Re: Towards better OAF/Bonobo integration

[Maciej Stachowiak <mjs noisehavoc org>]

On 29Jul2001 10:16AM (-0400), Owen Taylor wrote:
> 
> Maciej Stachowiak <mjs noisehavoc org> writes:
> 
> > For some time Michael Meeks and I have been discussing ways to make
> > OAF and Bonobo more closely integrated.
> > 
> > One key step that we feel would improve the consistency of the
> > platform is to make all of OAF's IDL interfaces (most particularly the
> > public Factory interface) inherit from Bonobo::Unknown. This is also a
> > critical step to developing a future public CORBA API to OAF
> > activation, another goal Michael and I agree on.
> 
> Does this mean that all servers activated from oaf have derive from
> Bonobo::Unknown? How does that affect servers whose life-cycle
> management isn't a good fit for Bonobo's refcounting?

No, it doesn't necessarily mean that. The Factory interface could
continue to create CORBA::Objects instead of Bonobo::Unknowns. I'm not
sure yet which approach is better.

For reference, in the Windows COM world, servers whose life-cycle
management does not fit well with refcounting still inherit from
IUnknown but completely ignore the refcounting methods and implement
their special life cycle management otherwise (no judement of whether
this is good or not is intended).

 
> If you are going to tight integration with Bonobo, shouldn't
> the activation daemon be packaged along with Bonobo? What's
> the advantage of adding the extra package?

As my email explained, the main reason not to do that is to avoid me
and Michael fighting all the time. I think it would be theoretically a
good thing to do but in practice personality issues would get in the
way. Perhaps if these fears prove to be ill-founded, the two packages
could be merged at some point.


Regards,

Maciej