--- Day changed Sat May 24 2008 00:49 -!- kanzure [n=bryan@cpe-70-113-54-112.austin.res.rr.com] has quit ["Leaving."] 00:50 -!- ybit [n=u1@c-71-207-240-143.hsd1.al.comcast.net] has joined #hplusroadmap 00:58 < ybit> http://www.facebook.com/video/video.php?v=3086075142 00:59 < ybit> thought this might stimulate no intellectual conversation whatsoever 01:01 < ybit> also located here: http://www.youtube.com/watch?v=kL_CVwqnn4k 01:28 -!- kramer3d [n=kramer3d@ip98-169-184-165.dc.dc.cox.net] has joined #hplusroadmap 05:06 -!- Vedestin [n=Vedestin@d58-111-90-12.sbr3.nsw.optusnet.com.au] has quit [Read error: 104 (Connection reset by peer)] 05:28 -!- kramer3d [n=kramer3d@unaffiliated/kramer3d] has quit [] 06:08 -!- Biopunk [n=p@h13n3c1o261.bredband.skanova.com] has quit [] 08:27 -!- Vedestin [n=Vedestin@d58-111-90-12.sbr3.nsw.optusnet.com.au] has joined #hplusroadmap 09:32 -!- Vedestin [n=Vedestin@d58-111-90-12.sbr3.nsw.optusnet.com.au] has quit [Read error: 104 (Connection reset by peer)] 09:58 -!- kanzure [n=bryan@cpe-70-113-54-112.austin.res.rr.com] has joined #hplusroadmap 10:00 < kanzure> CD3WD's clone: http://www.humaninfo.org/project_details.html 'The equivalent knowledge of 16 full university degrees available at fingertips' 10:00 < kanzure> 'An average university degree of four years requires integrating about 10.000 pages of information and knowledge. This HDL library 2.0 contains 160.000 pages, all accessible within seconds. This means the equivalent knowledge of about 16 full university degrees available at fingertips to any person.' 10:11 < kanzure> This is a terrible idea: http://nzdl.sadl.uleth.ca/cgi-bin/library?a=p&p=about&c=mhl 10:11 < kanzure> "Humanity Info Project" ... oh, but you must use our interface to the information 10:14 < kanzure> http://www.unisdr.org/eng/public_aware/world_camp/2008-2009/wdrc-2008-2009.html International Strategy for Disaster Reduction 10:14 < kanzure> Hm. 12:31 -!- Vedestin [n=Vedestin@d58-111-90-12.sbr3.nsw.optusnet.com.au] has joined #hplusroadmap 13:12 -!- kramer3d [n=kramer3d@ip98-169-184-165.dc.dc.cox.net] has joined #hplusroadmap 13:13 < kanzure> Hey kramer3d. 13:13 < kanzure> Hi Vedestin. 13:13 < kramer3d> hey kanzure whats up 13:13 < kanzure> I'm going through some international organizations, seeing some things that might be relevant to SKDB. 13:14 < Vedestin> hi kanzure 13:51 < kanzure> http://www.isi.edu/~salemi/ self-reconfigurable robotics for space exploration 13:54 < Vedestin> cool 13:57 -!- Phreedom [n=freedom@www.online.dn.ua] has joined #hplusroadmap 14:04 -!- amaruk [n=amaruk@p54B1B449.dip0.t-ipconnect.de] has joined #hplusroadmap 14:06 < kanzure> Hey amaruk. 14:08 -!- amaruk [n=amaruk@p54B1B449.dip0.t-ipconnect.de] has quit [Remote closed the connection] 14:28 < kanzure> It'd be nice if we could solve the browser problem. 14:54 < kanzure> http://www.nsf.gov/funding/pgm_summ.jsp?pims_id=13346 15:55 -!- kramer3d [n=kramer3d@unaffiliated/kramer3d] has quit [Read error: 110 (Connection timed out)] 15:57 -!- kramer3d [n=kramer3d@ip98-169-184-165.dc.dc.cox.net] has joined #hplusroadmap 16:20 < kanzure> http://www.cs.utexas.edu/users/novak/cgi/apserver.cgi automatic programming? 17:37 -!- kramer3d [n=kramer3d@unaffiliated/kramer3d] has quit [] 18:03 -!- kramer3d [n=kramer3d@ip98-169-184-165.dc.dc.cox.net] has joined #hplusroadmap 18:04 -!- kramer3d [n=kramer3d@unaffiliated/kramer3d] has quit [Remote closed the connection] 18:50 < fenn> its so automatic i dont even have to run the program to get equivalent results! 18:50 < kanzure> neat. 18:50 < kanzure> hehe 18:50 < kanzure> I think it means "intentional programming" 18:50 < fenn> well, 0=0 18:50 < kanzure> also, isoc.org should be interested in our projects 18:50 < kanzure> if not, I'm investigating the local manufacturing, uh, 'clubs' 18:50 < kanzure> sanctioned by the uni 18:51 < kanzure> also, to solve my browser problems 18:51 < kanzure> I'm considering hacking konqueror 18:51 < kanzure> I want to put it under the control of a messy perl script 18:51 < fenn> check out dcop 18:51 < fenn> run 'kdcop' 18:51 < kanzure> and this script would be what I control/navigate-with --- it would track link following behavior, back/forward/next calls, and close/open windows on demand and do my 'tabbing' 18:51 < kanzure> okay 18:52 < kanzure> what is this ? 18:52 < fenn> a list of methods callable through dcop 18:52 < kanzure> direct api access ? 18:53 * kanzure runs off to eat 18:53 -!- Biopunk [n=p@h194n2c1o261.bredband.skanova.com] has joined #hplusroadmap 18:55 < fenn> i use this one all the time, mapped to ctrl-alt-l in icewm: http://fenn.freeshell.org/gnurds/newkonq.sh 18:56 < fenn> its like hyperwords but less mouse-clicky 18:57 -!- wrldpc [n=ben@pool-71-174-89-247.bstnma.fios.verizon.net] has joined #hplusroadmap 18:57 < wrldpc> Hello. 18:58 < fenn> hi 18:59 < fenn> what brings you to our humble channel? 19:04 < kanzure> Hello wrldpc. 19:05 < kanzure> fenn: so your script is partially what I'm looking for. 19:05 < kanzure> I also want to interface with konq so that I get back/forward commands, stuff that's important to record like that 19:05 < kanzure> I could do logging via a proxy gateway I guess 19:05 < fenn> so you can record your 'path' and replay it? 19:05 < kanzure> yes, and other things 19:05 < kanzure> for example, my 300+ tabs in Opera from a few minutes ago 19:05 < kanzure> was making this thing sound like a jet airplane 19:06 < fenn> beeeeooooooowwwwpshshhhh 19:06 < kanzure> so, I want to be able to do tabbing sessions without the agonizing ridiculousness 19:06 < kanzure> nah, before takeoff 19:06 < kanzure> I'm trying to come up with a good idea for an interface 19:06 < kanzure> might just do bookmarking 19:07 < fenn> it all comes back to graph visualization it seems 19:07 < kanzure> right 19:07 < kanzure> but in this case I'm thinking of a small list of 'tabs' 19:07 < kanzure> this could be a page rendered by an app feeding off of my bookmarks db 19:07 < kanzure> simple XBEL -> HTML rendering or something 19:07 < kanzure> but it doesn't allow for easy tab manipulation like in Opera :-/ 19:08 < fenn> never having used opera, i can't provide any meaningful insight on that 19:08 < kanzure> oh, well 19:08 < kanzure> http://heybryan.org/shots/2008-05-24.png 19:09 < fenn> so what, it's just a list with little icons next to them 19:09 < fenn> easily done with frames 19:10 < kanzure> it's not just a list though, it's drag/droppable too 19:10 < kanzure> and a double click sends you to the tab 19:10 < kanzure> yeah, it's that stupid -- all of them are in immediate mem 19:10 < fenn> where do you drag them to? 19:10 < kanzure> bookmark manager, for example 19:10 < fenn> how is that useful? 19:10 < kanzure> well, it's the only way I can bookmark in Opera 19:10 < kanzure> without it crashing 19:11 < kanzure> I guess this could all be replicated with a JS interface 19:11 < kanzure> but frankly I'd like to avoid that. 19:11 < fenn> so. user interface stuff is hard to explain 19:11 < fenn> unless you have extensively studied HCI and know all the lingo 19:12 < kanzure> well, what would an alternative interface be like 19:12 < kanzure> still need a list of active tabs that you're working with, right? 19:12 < fenn> well, for one thing it'd never lose any input to the system (like elephant) 19:12 < kanzure> elephant? but yes. 19:12 < fenn> should be able to make both hierarchical and tagged sorting representations 19:13 < kanzure> should be able to scripthack the hell out of all of it 19:13 < fenn> semantically and graphically zoomable 19:13 < kanzure> eh? 19:13 < fenn> 'preview' icons are just the beginning 19:13 < fenn> but shrinking the rendered page down only goes so far.. everything ends up looking like white boxes with black text 19:14 < kanzure> right 19:14 < kanzure> back when I was using firefox I found this extension that was called 'webhistory' or something 19:14 < fenn> so we have these 'favico.ico' or whatever - the little pics next to the bookmark 19:14 < kanzure> it was sqlite + history + bookmarks + your browser trail session in one 19:14 < kanzure> on the side of the screen 19:14 < kanzure> the favico.ico doesn't matter much to me 19:14 < kanzure> I just need a list that can facilitate my selection of pages to open up in a timely manner, really 19:15 < kanzure> I don't care if it has to fetch it from disk 19:15 < kanzure> as long as I get to play around with different strategies for managing it 19:15 < fenn> well, with gigs of ram you shouldn't need to fetch it from disk :) 19:15 < kanzure> you'd think so, but still 19:15 < fenn> yes one would think so, but apparently the engineers at firefox have their own ideas 19:15 < kanzure> I have 1 GB RAM and Opera still whines like a whore with +300 tabs 19:16 < wrldpc> anyone know of a top secret tech project code named 'stonehenge' ? 19:16 < fenn> hmm was that the druid mainframe? 19:16 < wrldpc> heh 19:17 < wrldpc> got it 19:17 < wrldpc> Adobe's new CS4 19:17 < wrldpc> or CS 'whateva' 19:18 < wrldpc> Bryan you are a legend in Boston :) 19:18 * fenn stares at wrldpc blankly 19:18 < kanzure> Am I? 19:18 < kanzure> How's that? 19:18 < fenn> adobe? wasnt that some proprietary software company from the nineties? 19:19 < kanzure> fenn: no, it was a brick 19:19 < fenn> oh, right, like superadobe 19:19 < kanzure> used to make houses in the 0090s. 19:19 < wrldpc> At the DIY Bio meet up your name wasn't mentioned but you were referenced positively as "some high school genius doing everything himself" 19:19 < fenn> since obsolesced by compressed earth block 19:19 < kanzure> Awesome. 19:20 < kanzure> wrldpc: There's new updates to http://heybryan.org/ btw. 19:21 < kanzure> fenn: so, uh, interface. here's a good way to decide: 19:21 < kanzure> clicking required or no clicking? 19:21 < fenn> response time under 0.1 sec 19:21 < kanzure> this is mainly for the list of tabs 19:22 < kanzure> of 'current workspace' (whatever you're browsing about) 19:22 < kanzure> I think no clicking would be very nice 19:22 < kanzure> only time you should click is when konqueror is brought up 19:22 < kanzure> and you want to visit a link or something 19:22 < fenn> actually konqueror has these 'access keys' if you tap ctrl, gives you a letter for each link 19:23 < fenn> i never got the hang of it 19:24 < kanzure> oh, so that's what that is 19:24 < kanzure> hard to predict 19:26 < kanzure> seems to miss a few links on my sie 19:26 < fenn> i wonder how these humanitarian projects compare to wikipedia 19:27 < kanzure> fenn: hm? 19:27 < kanzure> the cd3wd projects? 19:27 < kanzure> or places like WHO? 19:27 < fenn> yes that and humaninfo.org 19:27 < kanzure> hm 19:27 < kanzure> well, they aren't wikis 19:27 < kanzure> and they probably should be 19:27 < kanzure> apparently it's just UN pamphlets dumped into a CD 19:28 < fenn> they are just agglomerations of various organizations' attempts to write a how-to encyclopedia 19:28 < kanzure> it's the UN ;-) 19:28 < fenn> ffs they could at least try to make the format consistent 19:28 < kanzure> making me download it from cnet is a stupid thing 19:28 < kanzure> I was going through each of the individual pages and clicking 'download' 19:28 < kanzure> so they suck. 19:29 < fenn> sure that's just cluelessness 19:29 < fenn> ideally there's be a torrent with all 16GB or whatever 19:29 < kanzure> right 19:29 < fenn> alex seems to come from a "commercial" programming background 19:30 < kanzure> peculiar :) 19:30 < fenn> buncha weirdoes :) 19:30 < kanzure> assbackwards 19:30 < fenn> no it makes perfect sense 19:30 < fenn> in the usual assbackwards way 19:30 < fenn> see, these relief organizations get a big chunk of money, they're like crap what do we do with this money! i know, we'll hire some consultants 19:30 < kanzure> guy who makes profit decides to turn around and liberate 3rd world development info in his spare time? 19:31 < kanzure> *steal steal steal* 19:31 < fenn> so then they use the same techniques as huge corporations with constant cash flows 19:31 < fenn> but they dont have a cash flow, they just have this big chunk at the beginning, then the data rots after that 19:32 < fenn> the methodology requires constant input of paid effort (since there's the commit access problem, no redistribution/modification etc) 19:32 < kanzure> maybe I don't need a list-view of tabs. just do a queue manager and go through it linearly. when done with a page, dump it in the bookmarks and write a few notes 19:34 < fenn> now i see why you want back-links to project maintainers 19:35 < fenn> i figured that it doesn't matter who the maintainer is, if all the documentation and source is there, and you have equal clout to modify and redistribute 19:35 < kanzure> sure you do, but .. the S in skdb is important :) 19:35 < fenn> S meaning social interaction? 19:35 < kanzure> yeah, it's a good bet that the guys who were doing the code might have a better clue than you 19:36 < kanzure> what did you realize about my wanting the backlinks anyway? 19:36 < fenn> oh, well i was browsing archive.org on the NIST hexapod project, and i realized that they didnt just delete all the data on purpose, it was some guy got laid off and his pet projects bit-rotted 19:37 < kanzure> oh, sure, that too 19:37 < fenn> and nobody even knew about their existence so they disappeared 19:37 < kanzure> I'm seeing a lot of organizations that *should* be very enthusiastic about skdb 19:38 < kanzure> IRTF, IETF, ISOC, WFP, ILO, FAO, UNESCO, UNICEF, UNEP, WMO, UNDP, WHO, NIST, WTA, NSF, etc. 19:38 < fenn> found this vrml model last night on archive.org (unfortunately not the pro-engineer files) http://fennetic.net/pub/irc/nist_hexapod.png 19:38 < kanzure> heh, blender 19:39 < kanzure> or is that you? 19:39 < fenn> that's me importing vrml into blender 19:39 < kanzure> ah, right, vrml model 19:39 < fenn> its nice to finally see it in 3d thogh 19:40 < kanzure> fenn: tell me why I need a visual list of tabs in my queue 19:40 < kanzure> much less a list that I can jump around in 19:40 < kanzure> do I really need that ? 19:40 < kanzure> do you? 19:40 < fenn> um, i'd like to "burn" a path through the desert and be able to look back on it 19:40 < fenn> but i dont know how to turn that into a computer interface exactly 19:41 < kanzure> path visualization can be done pretty easily, IMHO, at least as linear paths (retrospectively) 19:41 < kanzure> for example, in your bookmarks you could say "show session-pathings" 19:41 < kanzure> and then it would show what you were doing that got you to that page 19:41 < fenn> not so much jump around, but pull something from a couple minutes ago back into view, relating it to the current set of things i'm paying attention to 19:41 < kanzure> now, in cases where you didn't click to it (random insight to do a search versus being given a link), that's up to you - I guess you could always write down a few words, but I don't know any ML for that. 19:42 < kanzure> hm 19:42 < kanzure> yes, I do that too 19:42 < kanzure> I also jump between pages a lot 19:42 < kanzure> like when copying information in some stupid way 19:42 < fenn> also would like to tag/annotate/score a numerical "interestingness level" and possibly auto-archive over a certain level 19:43 < kanzure> sure 19:43 < fenn> but i think the whole bookmark idea is fubared 19:43 < kanzure> how's that? 19:43 < fenn> its just a bunch of links 19:43 < kanzure> oh, bookmarks-as-active-tabs ? 19:43 < kanzure> yes 19:43 < fenn> its supposed to be a memory augmentation system 19:43 < kanzure> right? 19:43 < kanzure> why can't it be links to local caches too? 19:43 < fenn> like.. i'm thinking of some page, its mostly red and has something to do with 1959 ford's 19:44 < fenn> can bookmarks do that? 19:44 < kanzure> uh, are they designed to? 19:44 < fenn> no fucking way 19:44 < kanzure> I don't think that's the idea 19:44 < fenn> its just a bunch of links 19:44 < kanzure> it's not a memory augmentation system 19:44 < kanzure> rather, 19:44 < kanzure> it's like an skdb package ;-) 19:44 < kanzure> these links can't really be intuited out of the divine void 19:44 < kanzure> that's why we have to use search engines 19:44 < kanzure> (and remember some particular servers, of course) 19:45 < kanzure> screw URLs, just imagine as if they are IP addresses only 19:45 < kanzure> because essentially the URLs are only for remembering 19:45 < fenn> back in the early days before search engines, there were these 'phone books' that had basically a hierarchical organization of pages by topic 19:45 < kanzure> but they are still essentially like IP addresses in that you can't really do the whole territory=map BS 19:45 < kanzure> yeah 19:45 < kanzure> but that's a MPOV thing 19:45 < kanzure> and that sounds like hierarchical bookmarks to me :-/ 19:46 < kanzure> the links were telephone numbers instead of hyperlinks 19:46 < fenn> oh, not phone numbers, real hyperlinks 19:47 < fenn> think gopher 19:47 < kanzure> ? 19:47 < fenn> anyway, you're right, that's no different from bookmarks 19:47 < kanzure> I don't know what you want, 19:48 < kanzure> this isn't my 'revolutionary web interface' stuff :) 19:48 < kanzure> this is just me trying to get through the day 19:48 < kanzure> IMHO, the next generation web interface will be my neurosci projects 19:48 < kanzure> the ones that exploit the information otherwise trapped in your head 19:49 < kanzure> ones where you can literally "import" information and so bookmarks are just a part of the underlying framework 19:50 < fenn> well, i've always been more of a fan of wearables than implants.. considering a PC a wearable in this context 19:50 < fenn> PC is an intelligence/memory/communication augmentation system no? 19:51 < kanzure> in a limited sense, yeah of course 19:51 < kanzure> I'm much more myself on a box than on paper, for example 19:56 < Biopunk> I've been thinking along the lines of 3d mindmaps 19:56 < kanzure> Eh? 19:56 < kanzure> You mean the 'write everything down as a map' strategy? 19:56 < Biopunk> for organizing large chunks of information in a way I can remember it 19:56 < kanzure> Anything that requires you to write, sucks. 19:56 < fenn> agreed 19:57 * Biopunk sucks 19:57 < kanzure> My brain compiler architecture is based off of the idea of auto-importing and integration of ideas, stimulation, and the automated processing of excess information. 19:57 < kanzure> Haha. 19:57 < kanzure> Biopunk: :) Don't take it that way. 19:57 < Biopunk> sob 19:57 < fenn> writing is about 100-1000 times slower than reading 19:57 < kanzure> Biopunk: I would like to see a 3D mindmap anyway, especially if it had those fancy navigation controls like characters flying through simulations. 19:57 < Biopunk> I meant for myself.. like to learn all the biology terminology 19:57 < kanzure> plus colors and fields of colors or something, I'm sure somebody's tried that out with fogs and so on 19:57 < kanzure> ah. 19:58 < fenn> mnemosyne? supermemo? 19:58 < kanzure> all of the bio terminology? that would overload supermemo 19:58 < kanzure> I mean, you could do it, if you could spend 14 hours a day 19:58 < Biopunk> i want fog 19:59 < fenn> even a hierarchical menu is faster than trying to fly through some VR environment 19:59 < Biopunk> i have been thinking about weather it would work... it adds the spacial tag to info... and level 19:59 < kanzure> while true, the supermemo strategy of repetitious and so on ... it's hard to download 20,000 terms into your head that way 19:59 < kanzure> I was spending way too much time with the tagging problem and writing questions / new content into the db 19:59 < kanzure> importing it all at once would have been nice, but 19:59 < Biopunk> maybe not if it's well organized 19:59 < kanzure> but ideally you're writing the content that you throw into it because that's how you're encoding it 19:59 < fenn> yeah supermemo doesn't give you a holistic view of things, which you need (at least initially) to build up a sense of what those definitions really _mean_ 20:00 < kanzure> no, trust me, organization only makes it worse 20:00 < kanzure> you need 'organization' (analysis) in terms of _processes_, not objects/hierarchies 20:00 < Biopunk> mm... i kinda think it's vital 20:00 < kanzure> how's that tree of life working out for ya 20:00 < kanzure> hehe :) 20:01 < Biopunk> ? the game? 20:01 < kanzure> there's a game? 20:01 < kanzure> that only complicates the issue 20:01 < fenn> Biopunk: kanzure was referring to the observer bias in taxonomy 20:01 < Biopunk> i was seeing some tree representation of the life game .. 'game of life'? 20:01 < kanzure> and, the taxonomical problems of there being multiple ways to approach a subject 20:02 < fenn> biologists have "solved" this problem by declaring "wah! i'm more senior than you! deal with it" 20:02 < Biopunk> yeah.. but i was talking about info that's alrady organized.. as names in biology 20:02 < kanzure> interesting solution, fenn 20:02 < kanzure> too bad it sucks. 20:03 < kanzure> I like that word today, apparently. 20:04 < kanzure> re: jumping around / going back to something from a few minutes ago (using active attention as opposed to whatever script for tab-queue-processing you wrote earlier) 20:04 < kanzure> what if you could have a log viewer for one monitor? 20:04 < kanzure> I don't know if I like that implementation though 20:04 < kanzure> something like that 20:05 < Biopunk> fenn: you have a point with the 'organize along the process' though 20:05 < Biopunk> t 20:05 < fenn> i do? 20:06 < Biopunk> you mean info has to be attached to a backbone 20:06 < fenn> no, there is no backbone 20:06 < Biopunk> hehehehe 20:06 < fenn> your brain works by relating lots of different blobs of information 20:06 < Biopunk> ok 20:06 < fenn> its not a hierarchical tree with one root 20:07 < fenn> this is why you can never get to the 'final' definition in a dictionary 20:07 < Biopunk> i called your process backbone i think.. cause thigs are attaced to it.. bad word maybe 20:08 < fenn> i dont understand kanzure's insistence on organization around processes - its just another type of bias imho 20:08 < fenn> some languages like inuit? use process as a basic unit of communication 20:08 < fenn> instead of nouns 20:08 < fenn> or haskell vs java, this argument's been thoroughly flogged 20:09 < Biopunk> do the inuits build nuclear subs? 20:09 < fenn> are you saying nuclear subs are a good thing? :) 20:09 < Biopunk> it's better to have them than not to 20:09 < fenn> why is that? 20:09 < Biopunk> mutual assured destruction? 20:10 < fenn> yes, so any of our petty squabbles can result in destroying the entire planet, great 20:10 < fenn> Biopunk: you're a poor representative for whitey 20:10 < Biopunk> ... it takes som organized effort to build them enyway.. so i used them as an example 20:11 < Biopunk> for 'the man' maybe 20:11 < fenn> honestly i dont know much about the inuit.. i'm sure it takes a lot of community organization to live in the arctic wilderness 20:12 < fenn> the point is almost moot. look at the lojban effort - people can't change their own bias even if they try 20:12 < Biopunk> i'm a bit sceptical of putting indigenous people on pedistals 20:13 < fenn> well. i'm a bit sceptical of putting invaders on pedestals 20:13 < Biopunk> ok... indigenous cultures then 20:13 < fenn> same diff 20:14 < fenn> unfortunately they're all dead so there's nobody to argue for their way of life 20:14 < fenn> or demonstrate it 20:14 < fenn> so i might as well be talking about how cool babylonianism is 20:15 < Biopunk> did their culture produce space flight and western medecine? 20:15 < fenn> well, fwiw space flight and 'western' medicine came out of nazi germany 20:16 < fenn> and with that, the thread has reached its natural conclusion :) 20:16 < fenn> nice talkin to ya 20:17 < Biopunk> there is a name for that somewhere... when the discussion ends with one calling the other a nazi i think? hehehe 20:17 < fenn> godwin's law 20:17 < Biopunk> ah ;) 20:18 < Biopunk> just heard about it maybe a year ago 20:24 < kanzure> Biopunk: It was I who was bringing up the 'backbone' process architecture stuff. 20:24 < kanzure> fenn: it's not just another type of bias :) process is where you have something that is actually *doing it*, whereas objects are just static and stale 20:25 < kanzure> re: effort to build submarines; a man in his backyard has built a nuclear submarine, IIRC. 20:25 < kanzure> however, the idea was to have it swim, so that didn't quite work out 20:25 < kanzure> my mom wants me to digg through trash 20:25 < kanzure> why is she here? 20:26 < Biopunk> kanzure: (yeah, I see that now) 20:26 < Biopunk> kanzure.. is this cannel logged by krebs? 20:27 < fenn> krebs is some bot that epitron left here 20:27 < Biopunk> so noone knows? 20:28 < fenn> i keep logs fwiw, they are not published anywhere 20:28 < Biopunk> thanks.. i like that 20:29 < Biopunk> there is some humor in 'krebs is some bot that epitron left here' ;) 20:30 * Biopunk examines krebs 20:31 < Biopunk> !help 20:31 < krebs> help topics: 6 core modules: auth, basics, config, httputil, remote, userdata; 73 plugins: alias, autoop, autorejoin, azgame, babel, bans, bash, cal, chanserv, chucknorris, debug, deepthoughts, delicious, dice, dict, digg, dns, eightball, excuse, factoids, figlet, forecast, fortune, freshmeat, grouphug, hl2, host, imdb, insult, iplookup, karma, keywords, lart, lastfm, linkbot, markov, math, modes, nickserv, q, quiz, quote, reaction, realm, remind, 20:31 < krebs> remotectl, ri, roshambo, rot, roulette, rss, salut, script, search, seen, shiritori, shortenurls, slashdot, spell, theyfightcrime, threat, time, topic, translator, tube, twitter, unicode, urban, url, usermodes, weather, wheeloffortune, wserver (help for more info) 20:31 < fenn> speaking of logs, i'm looking over some logs where kanzure is asking about a metalworking ontology; dmess answers 'its called the machinists handbook' which is true 20:32 < fenn> unfortunately it's 400MB: http://fennetic.net/pub/ebooks/bookwarez/machinerys-handbook-26.tar 20:33 < fenn> i wonder if that's accessible from outside 20:35 < fenn> http://en.wikipedia.org/wiki/Machinery's_Handbook 20:37 < Biopunk> help basics 20:37 < Biopunk> krebs help basics 20:37 < krebs> basics: quit, restart, join, part, hide, save, nick, say, action, topic, quiet, talk, ping, mode 20:37 < wrldpc> uh 20:37 < fenn> Because the information in the Handbook was so varied, little attempt was made to arrange it by subject matter. It was thought sufficient to include an extensive index of over 30 pages, with more than 4500 references, from which the location of any subject in the Handbook could quickly be found. 20:38 < kanzure> fenn: downloading to data server :) 20:40 < fenn> hmm there is a 34MB torrent, i wonder what the diff is 20:40 < kanzure> plaintext hopefully? 20:42 < fenn> there are seriously a lot of diagrams 20:42 < fenn> and tables and charts 20:44 < Biopunk> *.djv compression? 20:46 -!- wrldpc [n=ben@pool-71-174-89-247.bstnma.fios.verizon.net] has quit [] 20:47 -!- nsh [n=nsh@87-94-146-186.tampere.customers.dnainternet.fi] has quit [Connection timed out] 20:47 < kanzure> hm, djvu is just a pdf equiv IIRC 20:49 < fenn> djv just has a bunch of little pictures for each letter instead of letters 20:50 -!- krebs_ [n=krebs@208.106.116.196] has joined #hplusroadmap 20:50 < Biopunk> from what i've seen it gives a really good compression for documents like charts 20:51 -!- krebs [n=krebs@208.106.116.196] has quit [Read error: 104 (Connection reset by peer)] 21:15 < kanzure> so again, re: jumping around to what you were working on a few moments ago, seeing the whole view of 'working mem' 21:15 < kanzure> that's a general problem when making anything and zooming etc. 21:15 < kanzure> when you make a giant complex system, how do you identify the components that contribute to 'emergent' behavior anyway ? 21:15 < kanzure> I mean, when you 'make' something, just dropping a file in a folder doesn't do much 21:15 < kanzure> except allow you to forget about it except in as much as you consider the whole dir 21:16 < kanzure> as one single 'dir-file' 21:16 < fenn> yeah the whole 7 working items thing 21:20 < fenn> dmess (an air-force manufacturing contractor) claims the v-22 osprey has cost 200 billion in development - that's 10 years of NASA budget! 21:22 < kanzure> and supposedly the X43A cost a few billion too 21:23 < kanzure> why'd it cost that much to develop 21:23 < kanzure> I want to see a budget report 21:24 < kanzure> of the v-22 21:25 < fenn> dunno, wiki says $20 billion but he claims it's actually $200B 21:26 < fenn> political pork gone mad 21:30 < kanzure> I don't understand how it's $200B 21:30 < fenn> me either 21:31 < fenn> another open hardware machine tool project - watch this space: http://translate.google.com/translate?u=http%3A%2F%2Fwww1.eafit.edu.co%2Fwiki%2Findex.php%2FHerramientas_de_Manufactura_Dise%C3%B1o_Abierto&sl=es&tl=en&hl=en&ie=UTF-8 21:31 < fenn> we should contact mfg.com labs 21:31 < kanzure> ' Joint Project with MFG.com' 21:31 < kanzure> yes 21:32 < kanzure> http://64.233.179.104/translate_c?hl=en&sl=es&tl=en&u=http://www.mfgquote.com/about-mfg-management-team.cfm 21:33 < fenn> i dunno about their management but some "researchers" have posted to the emc list about open hardware, and are quite enthusiastic 21:35 < fenn> oh it was jorge barrera 21:35 < kanzure> good 21:35 < kanzure> I was going through ~50 tabs today on these topics 21:35 < kanzure> had mfg.com or whatever there 21:35 < kanzure> didn't get to it 21:35 < kanzure> opera kept on crashing 21:36 < fenn> does it preserve your tabs like firefox/konqueror do? 21:36 < kanzure> yes 21:36 < kanzure> opens/renders faster than both too 21:37 < kanzure> even with 400 tabs open, once you convince it to actually open a page it does it pretty quickly 21:37 < kanzure> but there are ... rules. 21:37 < kanzure> for example, at 200+ tabs, if you just opened up a page in the background, opening up another one would mean right-click and *immediately* go to open in background tab, if you don't do this then it will freeze with your mouse there selecting the option 21:38 < kanzure> or you can wait until the other tab is fully loaded 21:38 < kanzure> that sort of thing 21:38 < kanzure> it's kind of like teetering on an edge 21:38 < kanzure> I feel like that a lot with tech and gamehacking sometimes 21:38 < kanzure> it's retarded and shouldn't happen 21:38 < fenn> i think you're pushing it beyond the design point 21:38 < kanzure> oh, definitely 21:38 < kanzure> I've heard some Opera developers gasp in horror 21:39 < kanzure> I had to send in my old bookmarks file for their review 21:39 < kanzure> because Opera fscked up a write 21:39 < kanzure> I was hoping it was just a bitshift or something 21:39 < kanzure> but they never did get back to me ... 21:39 < Vedestin> why did you have 200 tabs open? 21:40 < fenn> dunno if you're interested, but apparently kazehakase is hot shit 21:41 < kanzure> but it uses gecko 21:41 < kanzure> aka it's firefox 21:41 < fenn> says you can switch to dillo/w3m 21:42 < Vedestin> do you use mouse gestures? 21:42 < Vedestin> i find them irritating 21:42 < kanzure> http://en.wikipedia.org/wiki/Comparison_of_layout_engines 21:42 < kanzure> I don't use gestures 21:43 < Vedestin> good 21:43 < Vedestin> gesture people are weird 21:43 < fenn> also GTKhtml and webkit, whatever those are 21:44 < Vedestin> i might go get a hard drive 21:44 < fenn> oh wait, those are planned, nevermind 21:44 < kanzure> hm? 21:44 < kanzure> planned? 21:44 < kanzure> ah, planned implementations 21:44 < kanzure> nevemrind 21:46 < kanzure> I've tried playing with gtkhtml and webkit before 21:46 < kanzure> I don't remember what my results were 21:47 < kanzure> webkit > khtml, seems to be the consensus 21:47 < kanzure> http://en.wikipedia.org/wiki/GtkHTML <-- a fork of khtml 21:49 < kanzure> http://arstechnica.com/journals/linux.ars/2007/07/23/the-unforking-of-kdes-khtml-and-webkit 21:51 < kanzure> I dunno, I always sort of dislike the differences between renderings between khtml v. pandora (Opera's engine) for example 21:51 < kanzure> it's just hard to get used to, it's almost like khtml is ignoring some stuff 21:51 < kanzure> apple webkit always looks like it's never readable at all 21:53 -!- kramer3d [n=kramer3d@ip98-169-184-165.dc.dc.cox.net] has joined #hplusroadmap 21:54 -!- krebs_ [n=krebs@208.106.116.196] has quit [Read error: 110 (Connection timed out)] 23:04 < kanzure> fenn: I got it :) 23:04 < kanzure> how about a tab list much like ALT+TAB 23:04 < kanzure> it'd be a hack to konqueror, and it'd just access a list from a flatfile somewhere or something 23:04 < kanzure> and then everything else is controlled by the underlying perl/shell script on a terminal shell 23:04 < fenn> doesnt seem practical for 200+ tabs 23:06 * fenn still stuck in revolutionary web-interface mode 23:07 < fenn> you need a ... tab manager! 23:07 < kanzure> hrm 23:07 < fenn> like a window manager, but instead of using the perfectly good window managers out there, we stick all the windows in the same window! 23:07 < fenn> snark 23:08 < fenn> i mean, um, narf 23:08 < kanzure> fenn: have you seen screenshots of my, uh, screen? 23:08 < kanzure> What are we going to do tonight, Brain? 23:08 < fenn> what are we going to do tonight, Bryan? 23:08 < kanzure> The same thing we do every night, Ben. 23:08 < kanzure> Try to take over the world. :) 23:09 < kanzure> The pinky and the brain, the pinky and the brain, one is a genius, the other is insane, 23:09 < kanzure> hehe :) 23:09 * fenn is still trying to figure out which one he is 23:09 < kanzure> batfucking insane 23:09 < kanzure> maybe not 23:09 < kanzure> http://heybryan.org/shots/2008-05-24_2.png 23:10 < kanzure> so, that's with only, what, 33 tabs displayed there 23:10 < kanzure> Opera does better density *and* a better scrollbar 23:10 < kanzure> this sucks :( 23:10 < fenn> i dont see any tabs? or do you mean tabs for the irc? 23:11 < kanzure> uh, windows to the left ? 23:11 < fenn> or do you mean the kde window manager? 23:11 < kanzure> you mentioned window managers 23:11 < kanzure> yes 23:11 < fenn> yeah, agreed kde's wm stinks 23:11 < fenn> i mean, it's ok, and much better than windows, but not "high performance" 23:12 < kanzure> personally windows has always been faster, but the hackability sucks immensely (zero) 23:12 < kanzure> just switching between two windows with alt+tab on kde sucks 23:13 < fenn> i guess i use the favico's because thats what happens in icewm when you've opened lots of windows 23:13 < fenn> then mouse-over or alt-tab will show the full title 23:14 < fenn> but the killer missing feature from kde is you can't ctl-alt-# to go to a specific desktop 23:15 < kanzure> this sucks. 23:17 < kanzure> so... 23:18 < fenn> heh you could try twm 23:19 < fenn> or a traditional unix-style 'collapse to icon' paradigm 23:19 < fenn> s/a/any/ 23:19 < fenn> the icons stack up on your desktop like tetris blocks 23:19 < kanzure> I need text 23:19 < kanzure> those might as well be .lnk files 23:19 < fenn> sure you can do text too, at least for some things 23:19 < kanzure> which is the whole idea of flat file bookmarking 23:20 < kanzure> so what happens when I have 12,000 bookmark files on my desktop 23:20 < kanzure> eh? 23:20 < fenn> no, not like that.. hmm how to explain the difference 23:20 < fenn> in say KDE when you save a page as .lnk it goes to a file, and the file is represented by an icon on the desktop 23:20 < fenn> in twm when you click 'minimize' on a window, it is represented as an icon (or text) on the desktop 23:21 < kanzure> right 23:21 < fenn> when you click on the minimized window it pops back up 23:21 < kanzure> I remember that. 23:21 < fenn> so these arent bookmarks, they're just windows that you're getting out of the way 23:21 < kanzure> it might as well be a bookmark to a program state 23:21 < kanzure> it's also a different implementation 23:21 < kanzure> than files being displayed on the desktop 23:21 < kanzure> I know that, but it might as well just do a write too at the same time, or load up a collection of bookmarks as tabs in that sense 23:22 < fenn> well, you dont want to bookmark every site you visit do you? 23:22 * kanzure ponders 23:22 < kanzure> if I don't want to bookmark it, I'll probably just close out of it anyway 23:22 < fenn> ponder the difference between history and bookmark 23:22 < kanzure> it should be stored in the history logs anyway 23:23 < fenn> yes one might hope that, but my experience is browsers delete your history after a certain period of time or they explode 23:23 < kanzure> so what 23:23 < kanzure> we can write an encapsulating script 23:24 < kanzure> for example, I've had 1000k webhits, it's all in my apache logs 23:24 < kanzure> that's +200 MB of logs 23:24 < kanzure> I've barely browsed that many websites methinks. 23:24 < kanzure> so it's not much to deal with. 23:24 < kanzure> as long as the logs are, uh, separated btw, none of this "all in one file" BS 23:24 < fenn> you could do logrotate 23:25 < fenn> i forget what the problem is now 23:25 < fenn> opera too slow? 23:27 < fenn> http://fluxbox.sourceforge.net/docbook/en/html/chap-tabs.html 23:32 < kanzure> opera whines like a beast 23:32 < kanzure> and then crashes when you start to do more tabs 23:32 < kanzure> also, it can't bookmark anymore, unless I do it manually 23:37 < Vedestin> do you like music kanzure? 23:38 < kanzure> Vedestin: sure 23:38 < kanzure> http://heybryan.org/music2/ 23:39 < kanzure> fenn: can we save a tabbing session and convert it into bookmarks 23:39 < kanzure> if this isn't scriptably viable then it's just as useless 23:40 < fenn> 'save a tabbing session' is basically browser history 23:40 < kanzure> no, in fluxbox 23:40 < kanzure> you have your working memory, 23:41 < kanzure> which is different from history 23:41 < kanzure> history is everything 23:41 < kanzure> working memory is task-oriented 23:41 < kanzure> (or can be grouped off into task-orientation) 23:41 < kanzure> uh 23:41 < kanzure> separate "no, in fluxbox" from the next message after that 23:43 < fenn> if you're using konqueror you can just go through all open konq processes and do some dcop call to get the url 23:43 < kanzure> maybe the idea of working memory and random browsing sucks 23:43 < Vedestin> weird mix kanzure 23:43 < kanzure> I doubt it though 23:43 < kanzure> Vedestin: yeah, it's true. 23:43 < kanzure> fenn: hm, okay 23:44 < fenn> that's window manager independent 23:44 < kanzure> also, I need a way to spawn a program call when closing konqueror 23:44 < kanzure> a 'bookmarking hook' 23:44 < fenn> hmm i'm not that kde-savvy 23:44 < kanzure> the logger is just some proxy 23:44 < kanzure> oh, I guess I could recompile konqueror 23:44 < fenn> i mean there's probably already some way to do it 23:47 < fenn> i dont think doing anything on program call is the right way to do it. instead you'd have some journalling system so things are preserved after a crash 23:47 < fenn> duh. s/program call/program close/ 23:48 < kanzure> mmhmm 23:48 < kanzure> yes 23:48 < kanzure> but at the same time there needs to be hooks for opening up new links easily 23:48 < kanzure> so opening a link should be done via calling an external script 23:48 < kanzure> surely *that* can be done 23:59 < kanzure> maybe I'll just split up konqueror into some shell scripts and perl scripts and see how it goes from there