Maya 3-29-07

Agenda

1. Obligatory administrivia
   1. Introductions, etc.
   2. Overview and status on Maya
2. Making the Maya project real
   1. Advice from community
3. Overview of the Cisco contribution
   1. Timing and components
4. Scope and Maya proposal feedback
   1. Clarifications, etc.
   2. Thoughts from community
5. Technology and touch point discussions
   1. Distinction from other provisioning projects
   2. Drill-down into technology details
6. Next steps for Maya proposal
   1. Next conference call, proposal changes, etc.

Attendees

Dave Stevenson – Rational, involved in provisioning related work Dimitry Voytenko -- Provisioning work with Java web start Pascal Rapicault – Equinox Team, investigation into provisioning Todd Williams -- Genuitec, interested in committing to Maya Thomas Hallgren – Buckminister, integration with Maya Elias Volanakis – Innoopract, interested in evaluating integration of Maya into Eclipse Jochen Krause – Innoopract, interested in extending Maya for commercial products Sharanya Doddapaneni – Cisco, contributing to Maya Jennifer Li – Cisco, contributing to Maya Dennis Vaughn – Cisco, contributing to Maya Tim Webb – Cisco, project lead for Maya

Discussed Topics

Introduction from project lead, Tim

Provided overview of project status including recap of discussion from newsgroup concerning role of Maya project as it relates to other provisioning activities.

Ganymede provisioning workshop

Maya team will have a representation at the workshop. Looking to work with others on long-term provisioning “story” covering multiple projects.

How do we most effectively get Maya up and running successfully?

Some suggested points during the discussion:

1. Well defined scope
2. Leverage community
3. Good and open communication
4. Strong committers

How do we reach people interested in provisioning not present at EclipseCon?

Well there is no easy way doing that. One of the solutions suggested was to publish a white paper to get momentum. Also, in general keeping people interested through active communications during the code contribution and approval phase is a key factor for a projects success.

One of the questions raised was regarding the availability of Maya code

Currently the code is not available in a public repository. The contribution will be made available as part of the parallel IP review process with Eclipse at the time of creation review. Assuming the creation review happens in a quick timeframe, no other plans will be put in place. Should the creation review be delayed, we may look at other ways to keep momentum.

Are there any timelines set for moving the code to Eclipse?

One of the action items pertaining to this to schedule a creation review and get the IP review process started.

Can there to be much diversity on a given project?

The success of the project depends on the whether there is a code base to start on. Also, proper distribution of code contributions among various committers or interested parties is a key for the progress of the project

Sample client of Maya on an external server

We are currently waiting to get everything EPL ready, before Maya can be made available to anyone externally. It is unlikely that a drop of Maya can be made available before Eclipse’s full IP review process is completed.

Things that need to be done for the creation review:

1. Switching hibernate out to EclipseLink
2. The majority of Maya is ready for contribution and will directly fall under EPL.

Is Maya moving towards JTA based solution to allow different persistence providers to be plugged-in?

Yes, Maya is aiming to provide this kind of solution

Questions or concerns on scope of the proposal

Is Maya going to provide extensions to plug-in other native installers or prohibit this functionality? There is an action item related to this currently Software licensing and control part of the proposal was not correctly phrased, as to whether Maya will provide some extension hooks to enable such extensibility

Maya Project Management

A plan needs to be put in place to structure the contributions from various committers

Is the Maya external software contribution stable?

There are things that need to be done like documentation etc., however the software is currently being used in a early fashion within Cisco with success.

Is it possible to have ongoing meetings to dig deeper into the technology?

We will schedule a follow-up architecture discussion for next week.

Action Items

• Cisco: Description or a Wiki on the bundles and extension points that are currently available (or planned) within Maya
• Tim: Request and schedule a creation review
• Tim: Call for overview of the architecture and extensibility hooks in Maya - 8:00 am PST Thursday April 5th
• Cisco: Follow-up with Oracle for timelines and investigate when is the best time to switch out hibernate to EclipseLink.
• Todd: Provide introduction to TopLink contact (Done)
• Tim: List Buckminister as an interested party / committer in the proposal
• Tim: Add another section to the scope highlighting the types of extensibility hooks that will be provided with Maya including referring to the out of scope section as areas that Maya could allow but will not directly support.