cbsmith at envise.com
Sat Dec 26 10:54:15 PST 1998
Bill Huey wrote:
> > objects system. For that you use CORBA. RMI's real value is that it
> > provides a very simple manner to provide distributed objects to small
> > scale applications.
> > Of course all that being said, once there's a GPL'd implementation out
> > there, there's nothing to stop somebody from writing an improved wire
> > protocol with better performance. ;-)
> So you're saying that the main limitation of RMI is the speed of the
> implementation ?
Not at all. All I was saying was that efficiency and performance were
not the real goals with RMI. Of course, one could easily argue that Sun
REALLY bombed in those areas, regardless of what their goals were. If
you'll notice my previous paragraph I said that if you wanted an
enterprise level distributed objects system, you should use CORBA, not
> What about the completeness of RMI relative to CORBA ?
> Sun has a graph denoting the RMI is somehow a technical superset of
> CORBA ?
I'm not sure what the context of that statement was, so I'm not sure
what they meant by that. Needless to say, however, CORBA provides tons
of facilities which aren't available in RMI. For starters, you have
language independance (which is a realy trick considering lots of
programming languages have different ideas about what objects are).
That's really only the tip of the iceberg though. CORBA is increadibly
sophisticated and includes all kinds of functionality. It's laughable
how MS tries to compare it to DCOM. The key advantage of RMI is it's
simplicity. Although it's more explicit than some distributed object
systems, it's still quite simple to use, whereas with CORBA you just
know all that sophistication is coming at a price in terms of
I think the best way to describe it is that it's actually possible for
me to conceive of writing an implementation of RMI by myself. The same
cannot be said of CORBA.
More information about the kaffe