[news.eclipse.technology.kepler] Re: Kepler and Buckminster

[Carlos Sanchez <carlos.sanchez@xxxxxxxxxxx>]

Hi Thomas,

I don't think we have the same objectives, related but different in essence.

In Kepler we try to handle the community aspects of a project,
independently from any other aspects like build information. Based in
our experience with Maven, forcing Kepler into Buckminster would be bad
for both, if we hadn't thought that, we would have developed Kepler as
part of Maven. We neither want our investigation to delay Buckminster's
progress, especially given it's importance to the Ganymede release process.

Buckminster obviously has a much larger scope, such as build system
abstraction, whereas for Kepler build info is out of scope. Again, in
other case we would have folded Kepler into Maven.

Although we have different objectives we definitely are going to
collaborate with Buckminster, as we already did by donating the
Buckminster Maven integration, and we will continue to look to reuse the
code and contribute to Buckminster again, and create extension points
where needed for a smooth integration between both.

I don't see any problem having related technologies in two different
projects, it allows focusing in a narrower objective. There are other
examples like P2 that are even closer to what Buckminster already does.

We believe that the metadata belongs as a separate project and we want
to get started in investigating the best way to move forward, and as
part of that we will explore convergence (perhaps over time) with the
cspec, or at the very least strong interoperation with it.

I think this is the time to go through the creation review and get some
room to try and experiment our ideas and look in the future how the
relations between the projects evolve.


Thomas Hallgren wrote:
> Reading and talking about the Kepler creation review, I've come to
> realize that the objectives of our two projects are 80+% similar. We
> both want to establish a common model of the component world. This is a
> great opportunity to join forces.
> 
> Kepler and Buckminster both have a common model that is:
> - agnostic to the type of component (Maven, OSGi, Eclipse Feature, etc.)
> - agnostic to the type of build system (Ant, PDE-build, Maven plugins,
> etc.)
> - extendable through extension points.
> 
> The Kepler and Buckminster platforms want an API that allows you to:
> - add new types of components
> - add new types of repositories
> - integrate tightly with Eclipse team providers such as CVS, SVN, and P4
> 
> Kepler and Buckminster objectives include:
> - consume the new p2 model
> - consume other component models
> - include community information
> - integrate with build server management
> 
> The Kepler team has additional good ideas, such as the "facets" concept.
> A facet is something that extends the model in some respect so that it
> can be extended with information from component models that goes beyond
> what the cspec can describe today. This is a good idea and, as far as I
> can see, the only real difference between the two projects.
> 
> So rather than duplicating each other's work, let's join forces in a
> single project and take advantage of the purpose of Eclipse: to reduce
> the cost of the common infrastructure through multiple contributors.
> Then we can each focus more of our efforts on our profit making
> extensions and services.
> 
> Kind Regards,
> Thomas Hallgren