2.0.0.Final Roadmap Planning Community Meeting | 2 March 2012

Version 1

    [10:31] <ALR> Welcome, all.

    [10:31] <ALR> We'll give the usual 5-minute leeway I suppose

    [10:31] <ALR> And then kick off the JSFUnit Community Roadmap meeting

    [10:31] <jose_freitas> okay

    [10:31] <lfryc> Hi guys!

    [10:31] <kenfinnigan> sounds good

    [10:31] <bleathem> 5-minutes - you're generous!

    [10:31] <lfryc> meeting over :-)

    [10:33] <jose_freitas> hehehe

    [10:33] <jose_freitas> hi lfryc

    [10:33] <ALR> lfryc: That's great!

    [10:33] <lfryc> hey Jose!

    [10:33] <ALR> Rollcall to aslak.

    [10:35] <ALR> And let's do this.

    [10:35] == lkrejci [~lkrejci@redhat/jboss/lkrejci] has joined #jbosstesting

    [10:35] <bleathem> I read that article you forwarded ALR

    [10:35] <ALR> So the inception of this meeting was triggered by a request from the AS team to update all components to "Final"-qualifier level.

    [10:36] <bleathem> ALR:  thanks for that

    [10:36] <ALR> Among them, JSFUnit.

    [10:36] <ALR> bleathem: Sure thing.

    [10:37] <ALR> We've last pushed a release: 2.0.0.Beta2 in September.

    [10:37] <ALR> So what I'd like to know is:

    [10:37] <jose_freitas> long time ago

    [10:37] <ALR> What are the project requirements for 2.0.0.Final?

    [10:37] <ALR> (If they're defined)

    [10:37] <ALR> If not, are there bugs or missing features blocking a final release?

    [10:37] <kenfinnigan> based on JIRA there's 30 odd issues assigned to Final

    [10:38] <ALR> I'm not sure I trust the state of JIRA at the moment.

    [10:38] <jose_freitas> yes

    [10:38] <kenfinnigan> fair enough

    [10:38] <jose_freitas> I guess that that's the only point we can count

    [10:38] <jose_freitas> I'm not sure about jsfunit usage too

    [10:39] <jose_freitas> lemme pick an email where stan stated some points to release a final

    [10:39] <ALR> jose_freitas: Are you in a position to determine which issues would actually block release?

    [10:39] <ALR> Or set a Project Roadmap?

    [10:39] <kenfinnigan> as you mentioned ALR, given there hasn't been any commits for 6 mths (since last release), does it make sense to release Final and then move everything else to subsequent version?

    [10:39] <ALR> Also I'll see if I can find Stan.

    [10:40] <jose_freitas> we need to push another beta working perfeclty with AS5 and 6

    [10:40] <jose_freitas> right now the beta 2 works with AS7 only.

    [10:40] <ALR> kenfinnigan: I believe that makes sense so long as there are no blocker-level bugs or missing features.  Otherwise we'd be advertising something as "Final" which is of inferior quality.

    [10:40] <kenfinnigan> jose_freitas: are there jira's for getting it working with AS5/6?

    [10:40] <bleathem> There is also the JSFUnit planned rewrite to consider

    [10:40] <kenfinnigan> bleathem: planned rewrite?

    [10:41] <lfryc> right

    [10:41] <lfryc> JSFUnit .NG

    [10:41] <ALR> bleathem: Explain that bit?

    [10:41] <bleathem> I'll defer to lfryc

    [10:41] <bleathem> it's his and aslak's brainchild

    [10:41] <lfryc> https://issues.jboss.org/browse/ARQ-578

    [10:41] <jbossbot> jira [ARQ-578] Create a Extension for handling Server State assertion during normal Servlet/JSF requests [Open (Unresolved) Feature Request, Major, Unassigned] https://issues.jboss.org/browse/ARQ-578

    [10:41] <lfryc> we are going to take this one, since it looks very nice for testing RichFaces

    [10:42] <lfryc> very promising

    [10:42] <lfryc> there is a gist for API proposal: https://gist.github.com/1184904

    [10:42] <bleathem> It's an effective replacement for JSFUnit

    [10:42] <ALR> lfryc: So this looks like a JSFUnit 3.x thing?

    [10:42] <ALR> Or an otherwise-rebranded project under the Arquillian Umbrella?

    [10:42] <lfryc> ALR: rather something-else-than-JSFUnit, since it should be more general

    [10:42] <ALR> OK, good.

    [10:42] <lfryc> for any HTTP client

    [10:42] <lfryc> but we will start with JSF

    [10:43] <lfryc> need to come with new name

    [10:43] <lfryc> calling it JSFUnit.NG in meantime

    [10:43] <ALR> Because looking forward after JSFUnit 2.x, we'll need a clear management for this project, or rather, its mission of testing component state

    [10:43] <kenfinnigan> lfryc: so it would allow snooping on lifecycle of JSF for instance?

    [10:43] <ALR> What is NG?

    [10:43] <lfryc> kenfinnigan: exactly

    [10:43] <kenfinnigan> cool

    [10:43] <kenfinnigan> Next Gen is my guess

    [10:43] <lfryc> Next Generation, yeah

    [10:43] <ALR> k

    [10:43] == lfryc [~lfryc@redhat/jboss/lfryc] has quit [Read error: Connection reset by peer]

    [10:44] <ALR> Good, that begins to answer my questions for Part II of this discussion.

    [10:44] <ALR> I'll wait for Lukas to reconnect.

    [10:44] <kenfinnigan> it sounds like ARQ-578 would lead to something outside JSFUnit

    [10:44] <jbossbot> jira [ARQ-578] Create a Extension for handling Server State assertion during normal Servlet/JSF requests [Open (Unresolved) Feature Request, Major, Unassigned] https://issues.jboss.org/browse/ARQ-578

    [10:44] <kenfinnigan> does this mean that JSFUnit would cease dev after 2.0.0.Final if this new generic project kicked off?

    [10:44] == lfryc [~lfryc@redhat/jboss/lfryc] has joined #jbosstesting

    [10:44] == jose_freitas [~spinner@186.215.116.36] has quit [Ping timeout: 260 seconds]

    [10:45] <bleathem> kenfinnigan: that's my take on it

    [10:45] <lfryc> sorry guys

    [10:45] <ALR> Yes, which I think is a fine step.  JSFUnit has been faltering a bit on its own.  Aligning it under the ARQ community gives it some greater support.

    [10:45] == jose_freitas [~spinner@186.215.116.36] has joined #jbosstesting

    [10:45] <lfryc> <lfryc> the big problem of JSFUnit is it can't verify anything else than FacesContext state

    [10:45] <lfryc> <lfryc> the .NG should test also CDI beans, or anything what can Arq provide (enrich)

    [10:45] <lfryc>  Tcl interface unloaded

    [10:45] <lfryc>  Python interface unloaded

    [10:45] <jose_freitas> sorry, network blackout

    [10:45] <bleathem> we should focus on gettting JSFUnit 2.0.0.Final out, then shift focus to JSFUnit.NG

    [10:45] <kenfinnigan> bleathem: agree

    [10:45] <ALR> Yes.

    [10:45] <lfryc> right

    [10:46] <lfryc> sorry for Tcl/Python lines ;-) wrong c&p

    [10:46] <ALR> As I said, Path II of this discussion looks like we have more answers than I was expecting, so great!

    [10:46] <ALR> So back to Part I for the time being.

    [10:46] <kenfinnigan> I'll be coming at it from a portal angle, so would be good to get a generic UI testing framework built

    [10:46] <ALR> Who is qualified to gauge the current state of 2.0.0 and brand it as "Final"?

    [10:47] <jose_freitas> there's a major issue with @Proxy in jetty

    [10:47] <bleathem> I'd say jose_freitas has the most experience with it out of those in attendance

    [10:47] <kenfinnigan> +1

    [10:47] <jose_freitas> but I have to look at jira to see others

    [10:48] <jose_freitas> there's definetelly the need to make it work with AS5 and AS6

    [10:48] <jose_freitas> (I don't know if you get my message earlier)

    [10:48] <ALR> jose_freitas: OK.  Do you have some time now to do a review of JIRA?

    [10:48] <jose_freitas> for beta3

    [10:48] <ALR> Hmm, AS5 and 6.

    [10:48] <kenfinnigan> does it need to work with AS5?

    [10:49] <jose_freitas> if you guys think that it's not needed, ok

    [10:49] <ALR> We're asking.

    [10:49] <jose_freitas> Actually, I'm not aware of the jsfunit users requirement

    [10:49] <jose_freitas> Maybe it isn't

    [10:49] == jpapouse [~jpapouse@redhat/jboss/jpapouse] has joined #jbosstesting

    [10:49] <jose_freitas> it would make things simple

    [10:49] * ALR pinged Stan.

    [10:49] <kenfinnigan> I guess the question is, does an older version of JSFUnit work fine with JSF 1.2, or is there something in the newer version that adds benefit to testing with JSF 1.2?

    [10:50] <ALR> Let's see if he can join us.

    [10:50] <jose_freitas> it'd be great

    [10:50] <kenfinnigan> If not, then AS5 probably not needed

    [10:50] <ALR> I think it depends.

    [10:50] <ALR> If we've said previously that the 2.x series would be compatible w/ AS5 and AS6, then it must be.

    [10:50] <kenfinnigan> ah, true

    [10:50] <ALR> "Final" is about meeting the project's stated goals.

    [10:50] <ALR> But I'm not aware of a 2.x written roadmap.  jose_freitas, are you?

    [10:51] <jose_freitas> Basically

    [10:51] <jose_freitas> I don't think that there's a formal roadmap

    [10:52] <jose_freitas> but there's some needs we've been talking about to make it more useful for richfaces and cdi

    [10:52] <ALR> jose_freitas: Then you, my friend, are in the enviable position of getting to define one.

    [10:53] <ALR> As I believe you're the most familiar with the gap between what it does, and what it needs to do to be called "Final".

    [10:53] <bleathem> The RichFaces requirements are taken care of in the .NG plan, we have not stake in the 2.0.0.Final

    [10:53] <ALR> Keeping in mind that "Final" does not mean *every* feature must be in place.  Just the ones essential to its mission, and no blocker bugs.

    [10:53] <bleathem> ^have no stake

    [10:53] <ALR> That's why there's 2.0.1 and 2.1.0 later.

    [10:53] <jose_freitas> as 2.x I understood after 2.0.0.Final

    [10:54] <kenfinnigan> I would agree that CDI and RF stuff doesn't need to be in Final, but rather .NG

    [10:54] <jose_freitas> ok

    [10:54] <jose_freitas> +1 to that

    [10:54] <ALR> bleathem: OK, so RF will block on JSFUnit.NG?

    [10:54] <jose_freitas> but I understood 2.x as the future pos 2.0.0.Final

    [10:54] <bleathem> not sure what you mean by block

    [10:54] <ALR> Good, that's what I call "efficient procrastination."

    [10:55] <bleathem> +1

    [10:55] <jose_freitas> hahah

    [10:55] <jose_freitas> so

    [10:56] <ALR> So: We need 1st a roadmap for 2.0.0.Final

    [10:56] <jose_freitas> bleathem, lfryc, so you have some idea for a roadmap for jsfunit.ng?

    [10:56] <ALR> It can be slim.

    [10:56] <jose_freitas> I can do that

    [10:56] <ALR> jose_freitas: Super!

    [10:56] <lfryc> jose_freitas: yes

    [10:56] <lfryc> the development will be started in a week or so

    [10:56] <jose_freitas> Awesome

    [10:56] <lfryc> on alpha version is planned for ~1,5 months after

    [10:56] <kenfinnigan> lfryc: will that be in arq github or somewhere else?

    [10:57] <lfryc> I would like to share the thoughts around it

    [10:57] <lfryc> on arq wiki

    [10:57] <ALR> Hang on guys, I know we're eager to talk about NG

    [10:57] <lfryc> :-) sure

    [10:57] <ALR> But I've gotta report back to AS team some resolution on 2.0.0 first, if we can.

    [10:58] <kenfinnigan> jose_freitas: do you have access to update JIRA?

    [10:58] <kenfinnigan> if so, you could update the roadmap there for 2.0.0.Final

    [10:58] <jose_freitas> yes

    [10:58] <ALR> That's a good place to put issues, yes, kenfinnigan .

    [10:58] <ALR> But also JIRA doesn't account for high-level objectives.

    [10:59] <ALR> I could also suggest simple Wiki, for instance:

    [10:59] <ALR> https://community.jboss.org/wiki/ShrinkWrapResolversReleasePlanningAndRoadmap

    [10:59] <ALR> Nothing crazy.

    [11:00] <ALR> Just a high-level overview of what we want to accomplish, and by when.

    [11:00] <kenfinnigan> need to steal that for PBR ;-)

    [11:00] <jose_freitas> heheeh

    [11:00] <ALR> PBR?

    [11:00] <kenfinnigan> portlet bridge

    [11:00] <jose_freitas> seems nice, after feeding the jira, we can create a public list like that

    [11:01] <ALR> (Where I come from that's Pabst Blue Ribbon, the beer choice of frats!_

    [11:01] <kenfinnigan> lol

    [11:01] <jose_freitas> I'll ask stan if it's possible to give me permission to change jsfunit wiki site

    [11:01] <ALR> Keep in mind ANY approach that works for you, jose_freitas, is the right one.

    [11:01] <ALR> So long as it communicates intent.

    [11:02] <kenfinnigan> ALR: do the AS team have a deadline of when they want Final out?

    [11:02] <ALR> jose_freitas: You should be able to just create a new article on the Wiki regardless

    [11:02] <jose_freitas> hm, nice

    [11:02] <ALR> kenfinnigan: Yesterday

    [11:02] <kenfinnigan> aha

    [11:02] <ALR> But honestly I'm not a fan of "promoting" components just because the downstream project wants it.

    [11:03] <ALR> It's a silly practice which kills the purpose of qualifiers in the first place.

    [11:03] <ALR> And also, JSFUnit is a test dependency only, so we might have some leeway there.

    [11:03] <ALR> That said, we need to go to Final anyway here to draw the line for 2.x and move forward with NG

    [11:03] <jose_freitas> ok

    [11:03] <ALR> So this makes, I think, a great opportunity to do that

    [11:04] <ALR> jose_freitas: When would you like to meet up again to discuss a roadmap?

    [11:05] <jose_freitas> I can prepare a roadmap proposal today at night, so we could meet again tomorrow maybe

    [11:06] <jose_freitas> or monday, when it better fits for you

    [11:06] <jose_freitas> for you all

    [11:06] <ALR> jose_freitas: Hehe, I love your enthusiasm.  And we each work weekends pretty often.  But can we keep from scheduling meetings then?

    [11:06] <ALR> Or does this conflict with your job on Monday?

    [11:07] <jose_freitas> Monday we can meet at the same time as today

    [11:07] <ALR> lfryc, kenfinnigan, bleathem ?


    [11:08] <kenfinnigan> monday should be fine, think I will have a 121 with theute at some point, but don't know when yet

    [11:08] <lfryc> +1

    [11:08] <bleathem> I have an 8am PST meeting Mondays

    [11:08] <kenfinnigan> though it would be morning as by then he'll be back in Switz

    [11:09] <bleathem> but it's cancelled for this monday

    [11:09] <bleathem> so I'm good

    [11:09] <bleathem> (sorry, had to check calendar)

    [11:10] == lkrejci [~lkrejci@redhat/jboss/lkrejci] has quit [Remote host closed the connection]

    [11:10] <ALR> kenfinnigan: Sorry, that's a +1 from you/

    [11:10] <ALR> ?

    [11:10] <kenfinnigan> ALR: yes

    [11:10] <ALR> Word up.

    [11:10] <ALR> Monday, same time it is.

    [11:10] <kenfinnigan> cool

    [11:11] <jose_freitas> perfect

    [11:11] <ALR> Thanks, all.  And a big hand to jose_freitas for picking up the JSFUnit Community Leadership

    [11:11] <kenfinnigan> jose_freitas: good work!

    [11:11] <jose_freitas> woot

    [11:11] <jose_freitas> thanks guys

    [11:11] <jose_freitas> I'll see you monday, then, with a roadmap proposal.