Going back to the original question. I completely agree that it is to late to introduce OSGi integration the 5.0.x cycle. Can we at least resolve the issue TAPESTRY-2519 before 5.1?
https://issues.apache.org/jira/browse/TAPESTRY-2519 It is just an improvement which will stay invisible for the most Tapestry users. It is really ugly to copy the ClassNameLocatorImpl to be able to override it. On Wed, Jul 16, 2008 at 7:53 AM, Howard Lewis Ship <[EMAIL PROTECTED]> wrote: > I think a good pattern may be to create a Tapestry commons that > contains smaller projects on their own schedule, separate from the > "core" modules. > > Further, I'm looking into setting up SVN, project home pages and > downloads at tapestry.formos.com, to support auxilliary projects that > (due to licensing or maturity reasons) can't be hosted at Apache. > > On Tue, Jul 15, 2008 at 12:09 PM, Igor Drobiazko > <[EMAIL PROTECTED]> wrote: > > I personally don't like the idea of "version-per-subproject". Managing > these > > versions without Maven can be very ugly. There are still a lot of > developers > > who download Tapestry binaries manually and put them into their version > > control system. > > > > Another idea of increasing the developer activity would be a so called > > "sandbox" or "incubator" project (see MyFaces or Eclipse). Some cool > stuff > > can be developed in the sandbox. Reaching maturity this stuff can be > moved > > into main production. > > > > > > -- > Howard M. Lewis Ship > > Creator Apache Tapestry and Apache HiveMind > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > > -- Best regards, Igor Drobiazko
