18:18:56 <mchua> #startmeeting
18:18:56 <zodbot> Meeting started Tue Jul 20 18:18:56 2010 UTC.  The chair is mchua. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:18:56 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
18:19:02 <mchua> #chair czwiebel sdziallas
18:19:02 <zodbot> Current chairs: czwiebel mchua sdziallas
18:19:09 <mchua> czwiebel: (do you know how to use the bot?)
18:19:16 <czwiebel> nope
18:19:23 <czwiebel> Damn, that was awesome!
18:20:09 * czwiebel imagines Mchua weilding some real robot. "Go meetbot, create meeting space." Executive board room with secretary unfolds and installs itself. =P
18:20:22 * czwiebel opens meetbot wiki
18:20:55 <mchua> czwiebel: the crash-course version is that if you want to call out something specifically for the meeting notes, you #info it, and if you want to assign a task you #action it, at the start of the line.
18:21:16 <mchua> #info Discussing next-actions for Etherpad FAD planning
18:21:17 <mchua> czwiebel: like that.
18:21:52 <sdziallas> Although this would probably be much more a #topic.
18:22:17 <czwiebel> #topic Overall preparation for FAD
18:22:31 <czwiebel> what needs to be done
18:22:42 <czwiebel> I can get space from Olin
18:22:56 <mchua> czwiebel: do you want to be event owner (and point person for physical-location logistics, basically)?
18:23:04 <mchua> drive the prep, make sure this happens?
18:23:25 <mchua> and delegate giant chunks of that to andyp and sdziallas and DJ and myself? ;)
18:23:28 * czwiebel looks for candidates
18:23:44 <czwiebel> um, I suppose
18:23:52 <mchua> w00t
18:23:56 <mchua> #info czwiebel is the FAD's point perso
18:23:58 <mchua> #undo
18:23:58 <zodbot> Removing item from minutes: <MeetBot.items.Info object at 0x1d31dcd0>
18:23:59 <mchua> #info czwiebel is the FAD's point person
18:24:02 <mchua> There we go.
18:24:10 <czwiebel> lets, see, just need to make sure there aren't conflicts here with Open
18:25:04 <mchua> I've already offered to be responsible for QA, and can also handle budget requests for travel for remote attendees (wrt looking for funding once we know who's asking for how much money to come).
18:25:21 <mchua> #info mchua is responsible for QA, and can also handle budget requests for travel for remote attendees (wrt looking for funding once we know who's asking for how much money to come).
18:25:22 <czwiebel> QA?
18:25:29 <czwiebel> you do testing on FADs?
18:25:36 <mchua> czwiebel: oh, no, on the etherpad code.
18:25:40 <czwiebel> Ahh, okay
18:25:55 <czwiebel> You usually have remotes for FADs?
18:25:57 <mchua> it's really hurting for scaleability testing, etc. and a way to figure out if the changes you've gotten in have just broken the build.
18:26:17 <mchua> czwiebel: yeah, the budget for a FAD is usually used almost entirely on travel/lodging for folks from out of town to come and hack.
18:26:27 <czwiebel> Yeah, I don't know about unit tests on EP, but I suspect it is lacking
18:26:37 <mchua> because the odds of having everyone in town you need to do the hacking are usually very low.
18:26:45 <czwiebel> Okay
18:26:56 <mchua> for instance, if you found one of the original dev team who wanted to come over and help hack, we could ask for funding for that person's plane ticket.
18:26:58 <czwiebel> Hold, on, perhaps we should discuss scope of this proj
18:27:09 <czwiebel> mchua: cool!
18:27:29 * mchua nods
18:27:37 <czwiebel> scope, don't know that I'll contriubute significantly outside the FAD
18:27:47 <czwiebel> sdz, I imagine, may work on packaging
18:27:56 <czwiebel> andyp, probably more a consumer
18:28:12 <mchua> czwiebel: no worries, a FAD is supposed to be a self-contained event anyhow.
18:28:18 <czwiebel> I almost feel we'd do best to do documentation, guiedes, etc surrounding EP
18:28:26 <czwiebel> mchua: Ahh, okay
18:28:54 * czwiebel wonders if we could work with people from the EP foundation
18:29:10 <mchua> czwiebel: Well, the fundamental problem as I see it is that the code isn't hackable - it's not a good FOSS project in terms of infra/community/being-an-upstream
18:29:16 * czwiebel worries they may be resistant to losing their meta-EP documentation (which really needs to go)
18:29:40 <mchua> czwiebel: I would hope they could come! Basically I feel like if those folks get hit by a raptor the EP code is going to be dead because nobody else knows how to hack on it.
18:29:49 <mchua> so things like "let us set up a bug tracker so we can track what work people are doing"
18:29:57 <mchua> and "I wonder if commenting some of this code would be good"
18:30:18 <mchua> or "maybe we should make some notes on code architecture so that folks can figure out how to navigate the code tree to make the changes they want"
18:30:24 <czwiebel> Hmm, github has bug tracking?
18:30:52 <sdziallas> Yup. Although their version looks somehow even more crappy-fied to me. But it's probably our best guess.
18:31:13 <mchua> Is anyone using the bugtracking in github?
18:31:36 <czwiebel> http://github.com/ether/pad/issues
18:31:40 <sdziallas> mchua: I think 124 issues have probably been reported by "somebody".
18:31:41 <czwiebel> dunno?
18:32:06 <mchua> ...so, kinda. Hrm.
18:32:11 <czwiebel> Frustrating, needs a community organizer
18:32:34 <czwiebel> I don't know that this redhog contrib person is good about that stuff
18:32:45 <mchua> Looks like they're finding their way through the code well, though.
18:32:46 <sdziallas> redhog@redhog.org
18:33:05 <mchua> Simon B, redhog, and Mikko might be interested in coming to the FAD.
18:33:12 <sdziallas> (lives in Norway.)
18:33:14 <mchua> Usually things just get done much faster in person.
18:33:21 <mchua> FAD budget is usually $3k per FAD, for the record.
18:33:36 * czwiebel takes note
18:33:46 <mchua> We were able to fly Henrik in from Germany for the Marketing FAD, and Neville came from Nicaragua, and Robyn and Ryan from Arizona (to Raleigh)
18:34:06 <mchua> and get hotel rooms (and a few others, like Paul and David, drove in and we spotted them for gas.)
18:34:09 * czwiebel is impressed
18:34:28 <sdziallas> It might make sense to talk to the Babson Executive Center.
18:35:07 <czwiebel> #info possible action: fly redhog in from Norway
18:35:25 <czwiebel> #info possible action: contact Babson Executive Center for lodging
18:35:30 <mchua> czwiebel: I think the first thing is to figure out what we want to accomplish during the FAD, and then who we might need to bring in
18:35:37 <mchua> hrm, the BEC is pretty expensive.
18:35:50 <mchua> I can crash with my aunt (or on a number of couches in the area) with no trouble.
18:36:12 <mchua> or, depending on how late hacking goes, sleep on someone's floor / suite lounge / etc. I've got a sleeping bag.
18:36:12 <czwiebel> I also don't think they'll want to "donate" or reduce, even for a "foundation"
18:36:16 <mchua> Yeeeeah.
18:36:21 <mchua> BEC is... just expensive all th etime.
18:36:23 <mchua> er, the time.
18:36:25 <mchua> Can't type.
18:36:32 * czwiebel woudl feel bad keeping redhog on a couch in his room
18:36:50 <mchua> Depends on what people want, some folks might want a hotel room, some might be fine with couchsurfing to keep costs down.
18:36:55 <czwiebel> Perhaps a good deal of pre-work needs to happen
18:36:59 <mchua> I'm happy anywhere I have some floor-space, for the record.
18:37:06 <mchua> Yeah.
18:37:06 * czwiebel takes note
18:37:20 <mchua> Okay, so - goal of FAD: what do we want to accomplish by the end of the event?
18:37:23 <czwiebel> I agree, EP could be come dead in the water in terms of hacking
18:37:36 <czwiebel> Think they needs some community infastructure
18:37:41 <mchua> To some extent we just kinda need to arbitrarily make these answers up so we have something to go from, you know how event planning goes. :)
18:37:45 <czwiebel> Mainly docs and mailling list
18:37:47 <mchua> czwiebel: that's exactly what I was hoping to teach you to do ;)
18:38:30 <czwiebel> Guess, I'll email them and check what they are open too
18:38:38 <czwiebel> #action find a mailing list for EP
18:38:47 <czwiebel> ^ shoudl it exist
18:39:12 <sdziallas> #link http://groups.google.com/group/etherpad-open-source-discuss
18:39:19 <sdziallas> #link http://groups.google.com/group/etherpad-dev
18:40:21 <mchua> czwiebel: zoom out a bit more first - what deliverables do you want to complete by the end of the FAD? Maybe that's one way to go.
18:40:50 <czwiebel> Hmm
18:40:57 <mchua> "have Etherpad deployed in Fedora's infrastructure" might be one (which would mean everything's packaged and deployable by others as well, for instance you could have an internal Olin instance up)
18:41:06 <czwiebel> Would like to see EP a nice OSS project
18:41:11 <mchua> or maybe there are some features we'd like to see in Etherpad that we could add to our deliverables list
18:41:16 <mchua> czwiebel: yes, but what does that mean exactly? :)
18:41:25 <czwiebel> Would like to have some of the EP live diffing separated for other projects
18:41:31 <czwiebel> (collaborative svg, anyone)
18:42:29 <mchua> ...ooh, okay, that's one potential.
18:42:45 <czwiebel> Basically, I see Etherpad as being snagged by a few people to make commercial sites--changes don't migrate back, this isn't how it should be
18:42:55 <mchua> Dunno if it's possible in this timescale or how difficult technically it would be, but basically "use Etherpad as a front-end for other things"?
18:42:58 * mchua nods
18:43:10 <mchua> czwiebel: yep, I agree (re: changes don't migrate back, this isn't how it should be)
18:43:12 <czwiebel> mchua: I was thinking of using it as a backend
18:43:45 <czwiebel> A packaging and use strategy would be good
18:43:57 <czwiebel> sdziallas has done a huge number on that front
18:44:00 * sdziallas waves.
18:44:20 <czwiebel> A framework for check in -> build -> packing -> distros would be cool
18:45:47 <czwiebel> Hmm, totally feel bad about forcing infastructure on EP when I'm an outsider on their community
18:45:56 <mchua> yeah, me too.
18:46:05 <mchua> Maybe we can think of this as a service we're offering as a way to start contributing?
18:46:30 <czwiebel> like, a promise to continue working with the EP community?
18:46:32 <mchua> "hey, we'd like to help, but we can't hack the code as fluently as you can yet... but we *can* organize a hackathon so y'all can get together in person and sprint on this, would that help?"
18:46:51 <mchua> nah, just "we can chip in by putting together a hackathon to which you're all invited, and we'll try to find a way to pay expenses too"
18:47:09 <mchua> hackathons are a (big) contribution. that's how I started with OLPC - the first Game Jam at Olin in June '07.
18:47:28 <czwiebel> ehhh, personally haven't put in much effort to try and hack. Couldn't argue that the code is "hack proof" (lol)
18:47:34 * mchua grins
18:47:43 <czwiebel> We could make it more approachable
18:47:46 <mchua> The barrier to code contribution is then too high for both of us, yeah. :)
18:47:48 <mchua> czwiebel: exactly.
18:47:59 <czwiebel> Bug submission may be an issue at present
18:48:07 <czwiebel> actually, probably not because of github
18:48:36 * mchua grumbles a bit about github being closed source, but if that's what upstream uses, that's what upstream uses.
18:48:42 <czwiebel> A tool to track the uptime of installations would be good
18:48:59 <czwiebel> I've had mixed success with them
18:49:23 <czwiebel> I still think delete functionality would be great, or authentication on the public facing pad
18:49:36 <czwiebel> (IE, you need some auth to let people safely delete)
18:50:13 * mchua nods
18:50:21 <czwiebel> Still think I would love to UI hacks
18:50:22 <mchua> #idea Deliverable: packaged and deployable in Fedora
18:50:38 <mchua> #idea Deliverable: user authentication
18:50:40 <czwiebel> That or modularize, so ppl can drop in own front end (which again, might not be doc based)
18:50:48 <mchua> #idea Deliverable: ability to delete pads
18:50:55 <czwiebel> Isn't it already packaged for Fedora?
18:51:02 <sdziallas> czwiebel: it's not in Fedora.
18:51:04 <mchua> #idea Deliverable: modularization - let people drop in different front-ends for editing
18:51:12 <sdziallas> czwiebel: I do have RPMs. They contain lots of bundled libs, though.
18:51:26 <sdziallas> (I ripped out 3/4 but we need to package large dependencies, still.)
18:52:19 * czwiebel checks for Ubuntu packages
18:52:34 <mchua> #note Most of the work we'll be doing during the FAD will probably be modularization/documentation/cleanup of the code, helping EP become a better upstream.
18:54:02 <czwiebel> #note Making a monolithic .war file would be awesome
18:54:24 <czwiebel> you can just drop .war files into tomcat and magic happens
18:54:41 <mchua> czwiebel: what I think we can do is have a few concrete deliverables to aim for, and use that as an excuse to help upstream clean up a bit. ;)
18:54:53 <czwiebel> #note a web-config tool for etherpad (as often found in J2EE applications) would be great
18:55:02 <mchua> i.e. the real point of the FAD is to modularize, document, bugtrack, etc. but we have a few specific goals to aim at so we have a direction to run towards.
18:55:05 <czwiebel> mchua: sounds good
18:55:06 <mchua> ...ooooh.
18:55:37 <czwiebel> mchua: For packaged J2EE apps (OpenGeo), that workflow is
18:55:40 <czwiebel> 1) grab .war
18:55:48 <czwiebel> 2) copy into tomcat
18:55:57 <czwiebel> 3) visit a URL in your browser and configure
18:56:17 <czwiebel> same(similar) URL is the actual application, its pretty sweet!
18:56:38 * czwiebel cool
18:56:38 <mchua> ...hrm
18:56:40 <mchua> nice.
18:56:51 <czwiebel> I think a packaging FAD would be a good goal
18:56:51 <mchua> That would be pretty sweet.
18:57:15 * czwiebel disclaims he doesn't know if too much of EP is PHP to be packaged as a .war
18:57:27 <mchua> czwiebel: so what I'd do (I should pop out soon, it's lunchtime at this conference and I have to meet people in a bit) is to note these possible deliverables on the event page
18:57:33 <mchua> #link https://fedoraproject.org/wiki/Etherpad_FAD
18:57:44 <mchua> and start poking the etherpad list with the idea of a hackathon and see how they respond
18:57:53 <mchua> and see if they can help us get a handle on how difficult some of this work would be to do.
18:58:07 <mchua> The #etherpad channel may also have some feedback, though it's usually super-quiet.
18:58:33 <czwiebel> mchua: FAD would be 1 day only?
18:58:37 <czwiebel> whole weekend?
18:59:58 <mchua> czwiebel: I'd say "start Friday morning, end Sunday afternoon"
19:00:12 <mchua> that way people fly in Thurs, only take one day off work, we have 3 days to hack.
19:00:20 <mchua> 3 days is usually a good FAD length.
19:00:43 <czwiebel> Sounds good
19:00:53 <mchua> Cool.
19:00:55 * mchua has to run
19:00:56 <czwiebel> man, this page is much more complete than I would have guessed
19:00:59 * mchua grins
19:01:08 <mchua> thanks for picking up on this, Colin - nag me incessantly if you need help. ;)
19:01:12 <czwiebel> mchua: you've been working on it, haven't you
19:01:20 * mchua whistles
19:01:27 <czwiebel> Oh, I'll keep bugging y'all
19:01:30 <czwiebel> cya
19:01:38 <mchua> czwiebel: I think a good deadline would be to try and request funding from Fedora before Aug 5th or so
19:01:41 * mchua waves
19:01:48 <mchua> czwiebel: (also, #endmeeting when you're done)
19:01:53 <czwiebel> will do
22:22:33 <czwiebel> hmm, need to get better w/ Meetbot, guess I'll
22:22:39 <czwiebel> #endmeeting