vuntz

as far as I see, we have only one (big) agenda item: Strategic roadmap for GNOME: long term goals

stormy

Anybody heard from gpoo? Since the earthquake?

diegoe

I haven't heard from people in Concepción yet

vuntz

I guess we'll have to wait for some online news :/

diegoe

but people in other regions is starting to msg in twitter

stormy

ok, sorry, vuntz, I didn't mean to derail.

vuntz

it's okay
so, does anybody want to start on the roadmap topic?

stormy

I think it would really help get individuals and companies involved if we had a longer vision roadmap.
We don't have to commit to specific dates but having a vision would greatly benefit us.

vuntz

okay, I wanted to wait a bit before giving my opinion, but let me go :-)
I didn't read the whole thread on foundation-list, so maybe I'll write things that were already mentioned there
the long-term roadmap topic is something that, in my opinion, the release team has been working on
and the first visible result of this was the GNOME 3 planning document: http://live.gnome.org/ThreePointZero/Plan
if you look at this, you'll see it mentions area that are worth exploring
the way I see it, most of what is mentioned there are items that are for GNOME 3
not GNOME 3.0, but GNOME 3 -- so for the next 2.5 years, let's say
and GNOME 3.0 is just the beginning of all this

diegoe

(actually that's a point to think: how long do we want our theorical major stable cycles to be?)

vuntz

yep, that's something the release team had initially discussed when doing the planning for 3.0, and one of our orinigal ideas was to have a cycle of around 3 years, iirc

diegoe

vuntz, kde 4.0 is not kde 4? :)
sounds good, LTS distros have similar ranges afaik

vuntz

diegoe: heh. To clarify: 3.0 is a specific version, 3 is a whole cycle

diegoe

yeah it makes sense ;)
a question might be how does recent UX hackfest results fit there

vuntz

I don't want to start chatting about GNOME 4 right now; I'd just like to hear from people here if all this was unclear, or if you feel it wasn't communicated at all
diegoe: from what I saw of the hackfest in the last two days, most things should be targetted for GNOME 3 (cycle, not 3.0)

diegoe

jjardon, any opinion from gtk land? :)

vuntz

(we can move to the GNOME 4 topic afterwards, of course; it's just that I'm interested in seeing how people about what were the release team plans)
how people feel

afranke

is the roadmap for GTK+ 3.0 available somewhere?

vuntz

afranke: there's been a few mails on gtk-devel-list this week or last week with the plans

diegoe

afranke, http://live.gnome.org/GTK+ there are two links there, but not sure how up to date they are

vuntz

http://mail.gnome.org/archives/gtk-devel-list/2010-February/msg00040.html

diegoe

jjardon should now

vuntz

and the few mails after that

jjardon

all the info is here: http://live.gnome.org/GTK+/3.0/
the work is coordinated here: http://live.gnome.org/GTK%2B/3.0/Tasks
Also, there already is a gtk-2-90 branch in GTK+ git

jjmarin

I agree this is a good strategical document. However, I miss a easy way to follow where we are

vuntz

jjmarin: I guess that people following the development closely know the current status; but they lack time to update the status on such a wiki page. So help would be welcome here

jjardon

jjmarin, GTK +3 document or Gnome 3 document?

jjmarin

vuntz: maybe to set some rules about this can help

vuntz

jjmarin: what kind of rules?

jjmarin

vunt: For example a page for every task to follow the status

vuntz

jjmarin: sure, but the issue is that somebody needs to update the status
jjmarin: and this is where we need help from people, I think
asking developers to update the wiki page just doesn't work, unfortunately

bkuhn

vuntz: (re devs updating wiki) Yeah, that makes sense. Perhaps a tutorial on how it should be updated. What would someone do? Watch merges, and update based on commit messages?

jjardon

As far as I know http://live.gnome.org/GTK%2B/3.0/Tasks is up-to-date

vuntz

jjmarin: I guess you're maintaining this page? Is this something where you need help?

jjmarin

vuntz: I guess mean jjardon :)

vuntz

fwiw, one issue with the page is that it doesn't list the features planned for integration -- but that's probably because the gtk+ team itself is not completely clear on this
jjmarin: yeah diegoe is right, it's confusing to have both of you ;-)

jjmarin

Another thing that we can discuss is: how to set strategical goals. If we need a LTS cycle.

vuntz

jjardon: do you plan to add the features that mclasen listed in http://mail.gnome.org/archives/gtk-devel-list/2010-February/msg00040.html on this tracking page?

jjardon

well, as far as I know, GTK+ 3.0 will not be a feature release, only a "cleaning" release (we are removing all the deprecated code, so we'll break the API and the ABI). Also, all the public members will be moved to private structures, so you should use accessor functions instead direct access
Some of these features can be implemented in 2.22 or 3.0
(If we want a 2.22 release)

jjmarin

A strategical roadmap is good for manage our planning, but as well is good fo giving a public message of the direction.

diegoe

agree

jjardon

vuntz, these features are already in the GTK+ roadmap: http://live.gnome.org/GTK%2B/Roadmap

stormy

It is really hard for someone not emeshed in GNOME to follow along.

jjmarin

A strategical roadmap can helps as well for planing a API and ABI breakage.

stormy

(And I agree, a strategic roadmap would very much help that.)

jjardon

Also, for GTK+ 3 work, all kind of hands are needed, even a newbie can help

diegoe

jjardon, could we perhaps promote that? with easy clear steps maybe?

vuntz

may I ask what you think should be in a strategical roadmap?

iven

This page has more information: http://live.gnome.org/GTK%2B/3.0/Roadmap

diegoe

maybe reviewing patches, like kalikiana and I used to do
that was easy, just a bit time consuming

stormy

vuntz: high level goals?

vuntz

stormy: okay, let me rephrase. Is http://live.gnome.org/ThreePointZero/Plan something that looks like what you'd expect?
or would you like to see something else?

stormy

Yes.
But I think it could be presented in a way that would make it easier for people to see highlights.

diegoe

vuntz, perhaps with an optional "bullet point summarised" summary

vuntz

fair enough

diegoe

yeah, highlights

vuntz

we all know release team people can't communicate well, though ;-)

stormy

And maybe some priority/timeline to it.
Are all these things for 3.0?

rubenv

vuntz: that's because of the amount of french speakers in there ;-)

vuntz

no, they are GNOME 3

stormy

In the GNOME 3 life span, when will they come out?

diegoe

one thing I might wonder if I read casually some blogs is "oh the task pooper, when will I see that? 3.0?"
so priority/rough "releaseline" would be good

vuntz

I agree with the priority/timeline thing, and that's actually an action item I took during the usability hackfest

stormy

cool

vuntz

it'd be nice to have other release team members here to have their opinion; but in my case, I don't think we could have put real timeline for this document until recently

rubenv

question: (I joined late so I don't know if this already came up) would it be desirable for gnome to adopt something like a major-version release cycle?

stormy

It's ok if they change ...

rubenv

something like the long term releases for ubuntu / rhel / ...

vuntz

stormy: yeah, could be
17:13 <@vuntz> yep, that's something the release team had initially discussed when doing the planning for 3.0, and one of our orinigal ideas was to have a cycle of around 3 years, iirc
rubenv: ^

jjardon

diegoe, sure

rubenv

vuntz: great, so will that idea be put into place or has it been dismissed?

vuntz

there's also the discussion on whether the release team is the most appropriate group to write those kind of documents. Some people feel it is, some people disagree
I'd love to hear what people think

lixem

i'm hungry

vuntz

rubenv: it's not dismissed, but not really adopted either
rubenv: lack of time, etc.

jjmarin

vuntz: Maybe it is necesary to open a period of petition of request of changes or something like that

vuntz

rubenv: my feeling is that we'll do something like that

rubenv

vuntz: might be good to keep it on the table though, I think both distro's and ISVs will like it when they know what to expect + developers need to spend less time on API/ABI conservancy and more time on innovative stuff
vuntz: great I'll stop hijacking the meeting!

vuntz

jjmarin: actually, we tried this

jjmarin

vuntz: OK :)

vuntz

http://live.gnome.org/RoadMap/Process

diegoe

jjmarin, debian has something like that, they propose goals with developers "backing up" it (just as proposing a module)

vuntz

this was a way to get feedback from maintainers for the next 6 months, but also for the long term
it worked to some extent, but required too much work
what happened during the GNOME 3 planning phase is that a few people stepped back and looked at all the interesting areas that were being explored
jjmarin: would you be interested in trying to make the roadmap process I linked to lighter, so we can restart it again?

jjmarin

Some nice suggestions for easy improving like http://live.gnome.org/GUADEC/2008/Slides?action=AttachFile&do=get&target=marketing_gtk_Guadec2008.pdf can be reconsider on the strategical roadmap.

vuntz

jjmarin: can you detail which suggestions? :-)
btw, my request for volunteers to improve the old roadmap process is valid for everybody, not just jjmarin ;-)

jjmarin

vuntz: basically, unify branding of GTK bindings and easy to install process on all OSes

vuntz

ah, bindings

diegoe

jjmarin, I'm working with aruiz on implementing part of his plan

vuntz

the release team is working on a plan to change the modulesets
and one part of this is to make the bindings part of the GNOME platform instead of having them in a separate moduleset
to clarify our message around them

jjmarin

vunzt: how well is working the roadmap process by far. I mean, people do their reports ?

vuntz

jjmarin: after a while, some maintainers didn't
and we stopped the roadmap process for with 2.28 since we were working on the 3.0 planning
(if I'm not mistaken)

jjmarin

This roadmap is good for the stable versions. Maybe for strategic roadmap is better to disccuss a general area of improvement and see how this affect to every component

vuntz

but that's something we'll want to restart for 3.2 and later anyway; we just need to improve this
jjmarin: yep. The question is how to find the general areas

diegoe

I feel those big areas would be related closely to UX targets
say integrating IM in these or those ways

jjardon

About the bindings: Maybe http://live.gnome.org/GnomeGoals/AddGObjectIntrospectionSupport should be a official GnomeGoal ?
Or the r-t has other plans?

vuntz

jjardon: I would say it should become official :-)
jjardon: but you've been doing a good job at pushing the goals, so don't think you need approval from the r-t Just propose it on d-d-l and see reactions there

jjardon

vuntz, thank you But maybe a "offcial" announcement from the release team will have more impact (inside your strategic for the bindings)

vuntz

jjardon: heh, if you think it helps... Can you write a draft mail for this? We can then send it

jjardon

vuntz, sure

vuntz

so, let's step back a bit
what I got from this discussion
- we need ways to identify project-wide goals for the long term. Usability hackfest can be a way to do this, but it's not the only one
- we need to put some timeline on the goals to help clarify our message

diegoe

yes

vuntz

- we need people to take what the release team (or another group) publishes and promote it in a understandable way
am I missing something?

diegoe

mmm, and that we should communicate clearly that gnome 3.0 does not aim to meet *all* the proposed goals

vuntz

and do people think the release team is the appropriate group to do this job?

pcutler

vuntz: when you talk about communication and promotion, I'd like to see collaboration with the marketing team

vuntz

pcutler: how would you like to see this happening?

jjmarin

About the where to start from. Maybe is good idea to write a SWOT analysis

vuntz

jjmarin: is this something you'd want to make happen?

pcutler

vuntz: i came to the meeting late, but you mentioned the release team is looking at the module sets, and andre had mentioned something similar a while back too. I don't know if I know the answer to "how" yet, but looking at the work we're doing with Plone, we have to have communication between the two teams for stuff like that

jjmarin

vuntz: I can help to create it if it helps
stormy: what do you think about a SWOT analysis for generating ideas for strategy

vuntz

jjmarin: the best to make it happen is to lead the effort to make it happen :-)
pcutler: the release team is working on the plan for the modulesets right now, and I guess it'll be proposed once we're happy with the plan

stormy

jjmarin: it's worth trying for sure.

jjmarin

vuntz: I don't have experience on that, but I can try it

vuntz

pcutler: the way I see it (and please tell me if it's wrong) is that the marketing team would see the plan when it's proposed (on d-a-l) and start to play with how to promote it

diegoe

jjmarin, would you like to assign yourself an task item to do that? :), maybe chat a bit with stormy about it

pcutler

vuntz: that's fair

vuntz

pcutler: sure, it's fair. But maybe it's not optimal for the marketing team. Is this something we should change a bit?

jjmarin

diegoe: OK, I have a new task then :)

diegoe

cool. then #task: jjmarin to discuss with stormy about a SWOT analysis

vuntz

anything else on this topic?

jjmarin

good enough by now :)

vuntz

any other topic people want to discuss?
or should we close the meeting?
okay, no reply means we're done

diegoe

:)

vuntz

thanks to everyone for joining!

stormy

:)

bkuhn

:)

jjmarin

;P

vuntz

diegoe: any idea if we'll do the next meeting next month? Or in two months?

diegoe

good question, maybe next month makes sense since we have a release, so opinions will be fresh

vuntz

okay
hm, does someone saved a log of the meeting?
we'll want to publish it and send minutes to foundation-list

jjardon

I have the log of the meeting

vuntz

jjardon: cool. Can you put it on the wiki?

jjardon

vuntz, sure

vuntz

jjardon: maybe linked from http://live.gnome.org/FoundationBoard/Minutes (see the bottom)
lovely