I have opted to chime in only once before, but I follow this list closely.
I have built many SDKs and frameworks over the years, and this might provide
a good distinction here.  MRI (and JRuby) are implementations of the Ruby
language.  It is always a give-and-take discussion about which features or
libraries to ship with a language implementation, but I have to side with
the 'lean' approach.  This keeps the runtime from becoming too large, and
helps avoid building dependencies on libraries that may be superseded in the
not too distant future.

It sounds like this would be a great ancillary project to JRuby - building
gems or all-in-one distributions that bundle JRuby with Rails and other
Java/Ruby libraries (an Eclipse distribution would be a natural as well).
There is growing interest in this "value-added" approach to bundling open
source projects into "install and run" solutions for developers and
end-users alike.  My time is somewhat limited, but I would certainly
consider initiating and contributing time to a project that creates and
distributes bundled JRuby deployments.  Especially if I can enlist the
support and maybe some time from others on this list.  From my reading on
this list, the actual mechanics of building is neither terribly difficult
nor (relatively) time-consuming.  The real trick is coming to agreement on
what to offer in terms of bundled solutions.  But the Rails package is a
no-brainer, and would represent a good way to kick off this project.  But I
don't want to step on anyone's feet - perhaps Charles/Thomas/Nick would
prefer to keep these distros under their control.

I would also add that the MS analogy is not a very good one.  MS Visual
Studio leverages functionality baked into the OS, and there are few willing
to take on the daunting task of offering alternatives.  The open source
approach is about offering choices, and choice is more difficult than one
size fits all solutions.  A JRuby package that meets one developer's needs
will not meet another's needs.  But I'm sure it is possible to find some
common denominators that -most- people will accept and work from there.

--
Michael

Reply via email to