00:22:37 has anyone used the template::util::multirow_to_list proc? 00:27:32 k2pts (~nkd@213.149.180.151) has joined #openacs 00:27:36 hey guys 00:27:39 hi k2pts 00:27:42 hey dave 00:27:53 I'm gonna send a friend here from cyprus...so be nice with him ;) 00:27:59 np :) 00:28:04 I'm trying to get more people involved in free software projects 00:28:48 great. you are the Cyprus LUG :) 00:29:12 no, there are no such things in cyprus 00:29:24 :) 00:30:58 Chris0000 (noBan@li-11-46.cytanet.com.cy) has joined #openacs 00:31:30 davb: Chris0000 is the guy I was tellin you about 00:32:08 hey guyz 00:32:17 Welcome Chris0000 00:32:26 thnx davb 00:32:28 I just told Christos about openacs 00:32:36 he is rookie but not for long... 00:33:01 davb: didn't have time to look at the rss stuff 00:33:31 I was thinking to use xslt to parse adp files and include them as somekind of portlets..what do you think? 00:34:43 interesting. how is that better than what we have now? also how will the xslt get the tcl variables? 00:35:01 the ns_xml module doesn't parse the xslt files as expected 00:35:15 davb: how do I get paje to log something for christos? 00:36:06 I was thinking that after we fetch the feeds, to store them in the file system...don't see why we need tcl vars...please elaborate 00:36:06 I am not sure what you mean. 00:36:16 paje: christos is a hell of a coder 00:36:16 OK, k2pts. 00:36:23 paje: who is christos? 00:36:23 well, christos is a hell of a coder 00:36:25 ah, you mean rss? 00:36:30 yes 00:36:34 ok. sorry :) 00:36:43 paje: spank talli 00:36:43 k2pts: what? 00:36:49 paje: again 00:36:49 k2pts: what? 00:36:57 paje: again? 00:36:57 wish i knew, k2pts 00:37:04 must have been removed from his memory. 00:37:23 yeap 00:37:32 paje: who is talli? 00:37:32 i think talli is the slayer of innocent domo-kuns the world over. 00:37:36 :) 00:37:44 k2pts: My idea has usually been to provide the data to a package/template and let the template decide how to display or use it. 00:38:36 we need the ns_xml stuff to work for that, I think... 00:39:06 storing it in the filesystem, is just some intermediate state for parsing it into adp files 00:39:29 and then just include them with templates...we can always modify 10-to-xhtml.xsl 00:39:56 right. but if I want to change the html of the display of the rss, I need to modify the stylesheet instead of my adp template. 00:40:19 right... 00:40:26 it's not the best solution but it's a start 00:40:47 do you know what happened with the guy who was going to work on improving the ns_xml stuff? 00:42:06 k2pts: that is why I parse the rss into a nsv variable. (on my old non-openacs aolserver site) 00:42:13 paje: seen hazmat? 00:42:13 hazmat was last seen on #openacs 2 hours, 42 minutes and 44 seconds ago, saying: thats a strange concept. [Sun Mar 17 15:01:30 2002] 00:42:17 I don't know who was planning on working on ns_xml. 00:42:27 let me get the url 00:42:37 it shouldn't be too hard. it is just a little wrapper on libxml2. hardly any code really. 00:43:11 hey k2pts! 00:43:13 http://openacs.org/new-file-storage/download/nsxml-2.html?version_id=477 00:43:13 A: http://openacs.org/new-file-storage/download/nsxml-2.html?version_id=477 from k2pts 00:43:16 hey rbm 00:43:25 rbm: let me introduce to Christos 00:43:53 k2pts: wow, I never saw that. 00:43:53 Chris0000: rbm is our documentation maintainer and one of the openacs gatekeepers... 00:44:12 Chris0000 is now known as Chris^ 00:44:32 ic 00:44:41 paje: who is rbm? 00:44:41 i heard rbm was Roberto, http://www.brasileiro.net/ 00:44:46 hi Chris^ 00:44:55 hey rbm 00:45:07 rbm: Chris is a 3rd year CS student at the university of cyprus 00:45:23 Nice. 00:45:25 I have known him for almost half a decade...he participated in the valkan olympiads in informatics 00:45:33 what u're trying to do is very nice u guyz 00:45:38 * rbm is also a 3rd year CS student and he is graduating in MAY!!!! 00:46:24 how's it going rbm ? 00:46:41 rbm: what's the word for 3rd year student (sophomore,etc) 00:47:02 k2pts: junior. But I'm a senior :) 00:47:07 :) 00:47:39 that ns_xml plan looks really good. A decent C hacker could probably handle it in a short time. 00:48:06 yeah it's pretty nice but I have already told you which one I prefer...to bad it's MPL 00:48:14 s/to/too 00:48:32 paje: insult Chris^ 00:48:33 Chris^ is nothing but an unoriginal thimbleful of bootless Stimpy-drool. 00:48:33 :) 00:48:40 !!! 00:48:46 right. I also would prefera real DOM parser instead of the pseudo dom one hacked in nsxml. 00:49:02 exactly 00:49:13 paje: insult talli 00:49:13 talli is nothing but an unoriginal thimbleful of beef-witted craptacular carpet droppings. 00:49:26 * rbm tries to finish his cheat sheet 00:49:27 rbm: I like paje a lot 00:49:28 TclDOM uses expat also. I wonder where the performance hit is. 00:49:38 I guess its in the TclDOM code :) 00:49:52 I don't know...but there was an article at developer works and Zoran is mentioned there 00:50:21 Yes, I recall that. 00:50:58 rbm: for the first time ever "theory of computation" had an easy exam last friday...everybody was smiling...first time ever 00:51:00 :) 00:52:25 k2pts: reportedly:" Just an update. I'm testing ns_xml 2.0 now, " 00:52:43 where did you get it from? :) 00:53:05 here: http://openacs.org/bboard/q-and-a-fetch-msg.tcl?msg_id=0003me&topic_id=OpenACS%204%2e0%20Design&topic=12 00:53:27 k2pts: Oh. I'm taking that class here too, and I have an exam wednesday. 00:53:43 if he needs someone to test I definitely will help him out. 00:53:51 no sweat rbm :) 00:54:08 :) 00:54:31 The cheat sheet I'm preparing is for my Multimedia Development's exam, which is tomorrow 00:54:49 s/'s// 00:54:52 davb: it's been a while since that post... 00:55:12 rbm: are you going to have a project ? 00:56:03 paje: who is GEM 00:56:03 GEM is installing on GEM's Debian system, and hate to sully it with non-aptable stuff if I can help it :-) 00:56:11 jim (~jim@12-233-187-5.client.attbi.com) has joined #openacs 00:56:15 hey jim 00:56:16 re 00:56:19 hi jim 00:56:21 hi k2pts 00:56:29 hya davb 00:56:48 hey jimy 00:56:50 davb, jim: how's the packages contract goes? 00:57:00 jim: Chris^ is from Cyprus 00:57:05 paje: who is jim? 00:57:06 jim is as jim does 00:57:10 :) 00:57:24 Chris^: jim...he's a lot into Debian stuff... 00:57:29 jim: right? 00:57:36 yeah, I'm a debian maint 00:58:04 * davb emails 00:58:18 davb: for the ns_xml 2.0 ? 00:59:17 yes. 00:59:45 would love to know what comes out of it? :) 00:59:55 k2pts: we have an initial plan on the contracting, involving the creation of a contract with a single operation (I just had another idea on that, davb) which takes a package instance id and removes all data associated with that instance 01:00:26 sounds good to me...I have read part of the chatlogs but not into the details 01:01:28 I was thinking something along the lines of this: 01:02:00 a second call that removes data associated with all instances of the package 01:02:42 so the first one could be a no-op if desired for certain packages 01:02:56 are you going to do a contract for removing stuff...I think a more general contract would be better IMHO... 01:03:06 s/stuff/stuff only 01:03:52 well, now's the time to throw out ideas on the contract... 01:04:07 we haven't even tried anything yet ;) 01:04:18 I have thought about aggregators in the past but never get around implementing one 01:04:29 * jim might need a new somethingorother (mobo? mem? video?) 01:04:32 aggregating packages that si 01:04:44 s/si/is 01:05:27 before I get too far, slipped my mind you just made an introduction... 01:05:34 Hiya Chris^ :) 01:05:46 n/p jim 01:05:47 s/n 01:05:52 what's Cyprus? 01:05:55 heh 01:05:58 s/n\/p/np/ 01:06:23 it's a country at the crossroads of Europe, Asia and Africa and a lovely home for Chris^and Neophytos 01:06:32 ahh :) 01:06:48 oh yes it is :) 01:07:10 though i'd rather be in new zealand :) 01:07:17 Chris^ has left #openacs 01:07:21 Chris^ (noBan@li-11-46.cytanet.com.cy) has joined #openacs 01:07:29 among other things, an aggregator can be used to instantiate/restore hierarchies of site-nodes...so you could have a pre-configured vertical app that it can be mount through the use of aggregators... 01:07:57 think of it as a connector of packages at the application level 01:08:18 that way you can develop new apps by just a few clicks 01:09:00 so like a struct whose members are package instances? 01:09:13 yeap 01:09:28 and packages could be refactored into mostly services? 01:09:37 that's really big 01:09:44 that's one way to think about it... 01:09:54 it's not as big as it seems in my opinion 01:10:05 edit-this-page does something similar 01:10:34 I use to say that edit-this-page is half an aggregator and half of a cms 01:11:09 well, could you come up with a contract that would include an instance deleter? 01:11:21 something that would facilitate what you' 01:11:25 re looking at? 01:12:05 I can try to prepare something for you... 01:12:13 I'm not looking at aggregators these days. 01:12:36 I look at the issue a few months back but I'm waiting for the new portal package unti I work on that again 01:12:49 I'm now working on a persistence layer but it's not moving very fast 01:13:06 Chris^: hazmat is a good friend and a great dev 01:13:17 hazmat: Chris^ is a friend from Cyprus 01:13:24 paje: who is hazmat? 01:13:24 i heard hazmat was somebody that drinks often from the river lethe 01:13:32 if someone does aggregator interface, generalizing it to a general purpose event channel would be cool. 01:13:35 heh, I knew it was that one 01:13:39 ;) 01:13:41 hehe 01:13:44 note that my reasoning behind the instance deleter is towards the appearance of stability in the ui, by completing the vision statement in the package mgr... 01:13:45 hi Chris^ 01:13:47 having said that... 01:13:54 hi hazmat 01:14:16 i'm wondering ... what if .... 01:14:25 paje: do something 01:14:25 Chris^: what? 01:14:33 oh :) 01:14:59 jim: go on 01:15:04 paje: do something is (wiggles|dances|runs|walks) 01:15:04 OK, jim. 01:15:15 paje: do something 01:15:16 * paje (wiggles 01:15:22 paje: do something 01:15:23 * paje (wiggles 01:15:33 paje: do something 01:15:34 somebody said do something was walks) 01:16:04 ok, so it doesnt' factor that way :) 01:18:22 k2pts: your ideas for increased factorability and increased admin control and designing of apps (and more infrastructure to allow that kind of thing to happen in the UI) is a definite vision, and the existing vision statement is interpretable in that way 01:19:00 I had a much more lowbrow idea, which was to make sure that more of the admin ui worked 01:19:25 right, I was just throwing ideas :) 01:19:33 jim: its not lowbrow, just a first step. read lars on focusing :) 01:20:02 well, ok, so lowbrow was the wrong choice of word :) 01:21:23 but I think the deleting thing could be done as a part of oacs-4.5.1, and the things you're talking about -seems- to me like oacs 5 or so 01:21:37 right 01:22:38 your expansion highly desirable to increase capabilities, my original thing necessary for basic stability 01:22:58 on the subject of ui, a suggestion, i think one of the biggest things needed from an end user perspective and a development perspective is to have /pvt/home turn into a user workspace with external api for packages/applications to plug in panels. to create a centralized user interface to the site functionality. 01:23:11 agreed 01:23:24 sheesh, I have a philg story about that :) 01:23:30 hazmat: hopefully dotlrn will have some of that 01:23:50 i would expect it would, but i wonder in what form. 01:23:54 I think that all of the ui should be available as api 01:24:07 we will see (i hope) 01:24:31 k2pts: I think so. so that it can be customized without digging into the tcl code? 01:24:34 for example a page could have been called/dispatched from another one...we can do that already using include statements but we need a better control 01:25:17 I want to easy way to link to pages in other packages or the url of an arbitrary acs_object. 01:25:25 I also want a pony. 01:25:32 lol 01:26:13 hehe 01:26:20 heh 01:26:57 I also want a pony with Steph Seymour riding it sidesaddle, going "c'mere, I'll show you entire VS line" 01:28:22 wow! I got rss-support working. at least on my dev site. now to move it to the production site. 01:28:32 no kidding... :) 01:28:34 what's that? 01:28:41 rss feeds... 01:28:49 links from other sites...automated 01:29:02 this is actually the generation of an RSS feed from my site. 01:29:34 oops, I lied, its still partially broken. 01:29:36 davb: I was over excited. I thought you were talking about the RSS stylesheets 01:29:44 sorry :( 01:29:50 But that is next! 01:30:55 k2pts: could you give a ballpark estimate of the operations you think should be in the contract (not number of, but actual functions) 01:30:56 ? 01:31:16 not off the top of my head. I have to think about it a bit 01:31:34 but I will get back to you and davb on this one 01:32:09 meanwhile we can do a refinement of the notes package with the delete contract 01:32:51 one request if it's possible 01:33:14 before you work on the notes package, please someone make it subsite specific by addin a package_id column 01:33:18 i think I'll be going u guyz. goodnight everybody ! 01:33:27 bye Chris^ 01:33:33 (with bags@ad.com's help, last feb, rewrote a notes delete script) 01:33:43 [ swarz. ] I'll be back :) 01:34:04 k2pts: but it is specific, as each note's context_id is the package instance id 01:34:33 yeah, I haven't checked it for a while. someone must have changed it 01:34:53 it worked that way back in Feb, iirc 01:35:04 bye Chris^ 01:35:19 cya :) 01:35:29 I could have green notes, red notes, and (for music lovers) blue notes... 01:37:42 jim: have to check it out. I was so sure that it was not package specific 01:42:19 k2pts has quit (devlin.openprojects.net irc.openprojects.net) 01:42:19 paje has quit (devlin.openprojects.net irc.openprojects.net) 01:42:19 hazmat has quit (devlin.openprojects.net irc.openprojects.net) 01:43:04 jim has quit ("[x]chat") 01:43:07 jim (~jim@12-233-187-5.client.attbi.com) has joined #openacs 01:43:18 jim has quit (Read error: 104 (Connection reset by peer)) 01:43:19 jim2 (~jim@12-233-187-5.client.attbi.com) has joined #openacs 01:43:24 jim2 has quit (Read error: 104 (Connection reset by peer)) 01:43:41 paje (~paje@slxwy.dorms.usu.edu) has joined #openacs 01:43:41 hazmat (~ender@adsl-66-123-57-58.dsl.lsan03.pacbell.net) has joined #openacs 01:44:22 jim (~jim@12-233-187-5.client.attbi.com) has joined #openacs 01:45:09 jim has left #openacs 01:45:49 jim (~jim@12-233-187-5.client.attbi.com) has joined #openacs 01:47:23 k2pts (~nkd@213.149.180.151) has joined #openacs 01:47:54 davb: I was reading the chatlogs and now I see why it didn't make sense about the RSS talk I did earlier 01:48:05 ok. 01:48:13 I meant use xsltproc and store the parsed files as adp 01:48:15 then use include 01:48:27 admitedly, my original post was confusing 01:48:38 00:33:31 I was thinking to use xslt to parse adp files and include them as somekind of portlets..what do you think? 01:48:48 I see. that is a neat idea. 01:48:50 it should have been: 01:49:17 use xsltproc to parse the rss feeds into adp files and then use include as somekind of portlets 01:49:38 that would be a start until we have a new xml package 01:50:03 does it make sense now? 01:50:09 sorry if I have confused you earlier 01:50:31 I see. 01:50:36 sure that makes sense. 01:51:11 ok, we talk l8r 01:51:14 have to head out 01:51:31 ok bye. 01:51:34 k2pts has left #openacs 01:53:52 is there a proc to get the main url of the site? 01:56:30 Chris^ has quit (Read error: 110 (Connection timed out)) 01:58:27 umm 01:59:03 can you look at the site return url (a kernel parameter)? 01:59:46 maybe. I want the URL of the main site. I thought there might be a built in way to get it without explicity querying the database. 02:01:40 ad_conn url? 02:01:42 here goes. I am installing my rss code into the live site. 02:01:50 that seems like cheating :) 02:02:59 that might have a better idea of the url than the parameter... the user could have made a mistake typing it in on install, and not installed any pkgs that depended on it being correct 02:03:10 well I have the ns_xml 2.0 code in it's current state. looks liek its time to learn C :) 02:03:14 true. 02:03:43 you gotta learn C fast? 02:03:59 not really. But I will try out the code and look at it. 02:04:13 if I really want to contribute I would need some knowledge of C. 02:04:48 well, either way, Holub's The C Companion is good as an advanced sidekick to a c primer 02:05:03 great. I'll put it on my list. 02:07:50 hmmmm. 02:08:35 oops, forgot to restart the server 02:15:32 does anyone know the command to start up the mozilla source code viewer? I want to register it as theapplication for xml files. 03:08:43 jim has quit (Remote closed the connection) 03:09:28 talilee (~talli@lti-4.dialup.access.net) has joined #openacs 03:09:47 davb, you're talking about john mileham's nsxml, right? 03:09:56 talli has quit (Read error: 113 (No route to host)) 03:10:20 talilee has quit (Remote closed the connection) 03:10:22 talilee (~talli@lti-4.dialup.access.net) has joined #openacs 03:46:53 talilee: yes 03:55:05 well bye 03:55:06 davb has quit ("Client Exiting") 04:08:32 talilee is now known as talli 05:05:49 Webseeker (Marcelo@200.181.243.218) has joined #OpenACS 05:51:52 Webseeker has quit ("Client Exiting") 06:59:25 dlk (dlk@walter.ita.chalmers.se) has joined #openacs 13:00:44 dlk has quit (Read error: 104 (Connection reset by peer)) 13:00:52 dlk (dlk@walter.ita.chalmers.se) has joined #openacs 13:38:53 davb (~dave@rrcs-nys-24-97-22-204.biz.rr.com) has joined #openacs 14:00:09 dlk is now known as dlk-meeting 14:56:25 dlk-meeting is now known as dlk 15:45:30 dlk has quit ("*puff*") 15:45:43 morning fellows 15:50:51 hi talli 15:51:04 hey davb 16:17:06 morning 16:17:12 hey rbm 16:17:24 anyone know of any good free icalendar projects? 16:20:55 sorry, no. 16:21:06 i fear one does not exist 16:21:53 lotsa talk about how to build one, not much action. i wonder if it's a hard problem 16:22:21 server or client? 16:23:12 server 16:24:26 have you looked at the one in StarOffice? I used it for a while. Supposedly it supports icalendar, but I never did much with it. The docs were really poor. 16:25:27 yeah, that's pretty much what every ical project is like 16:28:53 http://reefknot.org/whitepaper/whitepaper.pdf <-- maybe? 16:28:54 B: http://reefknot.org/whitepaper/whitepaper.pdf from rbm 16:29:22 i know that one 16:29:30 but i don't know how alive it is 16:30:32 any project that is a whitepaper and not a cvs archive does not inspire confidence :) 16:30:40 yeah 16:30:49 they released some things, but not in a while 16:31:06 and i am developing less and less trust in proejcts that use sourceforge 16:31:18 it seems sourceforge is where OSS projects go to die 16:32:02 as in, if you can;t maintain a webserver and cvs repository, you probably don't have the project management skills? 16:32:16 anyway, something just needs to be built. there's this data model at w3c which seems pretty nice: http://www.w3.org/2000/01/foo 16:32:25 davb, i didn't think of it that way, but good point ;) 16:34:05 talli: why? 16:34:28 sourceforge also has a horrible UI 16:34:44 so whenever i go there i can never find enough info about a proejct to see wheter it lives or not 16:35:13 it has really poor transparency, too 16:35:26 the mailing lists are too hidden 16:35:49 and probably i assign alot of guilt by association 16:36:16 so many projects are dead there since people think that opening a sourceforge project == a true open source project 16:36:53 hopefully the people who started those projects have moved on to real projects. 16:39:38 I don't like the UI much either. But before you'd have to search for things all over the web and only then find out if they were dead 16:39:51 yeah, that's true 16:39:52 so projects died everywhere, always. 16:40:05 no we have a central graveyard :) 16:40:32 But at least if you find something you like (and sourceforge makes that easier), at least you can pick it up and run with it. 16:40:54 true enough 16:41:27 that doesn't really fix my dislike of sourceforge, just point out the utility of a central project manager 16:41:35 true 16:42:06 savannah does a similar thing, but it's just a reverse engineered sourceforge 16:42:16 but it'll be distributed 16:42:28 that's a feature of GNU, not a bug, but it's not always the best feature 16:43:25 i don't think it's a reverse engineered sourceforge. That would mean savannah is sourceforge r-engineered and re-written. It _is_ sourceforge, without the proprietary features introduced in later versions. 16:43:36 oh yeah? 16:43:42 i thought that sourceforge wasn't OSS 16:43:49 for sure not GPl 16:43:57 it is. You can download it. 16:44:08 hmm... didn't know that 16:44:11 Not sure about the later versions. I think the license was changed. 16:44:22 ah 16:44:39 but sourceforge itself was kept at sourceforge. 16:45:11 and to be on sourceforge it has to be on a DFSG license 16:45:26 what's that? 16:45:42 when will people realize that GPL is the One True License? 16:46:12 Debian Free Software Guidelines. Defines guidelines for a free software project. The Open Source definition is basically DFSG without "Debian" in it. 16:46:24 ah, ok 16:47:19 If a project is DFSG-compliant, it cam be included in Debian's main tree. If it can't, it has to go on contrib (if it's DFSG but depends on non-free) or non-free (if it's not DFSG compliant) 16:52:39 One of the reasons why Debian rocks so much 16:54:10 true enough 17:23:13 talli has quit (Read error: 113 (No route to host)) 17:36:24 talli (~talli@xcdfddb76.ip.ggn.net) has joined #openacs 17:37:13 rbm: you still here? 17:37:55 talli: sorta. trying to do some studying and checking irc every so often 17:38:03 ok 17:38:12 what is the deb package for openssh? 17:38:22 for sid 17:39:59 ssh= 17:40:05 err, ssh 17:40:22 apt-cache search ssh 17:40:44 apt-cache search openssh 17:40:49 i tried that. it doesn't find anything for some strange reason 17:41:01 do you have non-US in your /etc/apt/sources.list? 17:41:09 ah, no i dont' 17:41:11 crypto stuff has to be in nos-US 17:41:13 thanks for the tip 17:41:15