[lsb-discuss] Conference Call Minutes (2012-02-08)

[Stew Benedict]

LSB teleconference

Wednesday,  8 Feb 2012; 11:00 to noon, US ET

Where:
     (605) 715-4920 Access Code: 512468

Gobby:
     gobby -j lsb-temp.pmman.net:6522 with passwd: LSB

IRC channel is:
     #lsb at freenode.net
     #lsb-meeting at freenode.net during meetings


Attendees:
in the form :[irc_nick ]name (organization)
===========================================
licquia Jeff Licquia (LF)
stewb Stew Benedict (LF)
AlanClark Alan Clark (SuSE)
mwichmann Mats Wichmann (Intel)
denis_silakov  Denis Silakov (ROSA Labs)
robjo Robert Schweikert (SuSE)
ktate Kay Tate (SuSE)

potentials:
Darren Davis (SuSE)
jdluhos Jiri Dluhos (SUSE)

Apologies:
==========
orc_emac Russ Herrold (Owl River)

Posted Agenda:
=================

  - Robert's LSB 5.0 roadmap.

  - Infrastructure bug review.

New business:
=============

Meeting opens at 11:00 with:

LSB 5.0 Roadmap:
https://wiki.linuxfoundation.org/en/ProjectPlan50

Robert: split up what we had before into separate documents. Main page 
(above) has the objectives for 5.0.

Other pages:
https://wiki.linuxfoundation.org/en/Uplift_Target
https://wiki.linuxfoundation.org/en/Parkinglot50

Robert: Any comments, does this work for everyone?
Alan: Maybe a section emphasising the business case for the 5.0 release. 
I'll try to work on that.
Alan: What are the things on the parking lot page? Out of scope?
Robert: Yes, out of scope, or not clear what to do with, or ongoing 
tasks. Some might get moved.

Robert: Pushed the soul-searching discussion out of the way, and 
focusing on the 5.0 goal.
Licquia: I think we had a good plan before the infrastructure issues. I 
think moving more towards the community distros is the right way to go.

Robert: Objective #1 - modularization of the standard - yes/no, which 
proposal?
(https://wiki.linuxfoundation.org/en/Modularization50)
Robert: I think we should pursue this. Current implementation is too big 
of a blob.
Denis: I agree.
Mats: I think we've been asked for this enough times that we would 
appear unresponsive to ignore it.

General agreement from the group
Review of the Modularization50 page to refresh folk's memory on the 
proposed approaches.

Some discussion of the Runtime Profiles proposal, where the SDK would 
allow an ISV to target newer versions of components/particular 
distributions.
Let people mull over the modularization proposals for next week.

Robert: Next test modularization - include in 5.0, yes/no? We've been 
relatively unsuccessful getting our tests accepted upstream. I don't 
know that we'll get them all the way upstream but we may be able to get 
them into the distribution QA process. The distribution people are more 
closely coupled to the upstream projects than LSB people.
Alan: I think this should be the premier objective of the release.
Jeff: Modularization could mean a couple of things: split them 
source-code wise, or keep the source code as it it, but allow running 
subsets of the tests against a particular library/component.
Robert: There's a task on the page, to develop a test modularization 
proposal, which would be the "how", but we need to decide initially 
whether we want to do this.
Denis: ROSA will be doing some work to integrate tests in our process, 
which should generate some feedback.
Robert: Anyone against test modularization as an objective?

No objection

Robert: Library uplifts are already confirmed.
Robert: Should we even consider new trial use modules?
Jeff: If we're going to try and track the community distros, I think 
we'll need to do some trial use modules. The question is which ones? 
(gtk3 was an example)
Robert: I'm OK with that, objections?
Alan: Fine with the idea, worried about the scope.
Robert: We can limit the scope when we go through the targets on the 
uplift page

4 Objectives discussed are all now confirmed

Robert: Anything I missed?
Jeff: There are things to be done with regard to deprecations
Robert: Tied to uplifts, I'll change the description.
Robert: For next week, we should talk about the modularization 
proposals, the test modularization proposal.

Tasks: Stew to work on a document describing the current test status.

Kate: Has there been any more discussion on validation?
Robert: Focusing for the moment on the tasks, putting the soul-searching 
argument aside. We know both distributions and ISVs have asked for 
modularization.
Kate: While we have selected input, statistical data would be useful.
Jeff: I think we've agreed we want to modularize. Looking at the 
schedule it looks like we have time to do some research like we talked 
about last week within the schedule.

Some discussion of how we'll move from these target items to tasks, to a 
release schedule.

End Of Call


Next meeting:
=============
   15 Feb 2012 11:00 to noon, US ET

=======================================
This document: LSB minutes 20120208.txt