00:00:59 <Spork> hey jerry
00:01:07 <Spork> * Spork sporks dave with a psychephylax
00:01:31 <jerryasher> Hi Spork.
00:02:02 <Spork> :)
00:02:10 <Spork> got your e-mail...
00:02:27 <Spork> hang on...on the phone 
00:04:03 <jerryasher> I got my car fixed.  Bad Saab design: lots of electrical switches (windows, roof, ignition) lay 
00:04:25 <Spork> i read your qualifications...."Daaaaaaaamn!"
00:04:25 <jerryasher> in the bottom of the car, right on top of the transmission.
00:04:26 <jerryasher> Whenever anyone drops/eats anything, the switches foul up.
00:04:32 <Spork> lol
00:04:46 <Spork> I never liked Saab cars (I almost typed caars)
00:05:01 <Spork> They look kinda funky
00:05:04 <Spork> like camels with humps on their backs
00:05:32 <jerryasher> Yes, I once thought they were like the ugliest car around.  But the manual, turbo 900 convertible,
00:05:56 <jerryasher> is fast (fast enough), handles very well, is considered very safe, and well, it's a convertible.
00:06:19 <jerryasher> So if I'm inside it, I don't have to look at it, I can lower the roof and look at the bay.
00:06:53 <Spork> heh, see...you live in Cali...not owning a convertible there is like a felony or something
00:07:21 <jerryasher> it should be.  It should be: own a convertible or a motorcycle or go to jail.
00:07:21 <Spork> You try that in NYC and you'll find yourself freezing your @$$ off by middle of october
00:07:39 <Spork> of course I wouldn't know...I drive a subaru with a broken gas tank sensor
00:07:40 <jerryasher> I bought it when I lived in Arizona.  That is an incredibly wonderful state to own a convertible.
00:08:07 <jerryasher> Middle of the night, it's still above 90 degrees.
00:08:07 <Spork> nice
00:08:08 <jerryasher> It's a beautiful state day or night.
00:08:30 <jerryasher> Anyway, I hate to run on you once more, but I now, have a meeting in SF to get to by 6 (or miss out
00:08:34 <Spork> I live in NY...it's got its moments...but usually it just sucks
00:08:41 <Spork> yikes
00:08:42 <jerryasher> on the munchies and um, networking.)
00:08:51 <Spork> I keep forgetting it's -3 there
00:09:07 <jerryasher> Hey I went Apple Picking in Bear Mountain Park, and toured the Hudson Valley.  Had a wonderful time!
00:09:09 <Spork> btw..I think you misunderstand the problem 
00:09:28 <jerryasher> Could be.  I think so too, but let's talk via email or tomorrow, okay?
00:09:34 <Spork> sure thing
00:09:38 <jerryasher> see ya.
00:09:43 <Spork> enjoy your meeting :)
00:09:50 <jerryasher> jerryasher has quit
00:14:31 <davb> Hi Spork
00:22:46 <Spork> yo ;)
00:33:05 <davb> Spork: did you old Tcl script poll all those routers?
00:35:12 <Spork> yes
00:35:25 <Spork> it used asyn communications
00:35:35 <Spork> the ns_snmp module does not have async mode
00:35:41 <Spork> Therefore I have to find a workaround
00:35:55 <Spork> jerry offered to help, but I think he wants to get paid ;(
00:37:51 <davb> Yeah. Well a C module to handle that would be an effort.
00:38:14 <davb> How did it do asyn communications in Tcl?
00:38:38 <davb> aha. I see. 
00:39:03 <davb> I think you should do it outside AOLserver.
00:39:11 <GEM> http://www.wired.com/news/technology/0,1282,47156,00.html
00:39:11 <chump> A: http://www.wired.com/news/technology/0,1282,47156,00.html from GEM
00:39:38 <davb> A: That's cool!
00:39:38 <chump> commented item A
00:39:49 <GEM> /openacs you know it?
00:40:37 <Spork> Well, the C library (Scotty) has a built in mode for it
00:40:44 <GEM> I hadn't actually read it yet :-)
00:41:38 <davb> Aha, but nssnmp uses a different library.
00:41:55 <Spork> correct
00:42:01 <davb> Probably easier to write a C AOLserver module for scotty. 
00:42:05 <Spork> that and he sort of chose not to implement async mode
00:42:11 <davb> Well yeah.
00:42:21 <davb> Oh its in libsnmp++ or whatever?
00:42:22 <Spork> Well, what I think might be the best
00:42:39 <Spork> from my understanding there's a number of SNMP libraries 
00:42:49 <Spork> SNMP++ and Scotty are two of the ones I have heard about
00:44:10 <davb> Right. But an AOLserver module is pretty simple. It just creates a couple Tcl Objects and creates the Tc commands and obviously calls the proper C code to implement the commands.
00:44:16 <davb> not that I can write one.
00:44:25 <davb> I just looked at ns_xml.
00:45:11 <Spork> It might be easier to extend the snmp++ based ns_snmp
00:46:32 <davb> Could be. I am not a C hacker. If your work paid to have it developed it would probably have to be GPLed though...
00:46:38 <docwolf> docwolf has joined #openacs
00:46:41 <davb> depending on the license of that stuff.
00:46:46 <davb> Hey docwolf!
00:46:52 <docwolf> howdy
00:49:50 <Spork> heh I think it's GPLed
00:51:14 <Spork> If i had money, I'd pay
00:51:26 <Spork> but I don't know how well my boss will like the idea of outsourcing something
00:59:19 <davb> do you have any C hackers there?
01:08:13 <hazmat> hazmat has joined #openacs
01:10:02 <docwolf> docwolf has quit
01:18:26 <davb> davb has quit
01:18:26 <rbm> rbm has quit
01:18:26 <Spork> Spork has quit
01:18:43 <rbm> rbm has joined #openacs
01:18:43 <Spork> Spork has joined #openacs
01:18:43 <davb> davb has joined #openacs
01:19:01 <Spork> no idea about C hackers
01:19:17 <Spork> I don't have enough experience and I don't think my boss wants to do it (if he can)
01:21:51 <Spork> most of the people are doing Java at my place
01:21:56 <Spork> That's what I would do this in anywya
01:22:03 <Spork> I'd have a lot more help from other emps
02:27:42 <hazmat> hazmat has quit
02:27:42 <davb> davb has quit
02:27:42 <rbm> rbm has quit
02:27:42 <Spork> Spork has quit
02:27:45 <ChanServ> ChanServ has changed the topic to: Welcome to the OpenACS Helpline | http://www.openacs.org/
02:27:45 <ChanServ> This channel is logged: http://www.blogspace.com/openacs/chatlogs/ and blogged: http://www.thedesignexperience.org/openacs/ircblog
02:28:05 <hazmat> hazmat has joined #openacs
02:28:44 <davb> davb has joined #OpenACS
02:28:52 <rbm> rbm has joined #openacs
02:44:36 <Klatu> [GLOBAL NOTICE] Apologies for the splits... the regular services technician was unavailable and we had to get creative...
02:45:14 <Spork> Spork has joined #openacs
02:50:45 <AaronSw> * AaronSw heads off
02:51:12 <AaronSw> * AaronSw returns
02:53:24 <Spork> * Spork enters
02:53:30 <Spork> ;-D
02:55:38 <hazmat> hazmat has quit
02:55:38 <abbaJ> abbaJ has quit
02:55:38 <rbm> rbm has quit
02:55:44 <ChanServ> ChanServ has changed the topic to: Welcome to the OpenACS Helpline | http://www.openacs.org/
02:55:45 <ChanServ> This channel is logged: http://www.blogspace.com/openacs/chatlogs/ and blogged: http://www.thedesignexperience.org/openacs/ircblog
02:56:00 <hazmat> hazmat has joined #openacs
02:56:03 <rbm> rbm has joined #openacs
02:56:04 <abbaJ> abbaJ has joined #openacs
03:09:18 <abbaJ> abbaJ has quit
03:10:11 <abbaJ> abbaJ has joined #openacs
03:11:52 <hazmat> hazmat has quit
03:11:52 <abbaJ> abbaJ has quit
03:11:52 <rbm> rbm has quit
03:11:58 <ChanServ> ChanServ has changed the topic to: Welcome to the OpenACS Helpline | http://www.openacs.org/
03:11:58 <ChanServ> This channel is logged: http://www.blogspace.com/openacs/chatlogs/ and blogged: http://www.thedesignexperience.org/openacs/ircblog
03:12:12 <rbm> rbm has joined #openacs
03:12:12 <abbaJ> abbaJ has joined #openacs
03:12:17 <hazmat> hazmat has joined #openacs
03:14:19 <AaronSw> AaronSw has quit
03:14:53 <AaronSw> AaronSw has joined #openacs
03:15:07 <AaronSw> * AaronSw returns
03:15:13 <Spork> * Spork enters
03:52:33 <abbaJ> abbaJ has quit
03:53:08 <abbaJ> abbaJ has joined #openacs
07:09:17 <hazmat> hazmat has quit
07:09:23 <hazmat> hazmat has joined #openacs
07:09:24 <abbaJ> abbaJ has quit
07:09:25 <abbaJ> abbaJ has joined #openacs
07:09:56 <abbaJ> abbaJ has quit
07:34:31 <GEM> GEM has quit
07:34:36 <GEM> GEM has joined #openacs
07:38:14 <abbaJ> abbaJ has joined #openacs
07:46:31 <hazmat> hazmat has quit
07:46:31 <abbaJ> abbaJ has quit
07:46:31 <rbm> rbm has quit
07:46:31 <AaronSw> AaronSw has quit
07:46:31 <davb> davb has quit
07:46:31 <chump> chump has quit
07:46:37 <irc.openprojects.net> Disconnected from irc.openprojects.net (Connection reset by peer)
07:46:47 <loggy> loggy has joined #openacs
07:46:47 <carter.openprojects.net> Users on #openacs: loggy GEM abbaJ hazmat AaronSw chump davb rbm 
08:25:07 <hazmat> hazmat has quit
08:25:07 <abbaJ> abbaJ has quit
08:25:07 <GEM> GEM has quit
08:25:30 <hazmat> hazmat has joined #openacs
08:25:30 <abbaJ> abbaJ has joined #openacs
08:25:30 <GEM> GEM has joined #openacs
08:48:12 <GEM> GEM has quit
09:01:41 <ola> ola has joined #openacs
10:46:32 <abbaJ> abbaJ has quit
10:49:44 <davb> davb has quit
11:33:58 <hazmat> hazmat has quit
11:33:58 <chump> chump has quit
11:34:18 <chump> chump has joined #openacs
11:34:24 <hazmat> hazmat has joined #openacs
11:34:25 <ChanServ> This channel is logged: http://www.blogspace.com/openacs/chatlogs/ and blogged: http://www.thedesignexperience.org/openacs/ircblog
12:27:02 <Spork> Spork has joined #openacs
12:32:17 <ola> hey Spork!
12:39:23 <chump> chump has quit
12:39:23 <Spork> Spork has quit
12:39:29 <chump> chump has joined #openacs
12:39:29 <Spork> Spork has joined #openacs
12:44:57 <Spork> brrrr
12:58:14 <chump> chump has quit
12:58:14 <Spork> Spork has quit
12:58:23 <chump> chump has joined #openacs
12:58:23 <Spork> Spork has joined #openacs
12:58:45 <chump> chump has quit
12:58:45 <Spork> Spork has quit
12:59:14 <chump> chump has joined #openacs
12:59:14 <Spork> Spork has joined #openacs
12:59:28 <chump> chump has quit
12:59:28 <Spork> Spork has quit
12:59:31 <chump> chump has joined #openacs
12:59:31 <Spork> Spork has joined #openacs
13:03:17 <chump> chump has quit
13:03:17 <Spork> Spork has quit
13:03:31 <chump> chump has joined #openacs
13:03:31 <Spork> Spork has joined #openacs
13:04:11 <ola> oh boy! jenny jones and oprah on TV at the same time. that's a tough one ;-)
13:07:37 <davb> davb has joined #openacs
13:19:53 <chump> chump has quit
13:19:53 <Spork> Spork has quit
13:19:58 <chump> chump has joined #openacs
13:19:58 <Spork> Spork has joined #openacs
13:54:17 <ola> morning davb!
13:58:10 <davb> Hi ola
14:07:53 <GEM> GEM has joined #openacs
14:08:03 <davb> Morning GEM
14:09:00 <docwolf> docwolf has joined #openacs
14:10:58 <davb> Hi docwolf
14:42:08 <idware> idware has joined #openacs
14:42:14 <rbm> moo
14:45:16 <rbm> uggh
14:46:20 <docwolf> hi
14:46:58 <hazmat> hazmat has quit
14:46:58 <GEM> GEM has quit
14:46:58 <chump> chump has quit
14:46:58 <Spork> Spork has quit
14:46:58 <docwolf> docwolf has quit
14:46:58 <idware> idware has quit
14:47:43 <docwolf> docwolf has joined #openacs
14:47:43 <Spork> Spork has joined #openacs
14:47:43 <chump> chump has joined #openacs
14:47:43 <idware> idware has joined #openacs
14:47:43 <GEM> GEM has joined #openacs
14:47:43 <hazmat> hazmat has joined #openacs
14:47:46 <ChanServ> This channel is logged: http://www.blogspace.com/openacs/chatlogs/ and blogged: http://www.thedesignexperience.org/openacs/ircblog
15:24:21 <davb> Yeah! when you don't accidentally delete the -- for comments in your SQL file, it loads correctly.
15:33:14 <ola> cool!
15:33:44 <talli> talli has joined #openacs
15:33:57 <talli> hey guys
15:34:11 <davb> Hi talli.
15:34:21 <davb> Of course I found something else thats broken.
15:37:51 <ola> hey talli
15:39:17 <davb> aha.
15:45:41 <hazmat> hello
15:46:16 <ola> are subsites really supposed to consist of the same members!?
15:46:23 <ola> hi hazmat.
15:46:53 <davb> No
15:47:11 <davb> They all com from the big users pool. But you can restrict subsite membershit by group I think.
15:47:40 <davb> see how easy it is to sound like I actually know what I am talking about? :)
15:48:01 <ola> I believe you..
15:48:03 <ola> :-)
15:48:10 <hazmat> :)
15:48:44 <davb> Hello hazmat
15:48:58 <davb> yippee! It still works
15:49:34 <hazmat> hi guys, i have a really stupid question, how do i do a cvs checkout as a committer?, i'm looking for the docs in file storage as i write...
15:49:39 <davb> now I have to get DanW to apply my new patches to content-folder and content-item
15:50:25 <hazmat> got it
15:50:29 <davb> cool.
15:50:36 <davb> I can't look at the file in new-file-storage.
15:50:52 <hazmat> hmm.. thats strange
15:51:20 <hazmat> if i checkout the whole openacs-4 from cvs via ssh, can  i commit on the relevant parts, or do i need to check out the modules i'm assigned?
15:51:35 <davb> don't co openacs-4 just acs-core
15:51:39 <davb> But yes.
15:51:52 <davb> If you are logged in with your openacs account to CVS
15:52:02 <hazmat> why not checkout openacs-4?
15:52:33 <davb> its ALL the packages. Takes a long time to install even though it doesn't install them.
15:52:42 <davb> Just co acs-core and the ones you are working on.
15:52:47 <davb> http://openacs.org/new-file-storage/one-file?file_id=205
15:52:48 <chump> A: http://openacs.org/new-file-storage/one-file?file_id=205 from davb
15:53:00 <davb> A:|Restart-Aolserver script for use with Daemontools
15:53:00 <chump> titled item A
15:53:26 <davb> http://openacs.org/new-file-storage/one-file?file_id=206
15:53:26 <chump> B: http://openacs.org/new-file-storage/one-file?file_id=206 from davb
15:53:36 <davb> B: RUN script for AOLserver and Daemontools
15:53:36 <chump> commented item B
15:55:22 <davb> OK. I have to store the path of static pages relative to the OpenACS install dir. Should I just do a string trimleft /web/openacs/ file-path ?
15:55:39 <davb> using the actual install dir of course.
15:56:20 <Psychephylax> Psychephylax has joined #openacs
15:56:27 <davb> hey Psychephylax
15:56:30 <Psychephylax> hidly ho neighborinos
15:57:32 <docwolf> tippety-top of the am to you
15:58:54 <Psychephylax> :D
15:59:47 <Psychephylax> Jerry is not around? 
15:59:48 <Psychephylax> shucks
16:00:20 <Psychephylax> I should be a QA guy or something
16:00:26 <Psychephylax> I found a bug in my cell phone software
16:00:59 <davb> heh
16:01:19 <Psychephylax> Yeah
16:01:39 <Psychephylax> my new  month started so I figured "Let's reset the total accrued time" 
16:01:51 <Psychephylax> Went to do it...only to find out there is no way of doing that
16:01:55 <Psychephylax> "D'oh!"
16:02:23 <Psychephylax> So I called up Sprint and they were also confused
16:02:30 <Psychephylax> He said he'll e-mail me
16:03:45 <Psychephylax> anyhoo
16:03:49 <Psychephylax> I gotta go to class :(
16:03:51 <Psychephylax> Moo!
16:05:11 <davb> anyone know what ad_conn file returns? I am guessing its the full path to the file in the filesystem.
16:05:34 <Psychephylax> I believe so
16:05:38 <Psychephylax> but don't quote me
16:06:25 <davb> aha, it is. API browser to the rescue.
16:06:43 <davb> I think I will create a new proc.
16:06:55 <Spork> ok
16:07:05 <Spork> if I have uptime;w
16:07:12 <Spork> it executes uptime and then w
16:07:23 <Spork> is there a way to do it so the output is all on one line?
16:07:49 <Spork> like I want to do something like echo #;uptime but have it on one line
16:07:55 <Psychephylax> Psychephylax has quit
16:09:06 <Spork> w00t
16:09:39 <Spork> Syntax error: end of file unexpected (expecting ")")
16:09:56 <Spork> Syntax error: end of file unexpected (expecting ")")
16:10:06 <Spork> {echo: not found
16:10:16 <Spork> hmmm
16:10:38 <Spork> awww yeah
16:10:50 <Spork> it's /exec -o echo "#" `uptime`
16:11:18 <Spork> Don't mind me...I'm just mentally challenged
16:21:51 <rbm> moo
16:24:11 <Spork> oom
16:24:13 <Spork> gotta go
16:29:50 <davb> if I do:
16:29:56 <davb> for v_stale_file row in
16:30:01 <davb> some complex query
16:30:03 <davb> loop
16:30:15 <davb> PERFORM static_page__delete(v_stale_file_row);
16:30:29 <davb> and the query returned nothing should I get this error?
16:30:42 <davb>  ERROR:  Attribute 'v_stale_file_row' not found
16:31:37 <idware> idware has quit
16:34:52 <hazmat> hmm... what happens if the query doesn't return any results?
16:35:41 <davb> thats my question. I guess I have to run it in psql to find out.
16:36:12 <davb> postgresql is non-specific when an error occurs in pl/pgsql function called inside a pl/pgsql function
16:38:18 <hazmat> sigh... sometimes i really like oracle..
16:44:23 <hazmat> yipee! i got zope embedded in inprocess in aolserver:)
16:47:13 <davb> wow.
16:47:51 <davb> So far, the time I save dropping and recreating a database makes up for the cryptic error messages.
16:51:26 <hazmat> but programming in pl/sql is a whole lot nicer than pl/pgsql esp when you get nice tools like www.sf.net/projects/tora
16:52:47 <davb> neat!
16:53:08 <davb> http://www.sf.net/projects/tora
16:53:09 <chump> C: http://www.sf.net/projects/tora from davb
16:53:15 <davb> C:|Toolkit For Oracle
16:53:15 <chump> titled item C
16:54:52 <davb> hmmm here is my DECLARE
16:54:54 <davb> declare
16:54:55 <davb>                 p_session_id            alias for $1;
16:54:55 <davb>                 p_package_id            alias for $2;
16:54:55 <davb>                 v_root_folder_id        sp_folders.folder_id%TYPE;
16:54:55 <davb>                 v_stale_file_row        RECORD;
16:54:55 <davb>                 v_stale_folder_row      RECORD;
16:54:57 <davb>         begin
16:55:31 <davb> I am assuming this is where the error is occuring:
16:55:36 <davb>  loop
16:55:36 <davb>                        PERFORM static_page__delete(v_stale_file_row);
16:55:36 <davb>                 end loop;
16:58:43 <rbm> when does this loop stop?
17:00:17 <davb> I left out the query. 
17:01:58 <rbm> You know that PERFORM discards the results of the query don't you?
17:02:07 <davb> yes. 
17:02:35 <davb> Postgresql is saying attribute v_stale_file_row not found
17:03:26 <rbm> Oh. You can't pass a whole record to the function. you have to do: static_page__delete(v_stale_file_row.COLUMN);
17:03:46 <davb> right!
17:03:59 <rbm> Where COLUMN is the column (assigned to the RECORD var) whose value you want to pass
17:04:11 <davb> thanks!
17:04:48 <rbm> np
17:05:09 <davb> the variable for "for" has to be a record or rowtype though right?
17:06:33 <davb> what if I have this:
17:06:36 <davb>  for v_stale_folder_row in
17:06:36 <davb>                         select dead.folder_id from
17:06:36 <davb>                         (select folder_id from sp_folders
17:06:36 <davb>                                 where (folder_id) not in (
17:06:36 <davb>                                         select folder_id
17:06:37 <davb>                                         from sp_extant_folders
17:06:39 <davb>                                         where session_id = p_session_id
17:06:41 <davb>                                 )
17:06:43 <davb>                         ) dead,
17:06:52 <davb> how do I refer to the result of that query in v_stale_folder_row?
17:07:47 <davb> or do I need a SELECT blah AS foo?
17:08:10 <rbm> davb: Yes. It has to be a RECORD or ROWTYPE (?)
17:08:31 <davb> OK. So every if I want one column I need to do it the same way.
17:08:36 <davb> s/every/even
17:08:37 <rbm> davb: I put an example in the PL/pgSQL. You should read it. Actually, 2 examples.
17:08:48 <davb> OK
17:11:09 <davb> oh no, the dreaded  ERROR:  parser: parse error at or near ""
17:15:38 <rbm> heheh
17:15:49 <rbm> Check your quotes. See if you escaped all of them
17:18:15 <talli> rbm: are you trying to spoil our little secret here?
17:18:21 <talli> this is holy ground!
17:18:30 <talli> we can't just let any infidel into the IRC channel!
17:18:45 <talli> christ, man! think before you speak! or post for that matter!
17:19:16 <rbm> heh :)
17:19:57 <rbm> * rbm sacrifices a lizzard for the IRC gods
17:21:09 <hazmat> ahh... i'm burning...
17:23:09 <talli> * talli launches an attack on afghanistan because they are threatening rbm's inventory of sacrificial lizards
17:29:27 <hazmat> talli: your cvs karma is dropping
17:33:18 <talli> i have cvs karma?
17:33:28 <talli> i thought i was jewish.
17:33:55 <talli> does cvs karma have anything to do with bin laden?
17:34:47 <hazmat> i'm not sure, only one way to find out...
17:37:54 <hazmat> so to get search working, you have to checkout openfts in addition to acs-core,  compile openfts for pg and aolserver, install the openfts and search packages, and say the magic word, am i missing anything?
17:39:18 <ola> no. I don't think so.
17:39:55 <ola> you'll need the openfts-driver...
17:40:50 <hazmat> ola: you mean the aolserver module?
17:41:01 <hazmat> nevermind
17:41:49 <ola> so far only notes is adapted to the new search, afaik.
17:47:42 <hazmat> and the cr
17:52:46 <davb> should RAISE NOTICE in pl/pgsql notice go into the server log?
17:54:35 <davb> Aha, yes they do.
17:54:46 <davb> If the function makes it that far...
17:59:01 <rbm> * rbm heads to class
18:01:55 <markd2> markd2 has joined #openacs
18:02:23 <ola> hey markd2.
18:02:30 <markd2> heya
18:07:15 <nkd> nkd has joined #openacs
18:08:00 <ola> hi nkd.
18:08:07 <davb> hello nkd
18:08:23 <nkd> hello everybody
18:09:31 <ola> how's cyprus today?
18:10:03 <nkd> I was just checking Roberto's post to the forums and thought it would be nice to join for a while...BTW, it's neophytosd
18:10:18 <nkd> the weather is good as usual
18:11:09 <ola> cool. Nicosia?
18:11:27 <hazmat> hi neophytos
18:11:32 <nkd> hi
18:11:35 <nkd> Yes Nicosia
18:11:43 <ola> :-)
18:11:51 <nkd> I am back to the University for the fall semester.
18:11:57 <hazmat> undergrad?
18:12:23 <nkd> I have just finished. I am working now as a special scientist (TA, Research) for a year.
18:12:55 <hazmat> sounds like fun, how is the workflow package going
18:13:05 <nkd> It's almost datamodel.
18:13:17 <hazmat> you're rewriting it? why?
18:13:34 <nkd> aD release 4.3 after Dan had finished porting 4.1.1
18:13:55 <nkd> it has some pretty good features (roles, better ui) so I thought it would be nice to port it
18:14:22 <hazmat> oh.. that makes sense, does the ui allow web configurable workflows?
18:14:27 <nkd> we are trying to have it ready for the first release but I am only able to work on it during the weekends
18:14:47 <nkd> what do you mean by "web configurable"?
18:15:27 <hazmat> i can define by states and transitions via web forms for a given object type, or something similiar, basically allow ttw customization of workflow
18:16:05 <nkd> The 4.1.1 provides a user friendly interface for designing workflows
18:16:35 <nkd> 4.3 allows the user to specify roles (e.g. editor) instead of static assignments
18:17:04 <hazmat> cool.
18:17:15 <markd2> Is this 4.3 Tcl, or the Workflow that's in ACS/Java?
18:17:23 <nkd> 4.3 tcl
18:17:25 <hazmat> tcl
18:17:35 <markd2> cool
18:17:44 <markd2> they took out a lot of stuff in acs/java
18:18:01 <nkd> I haven't checked it out to be honest
18:18:18 <hazmat> markd2:  you mean acsjava4.6 or 4
18:18:18 <markd2> it's DAGs instead of Petri nets
18:18:26 <markd2> 4.6
18:18:33 <markd2> the re-re-re-(re?)-write
18:18:56 <nkd> :) Well petri nets work well so far. I don't know why aD choose to change it
18:19:01 <hazmat> interesting, i kinda like the petri nets, but dags are almost impossible to do well in a purely rdbms
18:19:25 <markd2> I think it was one of those "woo hoo!  let's re-write it!  oh no!  it's harder than we thought to do.  Let's simplify it to just what CMS needs"
18:19:40 <nkd> I think, in the future, we should move some of the functionality to the workflow package (e.g. forum moderation, user approval, etc).
18:19:58 <hazmat> i just gave a talk yesterday at ad @ la, their closing office here and berkeley...
18:20:15 <nkd> There are a lot more uses for the workflow package (see previous post) than just what the CMS needs
18:20:44 <hazmat> workflow will be instrumental to alot of possible packages
18:20:50 <nkd> yes
18:22:51 <nkd> hazmat=???
18:23:12 <markd2> Hazardous Materials
18:23:17 <nkd> :)
18:23:23 <hazmat> :)
18:23:59 <hazmat> =kapil
18:24:27 <nkd> hey Kapil.
18:24:32 <talli> hey guys
18:24:36 <hazmat> hi neophytos
18:24:38 <nkd> hi Tally
18:24:47 <talli> i think you're opening a weird can a worms by hoping to use workflow more
18:24:54 <hazmat> why is that
18:25:08 <talli> lars is a massive ass kicker, and he did a great job, but it's a very complex package
18:25:17 <nkd> yes that's true
18:25:19 <nkd> but...
18:25:19 <talli> we're building a cool system with it, and it's what we need.
18:25:32 <hazmat> its better than alot of little explicits all over the place that require coding to fix
18:25:40 <nkd> agree
18:25:58 <talli> not really
18:26:10 <talli> how complex is forum moderation and user approval?
18:26:26 <nkd> not too complex, but..
18:26:31 <talli> do you really want to force people to download the acs-core and install workflow to do all that stuff?
18:26:38 <hazmat> the question is, do you want workflow to be end user configurable for a given package
18:26:50 <davb> hazmat: right
18:26:54 <nkd> if we use the workflow package the user will only have to check "my tasks"
18:27:18 <nkd> unified interface for every task that needs to be done
18:27:27 <hazmat> workflow is in the core btw.
18:27:33 <talli> hmm... that's not entirely clear to me
18:27:46 <talli> it may be in the core, but it's not used for the core functionality
18:28:16 <nkd> what's not clear (unified interface?)
18:28:24 <talli> yes, more or less
18:28:35 <talli> i don't see how that is explicity linked to workflow
18:28:38 <ola> * ola likes unified interfaces
18:29:21 <talli> also, the BIGGEST problem with the acs right now, from a user perspective, is the pathetic interface for user administration
18:29:27 <talli> assigning permissions and so forth
18:29:28 <nkd> sure
18:29:54 <hazmat> people have been bitching about it for ages, it only takes one person to fix it
18:30:07 <talli> no, not really
18:30:11 <talli> it's not an easy solution
18:30:32 <talli> it's a very complicated thing that has implications for ever package in the system
18:30:46 <talli> it consists of UI and functional code
18:31:10 <talli> workflow is a good solution for a lot of stuff, but mostly for systems that are truly "enterprise"
18:31:34 <talli> for instance, luke built ETP precisely because he couldn't figure out how to use the aD CMS which relied heavily on workflow
18:31:39 <nkd> the calendar package did something that it is close to what I consider as a solution
18:31:48 <hazmat> i disagree... any intranet that does project management can have a state machine that requires the flexibility of customized workflow
18:31:50 <talli> which is entirely uselsess for 99% of the systems out there
18:32:08 <talli> sure, but that is very different from requiring workflow
18:32:34 <hazmat> workflow is in the core, people get it with the acs, isn't it default install?
18:32:41 <talli> sure.
18:32:42 <nkd> It is important for companies pursuing iso quality standards (formalize their processes) what better way to do it
18:32:47 <talli> but that doesn't mean it's easy
18:32:55 <hazmat> so write some docs
18:33:19 <talli> that's no answer to why it should be required
18:33:42 <talli> 90% of the people who will use openacs will be using it for their boy scout troop
18:33:52 <cro> cro has joined #openacs
18:33:56 <talli> the last 10% will need workflow for enterprise systems
18:34:03 <davb> They don;t need to SEE workflow though.
18:34:31 <davb> only if they want to change the default processes.
18:34:32 <talli> if workflow could be built so that it's totally transparent, that's great.
18:35:15 <talli> but i still don't see why the docs should be written to implement workflow rather than to just figure out how to publish your html page
18:35:55 <nkd> transparent for user:yes, developer:no, workflow designer:yes
18:36:05 <davb> nkd: exactly
18:36:12 <nkd> e.g.user=moderator
18:36:21 <talli> that's really the big mistake aD made with 4tcl, then made it even worse with java acs
18:36:34 <davb> all the guts are exposed?
18:36:43 <hazmat> talli: ??
18:36:53 <hazmat> whats the big mistake
18:36:54 <talli> why make it more complex than it has to be? why not just make the it easy, then allow it to be flexible enough to increase the complexity?
18:37:09 <talli> for instance, what the fuck is bebop?
18:37:15 <nkd> don't know
18:37:19 <nkd> :)
18:37:35 <talli> why the hell did they spend so much time on the CMS workflow and fuck up the UI so badly?
18:37:39 <hazmat> talli:  i've worked with systems that try this from the other  end, ie try to make things better for the end user/manager it doesn't work, you have to make the system for the developer who will use it
18:37:46 <hazmat> bebop = xslt ui components
18:37:55 <nkd> thanks
18:38:06 <talli> no one ever got to workflow because the Ui was so bad
18:38:16 <markd2> bebop is bringing a Swing like UI fr web pages
18:38:17 <talli> bebop is that theoretically
18:38:29 <markd2> now, *why* swing for web pages is considered a good idea, that's the puzzler
18:38:38 <talli> markd: exactly
18:38:55 <hazmat> reuse is the name of the game
18:39:13 <talli> ah... just like zope, huh, kapil? :)
18:39:53 <talli> it's only reusable if it works and you never have to touch it. if it requires you to fix the system, then spend a lot of time configuring it, then the reuse is just over the top
18:39:56 <nkd> not for 4.3
18:39:56 <nkd> 4.3 has a nice ui
18:39:57 <nkd> you can specify, the panels, the workflow, the roles. Then, all you have to do is instantiate the workflow when you create a record (e.g. new bboard posting)
18:40:16 <nkd> BTW, I might be missing a few messages. My connection drops every while (about 5 minutes)
18:40:32 <hazmat> no not like zope
18:40:41 <hazmat> bebop that is
18:41:03 <nkd> panels=templates(tcl,adp)
18:41:07 <hazmat> zope went an xmlc ish route with page templates, using a atribute replacement language
18:41:13 <nkd> roles=moderator,editor,publisher,etc
18:41:17 <talli> i meant code reuse
18:41:34 <talli> nkd: i know, we're building a very fancy package using the workflow package
18:41:42 <talli> it's a double blind peer review system
18:41:48 <hazmat> openacs is all about code reuse as well, to me zope is about flexibility
18:42:04 <nkd> acs-sc, yeah :)
18:42:18 <hazmat> and speed
18:42:19 <talli> hey let's not get into an OOP argument in this channel :)
18:42:21 <hazmat> of development
18:42:53 <talli> anyway, workflow is very nice, and lars did a great job, but that doesn't mean it's easy
18:42:57 <hazmat> if it makes you feel any better i'm doing my latest project with openacs and python of course:)
18:43:07 <talli> using pywx? that's cool
18:43:28 <hazmat> yup, i got sidetracked and embedded zope in process in aolserver:) cool hack of the week
18:43:40 <talli> i could care less about the openacs code, to be honest. the real deal is aolserver
18:43:50 <talli> the other stuff is just candy
18:44:15 <hazmat> lost me there....
18:44:16 <talli> we're goign to do a project without any openacs code, just PG and aolserver. but we'll still enjoy rapid dev because of the power of aolserver
18:44:49 <talli> openacs is cool because of the data model, a lot of code, etc. but in fact, it wouldn't be nearly a nice system if it relied on apache
18:45:13 <talli> IOW, the only reason the system is in tcl is because of AOLserver.
18:45:14 <hazmat> aolserver is a really nice platform, but it still takes a good amount of time to build up a community infrastructure, and the acs db and templating calls are really  nice
18:45:31 <talli> right, but that stuff is not unique.
18:46:14 <talli> in other words, some system development is slowed or frustrated because of the limitations of the openacs core
18:46:16 <hazmat> the openacs hands you prebuilt functionality and extensibility that is hard to beat, otoh if you know requirements won't change do whats fastest.
18:46:24 <talli> exactly
18:47:10 <talli> if you have a project that needs X, Y and Z that is different than how OACS implements those pieces, that does not rule out aolserver at all
18:47:12 <hazmat> are u using the db api calls from acs in your development
18:47:30 <talli> well, we're bidding on the project
18:47:41 <talli> and we'll build a custom system rather than a customized ACS system
18:47:47 <hazmat> good luck.
18:47:55 <talli> using AOLserver, PG and something else
18:48:11 <talli> the reason is, btw, that we will need to work with a legacy data model
18:48:19 <nkd> something else=openfts?
18:48:23 <nkd> :)
18:48:47 <talli> if search is needed, then yes openfts will be a definite possibility
18:48:54 <talli> but this is a DB app, not realy a site
18:49:02 <nkd> ok
18:49:39 <nkd> Oleg and Teodor released tsearch as a contrib/module which replicates some of the functionality of openfts. In the future OpenFTS will make use of tsearch
18:50:11 <nkd> you can have something like the "contains" clause in Oracle, inside PG
18:51:09 <talli> who are these guys, oleg and teodor?
18:51:14 <talli> how do you know them?
18:51:20 <nkd> the russian developers
18:51:22 <nkd> of openfts
18:51:44 <talli> how do you know them?
18:51:53 <nkd> I have contact them when I was researching their search solution for rumbler.ru
18:52:06 <nkd> then we team up to release openfts under GPL
18:52:22 <nkd> and now we are good friends. Teodor visited Cyprus during the summer
18:52:31 <talli> cool
18:52:58 <nkd> yes, they a lot about searching. I have learned a lot from them
18:53:07 <nkd> ...they know a lot ...
18:53:23 <talli> are they doctoral students, or professors?
18:53:37 <nkd> I think Oleg is but I am not sure
18:54:53 <nkd> He certainly affiliated with Moscow University but I don't know the details
18:55:16 <nkd> (lost my connection, repeat if you asked something else)
18:58:18 <nkd> Does anyone know what exactly happened with Philip and aD? (after the legal/trial)
18:58:36 <talli> philip got $5 mil to shut up and go away
18:58:44 <markd2> exactly happened, nobdy but aD management and Philip
18:58:50 <markd2> accepted rumor is what talli said
18:59:03 <markd2> although I heard higer numbers
18:59:11 <nkd> :) ok, thank you. higer???
18:59:18 <nkd> bigger
18:59:21 <docwolf> yo
18:59:37 <nkd> docwolf???
18:59:38 <talli> docwolf heard money talk and came running :)
18:59:49 <nkd> is it roberto?
18:59:51 <docwolf> heh. it's an instinct
18:59:56 <hazmat> :)
18:59:58 <markd2> heh
19:00:29 <talli> roberto is rbm
19:00:44 <nkd> so, who's docwolf. BTW, I have a great time here.
19:00:50 <docwolf> I AM THE DOCWOLF
19:00:52 <docwolf> heh
19:00:56 <nkd> :))
19:00:56 <talli> docwolf is adam wolf farkas, md, mba, dds, esq, bbw
19:01:02 <nkd> :)
19:01:09 <docwolf> hey! no real names!
19:01:10 <docwolf> ;-)
19:01:15 <talli> sorry
19:01:20 <talli> but you are bbw
19:01:26 <talli> and swf
19:01:32 <talli> but not so much vip
19:01:33 <docwolf> rotfl.
19:01:43 <docwolf> am i 1337?
19:01:52 <talli> who, i need to get te 411 on that
19:01:54 <markd2> d0k w()lpH iz l33t
19:02:08 <docwolf> thank you. I feel validated.
19:02:09 <docwolf> :-)
19:02:16 <markd2> * markd2 thanks rzolf for teaching him l33tsp33k
19:02:23 <talli> oh christ.
19:02:25 <docwolf> haha
19:02:28 <talli> you learned something from rolf?
19:02:48 <nkd> ROTFL
19:02:53 <markd2> ROLFTL
19:03:01 <hazmat> its been fun guys, but i'm gonna try to make my day productive:), cheers
19:03:02 <markd2> actually, I learned a lot from rolf
19:03:08 <talli> the only thing i've learned from rzolf is that there's always a way to cynically comment on the world
19:03:18 <talli> wait - maybe i taught him that
19:03:28 <talli> whatever - it was a night of passion either way!
19:03:34 <talli> later kapil
19:03:59 <davb> now I am getting the informative: ,ERROR:  parser: parse error at or near "select"
19:04:05 <hazmat> hazmat has left #openacs
19:04:20 <talli> markd: what is ROLFTL?
19:04:24 <davb> I cut and pasted the query into psql and it works fine.
19:04:59 <markd2> talli: no idea.  'rotfl' makes my dyselxia read 'rolf'
19:05:23 <talli> ah good point.
19:05:25 <davb> * davb checks his email instead
19:05:51 <talli> i must admin i never noticed how close "rolf" is to "rotfl" i wouldn't be surprised if he was in on that one
19:06:09 <markd2> rzolf is omnipresent
19:06:31 <cro> cro has quit
19:12:16 <nkd> Tom has just posted about the nsjabber module he has been working on
19:12:57 <talli> yeah, 
19:13:22 <talli> his cronjob module sounds a bit cooler, though. or at least more immediately useful
19:15:05 <nkd> yes, I missed that
19:17:23 <nkd> have to go. have a nice day guys
19:17:33 <davb> bye
19:17:39 <davb> Spork : http://www.dqd.com/~mayoff/aolserver/dqd_threadpool-1.0.tar.gz
19:17:41 <davb> http://www.dqd.com/~mayoff/aolserver/dqd_threadpool-1.0.tar.gz
19:17:41 <chump> D: http://www.dqd.com/~mayoff/aolserver/dqd_threadpool-1.0.tar.gz from davb
19:18:02 <davb> D: Threadpool and Job Queue Manager for AOLServer
19:18:03 <chump> commented item D
19:18:14 <talli> later neo
19:18:25 <nkd> nkd has left #openacs
19:22:21 <markd2> markd2 has quit
19:30:15 <jerryasher> jerryasher has joined #openacs
19:33:20 <docwolf> docwolf has quit
19:39:30 <talli> talli has quit
20:26:10 <jerryasher> jerryasher has quit
20:26:10 <chump> chump has quit
20:26:10 <Spork> Spork has quit
20:26:56 <davb> davb has quit
20:27:11 <chump> chump has joined #openacs
20:29:43 <hazmat> hazmat has joined #openacs
20:31:40 <ola> ola has quit
21:43:47 <talli> talli has joined #openacs
21:43:53 <talli> hey
21:44:16 <talli> hey kapil, i spoke with lars about the workflow package being integrated throughout the system, as you were advocating
21:44:27 <talli> he said this is what the package was originally intended to do
21:44:53 <talli> he had to go before i could find out whether the package was ready for this to be done, but i'll follow up
21:45:13 <talli> so i will apologize and gladly support this move if it actually can work
21:52:29 <hazmat> cool, good to know.
21:52:56 <hazmat> don't worry about the apology, friendly discussion is good for the soul :) and karma too
21:52:58 <vinod> vinod has joined #openacs
21:56:08 <rbm> moo
21:58:11 <hazmat> you
21:58:28 <hazmat> cat
22:04:02 <talli> i'll speak more with lars about this stuff. 
22:07:49 <talli> hey vinod and kapil, are you guys working on SDM?
22:08:39 <talli> vinod: brett favre has a hairy back
22:14:10 <vinod> talli: you've seen brett favre's back?? you are now my idol
22:15:16 <vinod> i've thought about the SDM a little, but i haven't hacked much yet. unfortunately, i've been busy with other (non-openacs) stuff.
22:19:08 <talli> what? like someone's irregular bowel movements?
22:19:12 <hazmat> talli: i'm in the same boat as vinod, thought about it, been too busy with contracts to do much about making ideas reality/ why?
22:19:12 <talli> c'mon, man!
22:19:30 <talli> no prob.
22:19:43 <talli> well, we 're hoping to release the new site soon
22:20:05 <vinod> talli: have you been speaking to my gastroenterologist?
22:20:27 <talli> uh huh!!
22:20:54 <talli> hazmat: don't worry about it. working on projects is way more important
22:20:59 <vinod> don posted recently that he wasn't too impressed with the 4.x ticket tracker and would rather start from the 3.x sdm, so i'd have to see what his thoughts were
22:21:45 <talli> yeah, i asked him what was up and he said that you guys must be real busy
22:22:10 <talli> he said that it was importnat enough to him that he might pick it up himself. you might want to ping him to find out what's up
22:22:48 <vinod> i will real-soon-now (TM) :-)
22:22:53 <talli> cool
22:22:57 <hazmat> talli: actually, i'm getting ready to finish my last contract, but i've pro-bono commited to doing the python-repository site (cpan for python) (hence the openacs+python site) at the moment
22:23:18 <vinod> cool!
22:24:22 <rbm> hazmat: really? That's cool. 
22:24:27 <rbm> I'm starting into Python myself.
22:24:58 <rbm> I managed to convince my boss to stay away from VB and use Python + tkinter for a project we're going to go into.
22:25:05 <hazmat> i'm excited about it, rbm: python is hard not to like once you get over the indentation, enjoy!
22:25:18 <rbm> Oh, I love it already.
22:25:38 <hazmat> o... tkinter is not nice, you're better off using another graphics toolkit....
22:25:51 <hazmat> unless the app is simple
22:25:52 <rbm> I don't have a problem with the indentation (I'd do it anyways), although I'd be happier if I could use block delimiters.
22:26:15 <rbm> hazmat: It's the only free option we have AFAIK. It needs to run on UNIX, Win32 and Macs
22:26:29 <hazmat> oh
22:26:42 <hazmat> you said vb, and i thought win32
22:26:44 <talli> yes, even though tcl is great within the context of aolserver and there's no technical reason not to use it...
22:26:49 <rbm> I looked at wxWindows, and liked it. But wxPython on the mac is not working.
22:26:58 <talli> it would be real nice if it were in python, at lesat from a marketing perspective
22:27:04 <hazmat> there is qt, if you can afford it
22:27:13 <rbm> hazmat: That was one of the main reasons why I was able to convince them :)
22:27:26 <rbm> hazmat: we can't. It's a project to build software for schools.
22:27:34 <talli> rbm: do you think that the gatekeepers would be open to OACS modules written in python?
22:27:38 <rbm> I would like to use Qt actually.
22:27:38 <vinod> vinod has quit
22:27:46 <hazmat> i was wondering about that myself
22:28:05 <hazmat> cause i have some cool zope products sitting around that would be cool for openacs as well
22:28:31 <rbm> talli: I don't see why not. The role of the gatekeepers is to guard the core OpenACS. But I don't see why we would be against you (e.g.) writing something in Python.
22:29:19 <talli> there was some discussion about this originally about the difficulty in maintaining a toolkit written in different languages
22:29:24 <hazmat> don't forget openacs-dev-tools:) the extractor and friends, not the same as package for download...
22:29:33 <hazmat> not like there isn't perl in oacs
22:30:00 <talli> i believe that i raised the point that if someone paid us to build them an oacs in VB, i would do it, but it would be expensive
22:30:25 <talli> where is there perl in OACS? it's gotta be hidden pretty far in there
22:30:32 <hazmat> if someone paid me to do oacs in python, i would do it and love it:)
22:30:43 <talli> given the disdain people in the community have for perl
22:30:52 <hazmat> some of the shell scripts for mailing and updating indexes ( for oracle intermedia)
22:31:02 <talli> kapil, how do you find the performance of pywx to be?
22:31:15 <rbm> mp3jukebox used to have some perl
22:31:23 <rbm> Yeah, anyone used pywx here?
22:31:38 <hazmat> so far really good.
22:31:42 <hazmat> me
22:31:50 <rbm> cool
22:32:09 <hazmat> i should stop for a second and say what i'm using it for
22:32:47 <hazmat> python is mainly for doing graph structures (dependency trees) and xml parsing/generation for metadata and webservices soap/xml-rpc.
22:32:55 <hazmat> openacs handles most of the db work
22:33:07 <rbm> cool
22:33:14 <hazmat> where does don work?
22:33:17 <talli> is there anything that can be reused - released?
22:33:22 <talli> don works for furfly
22:33:22 <rbm> He works for flurfly
22:33:37 <talli> flurfly = furfly
22:33:40 <rbm> (which, for the longest time, I thought it was 'flurfy') :)
22:33:49 <hazmat> :)
22:33:59 <rbm> SEE? I still confuse myself.
22:34:12 <rbm> I think I associated the cats with flurfy
22:35:25 <hazmat> talli: i'm planning on releasing the zope integration stuff early next week, the repository, i'd guesstimate two-three weeks, i've written up some of my repository ideas in a wiki http://www.zope.org/Members/k_vertigo/Stories/Gideon
22:35:56 <talli> that sounds cool
22:36:16 <talli> so what kind of integration stuff is it? what does it provide? why would someone want to integrate the two?
22:39:01 <vinod> vinod has joined #openacs
22:39:17 <hazmat> i did it because i wanted to, what it does is remove zope's i/o framework ZServer and replaces it with aolserver, aolserver basically handles all the client net i/o and zope does processing, as for why.. again i did it because i wanted to, there are some possibilites for wrapping the db pools in a zope adaptor and getting better performance for relational interactivity, also the rich api of aolserver allows alot of flexibility th
22:40:31 <hazmat> iotw it depends if you've got a zope or aolserver perspective. not that this type of thing is nesc. greeted kindly because some people prefer for stability and security to keep the two things out of process.
22:41:15 <talli> so it's basically an experiment more than anything?
22:41:54 <hazmat> yes. i'm a geek i admit, just shoot me :)
22:42:28 <rbm> * rbm pees on people that send .doc attachments
22:42:36 <vinod> * vinod hides in case talli starts shooting geeks
22:42:38 <hazmat> it also gave me a chance to stress the capabilities of what pywx was capable of handling as zope has some fairly esoteric modules
22:45:33 <hazmat> i should clarify that zope has some c modules that provide some of the foundation for transparent persistence, there esoteric cause they mess with the type/object system in zope a little.
22:47:13 <talli> * talli decides not to shoot geeks, but just to pee on vinod until he becomes a .doc attachment
22:47:59 <talli> pywx is very tempting
22:48:23 <talli> we were thinking of trying to prototype some ACS modules, namely ETP, just to see how it might work
22:48:50 <talli> in fact, that project i was mentioning earlier, the one in which we won't use OACS, we may use pywx
22:49:25 <davb> davb has joined #OpenACS
22:54:02 <hazmat> what i'd like to see more of is people documenting the design process of the openacs, all this knowledge of the internals is being lost... that link above to the wiki, i think shows how they can be useful for this type of thing, low impediment colloborative documentation. the wiki referenced above lets anonymous people only comment, and registered users edit all minus the main-page... i got shot down last time i suggested this th
22:54:49 <hazmat> still time if people start
22:55:06 <hazmat> else we'll all be lethe drinkers
22:56:22 <hazmat> speaking of python in openacs, i was thinking that the easiest way to bring webmail back would be to do it in python...
22:57:12 <hazmat> speaking of python outside of the openacs in aolserver, i think thats great talli, if you don't mind i'd like to make two suggestions if you do this, pyunit.sf.net, and www.sf.net/project/pypgsql
22:57:47 <Spork> Spork has joined #openacs
22:57:57 <Spork> * Spork sporks
22:58:09 <vinod> vinod has quit
22:59:10 <hazmat> s/project/projects
22:59:37 <talli> ah, that pyunit thing looks dope
22:59:55 <talli> unfortunately the second link doesn't work
23:00:08 <hazmat> www.sf.net/projects/pypgsql
23:00:17 <talli> why would python help with webmail in the OACS?
23:01:18 <hazmat> cause it does imap/pop/nntp/smtp out of the box and has all the libraries for mime stuff and il8n.
23:01:53 <talli> what do you mean out of the box?
23:03:44 <hazmat> in the standard distribution, ie it comes with python
23:04:59 <talli> so you wouldn't need a mail server? or is there a simple connector to a MTA available in the pythong distro?
23:05:18 <hazmat> as an example, someone just sent me a link to a new messaging (so far only mail) client their doing for zope http://www.thingamy.com:11000/zope-messages/
23:05:34 <talli> it needs a pword
23:05:39 <talli> it's protected
23:05:42 <hazmat> talli: they call it smtp, it provides a simplified convience method
23:05:46 <hazmat> maitest, mailtest
23:06:25 <talli> hmmm... that's neat
23:06:31 <hazmat> .. method that allows you to just send the message without doing all the protocol nitty gritty (not that there is that much)
23:07:06 <talli> so the developer's responsibility is more or less to build the front end?
23:07:15 <talli> and stuff the emails into the server?
23:07:28 <talli> i mean DB?
23:08:06 <hazmat> not really, i don't know that its nesc. smart to record all messages inside of the db, it kinda defeats the purpose of imap.
23:09:14 <talli> true enough
23:09:29 <talli> so the responsibility is more or less to build the UI, yes?
23:11:02 <hazmat> mainly, but you need to think about architecture, ie what happens if your mail server goes down, than keep a queue for outgoing messages. etc. what it give a developer is the ability to focus on their application
23:11:14 <hazmat> s/give/gives
23:11:15 <davb> We can use ETP for editable pages in OpenACS.
23:11:39 <hazmat> lets do it, than.
23:11:57 <hazmat> talli: is there any project management stuff in the new openacs site?
23:12:16 <talli> no, not really
23:12:19 <talli> openacs.museatech.net
23:12:37 <talli> the project management stuff was to be built into developer.openacs.org
23:13:07 <talli> http://openacs.org/bboard/q-and-a-fetch-msg.tcl?msg_id=0002Qo&topic_id=12&topic=OpenACS%204%2e0%20Design
23:13:08 <chump> E: http://openacs.org/bboard/q-and-a-fetch-msg.tcl?msg_id=0002Qo&topic_id=12&topic=OpenACS%204%2e0%20Design from talli
23:13:19 <talli> E: link to discussion of developer.openacs.org
23:13:19 <chump> commented item E
23:14:00 <talli> i proposed an outline for what ought to be included on this site
23:14:15 <talli> more or less it's a project management site for the whole system
23:14:36 <talli> davb: i like the idea of using the ETP interface for the webmail. it would probably be the esaiest way to do it
23:15:09 <talli> although webmail does have a unique enough requiremnet list that it might be easy just to build an interface from scratch
23:15:24 <hazmat> oh etp for webmail... i thought it was for documentation?
23:17:43 <talli> arrrr...
23:18:02 <talli> etp is a complete CMS that will (hopefully) replace the aD CMS
23:18:20 <talli> it's built on the data model of the aD CMS, but rejects all teh crappy interface code
23:18:47 <talli> i only "arrrr" because many people thought that we built ETP just to enable wikis in the ACS
23:19:03 <talli> that's not it at all. it will be completely full featured. eventually
23:19:40 <rbm> ETP is a great app!
23:19:51 <rbm> Musea kicks butt
23:20:01 <talli> aw, shucks
23:20:17 <talli> * talli blushes like a little girl and toes the ground
23:20:25 <talli> thanks rbm
23:20:58 <vinod> vinod has joined #openacs
23:21:32 <cro> cro has joined #openacs
23:21:59 <cro> Greetings--I've installed openacs4 with Oracle 8.1.7.  I'm having trouble with bboard.
23:22:07 <cro> When I do a test insert, I get
23:22:24 <cro> > Who is your favorite bird character?
23:22:27 <cro> oops
23:22:30 <hazmat> etp is cool.
23:22:34 <cro> ora8.c:3568:ora_tcl_command: error in `OCIStmtExecute ()': ORA-29861: domain index is marked LOADING/FAILED/UNUSABLE
23:22:34 <cro> ORA-06512: at "OPENACS4.CONTENT_REVISION", line 35
23:22:34 <cro> ORA-06512: at "OPENACS4.ACS_MESSAGE", line 97
23:22:34 <cro> ORA-06512: at "OPENACS4.ACS_MESSAGE", line 63
23:22:34 <cro> ORA-06512: at "OPENACS4.BBOARD_MESSAGE", line 25
23:22:55 <cro> I didn't install site-wide-search.  Isn't that a message from Intermedia?
23:23:19 <hazmat> that doesn't look very good. can you use any other packages that use the content-repository
23:23:29 <cro> Let me see
23:24:21 <hazmat> is this a cvs checkout or a download or a download?
23:24:48 <cro> cvs checkout.  Haven't done an update in a couple of days,
23:24:58 <cro> Is there an easy way to test the CR somewhere else?
23:25:10 <cro> FAQ doesn't use it, right?
23:26:10 <vinod> vinod has quit
23:26:20 <hazmat> it doesn't look like
23:27:19 <hazmat> general comments does...
23:27:56 <hazmat> press does
23:29:42 <gaizka> gaizka has joined #openacs
23:30:18 <cro> hazmat: fails as well.  Same messge.
23:30:25 <cro> Same place, too.
23:31:10 <cro> ora8.c:3568:ora_tcl_command: error in `OCIStmtExecute ()': ORA-29861: domain index is marked LOADING/FAILED/UNUSABLE
23:31:10 <cro> ORA-06512: at "OPENACS4.CONTENT_REVISION", line 35
23:31:10 <cro> ORA-06512: at "OPENACS4.CONTENT_REVISION", line 68
23:31:10 <cro> ORA-06512: at "OPENACS4.PRESS", line 54
23:31:10 <cro> If I knew what index I could check it out, but I can't tell which it is.
23:33:21 <hazmat> i wish i had my handy oracle translator...
23:33:35 <cro> Think I figured it out--something went wrong with 
23:33:36 <cro> -- this index requires prefs created in content-search
23:33:36 <cro> create index cr_doc_filter_index on cr_doc_filter ( content )
23:33:36 <cro>   indextype is ctxsys.context
23:33:39 <cro>   parameters ('FILTER content_filter_pref' );
23:33:48 <cro> in content-create.sql.
23:34:04 <cro> OK, gotta drop the schema and start over. :-(
23:34:11 <cro> Job for tomorrow.
23:34:17 <hazmat> :(
23:34:18 <cro> Thanks for the suggestion hazmat.
23:34:45 <hazmat> cro: no problem, good luck
23:34:54 <cro> cro has quit
23:36:24 <talli> talli has quit
23:40:58 <gaizka> gaizka has left #openacs