00:00:02 OACS has its calendar system 00:00:04 etc 00:00:48 none of them are really based on a standard, most of them are pretty ad hoc 00:00:57 and they're all "web based" 00:01:12 so me and this mad scientist were thinking of just building an iCalendar server 00:01:43 it will have an extremely tight focus, just to implement the protocols of the iCalendar RFC 00:01:49 nothing else 00:01:56 no other "groupware" features 00:02:37 so in the same way that an MTA is focused solely on delivering mail, build a server that is just going to deliver the iCal protocols 00:02:55 That'd be pretty cool 00:03:00 then provide APIs to let other apps or scripting languages play with the iCal stuff 00:03:16 so you could build mod_xxx or ns_xxx to tie them into webservers 00:03:18 so have you defined the API? 00:03:37 or python or tcl interface to script the interface 00:03:52 no, the mad scientist and i have been looking into it the past couple of days 00:04:11 so much of this stuff is laid out, but i have yet to find anyoen just doing it 00:04:18 but i think that something like AOLserver is the model 00:04:31 a very tight piece of code with a limited API that solves one thing very well 00:04:39 are you talking about rfc2445? 00:04:41 yeah 00:04:52 http://www.imc.org/rfc2445 00:04:52 A: http://www.imc.org/rfc2445 from denshi 00:05:00 A: iCalendar RFC 00:05:00 commented item A 00:05:38 once this thing is built, let anyone who wants build clients or interfaces to it 00:06:18 but this piece of code should be extremely tightly focused, not like the apache project with all of its jakarta crap, or like jabber that seems to want to rebuild the system over and over 00:06:33 this spec is dated nov 1998. isn't there anything else out there? 00:06:53 i was thiking a C programmer because then it could be built on the apache runtime which runs everywhere 00:06:58 there has been work on it 00:07:02 lemme find something real quick 00:08:55 have you looked at the freeassociation project? 00:09:01 i can't find much else, but there is a lot out there 00:09:03 no, where is that? 00:09:13 They have had 0 commits in the past 3 months, but at least they have something there. 00:09:25 sf.net/projects/freeassociation/ 00:09:41 http://sourceforge.net/project/stats/index.php?report=months&group_id=16077 <-- free association project monthly stats 00:09:41 B: http://sourceforge.net/project/stats/index.php?report=months&group_id=16077 from rbm 00:10:45 "The Free Association project aims to develop an iCalendar based calendar user agent and calendar server. " 00:10:50 yeah 00:10:59 part of the problem is that they want ot build a user agent 00:11:03 And it's written in C, but I doubt it's written on the Apache foo 00:11:04 screw the user agent 00:11:15 focus on the server 00:11:17 i'll check it out 00:11:26 talli: You need at least one user agent if you're going to build the server. How else are you going to test your server? 00:11:41 and the home page link is broken 00:11:47 mozilla has a user agent 00:11:49 yeah 00:11:54 oh it has? 00:11:58 * rbm didn't know that 00:12:02 ximian is also supposed to work with ical 00:12:05 and outlook too 00:12:06 I guess I should know better. Mozilla has everything. 00:12:09 there are plenty of those out there 00:12:13 there's just no good server 00:12:20 reefknot.sf.net 00:12:28 talli: Isn't there a libical? 00:12:36 a perl toolkit for building ical systems. 00:13:46 reefknot seems to be dead, and it's also not the idea i'm looking for 00:13:53 reefknot is not a server 00:14:11 libical is of questionable life as well 00:15:06 i think the reefknot project may be dead as well 00:15:14 at least, it seems to be spread all over the place 00:21:45 Hmm, I can't find libical 00:22:40 talli: But free association has something done already. 00:23:33 Oh, it looks like free association _is_ libical 00:25:00 Not good when this happens: http://www.softwarestudio.org/libical/UsingLibical/node52.html 00:25:43 the oenome mozilla calendar ua is based off libical. 00:26:09 oenome? 00:26:14 paje: oenome? 00:26:15 rbm: wish i knew 00:26:38 a company that makes internet applicances running based on mozilla as the ui. 00:26:48 Is that the name of the project? 00:26:53 they donated the calendar project to mozilla 00:27:11 http://mozilla.org/projects/calendar/ 00:27:11 C: http://mozilla.org/projects/calendar/ from hazmat 00:27:18 google knows nothing about "oenome" 00:27:27 sorry oeone 00:27:36 oeone.com 00:27:46 ah 00:28:19 there is already a C API defined for vcalendar and vcard applications 00:28:42 so it's great if there is already something out there in liical 00:28:45 libical, sorry 00:28:58 but that's still not an application, right? 00:29:41 what's the motivation here, talli? 00:29:43 I'm not sure which parts of the spec does libical handle. 00:30:08 But it's funny that the mozilla guys are building an open source client when there's no open source server 00:30:22 vcal supercedes ical, libical does ical. afaik 00:31:29 * rbm goes back to his studies 00:31:37 the motivation is partly that there is nothing out there now 00:32:03 and partly that it would be great for clients 00:32:08 and i could really use it 00:32:48 is it something that is saleable to clients? 00:33:06 oh, no. of course not 00:33:25 i don't think that clients really care about scheduling or calendars. 00:33:39 ok, all the sarcasm is gone now 00:34:02 well, calendars aren't really my itch. So it's a fair question. 00:34:10 yes, of course. one could build a nice biz model off of an OSS calendaring system. but that's not all the motivation becuase i'm not going to try and raise money for this 00:34:17 i'm going to try and pull together a community 00:34:36 i don't think it needs a huge community of many hackers. just some good, disciplined C hackers 00:34:40 c ? 00:34:40 i think c is much more reliable than java... 00:34:41 why c? 00:34:57 but i could potentially get some big guys to kick in if it picks up steam 00:34:59 so it will run in apache, hazmat. 00:35:14 in the apache portable runtime, for one, hazmat 00:35:24 ah... ic. 00:35:31 then it becomes very portable 00:35:45 that's why i asked last night 00:35:49 why not use a higher level language? ruby, perl, python 00:36:17 i don't really care what language 00:36:31 that gives portability and is more RAD than c. 00:36:36 true enough 00:37:11 i'm not hte programmer. and a lot still needs to be worked out 00:37:18 but there's a lot of stuff laid out in C already 00:37:26 APIs and libraries, apparently 00:37:46 I think it'd be better done in a lower-level language. 00:38:15 why? 00:39:13 speed, no need for an interpreter running, other apps/langs could use a lib to have their own "server" 00:39:45 Python would be cool though, and a lot faster. 00:39:49 (to develop) 00:39:58 speed for a server is generally i/o bound, keep the data in a db, and you get access 00:40:14 and you can layer high level apis in a language of your choice. 00:40:15 generally, but not necessarily 00:41:13 the reason i would think C would be nice is the reason that AOLserver and Apache are nice 00:41:19 they do one thing, and then people build around it 00:41:34 build the foundation so it's rock solid, and then let people knock themselves out 00:42:35 to phrase it differently, unices, macs, and windows are all built on C, so C libraries have the simplest interface characteristics. 00:42:55 good point 00:44:28 but look at the development time and the amount of effort required to build those systems in c. 00:44:58 agreed, hazmat. 00:45:13 otoh i don't not know any serious c hackers. 00:46:30 davb (dave@alb-24-58-162-46.nycap.rr.com) has joined #openacs 00:47:42 paje: hazmat 00:47:42 rumour has it hazmat is somebody that drinks often from the river lethe 00:48:25 dlk has quit (Remote closed the connection) 00:51:07 anyway, i gotta go now guys. 00:51:12 talk to you more about this later 00:51:13 cheers 00:52:10 talli has quit ("Download Gaim [http://gaim.sourceforge.net/]") 01:44:16 denshi has quit () 02:31:20 http://www.boxesandarrows.com/ 02:31:20 D: http://www.boxesandarrows.com/ from davb 02:31:34 D:|Boxes and Arrows, peer-reviewed IA journal 02:31:34 titled item D 02:56:31 talli (~talli@lti-4.dialup.access.net) has joined #openacs 02:57:41 hey guys 02:58:01 here comes trouble 02:58:11 hi talli 02:58:17 and docwolf 02:58:34 so talli, whats the difference between ical and vcal? 02:58:37 hi 02:58:46 uh oh, talli's wacky calendar scheme. 02:58:54 how does he get involved with such projects? 02:59:02 not enough real work? 02:59:16 MY wacky calendar scheme? 02:59:25 :-) 02:59:40 i think that vcal is the "industry" version 02:59:42 i deny everything. true or not. 02:59:47 heh 02:59:50 i read through the vcal spec 02:59:53 its seriously tedious. 03:00:14 er.. the ical spec ;) 03:00:25 rfc2445 03:00:34 davb, ever hear of the ten commandments? 03:00:39 7 of them were docwolf's ideas 03:00:51 i suppose the most interesting part of this story is the mozilla calendar -- building a client with no server available. 03:01:05 andy_ has quit (devlin.openprojects.net irc.openprojects.net) 03:01:05 docwolf has quit (devlin.openprojects.net irc.openprojects.net) 03:01:05 rbm has quit (devlin.openprojects.net irc.openprojects.net) 03:01:07 rbm (rmello@fslc.usu.edu) has joined #openacs 03:01:21 maybe its on their list 03:01:21 "if you build it, they will come" 03:01:24 docwolf (~docwolf@adsl-34-182-127.bct.bellsouth.net) has joined #openacs 03:01:45 03:02:58 i have a name for the project, too 03:04:14 uh oh 03:04:40 anyone wanna hear it? or is everyone sick of docwolf by now and taking it out on me? 03:04:41 lets hear it!! 03:04:56 tallical? 03:05:38 talli has quit (Read error: 104 (Connection reset by peer)) 03:05:54 docwolf: you scared him away! 03:05:56 thats an exit ;) 03:06:06 andyn (~andy@12-254-190-230.client.attbi.com) has joined #openacs 03:06:11 has anyone seen the monty python in japanese with english subtitles? 03:07:02 hehe. another victory for docwolf. 03:08:47 talli (~talli@lti-4.dialup.access.net) has joined #openacs 03:08:48 talli has quit (Client Quit) 03:08:57 talli (~talli@lti-4.dialup.access.net) has joined #openacs 03:09:27 whoa, i kinda like tallical 03:09:29 it's done! 03:09:34 i bet i can find the URL too... 03:10:01 how tall is that ical? 03:10:19 that talli can be my cal anyday! 03:11:01 any, whaddya think about Momentum 03:11:12 momentum sounds good to me. 03:11:23 i don't know of any other project with the same name. 03:11:51 easy to pronounce, spell, and remember. 03:12:02 cool. 03:12:04 docwolf is the devil's child 03:12:07 i don't trust him 03:12:19 he only wants to manipulate others so he can use their software 03:12:21 hey, who else would buy your scrap metal? 03:12:44 good point 03:12:44 talli is becoming the fred sanford of the openacs community. 03:13:04 "becoming"? 03:13:05 actually it sounds like a product to keep you regular 03:13:05 i resent that 03:13:24 davb: it could be used for that purpose.... 03:13:26 davb, i think you're getting a little old 03:13:37 "2:00 -- boweling. Participants...." 03:13:44 oh man 03:13:47 heh 03:13:49 "very punny" 03:14:08 I am one of the youngest people ehere! 03:14:22 except for spork. 03:14:42 hey, you brought up regularity 03:14:57 anyway. this is the server. 03:15:36 we weren't gettimg enough email viruses at work, so we are switching to exchange/outlook 03:16:10 oh man 03:16:17 i'd love to meet the MS sales staff that actually has to pitch exchange 03:16:44 as far as I can tell, the scheduling is the only useful piece of the whole puzzle. 03:17:12 I can't imagine any other reason for introducing such garbage into an organization. 03:17:35 well all our other crap is MS, so... 03:22:02 davb: was it that they couldn't find expertise to admin a unix box, or did they actually need features of exchange not available elsewhere? 03:23:16 more that they would not even think of using a non-MS box. I am unfortunately not in a position to suggest things. We spent almost $10,000 on a server that would handle exchange, and file serving. 03:23:32 oh man 03:23:35 10K???? 03:23:35 but they do use scheduling. 03:24:10 its a really nice IBM with everything. 03:24:20 including IBM hard drives and RAM 03:24:39 heh 03:24:42 lame ;) 03:24:47 i agree with Spork 03:24:51 The IBM servers we have suck 03:24:52 yeah, me to. 03:25:08 The mouse or keyboard die on them through a KVM every other day 03:25:17 we had 4 drives fail within 24 hours one day 03:25:20 this one seems pretty good. it could probably serve a couple million web pages a day with openacs. 03:25:37 and they can't run Win2K better than my Athlon 750Mhz 03:25:41 is it a quad xeon? 03:25:46 no 03:25:52 Twin Xeon me thinks 03:25:56 nah, ours is dual pIII 03:26:01 holy 03:26:03 oh, right 03:26:03 maybe xeon 03:26:06 nad they spent 10K??? 03:26:07 our is a Dual PIII 03:26:14 I remember the post screen now 03:26:15 not quite. 03:26:22 Is it the IBM @server series? 03:26:28 probably. 03:26:30 we have the 420 or something in that range 03:26:36 We spent more than 10 :) 03:26:42 yours has more drives. 03:27:05 it has dual hot swappable power supplies, hard drives, hardware raid 03:27:29 yup 03:27:33 our old server running novell 4.11 is a PII 500 with 512MB ram and two 9 gig hard drives. 03:27:36 we have 4 SCSI drives 03:27:49 lucky for us 03:27:57 we had 2 drives fail on one box and not 3 03:28:03 It was able to rebuild the array 03:28:06 hopefully they are more reliable than the ide 03:28:10 yeah! 03:28:17 talilee (~talli@lti-4.dialup.access.net) has joined #openacs 03:29:10 * davb should be working 03:30:34 talli has quit (Read error: 54 (Connection reset by peer)) 03:30:42 talilee is now known as talli 03:32:38 * Spork should be sleeping 03:32:42 and he goes to sleep 03:32:43 :) 03:32:47 <--sleeping 03:46:26 davb has quit ("Client Exiting") 04:20:18 rbm, you around? 04:20:25 kind of 04:20:36 studying? 04:20:49 trying. 04:21:01 ok, we can chat tomorrow 04:21:22 great 04:21:23 tks :) 04:21:25 brb 04:21:26 np 04:24:13 ola talli 04:24:18 hey hazmat 04:24:26 how does that RFC look? 04:25:24 what's your take on the best way to implement something like this? 04:25:57 heinous, lots of contexts awareness needed, honestly i can understand why it hasn't been picked up by the os community. 04:26:06 in terms of implementing. 04:26:11 yeah, i sort of noticed that as well 04:26:17 a parser for the syntax and a date time library. 04:26:32 then a collection of objects for the various entities. 04:26:59 with marshalling via some sort of writer interface to format. 04:27:23 i've looked around there are a couple of nascent attempts at doing ical in xml, but they never seem to pick up speed. 04:27:34 yes 04:28:25 what is the bare minimum that can be done? 04:28:36 depends on what you want to do. 04:28:48 if you want a full interoperable server... 04:29:58 yes? 04:30:09 then implementing most of the three specs might be nesc. i'm not sure, i haven't read them in enough detail, plus the vagaries of client implementations have to be dealt with. 04:30:33 after seeing zope dealing with various webdav clients, i can say that it won't be fun :( 04:30:48 what i would hope is that the vagaries should be left for the clients to fix 04:30:49 esp. when ones from MS aren't compliant (a real shocker ;) ) 04:30:55 yeah, MS is the thing 04:31:00 they don't really care 04:31:12 nope, and their big enough that they can't be ignored. 04:31:16 would it be possible to sort of force compliance? 04:31:19 nope. 04:31:26 all of this is speculative 04:31:28 of course. 04:32:06 forcing compliance would involve possibly enlisting other large entities like ibm/lotus... 04:32:23 speculative since i don't know the state of compliance for ms ical usage. 04:32:24 one idea i had is that if one built the server logic, meaning built the system to interact with the DB, then you might provide a way for others to script the server to work with the client 04:32:33 bynari.com 04:32:48 so that you build the underlying mechanism, then let anyone build on top of that to fit it with their client 04:32:54 screw trying to work with every client 04:33:25 just provide a way for people to make the server talk to the client. perhaps embed a scripting language into the server to do that (python, perl, tcl, etc) 04:33:31 i saw bynari.com 04:34:20 does my idea just make things too complicated? 04:35:08 its hard to say. i haven't read any of protocol interaction/negotiation for exchange just the base level formatting of ical. 04:35:16 ok 04:35:31 i think some level abstraction would definitely be appropriate for the abstracting the client interaction. 04:39:06 http://www.imc.org/ietf-calendar/index.html 04:39:06 E: http://www.imc.org/ietf-calendar/index.html from talli 04:39:25 E: IETF Calendaring and Scheduling (calsch) Working Group 04:39:25 commented item E 04:41:37 talli: if you don't mind me asking how do you structure support contracts with clients? 04:41:49 good question ;) 04:42:01 only now have we begun getting clients that need support contract 04:42:03 contracts 04:42:08 so it's not so easy to lay out 04:42:15 what do you want to know? 04:43:21 not sure, i'm getting sucked into spending some time on a client that i worked with last year.. not sure whats appropriate to charge vs. just do as some obligation of consumer happiness... 04:44:14 i think that the best way to do it is to do it by incident 04:44:25 each month the client pays for X incidents 04:44:38 any incident over that they have to pay hourly 04:45:10 interesting.. 04:59:27 talli: thanks for the info 04:59:44 np 05:01:41 [Global Notice] Hi all. Europe is going to get a little crazy for a few moments as we upgrade the hub. Apologies for the inconvenience and thank you for using OPN. 05:03:35 docwolf has quit (devlin.openprojects.net irc.openprojects.net) 05:03:35 rbm has quit (devlin.openprojects.net irc.openprojects.net) 05:06:19 docwolf (~docwolf@adsl-34-182-127.bct.bellsouth.net) has joined #OpenACS 05:06:19 rbm (rmello@fslc.usu.edu) has joined #OpenACS 05:09:27 [Global Notice] That should be it for European rehubbing. Thanks again for your patience. 05:33:35 talli: still around? 05:48:37 talli: another thing to checkout syncml.org 05:57:08 [Global Notice] Hi all. One more server will be converted this morning, and this one has about 450 users, so it'll be a bit noisy. Please bear with us. 06:04:45 paje has quit (devlin.openprojects.net irc.openprojects.net) 06:04:45 hazmat has quit (devlin.openprojects.net irc.openprojects.net) 06:04:46 paje (~paje@slxwy.dorms.usu.edu) has joined #openacs 06:05:55 hazmat (~ender@adsl-66-123-57-58.dsl.lsan03.pacbell.net) has joined #openacs 06:07:43 jim (~jim@12-233-187-5.client.attbi.com) has joined #openacs 06:07:47 re 06:17:07 jim: Got my key signed. 06:17:18 hey hey :) 06:17:27 No AM yet though. 06:18:14 seems like they're progressing tho 06:18:33 Yeah. Remember I told you another guy here from USU had applied? 06:18:37 I think they take folks from the bottom of the list 06:18:43 umm I think so 06:19:04 Erik had signed his key. He had been waiting past the AM test for 8 months (!), and this week he was approved and the DAM created his account. 06:19:20 So I just had him sign my key instead :) 06:19:26 Much closer 06:19:55 cool :) did you sign his too? 06:20:55 yeah 06:21:33 wait, you mean this guy got approved shortly after he got his key signed? 06:22:35 No, he had had his key signed a long time ago. 06:23:07 was he on hold? 06:23:22 He was waiting for the AM, then for the DAM 06:23:42 No, he wasn't on hold. His AM just would take months to reply to his messages. 06:24:00 Why can't we just have more AMs? 27 seem insufficient. 06:24:03 who was the AM? 06:24:26 If we have 500 developers, there must be at least 10 others who would be qualified enough to help 06:24:35 oh, I don't remember. Let me see. 06:24:48 the NM people were originally 2-3 people who did all the nm stuff 06:25:10 (Martin Shulze (sp?) and James Troup) 06:25:35 at one point, they shut down nm for more than a year 06:29:40 wow 06:36:27 Well, i'm going to bed 06:36:29 g'nite 06:36:36 nite rbm 06:36:43 oh, b4 u go 06:36:54 (still here?) 06:37:26 yeah 06:38:02 are you? :) 06:38:04 I'm installing openacs on a debian, and trying to use daemontools, which isn't packaged that I know of 06:38:19 I don't think any of djb's tools can be packaged. 06:38:33 point, maybe we could add a section that reflects what I did 06:38:45 To the docs? Sure! 06:39:44 cool; meanwhile I have the aduni drive and a 100mb drive, gonna copy the aduni stuff to the 100mb, and have it also be a debian and an acs 06:40:04 You're going to make the aduni content available? That'd rock 06:40:28 available? I'm working on that part... 06:40:46 I'm not sure my isp will like me serving video content 06:40:58 however, I'll copy onto drives in my local area 06:41:21 and I'm also looking for mirrors for the video lectures 06:41:34 jim: How big are the lectures? 06:41:46 500mb apice 06:41:51 jim: I could talk to our CS department and find space in one of our servers. 06:41:55 well, one was 450 06:42:11