Developer Meeting Log - Oct 3 2007

18:05 [Users #foresight-devel]
18:05 [ doniphon        ] [ firstclown ] [ jtate     ] [ pcutler] [ specialK1vin ] 
18:05 [ DontFeedTheTroll] [ GnuKemist  ] [ kenvandine] [ pscott ] [ specialK3vin ] 
18:05 [ DreadPirateBob  ] [ groundstate] [ mr_ice    ] [ run_dfc] [ specialkevin_] 
18:05 [ dugan           ] [ int        ] [ OgMaciel  ] [ RyanK  ] [ stefw        ] 
18:05 [ elliot          ] [ jforbes    ] [ patoh     ] [ smithj ] [ tpfennig     ]

< pscott> ok, irc log for the meeting starts now
< pscott> ----
< pcutler> pscott: thanks
< pcutler> ken is going to be a little bit, and we will give doniphon a few minutes
< pscott> pcutler: your in charge of minutes right?
< pcutler> pscott: yes
< pscott> ok cool
-!- specialK3vin [n=Kevin@glug.id.iit.edu] has joined #foresight-devel
< pscott> item 0 covered then
< tpfennig> hi ho !!
-!- specialK2vin [n=Kevin@glug.id.iit.edu] has quit [Read error: 110 (Connection timed out)]
< tpfennig> gong
< tpfennig> If wiki is still not writeable I like to provide this provisional wiki page: http://gnome.wikiinfo.org/ForesightDeveloperMeeting
< tpfennig> for comments and minutes etc.
< specialkevin_> hey all I am here and ready to go
-!- specialkevin [n=kevin@foresight/developer/specialKevin] has quit [Read error: 110 (Connection timed out)]
< tpfennig> pcutler: Where we start?
< tpfennig> http://wiki.foresightlinux.com/confluence/display/DEV/Developer+Meeting+-+Oct+3+2007
< pcutler> i was hoping doniphon would be here, i was just giving him a minute, ken will be a few minutes late
-!- mr_ice [i=mr_icebl@213.233.70.4] has joined #foresight-devel
< pscott> ok
-!- doniphon [n=doniphon@foresight/developer/doniphon] has joined #foresight-devel
< tpfennig> should we use my wiki page to add notes or can we do in official wiki?
< pscott> sound off everyone thats here
< doniphon> (back)
< pcutler> here
< tpfennig> I can as Ive admin rights
< pscott> here (obviously)
< OgMaciel> ken is here now
< tpfennig> pcutler: you too?
< pcutler> tpfennig: i can't edit confluence until after it's upgraded
< kenvandine> woot!
< specialkevin_> here
< kenvandine> i made it
< pscott> tpfennig: might as well tomboy it and then post it afterwards
-!- stefw [n=stef@foresight/developer/stefw] has joined #foresight-devel
< pcutler> doniphon: can you take a minute and give an update on fl2 development
< doniphon> ok
< tpfennig> pscott: thats too complicated
< tpfennig> better have it online
< pcutler> what's working, major challenges, etc, then we can discuss testing timelines
< tpfennig> so anybody can look in notesd
< tpfennig> pcutler: We can also keep the logs of this chat if all are ok with that?
< pscott> tpfennig: already covered
< doniphon> tell me when to start plz.
< pcutler> tpfennig: yes, pscott has the log, and i'll recap it as meeting minutes
* jtate is lurking
< pcutler> doniphon: please go ahead
* doniphon looks for mkj and UP2 and don't see them...
< pscott> doniphon: your time on your choosen subject starts now
< doniphon> short story - things are moving at at speed again. atm 592 troves mook in a single rMake job.
< kenvandine> one sec... let me poke at people
< doniphon> against latest rpl:devel tolchain.
-!- SM2k [n=smg@rdu-nat.rpath.com] has joined #foresight-devel
-!- gxti [n=gxti@foresight/developer/gxti] has joined #foresight-devel
* doniphon waits.
-!- kelsin [n=kelsin@pool-96-233-22-88.bstnma.east.verizon.net] has joined #foresight-devel
< pscott> man, 10 minutes went fast
< specialkevin_> yea it did
< pscott> still only covered item 0
< kenvandine> mkj isn't coming
< doniphon> which was ?
< doniphon> so can i start ? or is more ppl coming ?/
< kenvandine> 0. Pick someone to create minutes for the meeting - easier to read than the IRC log
< specialkevin_> doniphon: how was taking minutes
< kenvandine> start
< doniphon> ok.
< tpfennig> I take some notes, too
-!- mkj_wk [n=johnsonm@rdu-nat.rpath.com] has joined #foresight-devel
* kenvandine is logging
-!- pscott changed the topic of #foresight-devel to: Developer meeting in progress - Agenda http://wiki.foresightlinux.com/confluence/display/DEV/Developer+Meeting+-+Oct+3+2007
< doniphon> basics of 2-devel (gnome and qt) along w/ some key apps are mooking ok against latest toolchain.
< doniphon> groups cook again.
< doniphon> gnome is already bump-ed to latest
< doniphon> will be commited tonight
< doniphon> xorg will be next
-!- Up2 [n=juphoff@66-207-95-55.cho.dmt.ntelos.net] has joined #foresight-devel
< doniphon> before starting to add new stuff at large some things need to be done
-!- joel [n=joel@ABordeaux-253-1-100-96.w86-196.abo.wanadoo.fr] has joined #foresight-devel
< doniphon> all pkgs need to be reviewd by hand to make sure that we don'r loose functionality from fl:1 or from rpl;devel  
-!- specialK1vin [n=Kevin@glug.id.iit.edu] has quit [Read error: 110 (Connection timed out)]
< doniphon> also buildReqs need to be checked. some pkgs mook ok but loose functionality when built inside rmake due to insuficient buildReqs. that is painfull thing that nees to be checked.  
< doniphon> also r.Configure args need to be looked. to make sure that we never miss functionality
< kenvandine> we should create a list of apps and expected behavior and give them a walk through
< doniphon> so. i expect to have an 'installable' 2-devel combo around mid/end of next week
-!- DNAfanboy [n=cesar@foresight/developer/DNAFanboy] has joined #foresight-devel
< doniphon> then, shortly after, is a matter of start extending it. meanwhile, i  did my best to comment both recipes and group-desktop-platform. lots of stuff there
< pcutler> so if we call that an alpha 1 by end of next week, each application needs to be manually reviewed by a testing team?
< mkj_wk> pcutler: in rPL land we create a wiki of things to test and testers mark things as in-progress or finished as they go
< kenvandine> doniphon: we really need to split out the gnome and kde stuff into new "dist" groups
< mkj_wk> pcutler: opening rITS tickets when things are found broken
* mkj_wk returns to lurking
< kenvandine> thx mkj_wk
< doniphon> kenvandine: 1 sec
< pscott> mkj_wk: is that public?
< mkj_wk> pscott: naw, we use that methodology for image testing
< mkj_wk> same idea, though
< doniphon> pcutler: yes and no.  i'm not worried atm about 'external' funcctionality from an user POV. moree about having complete (at recipe level of the options available (for those not covered by autoconf defaults) )
< kenvandine> doniphon: i feel the opposite
< kenvandine> we should look for "expected" behavior
< kenvandine> for example, does my ipod mount and work with banshee
< doniphon> kenvandine:  matter of wording. some ppl don 't expect some stuff because it twas never there.
< kenvandine> use case style
< kenvandine> mostly looking for regressions
< pscott> its use cases we care about
* kenvandine can lead that
< doniphon> i'm worried atm more with low level. i.e. what the apps allow, and expose that. (and look at what competition  (RH/suse/ubuntu) do. after that we carre about high level pkging.  
< kenvandine> i will start a wiki page collecting the use cases and how they apply in fl:1
< doniphon> kenvandine: regarding spliting groups IMHO it is too soon.  having it all together is proving valuable.
< pscott> if there is going to be a wiki page with tests then it needs to be 'my usb storage appeared on my desktop' not 'hal recognised the device and gnome-volume-manager mounted it'
< kenvandine> i don't want to start creating isos until we do that
< kenvandine> it isn't a good test if it installs both kde and gnome
< kenvandine> we just can't see the outcome
< kenvandine> and it should be trivial
< doniphon> kenvandine: will default to gnome atm.
< kenvandine> so not install any of the kde stuff?
< kenvandine> and why not split it?
< tpfennig> I can offer to create a wiki for just that. If you want to have.
< tpfennig> (testing)
< kenvandine> i see no point in not doing it
< kenvandine> tpfennig: we can use confluence
< doniphon> kenvandine: one group  = one single point of failure.
< pscott> kenvandine: err, no we can't
< doniphon> scalability issue, atm easier IMHO to track
< kenvandine> pscott: we will be able to
< tpfennig> I would also suggest that the use cases/goals have a hierarchy
< elliot> i'm with kenvandine on this...we should go ahead and split the group
< tpfennig> so that we can live on alpha when minor thing go wrong
< kenvandine> doniphon: it is a poor test if we don't split it
< tpfennig> so we only need core functionality
< tpfennig> on fl:1 many core things didnt work either
< tpfennig> so minor things are ok
< tpfennig> this would make testing shorter and easier
< kenvandine> the important thing is if something worked in fl:1, it should in 2
< doniphon> tpfennig: i don't like to repeat some fl:1 oddities.  
< tpfennig> kenvandine: why?
< kenvandine> doniphon: tracking shouldn't be any harder if we split it
< doniphon> kenvandine:  there is one extra point.
< kenvandine> tpfennig: we don't want regression in functionality
< doniphon> one goal i 'd like to achieve is to have perfect co-habitation of xfce/kde/kde
< kenvandine> if someone is used to being able to sync their ipod, they should still be able to do so
< tpfennig> kenvandine: well my gues is that alpha means you have to step backa  abit from eprfection
< tpfennig> perfection
< doniphon> so, for a start having them all not messing  with each other is a good test (IMHO)
< kenvandine> tpfennig: agreed
< kenvandine> that is the point of alpha
< kenvandine> but during alpha, we test that stuff
< tpfennig> kenvandine: so which things may not work?
< kenvandine> not blocker for release alpha
< tpfennig> ok
< tpfennig> understood
< kenvandine> doniphon: spliting the recipe doesn't affect that
< tpfennig> we need to make a list of what MUST work and what SHOULD maybe?
< kenvandine> group-gnome-dist would add group-desktop-platform and the other gnome bits
< doniphon> kenvandine: i need to rework all the kernel /kernel modules scheme, to make them uptodate/scalable
< kenvandine> tpfennig: right
< kenvandine> we need derived kernels
< kenvandine> and syslinux
< tpfennig> where as MUST we need to know before testing and needs to be a fixed list
< kenvandine> yes
< tpfennig> whereas SHOULD can be and endless list
< doniphon> kenvandine: afaik derived kernels need to be scrapped for now. unupdatable
< kenvandine> tpfennig: that is what i said
< tpfennig> fine
< kenvandine> doniphon: not with syslinux
< pcutler> let's hold off on discussing testing for a minute - back to FL:2 development:  let's resolve this group discussion and talk about blockers for getting to alpha.  Testing can come up in the second agenda item as a sub-team
< doniphon> kenvandine: need to look
< kenvandine> doniphon: but why not create the new top level groups?
< kenvandine> pcutler: thx
< pscott> pcutler: well moderated
< kenvandine> although... the group issue is the next thing we need to get done for fl:2
-!- mr_ice [i=mr_icebl@213.233.70.4] has quit []
< doniphon> kenvandine: frankly because i don't think is is clear (at least to  me) what the final structure of them  should be.
< doniphon> if you look at evolution of 2-devel master group things are evolving.
< kenvandine> we know there has to be other groups to define it... group-gnome can not have a parent of group-desktop-platform
< kenvandine> any way we look at it, there has to be groups to define what goes on the CD
< kenvandine> and we can't start testing until we do that
< doniphon> yes. ok.  
< doniphon> 1 sec
< kenvandine> perhaps they are in the same source trove
< doniphon> ah!!!!!!!!!!
< kenvandine> we have discussed that... but i would rather not
< kenvandine> my point is we can't do isos until top level groups are defined
< kenvandine> next topic
< doniphon> i'd prefer either that or a compromise - actual groups are in master group (group-desktop-platform) but are wrapped for world use in a separate source tyrove
< kenvandine> discuss that later
< pscott> i have one question about fl:2
< doniphon> ok.
< kenvandine> we agree we need to do this
< kenvandine> pscott: yes?
< pscott> did the qa process ever get fully agreed on?
< kenvandine> i think so...
< tpfennig> pscott: what is "fully"?
< doniphon> pscott: the only point is a technical one - how stuff flyes beteween repos.
< doniphon> workflow wise there is a consensus, i think.
< kenvandine> branches
< doniphon> devel -> QA -> public
< kenvandine> shadows vs. clones might be undecided
< kenvandine> but we should work that out quickly
< kenvandine> the QA sub-team can do that
< pscott> thats what i was thinking of i think
< tpfennig> I like to know if we want to define some time for a process? as an optimum
< tpfennig> things can take longer thats ok.
< pscott> i seem to remember doniphon posting his idea and mkj_wk commenting on it
< kenvandine> yeah
< kenvandine> a long thread
< kenvandine> that should be left to another time
< doniphon> pscott: yes. but i'll counter arg. but i needed rebuild done first to have a rea; test case
< pscott> it isn't clear to me how contrib packages are to be handled in fl:2
< kenvandine> let the QA subteam hash it out
< kenvandine> flattened into 2
< tpfennig> kenvandine: is there a list in wiki for qa team?
< kenvandine> i think is what was decided
< kenvandine> tpfennig: it doesn't exist yet
< kenvandine> we are creating it in this meeting
* tpfennig notes
< pscott> ok then, moving on
< pcutler> ok, back to fl:2.  Are there any other blockers for FL2?  I assume PackageKit is needed for beta, not alpha
< kenvandine> doniphon: any road blocks ?
< kenvandine> pcutler: it is ready for the alpha
< kenvandine> i will build it into 2-devel when doniphon gives me the thumbs up
< doniphon> a) time. tweaking groups. (add missing info-troves), etc
< doniphon> kenvandine: but should work. all buildReqs are there
< kenvandine> right
< kenvandine> just tell me when
< elliot> pcutler: there are some issues building anaconda-templates right now
< pcutler> no anaconda issues?
< pcutler> heh, good timing
< elliot> pcutler: think it may be due to bugs in conary
< kenvandine> that is a blocker
< doniphon> surprise.
< elliot> pcutler: don't know about anaconda issues yet....need to have templates to test anaconda
< pcutler> ok, anything else?  ready to move on to forming some subteams to tackle all this fun stuff?
< doniphon> pcutler: one more thing.
< pscott> doniphon: omg new ipod!
< doniphon> some stuff that diverges in x86/x86_64 (web browsers/multimedia) needs to be tweaked atm
< doniphon> also i need to get new oss java (iced tea) in
< pcutler> noted
< doniphon> and i need someone to do OpenOfice
< jforbes> doniphon: are you still planning 64bit firefox for fl:2?
< kenvandine> int: ?
< tpfennig> doniphon: why dont you use old one for now?
< doniphon> jforbes: off couse
< jforbes>
< jforbes> okay
< kenvandine> doniphon: why?
< kenvandine> why not having working flash?
< doniphon> kenvandine: why what ?
< doniphon> ndiswrapper works well.
< kenvandine> 64bit firefox...
< doniphon> blocker would be java
< doniphon> icvedtead has a 64b java plugin
< kenvandine> how about flash?
< doniphon> mplayer plugins work in ffox
< tpfennig> ok
< kenvandine> flash...
< pcutler> i would like to thank doniphon for all his hard work on fl:2 these last few weeks / months as well, we wouldn't be here without him
< doniphon> flash will work with with either mspluginwrapper
< jforbes> doniphon: mplayer works either way... I run the 32bit mplayer plugin, but it uses the 64bit mplayer binary
< doniphon> or with gnash )(next Ubuntu is using it)
< doniphon> so all covered
* tpfennig applauds doniphon
< jforbes> doniphon: since the plugin actually runs mplayer instead of linking against it...
< doniphon> lol
< pcutler> we might want to create a wiki page for x86 and x86_64 differences
< kenvandine> yeah... thanks for all the hard work doniphon!
* doniphon hides
* kenvandine would still vote for 32bit firefox
* jforbes too
< kenvandine> helps show off how multilib friendly we are
* doniphon also notes that this is about effort and fun. not rocket science. otherwise NASA would had been hired me already.
< kenvandine> hehe
< specialkevin_> haha
< jforbes> doniphon: the Russian program is better funded
< pcutler> ok, we'll table firefox for now, i've got it in the minutes
< doniphon> lol
< kenvandine> 2. Team Structure
< doniphon> kenvandine: jforbes: if we  can we should ship all x86_64 in x86_64. that 's the way others are going.
< jforbes> doniphon: they are doing it because their tools *suck* for multilib
< kenvandine> 2. Team Structure
< pcutler> here's my question for the second agenda item on forming some sub-teams:  we have two huge workstreams in front of us right now, building fl2 and then maintaining
< doniphon> jforbes: there are multilib issues on our side too. i need to fill some more bugs in rPath JIRA...
< kenvandine> pcutler: i would say create the teams as if it is just for building fl2
< jforbes> doniphon: there are, but we do fine shipping/running 32bit firefox and such.  I have never installed 64bit firefox as more than a test
< kenvandine> then re-evaluate after release to decide how we want to maintain
< kenvandine> we will learn allot in the process
< doniphon> ok. jforbes - we 'll talkk about it later.
< pcutler> all right, let's brainstorm for a couple minutes on what teams we need, i've already heard a QA team thrown out
< specialkevin_> QA, Marketing, Infrastructure
< specialkevin_> Bugs
< kenvandine> bugs and qa are one
< kenvandine> imho
< doniphon> i'd split QA in two levels - repo sanity/quality (build issues/API/ABI), and end user issues/tweaking
< pcutler> docs
< tpfennig> specialkevin_: I think we need to rethink
< pscott> kenvandine: in the sense its the same person with two hats?
< kenvandine> doniphon: i think the same team
< pscott> security team too
< tpfennig> I think maybe user/marketing is one team
< tpfennig> PR
< kenvandine> just covering both
< doniphon> kenvandine:  yes. but are != layers.
< elliot> release team
< tpfennig> user interaction, public appearance
< elliot> maybe
< doniphon> that must not be confused.
< tpfennig> we should define one team after the other
< kenvandine> qa, doc, marketing, infrastructure, release, development
< tpfennig> not just list all
< doniphon> kenvandine: +1
< kenvandine> we need an idea what the list looks like first
< tpfennig> and what they should do
< specialkevin_> kenvandine: +!
< kenvandine> to see where the overlap is
< tpfennig> I think the problem is with the flow of information
< pcutler> agree with ken's list, but would add security
< kenvandine> +security
< tpfennig> I dont see difference of qa and development
* doniphon volunteeres to qa, infrastructure, release (part-time), development
< kenvandine> very different
< kenvandine> one sec
< doniphon> QA picks stuff from devel and test it. then release do well (public) releases, to public repo.
< kenvandine> development - build, core technologies, platform, new technologies
< kenvandine> qa - test output of development, and push to release team
< tpfennig> kenvandine: I dont think you can dio that with a small team
< tpfennig> how many we got?
< kenvandine> release - promote to release team and manage groups
< tpfennig> thats for large companies?
< specialkevin_> tpfennig: we don't want to have to redo the team in a couple months after we get larger
< kenvandine> the rest are pretty  self explanitory
< doniphon> tpfennig: we 'be large. we just want to be prepared. for now it means some of us multitasking as usual.
< kenvandine> most of us will be on more than one team
< kenvandine> teams are more for splitting out focus
< kenvandine> to get things done
< tpfennig> specialkevin_:  I just think that many problems we got now come from development and qa being not really in one hand.
< kenvandine> less chaos
< tpfennig> and I fear this will get worse wit thios infrastruct
< doniphon> tpfennig: how so ?
< kenvandine> tpfennig: dev can't qa their own work
< kenvandine> that is never a good idea
< doniphon> kenvandine: QA is a whole label.
< tpfennig> because then things are done in one way by devel and the qa team does not know why
< doniphon> i mean tpfennig
< tpfennig> or devel dotn get feedback
< kenvandine> and really in our case, qa is more managing the flow
< kenvandine> making sure others test
< kenvandine> and validate before pushing to the release team
< tpfennig> ok anyway - go on I dotn want to block discussion.
< kenvandine> tpfennig: we still need to work together
< doniphon> tpfennig: QA ppl will look at what is in the 2-devel label, push it or not to 2-qa and them talk with release ppl to get it public
< pcutler> tpfennig: that was a good discussion, helped clarify for others listening
< elliot> tpfennig: just because we have different teams doesn't mean that can't talk to each other
< doniphon> it's an interactive process
< stefw> did someone already mention that changes are always made to devel and promoted to QA for testing, and that changes should be made on QA?
< tpfennig> elliot: yeah
< kenvandine> process is for later... decide on the teams
< pcutler> ok, so are we agreed on the following teams:
< stefw> sorry...should NOT be made on QA
< pcutler> QA / Bugs, Marketing / PR, Infrastructure, Docs, Security, Release and Development
< kenvandine> yes
< specialkevin_> yes
< tpfennig> one question
< pcutler> and volunteers?  i'd like to see a lead of some sorts, and then team members. we'll set up wiki pages and such later this weekend
* kenvandine volunteers for release and development
< pcutler> tpfennig: go ahead
< tpfennig> infrastructure and pr? infra=web site and?
< kenvandine> i would like to be lead on development and release
< tpfennig> also jira,...
* doniphon volunteeres to qa, infrastructure, release (part-time), development
< tpfennig> is web site more marketing or more infra?
< doniphon> both
* kenvandine nominates doniphon to lead qa
* elliot volunteers to work on release and devel
< specialkevin_> infrastructure is website maintaince, forum, jira, wiki
< doniphon> content is marketing.  sw is infra
< tpfennig> doniphon: so lets say ine needs to decide on software
< doniphon> kenvandine: no, plz no.
< kenvandine> doniphon: ok...
* specialkevin_ volunteers QA, Marketing, Devel
< doniphon> 1 sec
* pcutler nominates smithj for security
* elliot volunteers smithj to be the security team
< stefw> heh
< tpfennig> members of security team= gnome officers ?
< pcutler> i'll head up docs and marketing, and help with qa
< tpfennig> localization!
< tpfennig> we need that, too
* stefw can help pcutler with docs and marketing
< tpfennig> ot is that in a team already?
< tpfennig> part of docs
< tpfennig> bit not only
< tpfennig> also UI localization
< pcutler> i'm no expert on localisation, what do we recommend?
< tpfennig> maybe its part of qa also
< doniphon> pcutler: regardinf i18n. we 're making progresses
< tpfennig> like checking how localization works
< specialkevin_> where is OgMaciel when we need him
* kenvandine also volunteers for qa
< OgMaciel> right here
< doniphon> ffox/thunderbird/xulrunner are already fully localized in 2-devel
< doniphon> OO will be too.
< specialkevin_> OgMaciel: how should be handle localization
< kenvandine> do we need a team for it?
< specialkevin_> s/be/we
< tpfennig> Maybe we should have localization team.
-!- run_dfc [n=davec@voltron.atg.columbia.edu] has quit [Read error: 104 (Connection reset by peer)]
< doniphon> kenvandine: IMHO for docs
< doniphon> for apps we pick upstream
< OgMaciel> specialkevin_: as far as packages or docs?
-!- run_dfc [n=davec@voltron.atg.columbia.edu] has joined #foresight-devel
< kenvandine> doniphon: no... we need to help upstream too
< doniphon> no reasin to do custom localization except for pkgKit
< kenvandine> if an app we want isn't localized
< kenvandine> we should try to help them do that
* OgMaciel nods
< kenvandine> OgMaciel is good at that
< OgMaciel>
< doniphon> kenvandine: tricky. we help doing that upstream *not* ubuntu way that diverges from it.
< tpfennig> kenvandine: also our own apps
< kenvandine> right
< pcutler> doniphon: agreed
< kenvandine> it is more than docs though
< OgMaciel> doniphon: we would work directly with upstream
< doniphon> pcutler: kenvandine ^^ see OgMaciel
< OgMaciel> kenvandine: what ever happened to our localized site?
< kenvandine> so the i18n team would look at packages we include to make sure they meet our i18n requirements
< tpfennig> I would volunteer to be member of i18n team
* elliot thinks localization is a subset of qa
< doniphon> kenvandine: yep
< kenvandine> OgMaciel: the scripts are there
< elliot> well...qa/devel
< kenvandine> elliot: could be
< tpfennig> elliot: somehow yes but also dev and docs
< kenvandine> that could be a qa step
< tpfennig> i18n needs to write docs
< kenvandine> does it meet our requirements
< pcutler> let's make it a team for now, and use it as a check and a balance against qa, docs and devel
< tpfennig> and fix apps
< kenvandine> pcutler: +1
< stefw> i18n needs to be considered at devel, but a step in qa... but docs should be written at devel time, imo
< doniphon> pcutler: + 1
< tpfennig> also often only people of one language can check localization
< specialkevin_> pcutler: +1
< kenvandine> tpfennig: not true
< tpfennig> stefw: than dev team needs to write in all languages
< kenvandine> OgMaciel could look at the source and see if all the strings are empty
< OgMaciel> sure
* stefw nods
< tpfennig> hehe
< kenvandine> not if it is accurate
< kenvandine>
< tpfennig> kenvandine: thats what a script could do also
< kenvandine> just that it is there
< pcutler> doniphon: what help do you need with infrastructure?  you're the only team member atm, and i consider the appliances we need different than managing the content
< doniphon> pcutler: manpower mostly.
* kenvandine nominates either smithj or elliot
< doniphon> i'm not an expert in everything
< kenvandine> jack of all trades
* stefw sticks to the "I can just write about it and tell you what's wrong with it" status
< doniphon> i need help with jira/confluence. also i need a woorking website. django sometimes goes nus and wasts to much (dsk) i/O
* OgMaciel hates confluence
< doniphon> other than that, i have space now and provide xen VMs on  demand
< pcutler> we really need help with the web design for the new site in mercurial, no offense ken
< doniphon> OgMaciel: since we use JIRA confluence is for free. *exact* same eset of issues.
< OgMaciel> hehe
< pcutler> i'm happy to port the content over and manage that, but i suck at web design and coding it
* OgMaciel rather see a static web site managed in mercurial
* stefw pretends to know nothing of web design
< tpfennig> doniphon: I can support you with confluence/jira, still
< OgMaciel> different branches to be localized
< elliot> doniphon, kenvandine: I can help out some with infrastructure
< kenvandine> specialkevin_: you had volunteered to help with the site... right?
< doniphon> good. great. also we really need to get  rsayre to tweak our jira per rPath settings.
< OgMaciel> XFCE web site is maintained as static site
< OgMaciel> in SVN
< kenvandine> my scripts are in hg
< kenvandine>
< pcutler> doniphon: i emailed rsayre a week or two ago, he's ready to go, just needs to know what our requirments are
* tpfennig sighs
< kenvandine> and template i created
< specialkevin_> kenvandine: I don't think I will have the time but if nobody else does I might be able to bust something out
* OgMaciel will have to revisit kenvandine's scripts
< doniphon> pcutler: ok. i'll check with him, *after* i upgrade jira/confluence this weekend for new versions (due to new  licenses)
* kenvandine thinks OgMaciel should volunteer for the site...
< pcutler> ok, we're already segueing to the third topic
< kenvandine> someone that can prove the translations work
* OgMaciel takes on kenvandine's challenge
< pcutler> doniphon: agreed, i'll follow up with you and rsayre next week
< kenvandine> or localizing features...
< pcutler> let's talk about the web
< pcutler> we've previously discussed the following:
< pcutler> move to a static site
< OgMaciel> +1
< pcutler> implement forums
< doniphon> +1 less I/O issues w/ django
* tpfennig sighs
< pcutler> implement a venus appliances for planet functionality
< elliot> +1
< OgMaciel> +1
< tpfennig> pcutler: what is venus?
< elliot> pcutler: we already have a venus appliance
< tpfennig> love planet?
< OgMaciel> pretty cool planet engine
< elliot> pcutler: tpfennig a planet replacement
< pcutler> elliot: but not running of fl.org yet
< tpfennig> lol
< kenvandine> it is nice
< elliot> pcutler: no
< elliot> not yet
< tpfennig> with or wihout porn? lol
< pcutler> and i'd like request a wordpress or wordpress-mu appliance for the newsletter and other marketing
< kenvandine> static site is the only way to go... all the other cms systems are crap!
< kenvandine> </rant>
< elliot> pcutler: I would go for mu so that we can host developer blogs and such as well
< kenvandine> +1
* tpfennig shakes head
< kenvandine> has anyone volunteered for the docs team?
< elliot> +1 to kenvandine's rant
< pcutler> kenvandine: me and stefw
< kenvandine>
< specialkevin_> pcutler: I agree and then we could offer developer blogs
< elliot> kenvandine: I think stefw did
< kenvandine> ok
< kenvandine> any teams left unrepresented?
< pcutler> security is just smithj
< pcutler> and il8n is just tpfennig
< stefw> Agent Smith
-!- sharpshooter [n=sharp@adsl-69-221-98-212.dsl.akrnoh.ameritech.net] has joined #foresight-devel
< doniphon> just one note - *if* we offer dev blogs i need a commitement for ppl to have care with what they write there plz boxes are not mine...
* elliot thinks OgMaciel should probably be on i18n
< kenvandine> i suspect OgMaciel on i18n too
< OgMaciel> sure thing
< OgMaciel>
* stefw would like to see a forum rather than blogs
< tpfennig> pcutler: I think OgMaciel will be too
< pcutler> og's army of volunteers will be right behind him too i think
< tpfennig> yep
< OgMaciel> hehe
< specialkevin_> stefw: we would like to do both
< elliot> stefw: what about a forum in addition to blogs?
< kenvandine> yay vladimir
< OgMaciel> hahaha
< kenvandine> we need a forum
< pcutler> elliot: pscott and i have talked about SMF
< pcutler> thoughts?
< stefw> forum + blogs would be good
< kenvandine> but they can't replace blogs
< tpfennig> Orcs Army?
* OgMaciel would create a pt/pt_BR section
< pcutler> i'd recommend we start with wp-mu, but not implement dev blogs until we have all this other stuff up and running
* stefw would be willing to help implement the SMF, being that she maintains the rBO project for it
< pcutler> we have a lot to do
< elliot> pcutler: agreed
< specialkevin_> pcutler: +1
< doniphon> +1
< kenvandine> +1
< kenvandine> ok
< pcutler> just use it for the newsletter and other misc news
< kenvandine> so all teams represented?
< tpfennig> OgMaciel: I like to chat with you soon to discuss some thing about i18n
< pcutler> kenvandine: yes, all teams are represented
< specialkevin_> who are the leads for the teams
< OgMaciel> tpfennig: sure thing
< pcutler> biggest web needs right now are theming and design for wp-mu, the website, and smf
< stefw> yo yo represent yo
< pcutler> i have confidence we can get the appliances up and running fairly quickly
< doniphon> pcutler: +1. branding is critical.
* stefw is having to do some SMF theming for another project and can work on that
< kenvandine> ok... moving on...  we can leave this for the infra sub-team
< kenvandine> sprints...
< kenvandine> everyone familiar with the idea of sprints?
< elliot> pcutler: also a similar look and feel across web apps is going to be somewhat important
* stefw is... loosely
< doniphon> kenvandine: yep.
< tpfennig> pcutler: where is design  located? marketing?
< kenvandine> teams pick a group of tasks they can work on for a 1 week period of time
< pcutler> tpfennig: i hadn't thought about, not sure
< tpfennig> I guess marketing should be open to all kinds of graphic designer
< stefw> tpfennig: I would think marketing would be the placement for that
* OgMaciel has been doing sprints for the last 2 months
< tpfennig> pcutler: so they look for new logos and theming
< kenvandine> so we will pick issues in jira and mark them with a fixed version of the weeks sprint
< kenvandine> providing plenty of visibility into what is going on
< kenvandine> we just have to make sure there are issues for everything
< kenvandine> small chunks
< kenvandine> and work on it for the week
< kenvandine> i vote for wed-tues
< OgMaciel> +1
< kenvandine> we should also have a table or something in the wiki which lists all the stuff in each teams sprints
< specialkevin_> so sprints start on wed and end on tues
< pcutler> i like the idea of starting mid-week, if we can have them picked out by sat/ sund, we can do a mini-newsletter or blog letting ppl know to come help
< kenvandine> so we can quickly see what is going on
< tpfennig> what is a sprint?
< kenvandine> read back
< kenvandine>
< tpfennig> o k
< stefw> sprint = short-term goal with a short-term deadline
< tpfennig> I just new that from sports
< kenvandine> if tasks are to big... we split it up
< tpfennig> I thought I would need to do exercises now...
< kenvandine> hehe
* OgMaciel imagines tpfennig doing laps as we chat
< kenvandine> everyone understand?
< specialkevin_> kenvandine: what would be some issues for marketing team to work at sprints
< kenvandine> same thing
< stefw> and one and two and... lift those knees... and one more...
< kenvandine> we should use jira to manage marketing projects too
< doniphon> specialkevin_:  issues = tasks
< kenvandine> hehe
< tpfennig> specialkevin_: release stuff maybe
< kenvandine> remember... jira isn't just for tracking bugs
< elliot> kenvandine: we should use jira to manage everything
< kenvandine> it is a great project management tool
< tpfennig> liek next ween fl:2 release
< pcutler> elliot: +1
< tpfennig> week
< kenvandine> elliot: damn right!
< stefw> elliot: +1
< specialkevin_> elliot: +1
* tpfennig feels a bnit worried to depend on proprietary software
< kenvandine> why?
< tpfennig> like confluence today
< kenvandine> you think they are gonna mess with our data?
< tpfennig> that things dont work and we cant patch
< elliot> tpfennig: it is supported proprietary software
< kenvandine> they support us
< tpfennig> Hm, well lets go for it now
< tpfennig> but I wish in future rather use floss
< pscott> sorry - i had to step away and been reading up
< tpfennig> pscott: too late
< kenvandine> well
< tpfennig> you dwarf you...
< pscott> the only problem i have with jira is whats happened recently with users
< kenvandine> tpfennig: there is not any good alternatives to jira
< tpfennig> kenvandine: ok
< kenvandine> we have really looked
< kenvandine> jira is such a powerful tool
< kenvandine> and confluence has some nice hooks into jira
< pscott> which if i understand correctly is now sorted - we just need to update
< tpfennig> kenvandine: does rpath use it also for customer relations?
< kenvandine> like having a teams page
< kenvandine> tpfennig: yes
< kenvandine> and all project management
< kenvandine> and issues
< kenvandine> etc
< kenvandine> it support
< kenvandine> everything
< tpfennig> like we could have inbox for issues
< kenvandine> ir rocks
< kenvandine> yes
< kenvandine> i hate that feature though
< pscott> pcutler: i need to talk teams with you after
< pcutler> pscott: roger
< tpfennig> I use request tracker
< kenvandine> i used rt many years ago
< kenvandine> jira is much richer
< kenvandine> no pm integration
< kenvandine> anyway
< tpfennig> anyway. its good for now. nut i always prefer floss
< kenvandine> i do to
< kenvandine> :/
< gxti> jira beats the pants off anything i've ever used
< kenvandine> but no choice here
< gxti> i think we can forgive it
< tpfennig> go on
< kenvandine> yeah
< kenvandine> and they do support us for free
< gxti> they're FOSS-friendly, at least
< pscott> pcutler: whats the current topic?
< kenvandine> next agenda item pcutler
< kenvandine> PM?
* kenvandine introduces pcutler, our project manager
< pcutler> if we use jira, project management won't be a problem  
< kenvandine>
< pcutler> that's a wrap
* tpfennig does a little support in #foresight now read back later
< kenvandine> i think we covered #3 inliine
< pcutler> there's a couple action items we still need to re-visit, but i'll get the minutes and notes out later this afternoon
< gxti> what about LDAP?
* OgMaciel goes back to working on his presentation
< pcutler> thank you all for making the time

Generated by irclog2html.py 2.5 by Marius Gedminas - find it at mg.pov.lt!}}