glyn_normington at uk.ibm.com
Wed Sep 19 12:05:10 EDT 2007
Fascinating! Again on the JSR 277 front...
One important observation. I notice that your class loading search order
prefers a module's own contents to imported classes. The OSGi module
system, with which JSR 277 must interoperate, has a different search
order, preferring imported packages to a module's own contents. This isn't
necessarily a show-stopper for interoperation, but it might have some
nasty consequences. Some notes on interoperation are available (), but
the proposed solution is still being worked on in the JSR 277 Expert
You may also be interested in a draft formal model () of OSGi
modularity, which covers the class loading search order in section 9.
Alex/Andreas: are you going to formally add Rok to the Expert Group,
should we treat his research as "not a contribution", or what?
jsr294-modularity-eg-bounces at cs.oswego.edu wrote on 18/09/2007 07:47:50
> Dear Michal,
> > This looks very interesting!
> > A quick clarifying question: it's not clear to me from either the
> > paper or the web site or the paper what is the JAM referred to in the
> > paper. Is this the JSR-294 proposal? Or perhaps something else?
> JAM refers to the proposed Java Module System. That is, the
> combination of JSR-277 (its Early Draft) and JSR-294 (its Strawman
> p.s.: It appears that my posts to this mailing list are not being
> accepted. Can that be fixed, or should I rather post to another one?
> jsr294-modularity-eg mailing list
> jsr294-modularity-eg at cs.oswego.edu
Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the jsr294-modularity-eg