18:05:44 <stickster> #startmeeting Fedora Insight 18:05:44 <zodbot> Meeting started Thu Apr 29 18:05:44 2010 UTC. The chair is stickster. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:05:46 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic. 18:05:48 <stickster> #meetingname Fedora Insight 18:05:49 <zodbot> The meeting name has been set to 'fedora_insight' 18:05:54 <stickster> #topic Roll call! 18:05:58 * stickster 18:06:00 <pcalarco> pcalarco 18:06:14 * rbergeron 18:07:17 <stickster> OK then... 18:07:30 <stickster> #topic Agenda (5 min) 18:07:44 <stickster> I was not going to spend the agenda today on action items given our situation 18:08:17 <stickster> Since we had to stand down on staging because of some remaining blockers, I wanted to use the time for more discussion about what we should do next. 18:08:31 * stickster wonders if gwerra is about 18:09:01 <pcalarco> I was heartened to see Drak's message to the Logistics list, seems like they are commited to helping out with FI 18:09:08 * rbergeron agrees 18:09:20 <rbergeron> #link http://lists.fedoraproject.org/pipermail/logistics/2010-April/000516.html 18:09:33 <stickster> I was glad to see that too 18:09:41 <stickster> #topic Open floor 18:10:10 <stickster> #info Drak assures us participation is good for Zikula too 18:10:54 <stickster> At the same time, I've been thinking about our requirements and whether there are others we need to take into consideration 18:11:17 <rbergeron> I definitely think it would work out better for all involved if they have some of their team members work more closely / inside of Fedora 18:11:37 <stickster> rbergeron: Right. The most important factor, IMHO is having someone passionate and knowledgeable about the platform involved. 18:11:50 <stickster> Rather than being "service personnel" 18:12:20 <pcalarco> we need some knowledge transfer on Zikula internals and someone committed to keeping up with those as they evolve 18:13:15 <stickster> Let me ask this: pcalarco or rbergeron, are either of you intending to be a Zikula master capable of fixing things that break? 18:13:38 <stickster> Not whether you can do that *now*, but do you intend to do it in the future? 18:14:17 <rbergeron> well - let me put it this way: i'm committed to helping test things out - but when we're looking at modules like the auth fas piece - I have a bigger hill to climb there, which involves learning php. 18:14:41 <stickster> rbergeron: So basically "no" for you when it comes to fixing *internals* 18:14:44 <pcalarco> I actually don't know of the range of skill set that we would need. I am willing to learn and commit the time to be able to do this; I've done perl in the past but don't have current PHP skills 18:14:44 <rbergeron> So while I can read things - actually fixing those types of things and tracking issues down at that level is not my area of expertise. 18:15:03 <stickster> pcalarco: I think that PHP has a lot to do with it 18:15:08 <rbergeron> And i don't know that that would be a wise investment of my time, vs. someone who actually knows php. 18:15:25 <stickster> Just so I'm not misunderstood... I'm not begging you guys to do this. I just want to know what *you* feel passionate about doing with Zikula. 18:15:32 <pcalarco> rbergeon: that's how I feel too 18:15:39 * mchua as well. 18:16:17 <rbergeron> So my take is this: while I would love to work with Zikula the Project - as well as deploy FI 18:16:48 <rbergeron> I don't know that we have the expertise available to make Zikula the Project better - so that it's in a place where we think it's stable enough that we want to deploy permanently on it. That said, I'm not expert enough in code to even make that call. 18:17:19 <rbergeron> We haven't gotten far enough with Zikula to know how stable it is, how often we'd be needing updates and patches and so forth - and when there are those changes, how much effort it would take one person who was knowledgeable in php 18:17:23 <rbergeron> to make those fixes. 18:17:29 <pcalarco> I am most interested in the content end of things, editing, workflow and such; I am interested and willing to learn to do some troubleshooting, but I can't really commit time to learn PHP and it doesn't dovetail well with requirements for my $dayjob 18:17:40 <stickster> pcalarco: Understood and agreed. 18:17:52 <rbergeron> I'm in the same boat as pcalarco, minus the $dayjob part. 18:17:57 <stickster> I want to be very clear, I am so very, *very* grateful for the time you guys have put in here. 18:18:05 <stickster> And I am not asking for you guys to commit more to this project. 18:18:57 <stickster> I *am* interested in determining who is willing to be accountable for getting Zikula into running shape for a post-GA deployment. 18:19:13 <stickster> I think if Drak is willing and able to do that, we *all* want to work on the actual administration of the system. 18:19:20 * rbergeron agrees 18:19:23 <pcalarco> +1 18:19:38 <stickster> I have to admit, I've also taken the downtime to look at alternatives 18:19:51 <stickster> I pulled out a drupal package, because I found out it was in Fedora as well 18:20:13 * rbergeron nods 18:20:21 <stickster> And coincidentally, at the same time, I found out a colleague who had just rolled out a Drupal based solution had attended some big worldwide conference of theirs 18:20:41 <stickster> They have an *enormous* community apparently. 18:20:52 <rbergeron> in the end, though, we still boil down to the "we need customizations for FAS stuff" - and for all we know, that work could be as much as or more effort than we are putting in currently. 18:20:56 <pcalarco> yes, 4000+ modules 18:21:03 <smooge> ok sorry 18:21:05 <smooge> late 18:21:26 <stickster> Which is neither here nor there, I suppose, unless either (1) someone(s) in that community are here, in Fedora, passionate about working with us on Drupal every day; or (2) someone already in the Project who is a guru is willing to be that passionate linchpin 18:21:33 <rbergeron> exactly 18:21:46 <stickster> Otherwise, we could end up in exactly the same place in four months 18:22:50 * stickster waits for a moment so other people can talk... feeling like he's pontificating too much. 18:23:21 <rbergeron> well - here are my thoughts on drupal 18:23:28 <rbergeron> i agree it could wind us up in the same place in 4 months 18:23:42 <rbergeron> we also could, as you said, already have some magical local guru who knows things 18:23:57 <rbergeron> I think the FAS authentication thing is huge - no matter what we do 18:24:06 <pcalarco> one good thing: both Zikula and Drupal are PHP-based, so if we can find someone interested with PHP skills, we're good 18:24:14 <rbergeron> I think it's entirely possible that someone may want to use drupal for whatever on fp.o infrastructure 18:24:34 * rbergeron was thinking of trying it out for press.fp.o - but i still come back to the whole fas authentication thing 18:24:36 <stickster> pcalarco: Yes. If this were Python, we'd have likely got further on our own with help from Infrastructure folks. 18:25:01 <rbergeron> but - unlike zikula - I can hop on the googlez and search for drupal custom authentication and find a whole boatload of information 18:25:05 <stickster> But that means Plone, which has its own special set of crosses we don't want to (can't) bear. 18:25:55 <smooge> ok from an infrastructure point of view.. we need more PHP 18:26:11 <rbergeron> php expertise? 18:26:13 <smooge> people who are passionate about it and willing to deal with PHP's issues 18:26:14 <rbergeron> or just more php :) 18:26:15 <stickster> yes 18:26:16 <rbergeron> ah 18:26:35 <stickster> Nowadays, PHP rules the roost with Ruby zooming up to its tail 18:27:01 <stickster> In terms of having a lot of people out there in the expert base 18:27:02 <smooge> we currently have a mess of Python people who have to do heavy context switching in their head to deal with Ruby or PHP 18:27:07 <stickster> Right 18:27:21 <stickster> And we need to respect that we can't just dump a solution on that team and expect them to simply rev up 18:27:25 <smooge> this leads to "Why the hell would you do that in any sane language?" moments which doesn't help get things fixed 18:28:04 <smooge> on the other hand, we need to branch out somehow so we avoid NIH syndrome 18:28:14 <stickster> smooge: Agreed 18:28:23 <smooge> but by branching out it means we need to grow the team with expertise. 18:28:52 <stickster> I've only looked at PHP briefly, because I needed to for a WordPress theme I was customizing 18:29:15 <smooge> actually if you are doing anything 'serious' in webapps you need to know PHP :). 18:29:33 <smooge> Its one of those sad facts of life that makes real developers run away from web development :) 18:29:52 <smooge> there should be air quotes around real and web 18:29:52 <stickster> So here's a question I keep circling around mentally... are we more likely to grow that corps of people using Zikula? We have at least one, maybe two people who are willing to help in that capacity. 18:29:54 <rbergeron> So - just talking out my rear end here - are there any python-based frameworks we could look at? 18:30:03 <stickster> rbergeron: There's Zope + Plone 18:30:03 <rbergeron> is that a path we want to even think about? 18:30:19 <smooge> no 18:30:23 <stickster> But unfortunately they have (1) a very steep learning curve, compared to Zikula, Drupal, and others 18:30:27 * rbergeron nods 18:30:35 <stickster> (2) a very backward release model compared to Python itself 18:30:40 <pcalarco> there is less growth momentum in that direction too, methinks? 18:30:41 <smooge> I will say it nice and easy. no. you aren't going to get a solution out of any of them in 3-4 months 18:30:46 <stickster> which makes it a horrible fit for Fedorra 18:30:52 <stickster> *Fedora self.slap() 18:31:04 <stickster> smooge: I agree. We tried two years ago, and it went nowhere. 18:31:22 <stickster> It's not that Plone is bad at all, it's just ten times harder to do some things with it than it needs to be. 18:31:30 <rbergeron> okay - just double checking :) 18:31:33 <stickster> In the right hands, it's totally awesome. 18:32:00 <mchua> There are other python-based frameworks, like Turbogears/Django/pylons/etc. I'm sure the Python crew can tell us all about them. 18:32:06 <stickster> There are 18:32:13 <smooge> Grumpy old man with 3 hours of sleep time: We have to release at some point. Going to ground zero before release is not productive failure. Get zikula working, then re-evaluate what you want for 3.0 Insight. 18:32:16 <stickster> mchua: The problem is all of them require you to build the management solution on top. 18:32:39 <smooge> rbergeron, it was a good question.. and needed to be asked because otherwise I wouldn't get to pontificate 18:32:48 <stickster> smooge: That's what I'm thinking. If we can get our initial rollout done, we can think about 2.0 afterward. (We're still working on 1.0) :-) 18:32:50 * rbergeron is happy to serve your pontification needs! :) 18:32:51 <pcalarco> smooge: I agree 18:33:44 <smooge> stickster, what I meant was 1.0 is zikula. 2.0 is zikula.. in the background people who want to do something different can try variou sthings and by 3.0 we should have either another solution we like or zikula 18:33:58 <stickster> pcalarco: I think in terms of growth momentum, which you said before... Drupal is king 18:34:33 <stickster> It's getting penetration that may at some point make it the Mozilla Firefox of CMS 18:34:46 <stickster> But that's a 3.0 issue as Smooge says 18:34:47 <kushal> true 18:35:03 <stickster> So let me propose this, and then let's shoot at it if needed 18:35:24 <stickster> We should take Drak up on his offer, set up our weekly Insight meeting at a time where he can participate 18:35:40 <pcalarco> well, we can do FWN republishing right now with just a bit more effort; where we run into hurdles is when we want to open up content submission widely, for which we need FASAuth working; Could we ask Drak to help with FASAuth post-GA and work on remaining presentation layer issues? 18:36:06 <stickster> pcalarco: Yes, you skunked me :-) 18:36:17 <stickster> *: what pcalarco just said :-) 18:36:26 <stickster> And I would like to set a firm date for that to happen. 18:36:44 <stickster> Say, June 15, which seems totally reasonable to me 18:37:03 <stickster> But we can adjust that date at the first meeting based on what remains to be done. 18:37:06 <pcalarco> to have FASAuth in place? 18:37:21 <stickster> pcalarco: Everything that lets us publish the FWN on the new fp.o/insight site. 18:37:21 <mchua> and the target, for Drak to work towards for FI deployment, is basically *just* FWN right now? 18:37:45 <pcalarco> June 15 is very do-able 18:37:46 <stickster> mchua: Yes, and finish other content issues for Marketing for, say, 30 days after that 18:38:05 <stickster> Which *also* should be very reasonable if it's just some modules away as Drak intimated in his email. 18:38:12 * rbergeron nods 18:38:19 * rbergeron bobbleheads just for stickster 18:38:23 <stickster> heh 18:38:25 <rbergeron> lol 18:38:26 <pcalarco> :) 18:38:33 <stickster> Hang on, let me catch zoddie up here 18:39:35 <stickster> #agreed We should take Drak up on his offer, set up our weekly Insight meeting at a time where he can participate, and then work toward a June 15th milestone. At that point, anyone should be able to type in "fedoraproject.org/insight" into a web browser and see the latest FWN issue and get to any archives we have available at that point. 18:41:05 <stickster> The other point up for #agreed is... "30 days following we have a way to (1) promote a Planet story to a separate Insight feed without having to copy/paste or retype it, and (2) both this content stream and the FWN can be visible/RSS at once" 18:41:31 <stickster> That, to me, is a significant point of progress, but maybe not the complete picture yet. 18:41:55 <stickster> I wanted to propose a milestone without throwing *everything* into the "30 days from then" bin. 18:42:38 <stickster> I also think it goes without saying that we *can* put manual stories in as well, but that seems so elementary as not to need a special milestone. We can do that now, pretty much, modulo the FASauth problems. 18:42:49 <stickster> pcalarco: rbergeron: smooge: mchua: thoughts? 18:43:02 <smooge> one sec while I digest 18:43:12 <pcalarco> seems like a reasonable plan forward! 18:43:20 * rbergeron thinks it sounds good 18:43:22 <stickster> Does this coincide well enough with the larger plan for the content Insight should carry? 18:44:09 <rbergeron> i think so. 18:44:17 <stickster> mchua: You really helped write up that plan, so I'm interested in your opinion 18:44:39 <mchua> +1 18:44:45 <smooge> looks good as a good milestone 18:45:01 <pcalarco> I'd say so too 18:45:07 <mchua> I wonder about the other things that were looking at zikula. 18:45:09 <stickster> OK. 18:45:10 <mchua> Docs, for instance. 18:45:31 <mchua> And whether it makes sense to look at other options in parallel, just so we know what's out there. 18:45:40 <stickster> mchua: Precisely -- I think there are other facets of Insight we'll need to explore. 18:45:45 <pcalarco> I never understood the value added of Zikula for Docs. 18:46:03 <mchua> We went with zikula for FI because of the selection of zikula for docs. 18:46:07 <stickster> pcalarco: It's not clear there will be, because of recent enhancements to Publican 18:46:11 * mchua should have looked more closely at that rationale, tbh. 18:46:51 <stickster> mchua: Well, following that, to be fair most of the Docs people jumped out of the Zikula project. The people driving it are the ones who make the decisions, right? 18:47:43 <stickster> So yes, we have a situation here where we started with a set of assumptions, including (1) Docs would be using Zikula, and (2) Zikula was the right choice for the intended purpose. 18:48:11 <stickster> It's not clear Docs will need to use Zikula because of some enhancements that have gone on in the upstream for the tool they're using for publishing, Publican. 18:48:25 <stickster> Publican now has the ability to provide a pretty decent self-contained documentation site 18:51:02 <stickster> Which means their need for a different content management solution may have gone *poof*. 18:51:02 <stickster> OTOH, it remains to be seen whether that system can handle offering RPM packages. If it can't we could still end up helping them out, but that target is so easy to hit, I don't think it matters what we're using for an Insight platform. 18:51:03 <stickster> OK, let's get back to the agreement from befroe 18:51:04 <stickster> *before, even 18:51:14 <stickster> #agreed 30 days following we have a way to (1) promote a Planet story to a separate Insight feed without having to copy/paste or retype it, and (2) both this content stream and the FWN can be visible/RSS at once 18:51:28 * stickster feels like he blew up everyone's braaaaanes. 18:51:41 <stickster> Did what I wrote there look like nonsense to anyone? 18:51:42 <rbergeron> don't worry, mine was blown up hours ago. :) 18:51:50 <stickster> mchua: ? 18:52:13 * mchua haz brain? 18:52:14 <rbergeron> it makes sense to me. 18:52:20 <pcalarco> yes, me too 18:52:26 <mchua> same here. 18:52:27 <stickster> In short, I agree that we are dealing with choices made by others that might be different than the ones we'd make now. 18:52:42 <mchua> stickster, I think you're mostly summarizing concisely what we've all been thinking for a while. :) 18:52:56 <stickster> OTOH we have at least someone who claims to be passionately wanting to help us move forward, and we should respect and encourage that -- because we'd do it for any other contributor too. 18:53:05 <rbergeron> +1 18:53:09 <pcalarco> +1 18:53:19 <stickster> At the same time, we can plan for the future. I have to admit, I'm interested in seeing what Drupal can do, and I will start by putting my money where my mouth is 18:53:39 <stickster> If I can come up with a FASauth plugin for it, then I would say it's not that terribly hard to learn. 18:54:14 <stickster> And if Marketing wants to consider that as an alternative for Insight 3.0, so be it -- that will depend on having a passionate group of people who want to really plunge into the innards and Make Things Work. 18:54:33 <stickster> I don't mind being one of them, but I won't be the only one of them. 18:54:36 <stickster> That's all I have! 18:54:48 <stickster> Now you fine people say what's on your mind. I'm just one guy here. 18:55:20 <pcalarco> In the interim, shall I keep re-publishing FWN to FI? 18:55:26 * rbergeron is out of things to say. I am glad that Drak is willing to come out and help - I think it makes sense for both Fedora and Zikula. :) 18:55:30 <pcalarco> so we build up content 18:55:36 <stickster> pcalarco: Provided that it does not cause you grievous pain to do so, it would be very helpful 18:55:42 <rbergeron> pcalarco: I don't know that we can even log in to Zikula right now - but I think building up content is a good idea. 18:55:49 * rbergeron hasnt' tried since last week 18:56:04 <rbergeron> and yes, i added to that bug :) 18:56:12 <stickster> rbergeron: We can 18:56:24 <rbergeron> oh? 18:56:26 <stickster> and the unnecessary warning box is gone too. 18:56:30 <rbergeron> hrmmmmmmm 18:56:36 <pcalarco> I can get in too 18:56:47 <rbergeron> okay. i'll add that to my list of things that are good news to check out :) 18:57:00 <rbergeron> woot! 18:57:20 <stickster> I think that Drak fixed these shortly after our go/no-go... which is kind of unfortunate, but we had our decision point and called it. 18:57:32 <stickster> There may be other things still left to do 18:58:11 <stickster> Anything else here, then? 18:58:17 <stickster> What's our next action with Zikula? 18:58:18 <rbergeron> who is going to be writing back to drak to discuss a meeting time that works for him? 18:58:40 <stickster> rbergeron: I can do that since I've been running the meetings through now... seems fair to me. 18:58:45 <rbergeron> and what can we do to get him and/or whoever else quickly integrated with the things they need into fedora? 18:58:57 <rbergeron> or shall we see their response first, and go from there? 18:58:57 <stickster> I know he has the account he needs already. 18:59:33 <stickster> We let him catch up on the minutes first I think. 18:59:43 * rbergeron agrees 18:59:51 <stickster> If things go well over the next 45 days, we may end up perfectly happy to ditch any parallel efforts. 19:00:20 <stickster> The door is completely open. It really depends on whether we can build a corps of people who are interested in supporting the platform, down to the PHP level. 19:00:34 * stickster agrees with smooge that should be a side goal/benefit of this work. 19:00:50 <stickster> er, not "side" as in secondary -- I mean, complementary. 19:01:17 <stickster> #action stickster to make sure Drak gets a reply and sees minutes 19:01:24 <stickster> Anything else before we close? 19:01:36 <stickster> mchua: I know there's not a ton of action items here. Do you feel the meeting was useful anyway? 19:01:37 <pcalarco> nothing here; good meeting, all 19:02:21 * stickster says once again, thanks to YOU GUYS for helping keep us on track. 19:02:28 <stickster> OK, closing in 30 then. 19:02:55 <stickster> 10 19:02:59 <stickster> 5 19:03:04 <stickster> #endmeeting