exist?
23:07 < kanzure> the possible functions that you calculate and compile into your design, the 'constraints' and 'design/matter compiler' and so on
23:08 < fenn> the model could be wrong, of course
23:08 < fenn> the design might not actually work
23:08 < fenn> in that case, you must examine the process used to determine functionality (lower level functions)
23:09 < fenn> at some point they will have to be transcended, make a new scientific paradigm, but this doesn't happen often
23:09 < kanzure> let's say I have fabuntu running, assembled my chemical pipe line thingy and it has detected a leak in its unit testing procedures (hurray!), 
23:09 < kanzure> it then generates a quick bug report, stops operations immediately and then generates you a report
23:09 < kanzure> telling you that the implementation stuff is wrong
23:09 < kanzure> what then?
23:10 < kanzure> what do you mean by 'examine the process used to determine' -- do you mean, refactor the objects you have in your window ?
23:10 < fenn> what does the bug report say?
23:10 < kanzure> gas leak?
23:10 < fenn> how does it know there's a gas leak?
23:10 < kanzure> difference between gas on one end and the other of the pipe?
23:10 < kanzure> it's a modular unit test debugging procedure
23:11 < fenn> so, presumably one of the components had some geometry code that determined flow through the pipe, right?
23:11 < fenn> er, s/one/all/
23:11 < kanzure> I guess ... what if it's just a hole in the pipe, an error in the manufacturing?
23:11 < kanzure> well, I guess that's the geometry
23:12 < fenn> error in manufacturing, very bad! six sigma!
23:12 < kanzure> it's conceivably some other problem though, like the chemistry reacting with the inside tank walls
23:13 < fenn> so, if there's no code to calculate chemical corrosion, someone would have to write some
23:13 < fenn> and splice that into the unit testing procedures
23:13 < fenn> this gets horrendously complicated really fast
23:13 < kanzure> yeah ..
23:14 < fenn> this is why we have testing and proof of concept studies
23:14 < fenn> because the difference between theory and practice is greater in practice
23:16 < fenn> i'm going to venture a wild opinion and say that anything as stupid as a gas leak simply won't happen under normal circumstances
23:16 < kanzure> if I come in and play devil's advocate and bust up your tank or pipe, your system should very well detect that
23:17 < fenn> why should it?
23:17 < kanzure> unless you opted to exclude the sensor system
23:17 < kanzure> well, gas leaks are kinda deadly
23:17 < kanzure> also, the idea should be to have it operating :)
23:17 < fenn> what proportion of overall cost should be devoted to keeping the system from blowing up due to arsonists?
23:17 < fenn> eventually you will reach some gnarly security state with nukular lasers
23:18 < kanzure> and an ai overlord
23:18 < fenn> naturally
23:18 < kanzure> I see your point.
23:18 < kanzure> I thought he's on first
23:19 < kanzure> it occurs to me that it would be fun to be an industrial spy and stake out the containers on the backs of returning semi's.
23:19 < kanzure> to trace the supply chain.
23:20 < fenn> kurzweil (of wankal reputation) thinks that intelligence is the ability to achieve goals with limited resources
23:20 < kanzure> too bad nobody can prove the existence of goals from a biophysical perspective
23:20 < fenn> so you can't be 100% certain that it will work, you have to figure out what problems are worth spending time on
23:20 < kanzure> agreed
23:21 < fenn> nobody can prove anything
23:21 < fenn> goals are what i say they are
23:21 < fenn> unless i'm lying of course
23:21 < kanzure> how's that working out for you
23:21 < fenn> not very well really, nobody seems to share my goals really
23:23 < fenn> thanks for wanting to take over the galaxy, i appreciate it
23:23 < kanzure> heh
23:24 < kanzure> is that what I want?
23:24 < fenn> close enough
23:24 < kanzure> ah
23:24 < kanzure> galaxy domination isn't bad though, not many have done that
23:24 < kanzure> anyway, 
23:24 < kanzure> how would you get the object database to a usable status?
23:25 < fenn> by using it and fixing the parts that dont work in practice
23:25 < kanzure> kinda useless if you can only apt-get a few pieces of paper from your inventory
23:25 < kanzure> oh right
23:25 < kanzure> origami stuff
23:25 < kanzure> oh, we could use graphs for origami instead of silly blender
23:25 < kanzure> oh right, I looked into that
23:25 < kanzure> MIT closed source stuff
23:25 < fenn> origami sucks
23:25 -!- willPower__ [n=will@cpe-66-27-123-142.san.res.rr.com] has joined #hplusroadmap
23:25 < kanzure> crap
23:25 < kanzure> papakura too though
23:25 < fenn> its worse than reprap extruded plastic. the only advantage is that people already have laser printers
23:26 < kanzure> folds were edges or something
23:26 < fenn> what does origami *do*
23:26 < kanzure> waste our time?
23:28 < fenn> you can make an envelope, or a pinwheel, but they are so limited in shape and material performance that it really is a waste of time
23:28 < kanzure> the 'automated design lab' here had a little project that generated 10k permutations of a factory design for some soda coke bottle machine thingy 
23:28 < fenn> this is the same reason reprap is a waste of time, incidentally
23:28 < fenn> really, permutations? or variations?
23:29 < kanzure> what do you consider the difference to be ?
23:29 < fenn> permutation would be topologically different, i guess
23:29 < kanzure> permutations as in "a,b,c," and "a,b,d" and variations would be different - how?
23:29 < fenn> whereas a variation can change smoothly from one to the next
23:29 < fenn> btw were they "good" designs? actually funcional?
23:30 < kanzure> give me a few moments to check
23:30 < kanzure> it's worth reading or at least glancing at
23:30 < fenn> i looked at adl stuff a bit, but it kinda just seemed like boxes with words in them connected in a row
23:31 < kanzure> yeah, the software stuff just boggles me - doesn't look like it _does_ anything
23:31 < fenn> at least the constructal site had a pretty picture of some heat exchanger
23:32 < kanzure> http://www.me.utexas.edu/~adl/Dissertation_Research.htm
23:33 < kanzure> grr
23:33 < kanzure> it was already on the server
23:33 < kanzure> http://heybryan.org/~bbishop/docs/manufacturing/A%20computational%20approach%20to%20innovative%20conceptual%20design%20-%20Tolga%20Kurtoglu.pdf
23:33 < kanzure> page 105 was where I was at
23:34 < fenn> bleh.. do i really have to read that?
23:35 < kanzure> no, just scroll around and act like you care
23:36 < kanzure> I think they're just using simple components
23:36 < kanzure> since they don't plan to convert this into a real system 
23:37 < kanzure> page 88 (101 in the pdf thingy) has their graphsynth software with a screenshot
23:37 < kanzure> very confusing diagram
23:37 < fenn> well what the hell are they doing if it's not for real
23:37 < kanzure> "design"
23:39 -!- nsh [n=nsh@87-94-146-186.tampere.customers.dnainternet.fi] has quit [Read error: 110 (Connection timed out)]
23:39  * fenn is not impressed
23:40 -!- ybit [n=h@unaffiliated/ybit] has quit [Remote closed the connection]
23:40 -!- ybit [n=h@unaffiliated/ybit] has joined #hplusroadmap
23:41 < kanzure> it's kind of like my schedule generator, except a design generator or something, maybe
23:41 < kanzure> but I guess if they're not actually doing a thing
23:41 < kanzure> hm.
23:41 < fenn> i liked lipson's GA simulations -> rapid prototyped robots because they actually came up with automatically designed robots that worked
23:42 < fenn> the sand fish thing in the upper left: http://ccsl.mae.cornell.edu/
23:44 -!- willPow3r_ [n=will@cpe-66-27-118-88.san.res.rr.com] has quit [Read error: 110 (Connection timed out)]
23:44 < kanzure> hm
23:44 < kanzure> notbad
23:44 < fenn> surely there is a graph grammar going on somewhere in the simulation, but they dont make a huge deal about it
23:44 < kanzure> hod lipson
23:44 < kanzure> I've seen hod somewhere before
23:44 < kanzure> is he prolific or something?
23:44 < fenn> yes
23:47 < kanzure> oh, he's the block-arm-assembly-guy
23:48 < kanzure> http://www.molecubes.org/
23:48 < fenn> reminds me of superbot
23:49 < kanzure> 'SuperBot - User Friendly Web Downloader Software
23:49 < kanzure> SuperBot is the world's smartest, fastest, easiest website downloader. Download a free trial copy now!'
23:49 < kanzure> http://www.youtube.com/watch?v=5qz2lmUlZxs or that
23:49 < fenn> no, second result dummy
23:49 < kanzure> possibly http://www.alicebot.org/superbot.html
23:49 < kanzure> http://www.isi.edu/robots/superbot.htm ?
23:49 < fenn> http://www.isi.edu/robots/superbot.htm
23:50 < fenn> its the same cube-flexes-and-hooks-to-other-cubes
23:52 < fenn> its cool that they distribute the cad files and BOM
23:53 < kanzure> BOM?
23:53 < fenn> bill of materials, aka stuff you buy
23:53 < kanzure> oh, right
23:54 < fenn> i dont really get the fascination with rapid prototyped stuff.. CNC is so much better faster and cheaper
23:55 < kanzure> people are clueless, perhaps?
23:57 < kanzure> I mean that they genuinely don't know. Amplifying potentials of the internet ..
23:58 < fenn> i think its the amplifying potentials of the internet that made rapid protoypers the next big fad
23:58 < kanzure> right
23:58 < kanzure> what's the chances that the first freshman I'd talk to on campus would be "well versed" in reprap and such
23:59 < kanzure> (well, lots of bias on my end of course - maybe I just target these people)
23:59 < fenn> heh, definitely
23:59 < fenn> but i'm surprised to see something from a cornell robotics professor using squirted plastic goo where clearly anything else would be better