Actually, the way I see it, the "hardware" component of the CBI is
somewhat there - a big IBM box with all the services required (MySQL,
PHP, Apache, etc).
The current "code" behind the CBI doesn't exist, but Nick Boldt and
the EMFT team right now have code that we could potentially use... I
was thinking that when Nick got his projects to build on
build.eclipse.org, he could kindly "donate" that code to an Eclipse OSS
project, such as Phoenix, where Nick would become a committer
(actually, Mike Milinkovich had this idea, and I think it's a great
idea if Nick agrees). The foundation's IT team could then "install" an
instance of the CBI code for each project that wants to use the CBI.
Once installed, the project using it can customize it, or help maintain
it by submitting patches back to the owners (Nick via the Phoenix
project).
This would pretty much complete the circle of a CBI, without a need for
Foundation staff to code stuff - we'd all collaborate on the code and
evolve it as needed.
All this depends on whether or not Nick would like to donate his code
to a new Phoenix component on which he would become a committer.
Thoughts?
Andrew Overholt wrote:
On Mon, 2006-27-11 at 13:48 -0800, Bjorn Freeman-Benson wrote:
What would you like to say? Ward and Denis and I are actively reading
everything that goes by and I guess we are "the Foundation"...
I was referring to the work item we decided on of writing a report of
what we recommend for a CBI - something formal that could be presented
at a financial planning meeting.
If it were up to me, I'd recommend that the foundation undertake the
creation of something like I outlined in the previous image. This would
require at least one person, but in my dream world the foundation would
think this was useful and would hire someone with its buckets of cash :)
Of course, I realize that's a big expense and that the foundation has
typically rejected actually writing any code, but I honestly can't think
of any negatives that could come from implementing such a system. I
guess people could argue that they're already getting by with internal
machines and processes, but I see no reason those would have to go away.
In my dream world, I see the CBI becoming so good that over time every
team would *want* to use it :)
I realize, however, that it's not just up to me and that my dream world
doesn't exactly match reality. So I was questioning whether or not we
should write a report at all seeing as Nick's work is progressing and
could probably be extracted to a decent CBI. I guess I just think that
the foundation should undertake something like this as it does things
like facilitating project work and encouraging openness and all that
jazz :)
Andrew
_______________________________________________
europa-build-workshop mailing list
europa-build-workshop@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/europa-build-workshop
--
Denis Roy
Manager, IT Infrastructure
Eclipse Foundation, Inc. -- http://www.eclipse.org/
Office: 613.224.9461 x224
Cell: 819.210.6481
denis.roy@xxxxxxxxxxx
|
