00:12:53 tansaku_xr (~sam@n144-012.tokyu-net.catv.ne.jp) has joined #swhack 00:30:30 tansaku has quit (Read error: 110 (Connection timed out)) 00:34:07 monkinetic has quit ("cya") 00:53:20 davb (dave@alb-24-58-162-46.nycap.rr.com) has joined #swhack 01:14:05 Ash has quit (Read error: 104 (Connection reset by peer)) 01:17:15 * AaronSw listens to airline hold music 01:23:00 classical 02:05:59 davb has left #swhack 02:30:13 tansaku_xr has quit (Read error: 110 (Connection timed out)) 02:55:29 jeremiah has quit ("ChatZilla 0.8.5 [Mozilla rv:0.9.8/20020204]") 03:03:53 New Chimera's got an awesome logo. 03:14:12 mmm, anti-aliased web browsing 03:14:56 tansaku_xr (~sam@mtl10gw.mtl.t.u-tokyo.ac.jp) has joined #swhack 03:40:24 GabeW (~Gabe@12-236-237-100.client.attbi.com) has joined #swhack 03:48:22 This is a WALLOPS message for the Open Projects Network. These messages contain non-critical comments and announcements and detailed server admin information. To turn them off, turn off user mode 'w'. On most clients: '/mode -w'. Thanks. 03:49:30 Hints: If you absolutely need to pull a user off your channel, consider using /remove. Internal format is the same as KICK. It's a lot less disruptive though; it provides a PART with a fairly relaxed message. 03:54:10 cool nick: angelofentropy 04:25:18 redmonk (~steve@ip68-2-102-26.ph.ph.cox.net) has joined #swhack 04:25:46 we should run a turing test (type 1) sometime. 04:25:51 it'd be interesting to watch 04:28:43 hi all 04:28:56 of course type 2 might be more interesting, but it's probably really obvious right now 04:28:57 hey redmonk 04:31:29 s'appenin 04:32:51 think we should run a turing test? 04:35:34 Pat Hayes seems to be against it 04:36:20 Hm, looks like all the turing tests have involved machines... 04:40:23 "Studies show that 99%25 of people spell Hmm with 2 M's." 04:40:35 says A.L.I.C.E. when I say "Hm." 04:41:33 ha! it does the HAL speech 04:44:13 Turing Test one-liners, #4642 : Should David Beckham be captain of England football team? 04:44:30 number 3619 : Seriously, folks, how would you feel if you knew the true story that I was being held captive in an insame asylum in Pleasanton, Texas, and Governor George W Bush Junior and his wife the librarian com 04:45:05 talli has left #swhack 04:46:58 heh! 'Ms. Cynthia Clay, the Shakespeare aficionado, was thrice misclassified as a computer. At least one of the judges made her classification on the premise that ``[no] human would have that amount of knowledge about Shakespeare.''' 05:56:09 GabeW has quit ("Client Exiting") 07:38:20 redmonk has quit ("cya") 10:05:16 heh, heh 10:05:56 * sbp just wrote another song 10:13:22 aaaaaargh! doctype.org has gone! 10:13:33 crap! how did that happen? 10:14:48 s/doctype/doctypes/ 10:15:34 that's horrific. it was still up in January 10:16:26 aw, man... 10:17:21 Noooooooo! "0 pages found for http://www.doctypes.org/meta/NOTE-xhtml-augmeta.html" 10:21:21 hooray:- 10:21:22 [[[ 10:21:22 the HTML WG will not recommend to use 10:21:22 'text/html' for this kind of documents. 10:21:32 ]]] - Masayasu Ishikawa, on www-tag 10:22:30 - http://lists.w3.org/Archives/Public/www-tag/2002Apr/0087 10:26:39 * sbp searches for a local copy of augmeta 10:26:58 ooh, ooh, Aaron's HTTP archive! 10:27:32 oh, the Google cache has it. Phew 10:41:24 aw, man:- 10:41:25 [[[ 10:41:25 ----- The following addresses had permanent fatal errors ----- 10:41:26 10:41:26 ]]] 10:51:33 on his homepage: "Please note that the doctypes.org web site has been taken offline, though something always seems to be in the works." 11:54:52 tansaku_xr has quit (Read error: 110 (Connection timed out)) 12:02:51 .time 12:02:51 2002/04/09 12:02:51.6256 Universal 12:04:30 .time 12:04:30 2002/04/09 12:04:30.6221 Universal 12:04:50 pff, 1:39? is that it? 13:10:46 tomch (~lambda@modem-1934.panther.dialup.pol.co.uk) has joined #swhack 13:12:31 davb (~dave@rrcs-nys-24-97-22-204.biz.rr.com) has joined #swhack 13:16:10 Hey tomch, davb 13:17:31 * tomch waves to everyone and sbp 13:20:25 hiya 14:12:55 redmonk (~steve@63.149.73.20) has joined #swhack 14:27:01 tansaku_xr (~sam@n144-001.tokyu-net.catv.ne.jp) has joined #swhack 15:01:05 howdy sbp 15:02:34 hey aaronsw 15:02:45 hey redmonk 15:06:10 for the mathemetician in you, check this out: http://www.math.fau.edu/locke/graphthe.htm 15:06:17 graph theory 15:06:48 never studies it myself, but it's an interesting read, for a paragraph or so. then the formulae put me to sleep ;-) 15:06:54 er, (never studied) 15:06:58 heh 15:07:18 I've done a bit, it's not the most exciting thing in the world 15:07:54 no, but interesting from a software standpoint 15:07:56 to me at least... 15:08:06 i've been drawing state diagrams in OG lately 15:08:10 ah, cool 15:08:13 got me thinking about it 15:08:29 it's the olny way to keep my head around some of the business logic in our app 15:08:57 (this stuff is largely contained in an oracle database; we maintain a framework that encapsulates the business logic) 15:09:06 (ugly stuff) 15:09:30 hm. trees and graphs in oracle get messy, as i recall 15:09:51 well, the graph is only in my head 15:09:58 ah 15:10:30 for example we have a table that defines certain states that an entity can be in from month to month 15:10:48 in order to calculate the current balance for an entity, i have to account for each possible state 15:11:14 sometimes combining states or differentiating between them 15:11:22 hence the state diagram 15:11:38 just to help me make sure i've covered all the bases 15:12:40 nice 15:12:54 at my last job, one of our contractors wrote a java framework that let you write business logic as a state diagram. you would use a graphical tool to diagram it, then define the messages on your object that would move it from state to state. the framework kept track of what messages were valid ffor what state 15:13:06 oh, neat 15:13:15 yeah. pity we never used it in a project 15:13:24 we had a few it would have been great in, too 15:13:27 bad management 15:13:29 :-( 15:13:36 :-( 15:13:41 it was all XML based (the diagrams and stuff) 15:14:04 so they would also render them to html via xsl for documentation purposes 15:15:06 neat 15:15:07 neat 15:15:08 :) 15:15:26 heh 15:19:19 that sounds like acs-workflow, or at least what it was planned to be. 15:21:28 tomch has quit (Read error: 110 (Connection timed out)) 15:22:29 yeah, a while back i remember someone working on an xml-based workflow shema 15:22:32 schema 15:22:38 not sure what happened to it 15:29:09 tansaku_xj (~sam@h134-078.tokyu-net.catv.ne.jp) has joined #swhack 15:30:46 oh, it also had versioning built into the state diagrams. if you updated a schema, objects using one version would use it until they were done, and new objects would use the updated vesion. or you couldd msg all objects being managed to use the new schema 15:31:00 *sigh* 15:31:18 These objects sound pretty smart. 15:31:23 now it's trapped in a company that doesn't know what they've got 15:31:35 argh 15:32:20 well, the objects themselves were pretty dumb, but the system that managed your objects and the states was pretty smart 15:32:26 that was the nice bit - 15:32:54 your "business objects" only had to know how to do the proper calculations 15:33:38 the OSM (object state manager) would send msgs to your objects, and based on the results would set the object state in the diagram accordingly) 15:34:03 (state in this case referring to state in a diagram, not 'persistence') 15:34:36 if you tried to tell an object to do something that was not appropriate for it's current state, the OSM would yell at you. 15:34:54 Heh, cool. 15:35:11 Hm, 10:34... I really should get some work done today... 15:35:22 hehe 15:35:24 have fun 15:35:33 heh: *** tav|away has quit IRC (Killed (zelazny.openprojects.net (rowling.openprojects.net(tav) <- forward.openprojects.net))) 15:35:45 c'ya 15:36:48 Gotta run 15:45:54 tansaku_xr has quit (Read error: 110 (Connection timed out)) 16:12:17 @ http://www.domainmonger.com/ver_warn.shtml 16:12:54 A: http://www.domainmonger.com/ver_warn.shtml from AaronSw 16:13:11 A:|VeriSign Tries to Steal Domains 16:13:17 A::I got one of these letters, pretty nasty. 16:13:33 titled item A 16:13:55 commented item A 16:53:07 Morbus (~morbus@morbus.totalnetnh.net) has joined #swhack 16:53:28 Morbus has quit (Remote closed the connection) 16:56:49 anyone ever looked at this: http://www.opencyc.org/ ? 16:59:41 a little 17:00:00 davb, what's the syntax for foreign keys in pgsql? 17:00:11 column type references table? 17:00:51 ah, guess so 17:02:30 yeah :) 17:04:06 tomch (~lambda@modem-811.orangutan.dialup.pol.co.uk) has joined #swhack 17:05:27 * tomch wavelets 17:14:28 'lo tomch 17:35:46 :-) 17:36:48 * sbp wonders if tom is a wave or a particle 17:37:05 .google "46 72 69 6e 6b 20 72 75 6c 65 73 21" 17:37:06 "46 72 69 6e 6b 20 72 75 6c 65 73 21": http://www.eeggs.com/items/17723.html 17:37:32 a wavelet is a system used for categorizing images 17:38:03 I'm off for food anyway :) 17:39:08 are there any decent CGI modules for python? 17:39:11 the default one sucks... 17:39:29 i'm spoiled from the acs, i guess 17:58:02 heh 18:18:09 hazmat has quit (Connection timed out) 19:03:38 tomch has quit (Read error: 104 (Connection reset by peer)) 19:04:42 davb has quit () 19:08:13 davb (~dave@rrcs-nys-24-97-22-204.biz.rr.com) has joined #swhack 19:15:39 what do you guys think of collecting web site and page rankings from librarians? 19:17:04 what for? 19:17:10 sort of like the virtual www library? 19:17:53 kind of to augment page rankings such as google. 19:18:26 GabeW (~Gabe@12-236-237-100.client.attbi.com) has joined #swhack 19:18:39 got the idea from here: http://www.theshiftedlibrarian.com/2002/04/08.html#a1228 19:22:45 "we'll never trust pagerank enough to use it to verify authenticity." why? PageRank _is_ a trust metric! 19:23:17 set by magic definied by google's backend software. 19:23:43 I guess librarians would trust ratings by their fellow librarians, but then you only see what other people like you like. 19:23:47 it's hardly magic. it's been written up and published in scholarly journals, and studied by some of the best in the industry 19:23:47 which is kinda boring. 19:23:55 really? 19:24:03 ok. 19:24:07 * davb is wrong 19:24:14 heh 19:24:45 but anyway, would it be useful to have more than one way to rank page authenticity/usefulness? 19:24:53 sure 19:25:08 i think here that librarians are using a compleetely different trust metric 19:25:20 that's why they would not trust googles 19:25:20 yeah, that's probably true. 19:25:44 I just thought it would be interesting to capture the data. 19:25:46 they want to make sure that you get the RIGHT data... google can only tell you what everyone else THINSK is authoritative 19:25:52 actually it would be cool to compare it. 19:26:10 which is why dave w is a high hit on john doerr, just 'cuase he talks about him a lot 19:26:14 in the pagerank paper, they discuss running the pagerank algorithms using peoples homepages as a seed 19:26:18 it's pretty interesting 19:26:25 neat. 19:26:31 gotta run 19:26:34 bye 19:26:47 cya 19:26:48 comparing librarians to pagerank would have been a neat project for the google programming contest. 19:27:07 librarians have training in categorization and such 19:27:15 which google cannot take into account 19:27:21 necessarily 19:27:35 right. 19:27:59 pagerank could be abstracted i suppose and then run on a set of librarian-supplied data 19:28:13 librank ;-) 19:28:44 hey, this is what swhack is all about right? interesting projects that distract us from the "real job" 19:28:55 of course 20:10:28 Morbus (~morbus@morbus.totalnetnh.net) has joined #swhack 20:18:10 Morbus has quit (Read error: 104 (Connection reset by peer)) 20:19:55 Morbus (~morbus@morbus.totalnetnh.net) has joined #swhack 20:22:36 Iffy 20:22:42 hey there. 20:22:45 got a new computer at work. 20:22:49 gonna be in and out all day 20:22:50 stats? 20:23:11 i dunno. 800mhz, p3, 196ram, 16meg video, sound card (finally), win2k 20:23:33 cool (roughly the same as this box) 20:23:47 my old machine was a 233 :) 20:23:54 heh, yeah - I remember 20:23:59 dunno how you'd put up with it 20:24:24 easily, actually. 20:24:26 reboot. 20:24:31 Morbus has quit (Client Quit) 20:24:44 CygBot (~sbp@m207-mp1-cvx4c.pop.ntl.com) has joined #swhack 20:25:13 $ python -c "import md5; print md5.new('b11242799d8e4de1052e3c7e3cb037ac').hexdigest()" 20:25:21 > b11242bc1ac79a78620720e3539388c9 20:25:21 > [end] 20:25:42 the importance of checking more than the first few characters of a hash... 20:25:48 CygBot has quit (Read error: 104 (Connection reset by peer)) 20:26:36 alright, even that only happens about once in 10 million hashes, but still 20:28:01 Morbus (~morbus@morbus.totalnetnh.net) has joined #swhack 20:29:52 davb has quit () 20:41:24 High water mark for tonight is 6,665 clients. Say, that's kind of neat. :) 20:41:48 high tide? 20:48:08 Morbus has quit ("http://www.disobey.com/") 20:56:30 Morbus (~morbus@morbus.totalnetnh.net) has joined #swhack 21:01:57 This is a WALLOPS message for the Open Projects Network. These messages contain non-critical comments and announcements and detailed server admin information. To turn them off, turn off user mode 'w'. On most clients: '/mode -w'. Thanks. 21:02:19 Hint: If you see a problem with the server code, please email as much detail as possible to bugs@openprojects.net . Thanks! 21:07:44 Morbus has quit (Read error: 104 (Connection reset by peer)) 21:10:05 Morbus (~morbus@morbus.totalnetnh.net) has joined #swhack 21:19:49 GabeW has quit ("Client Exiting") 21:31:32 monokrom (mekanik100@208-58-240-202.s456.tnt1.atnnj.pa.dialup.rcn.com) has joined #swhack 21:31:40 Morbus has quit ("http://www.disobey.com/") 21:38:05 Morbus (~morbus@morbus.totalnetnh.net) has joined #swhack 21:49:15 Morbus has quit (Read error: 104 (Connection reset by peer)) 21:49:19 redmonk has quit ("cya") 21:49:24 Morbus (~morbus@morbus.totalnetnh.net) has joined #swhack 21:49:34 Morbus has quit (Remote closed the connection) 21:51:28 MorbusIff (~morbus@morbus.totalnetnh.net) has joined #swhack 21:57:53 MorbusIff has quit (Read error: 104 (Connection reset by peer)) 22:26:32 monokrom has quit ("It ain't easy bein greasy") 22:33:04 heh, that's quite awesome, really. 22:33:24 what is? 22:33:31 re: b11242799d8e4de1052e3c7e3cb037ac and b11242bc1ac79a78620720e3539388c9 22:33:40 aha :-) 22:38:05 we were wondering if there was a document which included its hash 22:38:16 ah! I was wondering the same thing 22:38:17 or rather, could you make one that did 22:38:22 in fact, that's what I was trying to do! 22:38:29 and we figured out how to 22:38:31 it's practically impossible 22:38:38 I worked out that it'd take me 5 billion years 22:38:43 figured out how to: easy 22:38:46 (this was after dinner at p2pcon2, as we walked back to the hotel) 22:38:52 yeah, it's pretty easy 22:39:01 you just list all the possible hashes, and then hash that. 22:39:04 heh. weird. People all over the world must have had the same idea :-) 22:39:20 i was thinking about this yesterday for some reason... or was it this morning 22:39:22 Hmm... actually, I wasn't thinking about that 22:39:23 dirk gently and all... 22:39:33 listing all of the hashes and then hasing it is cheating 22:39:37 heh. 22:39:42 you want to be able to control the content of the document 22:39:56 I just grabed the latest Dirk book from the library, been reading it on the way downtown and back. 22:40:25 so I was thinking that you have your string x, and your intial hash y. then you hash x+y. if the hash of x+y == y, then you've got a result, otherwise you take the new hash and use that for the value of y 22:40:33 repeat until you find a result 22:40:45 er... 5 million years, not 5 billion 22:40:55 it's pretty easy to generate a billion hashes that way 22:41:05 What if none of them work? 22:41:09 in fact, I've already sifted through over a third of a billion tonight 22:41:20 if none of the work, so what? you just carry on 22:41:29 I also came up with something to make sure that they don't loop, too 22:41:36 because that'd be a disaster 22:41:37 But there are a fininte number of potential y values. 22:41:47 And it's highly unlikely that any of them will work. 22:41:52 yeah? 22:42:10 So your method sucks. A birthday-style attack is much better. 22:42:11 got a better way of doing it? you may as well just iterate through the list of hashes 22:42:33 how would you use the birthday attack on this? 22:42:45 IOW, pick a document x and a hash y. then keep on adding spaces to the end of them until the hash of x+y+spaces = y 22:43:00 i forget the name of this process, but it's in AC 22:43:07 ugh, but then you'll end up with a document that's GBs long! 22:43:16 you don't even need to add spaces onto the end, you can add spaces to the end of each line 22:43:16 which is quite besides the point 22:43:22 that'll give you 2^n possibilities 22:43:29 so you only need a document of 512 lines long 22:43:38 and you can halve that pretty easily 22:43:49 ugh. whitespace 22:44:02 or get some funky unicode character that doesn't appear. i don't care 22:44:18 heh 22:44:28 anyway, if you can do this you can break digital signatures 22:44:37 since you can just make people sign whatever you want 22:44:55 Hmm... I really don't see how this is any faster 22:44:56 since you can pick the hash 22:45:08 It terminates! That's how it's faster! 22:45:16 Yours has no guarantee of success, mine does. 22:48:00 ah, right. I can make mine terminate if I go through the list of hash values starting from 00000000... 22:48:31 so your method is still going to take a million years, and you'll end up with gigabytes of whitespace 22:48:49 huh? yours still won't terminante 22:49:01 why not? is has to have a hash 22:49:09 you won't end up with gigabytes of whitespace if your messages has more than 2^512 lines 22:49:25 you'll have a max of one on each line, and that's a very rare case 22:49:31 and even so it's 512 chars 22:49:41 Let me see if I understand your method: 22:50:05 Take a document x and a hash y (initially 000000...) hash(x+y) 22:50:12 increment y until hash(x+y) = y, right? 22:50:30 that's not going to work, since y changes each time 22:50:36 no. take a document x, with a hash y in it, and guess the hash 22:50:46 guess the hash? 22:51:04 i.e. for z in range(ffffffff...): if y = z: break 22:51:15 s/=/==/ 22:51:24 what's y? 22:51:39 argh, hang on, that won't work 22:51:42 * sbp smacks self 22:51:44 :-) 22:52:19 I still don't understand the whitespace approach, though 22:52:50 OK, so you have a document x that you want to hash to y. 22:53:22 and if you have 2^n (where n is the size of the hash) different documents, one of them must hash to y 22:53:26 with me so far? 22:53:30 uh huh 22:53:51 so whehther a line has a space on the end is one bit, 2 possibilities 22:54:12 pardon? 22:54:29 a line of text 22:54:33 is there a space at the end or not? 22:54:42 O.K., boolean, gotcha 22:54:43 humans don't notice these things, which is why it's good 22:54:58 oh! right, I've got it 22:54:59 ping 22:55:04 so by combinatorics, multiply the possibilities together: 2 *2 * ... 22:55:05 yeah there 22:55:12 heh, quite ingenious 22:55:30 heh, Dave's got access to the Google SOAP interface 22:56:14 Hmph, the amount of Daves that I know seems to rise every day. it's getting so confusing 22:56:45 ah, that must be why Trigger always calls Rodney "Dave" 22:56:58 Rodney? 22:57:01 it's just easier 22:57:09 .google Trigger Rodney Dave 22:57:09 Trigger Rodney Dave: http://www.pubnetwork.co.uk/TVpubs/NagsHead/NagsHead.htm 22:57:33 We heard the people of Brazil complaining about the Simpsons on the Radio today. 22:57:44 cf. http://www.azcentral.com/news/articles/0408Simpsons-ON.html 22:58:03 The Simpsons are on the radio now? 22:58:20 ah, right, I heard about that episode 22:58:30 No, the people of Brazil were on the radio 23:01:13 xoot (xoot@177.sanjose-01-02rs16rt.ca.dial-access.att.net) has joined #swhack 23:01:17 xoot! 23:01:21 hello 23:01:41 I guess someone is alive here :D 23:01:41 what a cool nickname. I'm going to be saying that for days now 23:01:48 yeah 23:01:54 I post at macosx.com 23:01:59 same nick 23:02:00 Uh oh 23:02:01 pronounced zoot, soot, or scoot? 23:02:29 ksoot 23:03:07 GabeW (~Gabe@12-236-237-100.client.attbi.com) has joined #swhack 23:03:11 yo 23:03:58 so, what's the adjectival form of your nickname? xootic? xootal? xooty? 23:04:12 .google xootic 23:04:12 xootic: http://www.win.tue.nl/xootic/symposium2001 23:04:28 xooty 23:04:37 and I am an insane poster at macosx.com 23:04:44 not to advertise the site :) 23:04:49 it's not mine ;) 23:04:58 heh, right. I noticed 23:05:39 my site's at: http://osx.blogspot.com/ 23:06:17 aha: "Once I found out about Mac OS X, I replaced the r in root with x. The result: xoot." - http://www.macosx.com/forums/showthread.php?s=d4c63285f808af0553362093a2a5af0c&threadid=15842&perpage=15&pagenumber=2 23:06:25 yeah 23:06:30 hazmat (~ender@adsl-66-123-57-58.dsl.lsan03.pacbell.net) has joined #swhack 23:06:46 man, as if I didn't already have enough of OS X people with Morbus and Aaron 23:06:59 hehehe 23:07:03 I don't know if I can take a third person saying "isn't OS X so great!!!" without throwing up 23:07:12 isn't OS X so great!!! 23:07:13 and deus_x, and rillian... 23:07:18 :D 23:07:21 * sbp feels the vomit rising 23:08:05 What version of #swhack are *you* running? :-) 23:08:07 i normally chat at irc.press3.com, but I decided to visit here for a while 23:08:14 swhack? 23:08:17 swhack! 23:08:21 huh? 23:08:24 :D 23:08:47 if you didn't notice, that's the channel name, and Aaron just VERSION'd us all. It looks a bit funny in my client 23:08:57 [AaronSw:#swhack VERSION] 23:09:16