Re: [rt-pmc] Virgo tooling

[Jeff McAffer <jeff@xxxxxxxxxxxxxxxxx>]

Door #3 sounds good. There have been several discussions around OSGi improvements in PDE and WTP being able to deal with bundle based servers.  Perhaps this is an opportunity to drive both of those and create a base powerful enough to support Virgo tooling needs.

There is one caveat though: this WTP and PDE spinoff efforts should happen reasonably promptly.  While the charter snippet you point out does open the door to tooling, the first "out of scope" element ("Major tooling efforts") limits how wide it can open.  From what I can see STS is a reasonably major effort and thus should not "live" in RT.  RT as a halfway house sure.

The approached used with Virgo itself might make sense.  Get it in an stabilized as part of Eclipse and then the second release should have the refactoring, deeper integration, ...

Jeff
 
On 2010-10-14, at 5:28 AM, Glyn Normington wrote:

> We are considering how best to donate the Virgo tooling to Eclipse. Martin Lippert will lead this activity with support from the Virgo team. We are currently considering the options for which project to donate the tooling into.
> 
> WTP seemed like a reasonable option but the WTP charter [1] limits the content to standards-based tooling. This is not appropriate for the Virgo tooling which supports a number of non-standard features (e.g. PARs and plans). After donation to some other place, it would make sense for the standards-based features of the Virgo tooling (e.g. graphical exploration of package and service wiring) to be factored out into WTP. This makes particular sense in view of SAP's recent OSGi Enterprise Tools submission [2] to WTP and SAP's active involvement in Virgo.
> 
> Another option is to create a separate subproject under the top level Tools project, but that feels a bit too high profile since the Virgo tooling is so Virgo specific. Also, we don't want to position the Virgo tooling as a competitor to WTP or PDE, so something lower profile would be more suitable.
> 
> A third option is to donate the Virgo tooling into Virgo and then follow this up with appropriate rationalisation with respect to the OSGi Enterprise Tools. The goal for this follow up work would be to factor out enterprise OSGi specific tooling into WTP, maybe some general OSGi parts into PDE, and leave only Virgo specific features within the Virgo project.
> 
> We prefer the third option as this would bring the tooling into Eclipse relatively quickly rather than having Virgo application developers use SpringSource tooling. It will also be easier to drive the refactoring and rationalisation with respect to the OSGi Enterprise Tools from inside Eclipse (e.g. by linking bugzillas together and having visible discussion on Eclipse communication channels).
> 
> The RT charter [3] seems to cater for just this situation in its scope:
> 
> "Incidental tooling efforts to enable or facilitate particular runtime functions in conjunction with (e.g., as a component of) a sub-project."
> 
> but before proceeding along those lines, I wanted to check with the PMC. Thoughts?
> 
> Regards,
> Glyn
> [1] http://www.eclipse.org/webtools/project-charter.html
> [2] http://eclipse.org/proposals/osgi-enterprise-tools/
> [3] http://www.eclipse.org/rt/charter.php
> _______________________________________________
> rt-pmc mailing list
> rt-pmc@xxxxxxxxxxx
> https://dev.eclipse.org/mailman/listinfo/rt-pmc