Hi to all,
some reflexions on this, just my opinion ... to discuss:

given the complexity of eclipse and its dependency and requirements, I
fear to add it in the core pivot project.
But don't get me wrong, I think this could be a great addition to Pivot.

Only I think it not belongs to the core platform, and the same for Bnd
and other OSGi related tools (at least at the moment). As Todd said,
we could put all this stuff under out project on Google Code, also the
OSGi-fy of Pivot jars.

Some time ago I proposed to change that project to something like
Pivot-stuff (like Wicket ...), a place where to put all Pivot-related
things but that not belongs to the core platform.


Under tools I'd put all our utilities, but without external
dependencies (if possible, or to Apache libraries, to avoid license
problems), and maybe for the 2.0 also a GUI Builder, but all done in
our code.

> How essential is it that the build script we have is the main mechanism for 
> building the release?
It's the standard way of work, as most of us (and others, don't forget
our current and future users).
> Is there scope for any manual process?
Yes, i think for all this stuff, but not for Pivot itself.


Again, don't get me wrong ... my point of view here is only to keep
Pivot as simple as possible, also in its build and dependencies.

Comments ?


Bye,
Sandro

Reply via email to