00:01:01 apparently not listening. 00:06:51 heh 00:06:58 Too much partying for rbm 00:07:03 and paje is still not around 00:29:09 rbm should share the love. I certainly could have used some partying this weekend. 00:37:35 me too 03:23:35 anyone home? 05:22:01 denshi has quit (bear.openprojects.net irc.openprojects.net) 10:00:39 davb has quit ("Client Exiting") 10:59:39 hazAway is now known as hazmat 11:01:16 hazmat has left #openacs 11:32:48 nsadmin has quit (Remote closed the connection) 11:33:27 jim has quit (Remote closed the connection) 11:37:31 davb (~chatzilla@rrcs-nys-24-97-22-204.biz.rr.com) has joined #openacs 11:42:02 hi 11:44:21 hey 11:45:10 read this: http://www.geekissues.org/quotes/?top 11:55:06 hey guys 11:59:09 Psychephylax: that is wacky 11:59:21 at least we know some people have even less to do than us. 11:59:59 hehe 11:59:59 no sign of paje or rbm 12:13:23 bah 12:13:27 Jury duty today 12:23:28 * Psychephylax kicks dave 12:27:46 what?? 12:28:02 have fun at jury duty 12:28:20 lol 12:28:36 look at this weird picture http://www.pbase.com/image/2823883/original 12:28:51 Look in the center for what I'm talking about (besides the white dot) 12:29:24 ok. 12:29:42 weird 12:30:45 heh 12:30:53 someone just posted the reason 12:30:57 I forgot about it too 12:31:18 hrm 12:32:02 499 messages in a month 12:32:08 i'm not that active in forums usually 12:32:24 really? what is the reason? 12:32:45 i like the forum? 12:33:35 someone just posted the reason 12:33:52 oh 12:34:04 the viewfinder is supposed to be closed so the light doesn't leak in 12:34:08 damn SLR cameras :0 12:36:00 really? 12:36:19 on a real SLR, the mirror flips up, and closes off the viewfinder. 12:36:33 markd2 (~Snak@h166-102-041-088.ip.alltel.net) has joined #openacs 12:36:42 hi markd2! 12:36:57 yo davb 12:37:16 the weekend's over. now I can relax :-) 12:37:33 LOL 12:37:37 * Psychephylax thwaps markd2 12:37:45 yay! a thwapping! 12:37:47 © 12:38:03 denshi (~chatzilla@cs6625161-163.austin.rr.com) has joined #openacs 12:38:04 but of course 12:38:04 paje (~paje@slxwy.dorms.usu.edu) has joined #openacs 12:38:04 oom 12:38:10 paje! 12:38:11 wow! 12:38:15 hi everyone 12:38:29 hi all 12:38:30 paje, status 12:38:31 Since Mon Jul 1 07:37:32 2002, there have been 0 modifications and 0 questions. I have been awake for 28 seconds this session, and currently reference 3199 factoids. Addressing is in optional mode. 12:38:36 PAJE!!!!!!!!! 12:38:36 i am acannibal 12:38:40 * Psychephylax hugs paje 12:38:46 oh good, rbm found paje's brain 12:38:51 gitdown! 12:38:51 gitfunky 12:38:51 mark, can you fix the acannibal to a cannibal? 12:38:57 paje, botsmack! 12:38:57 * paje slumps into a purple haze 12:39:24 Roberto!!!! 12:39:24 roberto is probably splurging on a digicam :) 12:39:33 paje, forget roberto 12:39:33 Psychephylax: I forgot roberto 12:39:36 markd2: do you know how to move an infobot to a different server without it losing its factoids? I tried moving paje to the openacs.org box so it wouldn't come down so often, but it forgot everything when I did. 12:39:59 Psychephylax!! 12:39:59 it has been said that psychephylax is is the king of newegg and the oracle of digital photography and consumer electronics 12:40:05 heh 12:40:21 It probably needs the permissions transferred with it 12:40:31 rbm: I've not done it. You might try leaving a memoserv memo for "Gerr". He's migrated the wplug infobot I think 12:40:36 rbm, party hardy that Brazil won? 12:42:09 heh...Fleet bank is after me with the student loan I have to pay 12:42:13 bastards 12:44:07 fascists 12:44:51 Psychephylax: I missed all but 10 secunds of the final game :( 12:45:15 LOL 12:45:27 markd2: Did you have to do anything special to get it moved? I thought I'd just have to scp the files then change te config files. 12:45:53 I dunno. I haven't moved one before 12:46:00 I woulda thunk that would have worked 12:46:15 markd2: But apparently I need to do something else. I'm not sure where it associates the current hostname/IP (I presume) with the logs 12:46:16 did you tar everything before scp, so that symlinks are preserved? (dunno if infobot install has symlinks, but ya never know) 12:46:26 markd2: Oh, sorry. I misread you that you had done it before 12:46:59 Gerr (geremy) over in #wplug runs our infobot, and everything got moved about a month or two ago 12:47:03 so it can be done :-) 12:47:48 k. I'll leave him a memo 12:50:37 * Psychephylax goes to convict 12:50:46 K, memo left 12:52:00 cool 12:52:07 Psychephylax is in prison? 12:59:55 again? 13:04:28 rbm: how did you miss the game??? 13:16:16 talli: I went to wal-mart at 10:30 PM to buy stuff for the BBQ my wife is putting together for her area (she's a RA), and the wal-mart lady made us scan everything _3_ times because she kept screwing up with the tax exemption... 13:16:37 talli: ...so we got home at around 1 AM and went to bed at 2, and I couldn't wake up at 4 for the game 13:19:00 ouch 13:20:42 cro (~cro@defiant.nca.asu.edu) has joined #openacs 13:26:53 cro has quit (Read error: 104 (Connection reset by peer)) 13:33:22 rbm: that truly sucks. 13:33:26 shagster (~mkovach@web1.alal.com) has joined #openacs 13:33:32 were you able to party for the rest of the day? 13:33:41 shagster lacks pants 13:33:58 I have pants on, just no underwear :) 13:34:47 that's ok. it's not like you have anything to cover 13:35:04 Hey now :) 13:37:22 I think we should setup a 'Cloning Popular Sites with OpenACS' in the near future 13:37:24 jim (~jim@12.155.172.152) has joined #openacs 13:37:31 re... 13:38:04 ...er 13:38:25 ...er 13:39:23 anyone define plain sql functions in acs using pg? (just curious to know if it's prevelent) 13:51:57 i have never used a SQL function. 13:52:39 same here... been poking around the pg programmers guide, happened to see them in there 13:53:41 i am not sure if there is any advandate/disadvantage over pl/pgsql 13:54:22 I'm looking to index many tunes from jazz fake books, and I came up with a start that's very notes-like so far... 13:54:53 (unlike notes, the package drop script works every time :) 13:55:14 a working drop script? is that allowed? 13:55:21 jim: are you using the CR? 13:55:28 nonoNO :) 13:55:33 markd2, they work as long as you don't actually _use_ the package. 13:55:43 ah. ok. 13:56:20 well, I'm thinking about doing so, the data model will be going thru many changes before it's actually done 13:56:21 cro (~cro@defiant.nca.asu.edu) has joined #openacs 13:56:26 actually we fixed alot of them in the porting. 13:56:34 using the CR? 13:56:40 yeah 13:57:59 so, my present idea, is to have a table of tunes, with composer, genre and approx date of composition, and general-comments 13:58:09 on each tune 13:58:18 ok. 13:58:46 BTW, general comments needs to be fixed to use the CR properly :) 13:58:48 that part is started, ui is basically working, add/edit/drop of songs work 13:59:39 sure. did you see the formwizard.tcl in new-filestorage? feed it a table name, and it extracts he metadata from the database and makes forms and add/edit/delete functions. 13:59:59 although I haven't actually used it yet... 14:00:20 no, not yet... 14:00:52 but I'll definitely look; seems promising... I did see some wizardlike stuff in the template form demos 14:02:51 I guess there is work more in this direction. it would be nice to have it built into the toolkit. 14:03:10 once I have all those things, I want to be able to form sort criteria sorta on the fly, seems like tune-within-composer-within-genre can be as useful as tune-within-genre-within-composer 14:03:55 then, that forms the songs that can go into any number of lists 14:04:05 Ah. you want multiple sort fields? 14:04:32 yeah, thinking about it... 14:04:37 news has some good examples of sorting on one field. 14:04:48 there is some code built in to openacs to help. 14:04:55 but its in the wrong place :) 14:04:58 but I want to go in the direction of simplest-possible 14:05:37 jim: sounds like you want two things: limit by foo, sort by bar.... 14:05:47 or sort first, then limit results. 14:06:11 which if it isn't built in, probably should be. 14:06:44 well, yes. but more than that: I want lists of tunes, and the ability to form types of lists that add attributes 14:07:12 forexample 14:08:30 take three tunes, Blue Monk (Monk), Blue in Green (Miles) and Witchunt (Wayne Shorter) 14:08:58 these tunes would be in the base system I just described 14:09:46 and I want to let the user say, "I want to make my own list, containing those three tunes" 14:10:02 ah. kinda like amazon.com :) 14:12:19 Further, I want the user to be able to say "My list is based on pages of The Real Book, 5th edition, so I want to add the fields 'page number in that book', 'recording that the transcriber listened to to produce the chart' and 'key the song is transposed to in that chart'" 14:13:23 I want other users to be able to say, "Oh, cool! the real book! I have that book, so I'll dump these tunes into it with the page numbers, keys and recording references" 14:13:44 BTW, it looks like formwizard.tcl breaks on postgresql 7.2 14:13:59 maybe he has a half hour, so he only does the S-es. 14:15:08 Then, I want someone doing a gig to be able to say, "I want a list of tunes out of the real book, 5th ed for the sets I'm going to play." 14:15:51 now, switching to jamsession mode... 14:17:40 maybe there are groups of musicians forming... perhaps a bunch of people go to wander over to a particular jam session, seeing its listing on the jamsession site... 14:19:39 and on the site with it, are a single list of tunes taken from multiple lists... some from Real Book 5th ed, some from Real Book Vol II, some are originals by one of the session players, so he would supply the charts (hence, maybe they're not in any fake books) 14:20:42 so, as you can see, I want ordered and unordered lists which are heterogeneous 14:28:58 another part, the Real Books are widely known for their chord progression errors... so the same tune by the same composer could have charts in several books, with the given charts having varying degrees of accuracy 14:29:39 http://kevin.watson.name/research/openoffice.html 14:29:40 A: http://kevin.watson.name/research/openoffice.html from rbm 14:29:54 A: "OpenOffice.org in an Academic Environment: A Viable Alternative to Microsoft Office 2000?" 14:29:55 added comment A1 14:30:03 A:| "OpenOffice.org in an Academic Environment: A Viable Alternative to Microsoft Office 2000?" 14:30:04 titled item A 14:30:16 A: Study done at Utah State University 14:30:17 added comment A2 14:30:27 A: (qualitative, rather than quantitive, I should add) 14:30:28 added comment A3 14:31:05 jim: sounds like be need a "list of other objects" package. 14:31:22 yeah 14:31:29 or, you can use acs_rels 14:31:35 now I have some questions about rels 14:31:39 :) 14:31:57 basically, all the packages I have seen have created their own mapping tables insteadof using acs_rels 14:32:24 if these toolkit wide tools are too slow to be used for what they were designed for, fix them or get rid of them :) 14:33:56 I think that people are coming up with really good ideas... but they're often hard to understand (sc, cr, portal come to mind) 14:34:18 so not many want to or are able to just jump right in 14:34:39 but I'm getting away from the questions I had... 14:34:47 could be. i meant the existing packages too. of course the toolkit was never finished so the packages were built to work with whatever was working at that time. 14:34:54 ok jim. back to the real world. 14:37:06 when I look at the admin page for a subsite, I see these 4 links that I never understood completely why they were all there: group, group type, relational segment, relationship type 14:37:36 I understand that a group is implemented by a table of graph edges 14:37:57 * davb needs to learn this graph stuff... 14:39:12 * rbm hands the travelling salesman problem to davb :) 14:39:24 thanks. is that in SICP? 14:39:34 I think I am missing some math here. 14:40:50 is the farmer's daughter involved in this problem? :) 14:41:57 and I understand you can create types of groups by specifying the object types for the ends of the graph edges 14:42:03 but that's about it 14:42:42 jim: i think that most of the time, you can ignore that stuff. the interface is for engineers, not users. 14:43:06 I have a engine in my car... 14:43:38 weird. according to postgresql the pg_attribute table does not have OID. i though all tables had oid. 14:43:40 but seriously, what are those things, and how do they relate to each other? 14:44:07 a relationship type is how someone is a member of a group. 14:44:16 not sure on relationship segment. 14:44:33 I am not sure how they are related to subsites. 14:44:45 member versus component? 14:44:47 we need someone from dotlrn to explain how they built it using that. 14:44:51 jim: yes. 14:45:39 direct member versus member of a group which itself is a component of the group in question? 14:45:51 maybe :) 14:46:08 anyway. for example Main Site has a predefined relational segment: Main Site Member 14:46:38 so a relationship type defines the type of relationship. 14:47:07 apparently, new relationship types currently need to be created by code. 14:47:16 so forget that. complete useless. 14:48:05 just means I'd have to provide UI for it 14:48:15 (if I were to go in that direction) 14:48:20 relation segments are where you define the actual relationships. Ie member of main site. then you can assign that relationship to a group. 14:48:45 jim: i doubt you need anything fancier than member or component. just create groups with member groups should handle most cases. 14:48:46 I see 14:49:14 note: i still don't know how you would use this to make a useful application out of it. 14:49:23 so I'm looking at creating "groups of songs" 14:50:13 interesting. 14:50:39 but I will need to add attributes somewhere... to the graph edges? 14:51:14 i don't think you want to use groups. 14:51:18 so on one end of the edge is the group ("song list"), on the other is the song... 14:51:23 that seems to be specifically for users. 14:51:29 well, an acs rel 14:51:31 acs_rels is the generic version. 14:52:00 and then, piled onto the middle of the edge, is extra attributes 14:52:07 which are objects themsevles. so you can have a song_group rel with a song_groups attribute table to store them. 14:52:52 see, I want users to be able to add rel types, with arbitrary numbers of attributes 14:53:04 ah. 14:53:06 hmmmm. 14:53:33 you could possible use acs_attributes for that. acs_objects can use a generic storage for object attributes. 14:53:56 when you create an object type you can specify the custom table to store attributes or use the generic storage. 14:54:04 I don't want them to -know- they're doing that... as far as they know, they're making types of song lists (book lists, gig lists, list of arranged tunes, etc) 14:55:14 so you could create the new object type based on their input. 14:55:25 and I want them to be able to decide, "I want this kind of song list to have a reference to a book of tunes and a page number in whichever book that is" 14:55:37 now I see why you like the form generation idea. 14:55:46 yep 14:56:08 that kind of thing is going to play a role 14:56:31 i don't know anywhere in the toolkit thar previously did this. but you might want to look at CMS. It might have some dynamic content-type features. 14:57:09 test=# select oid from pg_attribute; 14:57:16 ERROR: Attribute 'oid' not found 14:57:16 Label ERROR not found. 14:57:26 anyone else see this? PG 7.2 14:57:44 or can anyone confirm this does work on 7.1? 14:57:57 oid's are optional in PG 7.2 14:58:46 so here's a guy with a fake book, the first one on the site (and I didn't predefine such), he should be able to go, "I want to represent fake books, so I'm creating a type of list called a 'Fake Book', which is a list of tunes, each of which know their book and page number" 14:59:03 ok. :) pg_class and pg_type have oids, but pg_attribute does not. maybe I don't need it. 14:59:22 oids better not be catching... 15:00:04 jim: that is pretty complex. 15:00:35 first you need to dynamically generate the tunes objects. 15:00:40 then the list objects. 15:01:57 maybe I can hide the complexibility by letting them just define lists, let them think that's as far as it goes, then let them define attributes 15:01:59 have you ever looked at the open/acs 3.2 photo gallery code? it lets users define new fields for photos and stuff. this can get pretty complex on the UI. 15:02:34 jim: maybe. 15:02:46 * jim thinks about that... 15:03:24 I wonder if your magic make-you-own-object-type code could be mergerd with the enter-data-about-object-objects package (ETP2) 15:04:03 I'd almost say, get it working with predefined tunes and lists. then go from there. 15:04:19 thing is, I can let them -really- do that... and not be bothered with -types- of lists (there's only one kind, and every list can have its own set of attribs) 15:04:46 jim: sure. 15:04:53 the problem with that being there exist types of lists of songs in the real world 15:05:51 here's the original thought: me and this piano player sometimes get together to play, and we're thinking of taking it out and getting some gigs 15:06:02 so we're creating lists of tunes 15:06:27 all these tunes are "tunes we both want to be able to play" 15:06:47 one is a tune neither of us knows well or at all 15:06:52 (Lush Life) 15:07:06 some are tunes we both know 15:07:10 some I just know 15:07:13 some he just knows 15:07:51 so our goal is to arrange lists of tunes we know to form gig lists 15:09:09 and to move the tunes either or both of us dont' know well over to be part of the tunes we do know well together 15:10:15 ok so you want to create a list of tunes with several attributes, I know this song, I would like to learn this song, etc. or maybe a rating of how well you know a song. 15:10:26 yes 15:10:37 you both create a list, then you want to be able to compare them to see which songs you both know etc... 15:10:55 him and I both own Real Book 5th ed and Real Book vol 2 15:11:26 yeah, that's the original idea 15:11:30 ah, so that needs to go in there too? so you can say, if its in Real Book 5th Ed. I can learn it? 15:12:04 and it's on page 204 (out of like 700, these books are not small) 15:12:58 plus, I play bass and learning to represent harmony in new ways on its fingerboard, so I need some list attribs 15:13:50 maybe I need also a way to say "Hey, the complete 5th edition is right there, I want to copy that list and add personal attribs" 15:14:21 ah. so you want to annotate an existing list. 15:14:31 could be, yes 15:15:10 plus, I want to have each tune in certain lists commentable 15:15:25 that could be accomplished by creating a related list. so jims-list-related-to-the-other-list and your attributes are stored in the parallel list 15:15:40 and still have general comments on the list, or list items. 15:16:07 ack. but. do the general comments on a list-item apply to the item, as it appears in any list, or only the item in that particular list? 15:16:09 :) 15:17:04 "Beautiful tune, I went to Yoshi's and it was Craig Handy, Dave Holland, Jeff "Tain" Watts and Herbie Hancock, and they did such a beautiful rendition, and they reharmonized the bridge... the new chords are ... " 15:17:43 (and how would I know what chords they were playing? I got a really good ear :) 15:19:19 ok. brb, another crisis :) 15:19:35 ok... I'm wandering out for awhile 15:19:41 bblater 15:20:38 paje has quit ("regrouping; bbiab") 15:20:49 paje (~paje@208.184.248.88) has joined #openacs 15:20:58 paje: the problem? 15:20:58 rbm: wish i knew 15:21:00 Crap 15:21:05 paje has quit (Client Quit) 15:24:38 paje (~paje@208.184.248.88) has joined #openacs 15:25:19 paje, status? 15:25:19 Since Mon Jul 1 12:06:30 2002, there have been 0 modifications and 3 questions. I have been awake for 43 seconds this session, and currently reference 18 factoids. Addressing is in optional mode. 15:25:30 crap. he's lost his memory again 15:26:00 I'm testing the move again. paje's memory is backed up. 15:26:50 paje has quit (Client Quit) 15:29:35 does paje use pg? 15:29:45 dbm 15:31:21 paje, dbm? 15:31:56 paje, are you a dba? 15:32:05 or maybe doa 15:32:12 :) 15:37:01 docwolf (~wolf@adsl-34-224-116.bct.bellsouth.net) has joined #openacs 15:39:33 paje (~paje@208.184.248.88) has joined #openacs 15:39:37 paje: who are you? 15:39:37 i am a slut 15:39:43 paje: literal paje 15:39:44 rbm: paje =is= the medicine man and oracle in Amazonian tribes | a silly bot|a Peson Assembled for Judo and Exploration|a cannibal|acannibal|a slut|a wanker! 15:39:52 paje: rbm? 15:39:53 you are a CS messiah who will crush all the EJB posers and their kindred and lead us all to a new era of coder enlightenment. 15:39:56 alltel? 15:39:57 alltel are a bunch of wankers 15:40:00 YES! 15:40:11 woot! 15:40:50 cablevision? 15:40:50 somebody said cablevision was almost a bunch of wankers but they managed their ISP pretty well 15:41:55 I'm going to hibernate paje for a minute more, so I can run him on a screen session 15:42:18 paje has quit (Client Quit) 15:42:50 paje (~paje@208.184.248.88) has joined #openacs 15:43:05 paje: you're now back for good. 15:43:05 rbm: excuse me? 15:43:15 paje: your back! watch out! 15:43:15 markd2: sorry... 15:44:55 we need to teach paje to hack servers to store himself in 15:52:13 any of you ever do skeet shooting? 15:52:27 I have made my peace with skeets 15:52:32 til has quit (bear.openprojects.net irc.openprojects.net) 15:52:32 Spork has quit (bear.openprojects.net irc.openprojects.net) 15:53:12 paje, netsplit is | NETSPLIT! RUN FOR YOUR LIVES! 15:53:13 OK, denshi. 15:53:23 paje, netsplit! 15:53:23 denshi: what? 15:53:27 paje, netsplit 15:53:27 * paje NETSPLIT! RUN FOR paje's LIVES! 15:53:41 heh 15:53:43 ctpat3r (~ctpat3r@129.252.42.9) has joined #openacs 15:53:43 til (til@port-212-202-128-197.reverse.qsc.de) has joined #openacs 15:53:43 Spork (~nblyumbe@bosco.eng.cv.net) has joined #openacs 15:53:45 hey Roberto 15:53:46 is he part cat? 15:53:57 how far are you from Ogden Utah? 15:55:10 [GlobalNotice] Hi all. We've just experienced a hubbing problem affecting one main rotation server. We have rehubbed it and are waiting to see if the new configuration is stable. Apologies for the problems and thank you for using OPN! 15:55:58 Psychephylax: About 1 hour away by car 15:57:11 jun_ (~jun@AP-203.167.6.193.sysads.com) has joined #openacs 15:58:16 ok 15:58:26 You have to go buy me a machine gun from them...I'll paypal you the money 15:58:27 why? 15:58:43 a machine gun? Like a real machine gun? 15:58:43 hi guys ... seems to be quite now 15:58:49 like yeah... 15:58:56 You know..one that goes bang bang bang? 15:58:58 rbm, when has *phylax made sense before? 15:59:00 And all the skeet die 15:59:11 well, not die..shatter 15:59:12 hi jun_ 15:59:20 hi dave 15:59:31 i will just be here for a short time 15:59:37 Psychephylax: err, okay. as long as guys with black suits and sun glasses don't come after me because of that :) 15:59:40 really need to get some ZZZZs 15:59:47 *phylax, it takes a two-man team to operate a machine gun 15:59:59 jun_: ok. what TZ are you BTW? 16:00:00 you're right... 16:00:07 one to pull the skeet...and one to shoot 16:00:13 I don't think skeet shooting has a 'heavy weapons squad' division in competition 16:00:20 that seems rather unsporting. 16:00:30 :-/ 16:00:34 ah, that would be a 3 person team. probably 6, to get enough skeets in the air to count 16:00:45 are you all trying to ruin my dreams? are you? ARE YOU?!! Hmmm...fess up! 16:00:57 GMT +8 16:01:07 1 person to shoot, 1 person to feed ammo & swap out barrels, 4 people to pull skeets 16:01:16 ah, practially opposite me. 16:01:22 i am working on a project now using cofax 16:01:33 cool. is it better than ACS Java? 16:01:46 maybe after this I will bring in some concept into etp 2... etp 2.5 16:01:54 well i believe its better 16:01:57 great. 16:01:58 its more simpler 16:02:05 not mumbo jumbo java 16:02:08 it uses java 16:02:26 but no xml, il8n, blah blah 16:02:33