News/MinutesSept10Meeting

Minutes of the 2009-09-10 IRC meeting

Following the team lead change, an IRC meeting was lead to discuss the 1.x and 2.0 roadmap, as well as issues related to the Propel website and symfony integration. Here are the minutes of the meeting, as recorded on the #propel IRC channel.

20:00 < everplays> it seems hearts are beating in #propel :P
20:24 <       jrw> A welcome change - it's been a quiet year.
20:27 < stephenli> I'm not even sure most developers know propel exists :P
20:28 < stephenli> I had been writing my own database stuff for almost three years before stumbling across this accidentally (I think I was looking into frameworks for one of my clients)..
20:29 <       jrw> Well if you start digging around for a PHP ORM you kind of can't miss Propel, but yeah you have to know what to look for first.
20:29 <       jrw> I don't think the issue's been a lack of visibility, but the lack of activity.
20:30 < stephenli> yeah, thats a good point too.. it'd be hard to sell a company on using something that looks like its dying off..
20:48 < guilherme> propel meeting already happened?
20:48 <  francois> hi everybody
20:49 < guilherme> hi francois 
20:49 < stephenli> *waves*
20:50 <  francois> guilhermeblanco, jwage: you guys are interested by Propel's future ? :)
20:50 < guilherme> francois: of course we are
20:50 < guilherme> =)
20:50 < guilherme> francois: romanb is also present
20:50 < guilherme> =)
20:50 < alanpinst> hi all -- Alan checking in here. be back in a few when things kick off. looks to be a great meeting@!
20:50 <  francois> hi
20:51 <  francois> I just hope there will be more Propel users than Doctrine users
20:51 <  francois> Are there some Propel users reading this?
20:51 <    davide> one here!
20:52 < stephenli> yep
20:52 <     Jarda> o/
20:52 <    v-dogg> +1
20:53 < guilherme> francois: I';; try to find you an old email full of things from a very old list of ORM developers
20:53 <  francois> guilhermeblanco: fine, thanks
20:53 < guilherme> francois: can you pvt me our email addr?
20:54 <  francois> so, for the four guys actually using Propel, for the four guys using Doctrine, for the bots and all the ones that are just not in front of their screen...
20:54 <  francois> the meeting is starting
20:54 < christoph> propel +1
20:54 < christoph> :-)
20:54 <   benschi> Yay
20:55 < alanpinst> 4-5 year propel veteran here ;)
20:55 <  francois> welcome everybody
20:55 <    SidGBF> propel here too
20:55 <  francois> I'm glad that there are som many people interested in Propel's future
20:55 <  francois> we have a lot to talk about today
20:55 < alanpinst> thanks! and thanks @francois for taking over the lead. we needed it!
20:55 <  francois> and I'd like to hear each of you talking
20:56 <    v-dogg> do we have an agenda?
20:56 <    v-dogg> for this meat, that is
20:56 <     tiddy> hi
20:56 <  francois> v-dogg: I suggest
20:56 <  francois> 1 - 1.4 and 2.0 timeframes
20:56 <  francois> 2 - 1.4 roadmap
20:56 <  francois> 3 - websites improvements
20:57 <  francois> 4 - miscellaneous
20:57 <  francois> (there always seems to be a - miscellaneous)
20:57 <  francois> anything else you want to talk about ?
20:57 <    v-dogg> 5 - infra like SVN etc. server stuff
20:57 < alanpinst> sounds like a good plan yeah liek #5 too (same as #3 maybe)
20:57 <   benschi> Sounds like a Plan to me
20:58 <     tiddy> ok, i`ll think francois will be discussion leader
20:58 <  francois> v-dogg: 3 = 5
20:58 <     tiddy> and keep focus on roadmap
20:58 <    davide> maybe some stuff related to symfony?
20:58 <  francois> tiddy: do you wanna lead?
20:58 <  francois> davide: ok, 4 - symfony related stuff
20:58 <       jrw> Umm, won't there be anything on 2.0 except a timeframe? :o
20:58 <  francois> and 5 - miscellaneous
20:58 <  francois> ah, yes
20:58 <  francois> 1 - 1.4 and 2.0 timeframes
20:58 <  francois> 2 - 1.4 roadmap
20:59 <       jrw> Because that's why I'm here for, mostly. :P
20:59 <  francois> 3 - 2.0 roadmap
20:59 <       jrw> s/why/what/
20:59 <  francois> 4 - websites improvements
20:59 <  francois> 5 - symfony stuff
21:00 <  francois> but first let me start by saying that I'm glad to work for the Propel project and I hope that we can all make it better together
21:00 <       jrw> Thank you for taking point.
21:00 <  francois> we'll need all of you guys for that
21:00 <  francois> so, back to timeframe
21:01 <  francois> 1.4 timeframe is pretty easy
21:01 <  francois> symfony is the main project using Propel, and symfony 1.3 will release end of this year
21:01 <  francois> if Propel 1.4 want to get some exposure and balance the idea that the project is abandoned,
21:02 <  francois> propel 1.4 has to be in symfony 1.3
21:02 <  francois> that means feature freeze in mid october
21:02 <  francois>  (that's in about one month)
21:02 <    davide> that's soon
21:02 < alanpinst> what is the 1.4 roadmap? what's new?
21:02 <  francois> well, the timeframe will decide the roadmap
21:02 <       jrw> Just a whole lot of smaller stuff + namespaces, afaik.
21:02 <  francois> since we only have one month, we won't be able to put a lot of stuff in there
21:03 < alanpinst> oh ok (fyi it isn't listed on the wiki)
21:03 <  francois> but there will be a 1.5
21:03 <  francois> (which we'll discuss in a meeting after 1.4 is over ;)
21:03 < alanpinst> is the stuff I've been committing to 1.3 branch going to make it into 1.4?
21:03 <  francois> alanpinstein: the 1.4 roadmap isn't decided, that's the second point we have to discuss
21:03 < alanpinst> ok
21:04 <    davide> so just little enhancements for 1.4
21:04 <  francois> alanpinstein: what are you talking about exactly?
21:04 < alanpinst> just a few bugs I have fixed. can't remember them all now
21:04 < alanpinst> anything i committed should be in trac... want me to look?
21:04 <  francois> alanpinstein: probably. but first, let's listen tiddy about 2.0 timeframe
21:05 <    SidGBF> Guys, let Francois talk and let the thing go...
21:05 < alanpinst> ok
21:05 <  francois> tiddy: are you there?
21:06 < guilherme> francois: you hsould at least point to something and/or feature you plan to get done until mid-october
21:06 <  francois> maybe not
21:06 < guilherme> what do you plan to get done?
21:06 <  francois> guilhermeblanco: it's #2 in the agenda
21:07 <     tiddy> to give 1.x a future and a way.
21:07 <     tiddy> i think propel 2.0 should be a complete rewrite
21:07 < alanpinst> +1
21:07 <     tiddy> from the scratch.
21:07 <       jrw> +1
21:07 <     Jarda> +1
21:07 <    davide> +1
21:07 <     tiddy> ok :)
21:07 <  francois> tiddy: errr... -1, if I may
21:08 <     tiddy> of course!
21:08 <  francois> complete rewrite = not much to be used before a year and a half
21:08 < christoph> depends on features
21:08 <  francois> are you guys willing to work that long before a first release?
21:08 <    SidGBF> I think we should stay at the agenda... so far Francois spoke about 1.4 and symmfony wich means 30 days to build something usefull. So, let's stop wasting time ;)
21:08 <     Jarda> francois: that's why we would develop 1.x branch also
21:09 <  francois> SidGBF: :)
21:09 <    davide> francois: that's what 1.x is here for :)
21:09 <  francois> ok then
21:09 < alanpinst> my opinion: 1.x is great and works, but has many weaknesses that can't be fixed w/o BC-breaking freedom
21:09 <     Jarda> to have new features coming, but 2.0 would be a rewrite, allowing us to do things the right way
21:09 <  francois> I agree with rewriting some parts, just not going from scratch
21:09 < benschi__> Sorry for the disconnects but i'm on the Train, so Connection sucks
21:09 < alanpinst> so continue 1.x for legacy projects, re-write 2.0 with lessons-learned and try to end up with an API that isn't brutally different so that at least updating to 2.0 won't be totally horrible
21:10 <    davide> francois: what do you suggent we can keep from 1.x?
21:10 <    SidGBF> What about moderate the channel and give Francois +v?
21:11 <  francois> dunno, but make the change progressive
21:11 <  francois> otherwise there is a risk of Duke Nukem Forever Syndrome
21:11 <       jrw> I too think 2.0 should be a complete rewrite, but that isn't to say a lot of working code can't be carried over, if it still serves its purpose.
21:12 <  benschi_> To ask the question another Way what would be the right way to do things?
21:12 <       jrw> I mean, it's a LOT easier to rewrite something when you already have a working implementation to look at.
21:12 < alanpinst> +1 to jrw's comment
21:12 <     tiddy> i think a complete rewrite is absolutely necessary
21:13 <     tiddy> to get a better and more flexible build && plugin system.... 
21:13 <     tiddy> it`s a hard to do it by refactoring
21:13 <    SidGBF> Create 3.0 from scratch, use 2.0 as a step so don't waste what have been done so far.
21:14 <    SidGBF> what I want really is 1.x / symfony... let the guy talk!
21:14 < alanpinst> i am not sure there's much there to 2.0... what's there?
21:14 <  francois> tiddy: when do you think you can have a beta release ?
21:14 <    SidGBF> alanpinstein: wiki
21:14 < christoph> whats about refactoring the builder but keep the runtime code in the first step?
21:14 <  francois> christoph_: +1
21:14 < alanpinst> wiki roadmap != implemented code.. i meant what's been BULIT
21:14 < alanpinst> eesh built
21:15 <    v-dogg> (<- back, sorry had to go check up on our two little girls)
21:15 <     Jarda> propel 2.0 is no good, if we don't rewrite Criteria totally
21:15 <     Jarda> it's way too limiting atm
21:15 <     Jarda> doesn't serve it's purpose
21:15 <  francois> Jarda: so rewrite runtime and keep build :)
21:15 <     Jarda> francois: jrw has great ideas of the build system
21:16 <    v-dogg> rewrite both for 2.0
21:16 < alanpinst> build is pretty flawed too... very hard to use custom data types like GIS and ltree from postgres. i have to do way too much hacking
21:16 <  francois> ok then. tiddy, any timeframe ?
21:16 <     tiddy> timeframe is hard to discuss. first of all, we have to collect all ideas
21:17 <    v-dogg> ++
21:17 <    davide> IMHO, we actually can't set a right timeframe for 2.0, if we don't really decide what it's going to be, whether a refactoring or a complete rewrite. And I think this belongs to point 3 of the agenda :)
21:17 < alanpinst> yeah i think timeframe doesn't really matter much (for 2.0) it's not like anyone is waiting on it to do a project!
21:17 <       jrw> +1 for idea harvesting
21:17 <     tiddy> then give priorities to feature
21:17 <    SidGBF> davide: +1
21:17 <    v-dogg> 2.0 timeframe can be vague now, only the feature list is a must
21:17 <     tiddy> alanpinstein: +1
21:18 <  francois> ok, so if we agree on that shall we switch to #2?
21:18 < alanpinst> y
21:18 <    v-dogg> +1
21:18 <    davide> y
21:18 < christoph> +1
21:18 <  francois> so, bcak to the 1.x branch and the features we could stuff in one month
21:18 <  francois> s/bcak/back
21:19 <  francois> I have some ideas
21:19 <  francois> but I also want to hear yours
21:19 <  francois> let's start with what I believe can fit in
21:19 <    v-dogg> we could push 1.3.1 in one month but I reckon trying to get 1.4 (with new stuff) is not going to happen in one month so we shouldn't even try
21:20 <    v-dogg> 1.3.1 being a bug fix release
21:20 <    davide> closing some tickets would be something
21:20 <  francois> v-dogg: well, I'll take the challenge
21:20 <    v-dogg> there aren't too many simple ones
21:20 <  francois> what could easily fit in
21:20 <  francois> 1 - complex hydration
21:21 <  francois> 2 - a pager object
21:21 < alanpinst> i have a few simple open tickets I've reported: http://propel.phpdb.org/trac/search?q=apinstein&noquickjump=1&ticket=on
21:21 <     Jarda> PropelPager?
21:21 <  francois> 3 - code coverage
21:21 < alanpinst> why is  a pager needed for Propel? paging is a UI function...
21:21 <  francois> 4 - optional events
21:21 <     Jarda> we already have a pager..
21:21 <  francois> and 4 opens the door for 5 - behaviors
21:21 <     Jarda> http://propel.phpdb.org/docs/api/1.3/runtime/propel-util/PropelPager.html
21:22 <  francois> Jarda: my fault
21:22 <     tiddy> pager is not propel-stuff i think. we should propel keep clean in the future
21:22 <  francois> then 2 - pager object documentation
21:22 <  francois> tiddy: but it's already there ;)
21:22 <    v-dogg> francois: please elaborate on 3
21:22 <  francois> ah, I forgot: 1.4 should be totally BC with 1.3
21:22 <    v-dogg> agreed
21:23 <  francois> and of course, the wiki has a lot of tickets waiting to be fixed
21:23 <    davide> yes, this is for sure
21:23 <       jrw> tiddy: Sure a 2.0 could ship with a Pager and all sorts of bells and whistles, but they should be "extensions" of some sort, the core should be tight.
21:23 <    v-dogg> +1
21:23 <  francois> v-dogg: sorry, I wanted to write : unit test coverage
21:23 <    v-dogg> ah, roger that
21:23 < alanpinst> if we are re-writing 2.0, what is the benefit of doing test infrastructure for the 1.x branch?
21:23 <     Jarda> pager documentation is easy, it's not that complicated :)
21:23 <    davide> by now, what percenteage of Propel is covered by tests?
21:24 <     tiddy> jrw: +1
21:24 <       jrw> alanpinstein: There already is a plethora of tests.
21:24 <     tiddy> extension sounds cool.
21:24 <       jrw> alanpinstein: And a ready infrastructure for them, the point was there should be more of the tests themselves. :)
21:24 < alanpinst> stuff like "pager" is typically put into a "contrib" folder
21:25 <  francois> so, for nay other proposal / desire for Propel 1.4, please shoot
21:25 <  francois> s/nay/any
21:25 <     Jarda> francois: optional namespacesupport?
21:25 <    davide> for 1.4 is too soon!
21:25 <  francois> Jarda: like PHP 5.3 namespace?
21:25 <     tiddy> i think we had some discussion about the resultsetiterator on dev-list
21:25 <     Jarda> francois: yeah
21:25 < guilherme> francois: item 1 may require quite a lot fo efforts
21:25 <    v-dogg> francois: multiple joins on the same table
21:25 < alanpinst> ok yeah of course more tests are always better, but as a community we have limited resources to apply to coding and I am not sure the benefit of "focusing" on improving code coverage for a branch that will be deprecated. clearly tests should be added to cover new feature development, but flushing out coverage isn't a good use of time IMHO
21:25 <  francois> Jarda: then no. It's too soon.
21:25 <  francois> v-dogg: +1
21:26 <    v-dogg> might not be easy, so 1.4 might come too soon
21:26 <     Jarda> francois: ok, it's been developed in trunk, but not tested..
21:26 <    fabpot> better introspection at runtime. One of the most limiting factor using Propel for advanced stuff is the lack of a good introspection API.
21:26 <  francois> guilhermeblanco: there is a patch pending in the wiki to allow just that
21:26 < alanpinst> +1 on introspection!!! i run into that problem frequently
21:26 <  francois> guilhermeblanco: requires unit tests, documentation, and it could do most of the job
21:26 < alanpinst> fabpot is in the same boat as me; integrating a framework with propel
21:26 <       jrw> +1 introspection
21:27 < guilherme> francois: ya... it's a hard component, may require too much efforts... just to notice
21:27 <  francois> fabpot: +1
21:27 < christoph> fabpot +1
21:27 <  francois> does that mean that most of the schema information should be copied in the Peer classes?
21:28 <       jrw> But I think we should be VERY careful with what we do with 1.4 - a hastily compiled 1.4 that ships with sf 1.3 is sure to convince anyone still wondering if they should abandon Propel for Doctrine.
21:28 < alanpinst> well Map already has some
21:28 <     tiddy> with linguafranca i started an inscpection tool some long time ago
21:28 <    fabpot> we already have Map classes, but they lack information available from the schema
21:28 < alanpinst> but stuff is spread everywhere
21:28 < alanpinst> i was very interested in linguafranca but it dies
21:28 < alanpinst> died
21:28 <    v-dogg> jrw: that's why I think one month is not enuf
21:28 <     tiddy> but propel 2.0 did not move forward, so i just stopped it
21:29 <  francois> ok. Any more ideas?
21:29 < guilherme> francois: details events
21:29 < guilherme> what's the idea?
21:29 <    fabpot> also,  a __call() method to avoid all the getAll...() generated methods
21:29 <    v-dogg> francois: you keeping track of the ideas or do you go thru the logs afterwards?
21:29 < guilherme> similar to what we have in Doctrine? pre/postHydrate, pre/postDelete?
21:30 <  francois> guilhermeblanco: add events to the classes that can be listened to at runtime by listeners
21:30 <  francois> v-dogg: I record the discussion
21:30 <  francois> guilhermeblanco: yes, similar to what symfony added to Propel
21:30 <       jrw> I think events are wayyy too much for 1.4.  They're 2.0 stuff IMO.
21:30 <  francois> but usig an existing event library, to avoid reinventing the wheel
21:31 < guilherme> francois: hm... we implemented DOM 2 Events
21:31 < christoph> fabpot: i dont see a real adventage from this
21:31 < alanpinst> why events vs callbacks (delegation)?
21:31 <     tiddy> which one do you think of?
21:31 < guilherme> internally to Doctrine
21:31 <    fabpot> as already discussed with Hans, using the sfMixer from symfony is not a good idea. Better go with the sfEventDispatcher
21:31 < guilherme> won't be difficult to implement your own
21:31 < alanpinst> delegation is lighter-weight and easier to use as a client
21:31 <  francois> fabpot: I totally agree with that
21:31 <    fabpot> which means that it will be difficult to be BC (as far as symfony is concerned of course)
21:32 <  francois> fabpot: not if it's optional
21:32 <    fabpot> francois: fair enough
21:32 <  francois> and activated globally (in propel.ini), and overriden at the table level
21:32 <    v-dogg> (no .inis please, xml :)
21:33 <    davide> (no .xmls please, yml :)
21:33 <    v-dogg> (yackml)
21:33 <  francois> ok. Any other idea for 1.4?
21:33 <     tiddy> why not implement normal SPL-Observer
21:33 <  francois> tiddy: not reinventing the wheel
21:33 <   benschi> Xml +1
21:34 <  francois> anyway, we'll have other meeting to discuss the implementation details I suppose...
21:34 <  francois> who is on to give a hand for propel 1.4 ?
21:34 <    v-dogg> which brings up another topic - too many options scattered around build.properties and runtime-conf
21:34 <    v-dogg> and schema.xml
21:35 < stephenli> *raises hand* I can help (I think)
21:35 < alanpinst> i can work on the bugs I reported -- though normally i fix any bug that I can (that doesn't require internals knowledge)
21:35 <  francois> stephenliberty, alanpinstein : you're in
21:36 <  francois> just give Propel 50% of your time, and give me 50% of your income ;)
21:36 <    v-dogg> me too, although - again - one month is too soon :)
21:36 < stephenli> ha, 50% of nothing is still nothing..
21:36 <  francois> all right
21:36 <  francois> shall we talk about Propel 2.0?
21:37 < stephenli> should I bug you later about what I'll be helping with specifically?
21:37 <  francois> stephenliberty: yes, please
21:37 < stephenli> k
21:37 < christoph> will help to
21:38 <  francois> tiddy: do you want to discuss Propel 2.0?
21:38 <     Jarda> I guess the discussion of 2.0 is, whether or not one is to be started
21:38 <       jrw> Well yeah, obviously.
21:38 <     Jarda> if yes, then it's just mindstorming etc (maybe in the wiki or later here..)
21:39 <    v-dogg> do we all agree that it would be a total rewrite more or less?
21:39 <     tiddy> yes, i`d like to do some brainstorming
21:39 < alanpinst> +1 mostly total rewrite
21:39 <       jrw> Didn't we get all +1's except for francois' last time?
21:39 <    v-dogg> keeping the principles but ditching the code?
21:39 <     tiddy> let`s put all ideas together
21:39 < guilherme> tiddy: you're the project lead of Propel 2.0, right?
21:39 <     tiddy> without giving comments, only collet
21:39 <    davide> I agree. If we can get something from 1.x it would be better, but it's not necessary
21:39 <     tiddy> yes
21:40 <       jrw> IDEA: A new, modular build system, making hacking in new features and testing the default ones easier
21:40 < guilherme> first you have to decide? what route you want to go? ActiveRecord? DataMapper?
21:40 <       jrw> IDEA: An event system (observer/callback/something else)
21:40 <     tiddy> if we get all feature wishes, we`ll discuss and evaluate.
21:40 <       jrw> IDEA: Separating properly build properties from schema data
21:41 < christoph> use of late static binding
21:41 <    davide> guilhermeblanco: With 5.3 we can get a good ActiveRecord implemetantion, so I think that's the way to go
21:41 < alanpinst> there are 4 major problem areas for me: 1) build process/workflow 2) runtime introspection of obj model 3) ability to deal with build tweaks (like custom data types) 4) problems using in a DI environment
21:41 <     tiddy> stiop.!
21:41 <     tiddy> stop!
21:41 <       jrw> :D
21:41 <     tiddy> let`s write it all down in wiki!
21:41 <       jrw> ... didn't I ask for us to do that last week? :)
21:41 < alanpinst> now? can we all edit the wiki at the same time?
21:42 <    v-dogg> no
21:42 <       jrw> Yeah, it'll just be pretty nerve wrecking as Trac will reject anyone's change if the document changed while you edited it. :)
21:42 <     tiddy> ok, i think we`ll should set a date for all ideas
21:42 <     tiddy> we`ll not collect all ideas todays.
21:42 <     Jarda> I think this isn't the time of discussing 2.0 features
21:43 < alanpinst> ok
21:43 <     Jarda> as they need a lot of thinking and rethinking and planning.. we can't start coding right away
21:44 < alanpinst> yeah i would imagine we will have 2-3 months of discussiosn
21:44 <     tiddy> alan: +1
21:45 <    davide> One thing I'd really appreciate would be to simplify Propel installation without symfony. A collegue tried it, but it was too complicated, and then he switched to Doctrine :)
21:45 <     Jarda> davide: well, maybe better documentation
21:45 <  francois> davide: the problem lies in te dependencies
21:45 <       jrw> davide has a point.
21:45 <     Jarda> and not instructing people to use pear
21:45 <       jrw> francois an even better one.
21:46 < christoph> i remeber the first time the ini stuff was confusing
21:46 <    v-dogg> davide: http://blog.veikko.fi/post/82466126/installing-a-project-local-propel
21:46 <     Jarda> propel has no ini
21:46 <     tiddy> dvide: +1
21:46 <     Jarda> or at least I have never used one..
21:46 <  benschi_> I would volunteer to write an installation guide
21:46 < stephenli> i've done at least five propel installations now without symphony, all of them going smooth as silk - the documentation was fine for me (but then again I'm pretty persistent)
21:46 <    v-dogg> installing propel is not difficult, it's just our documentation that makes it look that way :)
21:47 <  francois> benschi_: you got it. 
21:47 <     Jarda> I could maybe write pager documentation, as the api docs already has examples etc..
21:47 <     Jarda> if the pager would be introduces as a "feature" in 1.4
21:47 <    v-dogg> benschi_, francois: please read my blog post and don't learn people install globally via pear
21:48 < francois_> makes bundling code+tests+doc easier
21:49 <    davide> stephenliberty: I know it's not hard, but with configuration files etc is a lot harder than Doctine
21:49 <       jrw> Yeah, having people go around looking for the dependencies is just plain silly when we could include them IN A READY WORKING ORDER as externals from our repo.
21:49 <    v-dogg> what about the documentation tools? wiki has so many issues I really like to see something else being used
21:49 < alanpinst> what's wrong with the pear install? how are you installing it w/o pear?
21:49 <       jrw> I for one love Trac and its wiki.
21:49 <     Jarda> alanpinstein: read the blog post from veikko..
21:49  * everplays lost the meeting :(
21:50 <    SidGBF> everplays: it is kinda happening
21:50 <    SidGBF> but Im totally lost...
21:50 <    v-dogg> alanpinstein: I have 5+ projects in my workspace and all of them might use different propel version
21:50 < alanpinst> where is that blog post link?
21:50 <    v-dogg>  http://blog.veikko.fi/post/82466126/installing-a-project-local-propel
21:51 < alanpinst> oh heh duh it was in my scroll buffer i looked away for a bit and never saw it
21:51 < alanpinst> ah ok yeah.
21:51 < francois_> hall we talk about Propel website improvements?
21:51 < alanpinst> i pretty much do taht myself now :) except I still use pear to install
21:51 <     Jarda> it doesn't make sense to have propel installed "globally" on you dev computer as you can't have the same version in production if you don't check periodically if curren propel is wroking with the legacy-project
21:51 <    v-dogg> francois___: go go
21:51 <  benschi_> v-dogg: I'll use your Blog post As Reference when writing
21:51 < alanpinst> anyone else interested in moving the source to git besides me?
21:52 < stephenli> *raises hand*
21:52 <     Jarda> no git..
21:52 <    v-dogg> -1 for git
21:52 <     Jarda> svn:externals is pure love..
21:52 < alanpinst> i use a *local* pear for each project - so that isn't a problem
21:52 <  benschi_> -1 for git
21:52 <     tiddy> what are the advantages of git?
21:52 < alanpinst> well for open-source projects the stuff that GitHub offers are *incredible*
21:52 < alanpinst> the ability for anyone to fork a project and make changes and do pull requests is amazing
21:53 <    v-dogg> what does it offer that we would need?
21:53 < alanpinst> it makes it much easier to have dozens of feature and bug fix branches going on at the same time w/o it being a pain to administer
21:53 <     tiddy> is there a nice plugin for eclipse :)
21:53 <     tiddy> ?
21:53 < stephenli> i'd agree with alanpinstein - I've got a few projects based off the same code base. they're all forks and its very easy to merge in and out
21:53 < alanpinst> heh probably
21:53 <    davide> +1 for git
21:53 < everplays> git rocks :)
21:53 <  benschi_> alanpinstein: i don't think so many People will acually fork
21:54 < alanpinst> plus, if you make it "normal" for people to install from git instead then *everyone* is a dev
21:54 < alanpinst> and you don't have to mess with managing who can commit
21:54 < alanpinst> *anyone* can fork and "submit" a bugfix
21:54 < alanpinst> and you only merge it into trunk if you like it
21:54 <     Jarda> alanpinstein: and someone has to accept the fices?
21:54 <     Jarda> ok.
21:54 < francois_> sorry guys, bad connection
21:54 <    v-dogg> really bad :)
21:55 < francois_> did I miss a consensus about going to git?
21:55 < alanpinst> yes.. github has an entire infrastructure for managint that
21:55 <     Jarda> well, +-0 for git then maybe...
21:55 < alanpinst> lemme show you something (sec..)
21:55 <     Jarda> I would like to have post commit hooks for automated unit testing
21:55 <    v-dogg> francois___: a few votes against, a few in favour
21:55 <     Jarda> dunno if that's possible with or without git
21:55 <    v-dogg> does trac have git integration?
21:55 < stephenli> i think it is
21:56 < francois_> ok. What about... a real blog in the Propel website?
21:56 <    v-dogg> francois___: yes
21:56 <     Jarda> what about a real website..
21:56 <     Jarda> trac isn't a website
21:56 < everplays> +1 francois____
21:56 <    v-dogg> was just going to say.. :)
21:56 <    v-dogg> yeah, trac is for project management, we need a website
21:56 <     Jarda> when I first came to propel I tried to search for the website, found only a wiki and went away
21:57 <     Jarda> it's ugly and non-functional
21:57 < alanpinst> github also has full infrasatructure for issue management, wikis, etc
21:57 < stephenli> Jarda: took a second: http://www.kernel.org/pub/software/scm/git/docs/v1.3.3/hooks.html (sorry, convo seems to have moved elsewhere)
21:57 < alanpinst> it is light years ahead of track and moving fast
21:57 < alanpinst> so ok look at this: http://github.com/rails/rails/network
21:57 < alanpinst> note that you can drag the flash thingie up/down/left/right
21:57 < alanpinst> the power of git + github is how easy it makes involvement
21:58 <  benschi_> You can actually Themen trac quiet good
21:58 <    v-dogg> benschi_: but it's still not a website
21:58 <    v-dogg> it's an issue tracker + wiki
21:58 < everplays> what about propel it self? these are just version control, blog, ...
21:58 <     tiddy> ever heard of http://arbitracker.org/news.html
21:58 < francois_> if anyone else has the complete logs, I'll appreciate :(
21:58 <     tiddy> written in php 
21:58 < everplays> did i miss main part?
21:59 <  benschi_> This is true, but atleast it aint ugly anymore
21:59 < stephenli> everplays: yeah, i think so
21:59 <     Jarda> francois_with_wi: I can give you the logs, just gimme your e-mail
22:00 < francois_> ok, better website could be with or without trac
22:00 < francois_> but it needs some design skills
22:00 <       jrw> -1 for git
22:00 < francois_> any volunteer?
22:00 <       jrw> (yeah I was away for a bit :)
22:00 <     Jarda> francois_with_wi: I guess v-dogg has a draft already
22:00 <    v-dogg> francois___: we could start with something pretty simple (anything is better than the current front page)
22:01 < everplays> who count votes?
22:01 < everplays> \o/
22:01 <    v-dogg> francois___: http://propel.phpdb.org/trac/wiki/Development/NewSite
22:01 < francois_> everplays: after counting, there doesn't seem to be a consensus about git. So no migration for now. 
22:02 < stephenli> I can help with design. I'm not mr. magical when it comes to design stuff, but I can help out if no one else steps up
22:02 <    v-dogg> francois___: "Needs candy of course (I'm no designer) but I'd like to keep it minimalistic - very little text, max 5 items in the main menu and a few code examples (in the end that's what we all want see straight away) "
22:02 < everplays> :(
22:02 <    v-dogg> (from the wiki page)
22:02 < alanpinst> i was about to ask about git... did anyone have any thoughts about it now that we discussed?
22:02 < francois_> v-dogg: well, if noone else comes with anything better, we've got a design
22:03 < everplays> alanpinstein: i think git can handle propel better
22:03 < everplays> who will be project leader?
22:03 < benschi__> re
22:03 < francois_> stephenliberty: can you start frmo v-dogg's draft?
22:03 < stephenli> sure.. where is it?
22:03 <    v-dogg>  http://propel.phpdb.org/trac/wiki/Development/NewSite
22:03 <     Jarda> http://veikko.fi/propelorm.org/ && http://propel.phpdb.org/trac/wiki/Development/NewSite
22:03 < stephenli> er
22:03 < stephenli> yeah, just saw it, sorry
22:03 < francois_> everplays: ?
22:04 < alanpinst> maybe a better git question is "does anyone that *understands* git think we shouldn't move to it?"
22:04 < stephenli> francois_with_wi: can I get a list after the meeting of folks who are responsible for site stuff, so I can bug them about design revisions?
22:04 < benschi__> v-dogg: I could try to get your layout ported to trac
22:05 < everplays> alanpinstein: yeah, many project are migrating, so day after day people understand it better
22:05 < francois_> stephenliberty: yes
22:05 <    v-dogg> alanpinstein: is there a git-to-svn bridge?
22:05 <     Jarda> no no, no trac for website
22:05 <    v-dogg> I like my svn:external
22:06 <    davide> v-dogg: yes, you can even use git locally with a remote svn repo
22:06 < stephenli> v-dogg: git can commit and pull to/from svn - we used it at my old work before moving completely over
22:07 <    v-dogg> can github and an svn repo work side by side?
22:07 < francois_> ok guys, let's cnotinbue wit hthe agenda
22:07 <    v-dogg> francois_with_wi: one more website issue - the domain
22:07 < francois_> propel will not move to git now, it doesn't mean it won't move to git in the future
22:08 < francois_> v-dogg: yes. Do you own a better one?
22:08 < everplays> francois_with_wi: why not?
22:08 <    v-dogg> francois_with_wi: propelorm.org (and .com)
22:08 <    davide> propelorm.org, propel-project.org...
22:08 < alanpinst> we used the git-svn bridge too (use git to commit to svn essentially). but that is realy only a solution for an individual coder, *not* for the project
22:08  * everplays likes propelrom.org
22:09 < francois_> someone owns this?
22:09 < everplays> i think it's better to discuss about features propel needs & we **faced** in our projects, then vote for project leader
22:09 <    v-dogg> <-
22:09 <       jrw> +1 propelorm.org
22:09 <   benschi> I would first keep the structure intact, evaluate the alternatives and then discuss it again
22:10 < francois_> everplays: didn't we just discuss the features?
22:10 <     tiddy> i think we should stop dicussion about git, the version control system is the less problem.
22:10 < everplays> francois_with_wi: it's all about wiki, blog, ...
22:10 <    v-dogg> and don't we have a leader?-)
22:10 < everplays> it's not directly related to propel
22:10 < everplays> v-dogg: who's the leader?
22:10 < francois_> everplays: do you want to become the leader?
22:11 < everplays> nope
22:11 < francois_> well, then it's tiddy and me until someone else asks for it
22:11 <    v-dogg> :D
22:11 <     tiddy> :)
22:11 < everplays> :P
22:11 < stephenli> can I suggest something regarding the site (specifically a very new thing that might make meetings easier)?
22:11 < francois_> stephenliberty: shhot
22:11 < everplays> nice to meet you tiddy & francois_with_wi :)
22:11 < everplays> \o/
22:12 <     Jarda> francois_with_wi: can we have a better environment than phpdb.org..?
22:12 < francois_> Jarda: environment?
22:12 < everplays> stephenliberty: ...
22:12 <    SidGBF> is the meeting over?
22:12 < francois_> SidGBF: nope
22:12 <     Jarda> I mean, something like a svn that isn't down half the time..
22:12 < stephenli> I think that we should have an actual on-site meeting area.. with a chat and whiteboard. I'm loosing track of whats going on, and I'm sure its going to be hell to try and get all of the ideas out of this massive log
22:12 < francois_> Jarda: +100
22:12 -!- v-dogg changed the topic of #propel to: Meeting is: ON. Current topic: website and servers etc.
22:13 < francois_> benschi volunteered on this one
22:13 <     tiddy> what hardeware is actually behind propel-site / svn and so on?
22:13 <   benschi> As I already posted on the mailing list, I have plenty of capacity free on a server of mine
22:13 <     tiddy> i didnt have time to look at it until now.
22:13 < francois_> stephenliberty: I'll add track tickets for features we agreed on for 1.4
22:14 <     Jarda> yeah, I also do have a virtual server that has capacity (at least to work as a backup svn-mirror)
22:14 < everplays> francois_with_wi: what about idea's witch is not in track?
22:14 < alanpinst> ok guys I am outta here. thanks to the new leaders for taking over -- look forward to seeing hte de-brief and moving forward...
22:14 < francois_> tiddy: the problem is not the hardware, it's the time required to administer the serever
22:14 < stephenli> francois_with_wi: meh, I didn't exactly explain it right. I'll bug you later about it
22:14 < francois_> everplays: add it as a ticket, we'll discuss it in the ticket's comments
22:14 <   benschi> francois_with_wi: but once eveything is setup right, it should just run
22:15 < francois_> everplays: or send amail to the devs mailing-list
22:15 < everplays> francois_with_wi: ok
22:15 <     tiddy> sry... i ask for hardware. perhaps we can find a new sponsor for hardware or a nice vm
22:15 < francois_> tiddy: benschi offers hardware and administration
22:16 < francois_> benschi: is that correct?
22:16 <   benschi> that is correct
22:16 <    SidGBF> francois_with_wi: I really expected a bit more of focus to 1.4/symfony once is the closest deadline we have
22:16 < stephenli> (and I can help if need be - I'm hosting/administering 13 servers right now.. whats another one?)
22:17 <    v-dogg> benschi: are you able to give sudo rights to others as well? currently the problem has been that none of us here have the rights to manage the server and when it's down it's down
22:17 <    v-dogg> here == irc
22:17 <    v-dogg> == pretty active and often awake & available
22:17 < francois_> v-dogg: I guess 2 sudoers is already a lot, but you can tell me
22:18 <   benschi> I'm not very pleased with it, because I don't know you guys in persona, but we can agree on that somehow
22:18 <    v-dogg> don't know if sudo is actually needed but some administrative rights 
22:18 < francois_> SidGBF: we can go on speaking about 1.4 after talking about symfony integration if you want
22:18 <   benschi> basicly I could offer you two optinions. 
22:18 <    v-dogg> to restart dead processes etc.
22:18 < francois_> but we have to move forward. It's already 1:30 of meeting!
22:19 <    SidGBF> francois_with_wi: ignore me, just seen what's the current topic... my bad :p
22:19 <   benschi> first host it on the main server which gives you more power, or second own vServer with less power
22:19 <    v-dogg> francois_with_wi: next topic?
22:19 < francois_> v-dogg: symfony integration
22:19 -!- v-dogg changed the topic of #propel to: Meeting is: ON. Current topic: Symfony integration
22:19 < francois_> :)
22:20 <    v-dogg> go go
22:20 < francois_> benschi: you can go on whil we talk
22:20 < francois_> so, symfony's Propel plugin is the first contact many people have with Propel
22:20 < francois_> it should be as polished as Propel itself
22:21 < francois_> and we should be commited to its features as well
22:21 < francois_> Unfortunately, the sfProplePlugin dev team is pretty much dealing with Doctrine
22:22 < francois_> so we need somebody to port sfDoctrinePlugin features to sfPropelPlugin
22:22 < francois_> any other suggestions about symfony?
22:23 < francois_> fabpot: any suggestion?
22:23 <    davide> A little question, with sfPropelPlugin you mean the one shipped with symfony or the one in the plugins list of the website?
22:23 < francois_> davide: it's the same, I believe
22:24 <    davide> "sfPropelPlugin adds support for Propel 1.3 for symfony 1.1."
22:24 < francois_> who else wanted to talk about symfony nitegration?
22:24 < francois_> davide: ah, ok
22:25 < francois_> I proposed to help maintain sfPRopelPlugin for symfony 1.3
22:25 < francois_> ok then
22:25 <    davide> So it will be basically a port of Propel 1.4 for symfony < 1.3?
22:25 < christoph> since im using symfony i will contribute patchs
22:25 -!- francois_with_wi changed the topic of #propel to: Meeting is: O?. Current topic: Miscellaneous
22:26 <    SidGBF> that was quick...
22:26 -!- francois_with_wi changed the topic of #propel to: Meeting is: ON. Current topic: Miscellaneous
22:26 < francois_> SidGBF: we can come back to symfony if you peaople have more to say...
22:26 <       jrw> So we went over the symfony stuff without hearing a word from fabpot for example..?
22:26 <    SidGBF> Well, my main question is what we should expect to sf 1.3/1.4?
22:27 < francois_> SidGBF: do you mean: will they drop Propel?
22:27 <    SidGBF> no, I mean 'what will happen to propel to 1.4/1.4'?
22:28 <    davide> On the sf blog it was posted that propel will be dropped in sf 2, and deprecated in sf 1.3
22:28 <  francois> well, that's why it's important that we ship soimething new for mid october
22:28 <    davide> it was before francois and tiddy took the keys of the project, right?
22:28 <  francois> SidGBF: I'm not getting you
22:28 <    SidGBF> IMHO I think that the problem is exactly that Propel team didn't supported sf too much. As you said, many prople meets propel with it, so, would be a difference if Propel team handle that
22:29 <    SidGBF> Will be shipped something to 1.3?
22:29 <  francois> SidGBF: I agree, and I have nothing against symfony integration
22:29 <    SidGBF> sf 1.3, I mean
22:29 <  francois> SidGBF: yes
22:29 <     tiddy> i think propel must be that flexible that it can be shipped with nearly other framework, components or something else.
22:30 <    SidGBF> will it be propel 1.3.1 or 1.4?
22:30 <  francois> SidGBF: Symfony 1.3 - Propel 1.4
22:30 <     tiddy> so, I`ll focus on it for propel 2.0
22:30 <    v-dogg> propel team has been unable to support sf users because sf has made so many tweaks to it (or so it seems)
22:30 <    davide> tiddy: +1 I've seen tutorials on how to integrate Doctrine with ZF, but as far as i know, nothing about propel
22:31 <  francois> v-dogg: well, some of those tweaks deserve backporting to the Propel core IMHO
22:31 <  francois> davide: would you be willing to write one?
22:31 <    v-dogg> francois: sure
22:31 <    SidGBF> In symfony propel and doctrine are already plugins, so, there's no problem have 1.3/1.3.1/1.4 for now and later a 2.0, of course it hit harder when launched together, but the point now is to at least hit the target :p
22:31  * everplays thinks propel needs more documents & blog posts
22:31 <     tiddy> so, propel is not a part of symfony. propel is interested in being part of symfony, agavi, zend ... or to be used in combination
22:32 <     tiddy> that must be target!
22:32 <    davide> francois: no I'm sorry, it was an example, I don't really know the zend framework :)
22:32 <  francois> tiddy: yes of course
22:32 < stephenli> tiddy +1
22:32 <    v-dogg> propel is interested in being easily used with any framework
22:32 <    v-dogg> personally I dislike bolting an ORM tightly into a framework
22:33 <    v-dogg> but that's just me and it doesn't matter now :)
22:33 <    davide> v-dogg: why? It's the M of MVC
22:33 <       jrw> tiddy +1
22:33 <    v-dogg> nonono
22:33 <     tiddy> so, lets try to get out a signal to all people out there by releasing propel 1.3.x and 1.4 as soons as possible as francois proposed.
22:33 <    v-dogg> it IS NOT the M in MVC
22:33 <    SidGBF> tiddy: Except that Symfony offered a hand to propel, giving it a prefference for almost 2 years. Symfony will be launched in 2 month so, this is the closest target
22:33 <    v-dogg> that's what pretty much every framework do wrong
22:33 < everplays> v-dogg: +1
22:34 <  francois> a little word about Propel 1.3.1
22:34 <  francois> I'm not sure if there will be any
22:34 <  francois> especially, some of the commits to the 1.3 branch are not entirely BC
22:34 <  francois> and will probably be reverted in the 1.3 branch
22:34 <     tiddy> i know that symfony is important to propel and has been important to propel.
22:35 <  francois> but will remain in the 1.4
22:35 <    davide> francois: if we need to release a 1.4 soon, I don't think releasing a 1.3.1 would be needed
22:35 <  francois> davide: I agree
22:35 < everplays> davide: +1
22:35 <  francois> so please, focus your efforts on the 1.4 branch
22:36 <  francois> (unless you want to work in 2.0, of course :))
22:36 < christoph> looks like 1.4 will only be a little step anyway
22:36 <  francois> christoph_: one little step for man...
22:36 <    davide> "one little step for symfony, one giant leap for propel" :)
22:36 < Crafty_Sh> I came into the conversation a little late, have the milestones for 1.4 / 2.0 been set, and is there a place I can see them?
22:36 <  francois> davide: great minds
22:36 < christoph> yea:-)
22:37 <    davide> one last question, then I have to leave
22:37 <  francois> Crafty_Shadow: we've discussed 1.4 a little, I will be posting tickets to sum all that up
22:37 <    davide> francois: you won't move to python, right? :)
22:37 <  francois> and the discussion remains open in the devs mailing-list
22:37 <  francois> davide: you never know...
22:38 <    davide> bye everybody, see on the ML
22:38 <  francois> davide: not in the near future anyway
22:38 <     tiddy> ok, guys, i have to leave and will keep stick to the list.
22:39 < christoph> so hans isnt here, right?
22:39 <  francois> ok, it's been 1h and 45m; anyone wants to add something?
22:40 < stephenli> yep
22:40 < stephenli> whats the plan with 'spreading the word'?
22:40 <    v-dogg> christoph_: I don't think hans has ever been here :)
22:40 < stephenli> i.e. trying to get a larger user base?
22:41 <  francois> stephenliberty: don't get kicked of symfony - 
22:41 < stephenli> besides making it easier to use with frameworks
22:41 <  francois> stephenliberty: make a kick ass product
22:41 <  francois> stephenliberty: write tutorials in blogs
22:41 <  francois> stephenliberty: and make conferences
22:41 < stephenli> good!
22:41 <    v-dogg> stephenliberty: development activity, new web site, dev blog etc.
22:41 < stephenli> just making sure :P
22:41 < stephenli> relying on one method always ends up with 'very bad things'
22:42 < Crafty_Sh> This might be a bit on the symfony side, but I think the schema definitions should allow the generation of form/filter classes to be disabled
22:42 <  francois> stephenliberty: your help is welcome if you have some writing / marketing skills
22:42 <   benschi> Okay I outta here for now.
22:42 <   benschi> see you guys
22:42 < Crafty_Sh> however the way it's done right now requires changes to the Propel plugin
22:42 <     Jarda> I have to get to bed, should I post you the logs francois now or in the morning?
22:42 <  francois> and of course, reactivity in the mailing-list and IRC channel is a bonus
22:43 <  francois> Crafty_Shadow: this should be handled by the sf Propel plugin
22:43 < stephenli> well, now that I've learned how to use IRC, I should be around almost 24/7.. I've got it hooked into pidgin :P
22:44 <  francois> ok, then
22:44 <  francois> anything else?
22:44 < Crafty_Sh> ok
22:44 <  francois> please voice your opinion
22:44 <  francois> (or be quiet forever ;) )
22:44 <     tiddy> sry, guys have to leave tomorrow early. bye
22:44 < christoph> im glad something is happening
22:44 -!- v-dogg changed the topic of #propel to: Propel PHP5 ORM :: Have a question? Just ask it, and wait patiently (really, be very patient, a couple of hours is no exception) :: Creole and Phing question friendly channel, too :: Alternate svn mirror: http://propel.mirror.svn.symfony-project.com/
22:45 <  francois> v-dogg: you're quick
22:45 < christoph> hehe
22:45 <  francois> so, the meting is officially over
22:45 <  francois> thanls you all for coming
22:45 <    v-dogg> just making sure the topic gets back, do continue if you still have something to say :)
22:45 < christoph> ok, thank you all
22:45 <  francois> and please tell it to your friends: Propel is alive and doing well
22:46 <  francois> It's got a new liver and all that