Re: no 'object architecture' in URI working group

Larry Masinter wrote:
> Whether or not the IETF should standardize on an object system
> architecture (CORBA or no):
The IETF _should_not_ standardise on an object system architecture.
It should develop architectural frameworks within which future object
system architectures can exist and be exploited.

> a) it won't happen in the URI working group

It shouldn't happen in the URI working group.  It's an accident of
history that the subject has been raised again.

BUT: as a matter of principle, the URI group should not ignore the
existing standards, neither should it close off desirable future
development pathways.

> b) the URI working group should not wait until it happens

Strongly agreed.

> so I'd like to close off this discussion, except insofar as it affects
> the topics at hand:
> Are there URL or URN schemes that would be used for naming CORBA or
> SOM or OLE objects? Is there a CORBA/SOM/OLE equivalent for URCs? 

As I've said already, SOM is supposed to be CORBA compliant.  IBM
signed the CORBA 2.0 spec.  I can't speak for Microsoft, but they have
said that OLE will be CORBA interoperable.  Sticking with CORBA, there
isn't a problem in defining URCs, except that there is as yet no
specified mapping of CORBA to SGML.

> Conversely, are there ways that URLs, URNs and URCs are reasonably
> expressed inside CORBA or OLE etc? I vaguely remember something (a
> conjecture?) that Microsoft's Internet Assistant embedded URLs inside
> OLE links, for example.
This way round, it seems that one could define URNs to be understood
as CORBA tagged profiles and use an ORB to provide a resolution

> These interoperability mechanisms might well make topics for
> useful RFCs.

Strongly agreed.

Mark Madsen: <> <URL:>
Information Services Framework, The ANSA Project, APM Ltd., Castle Park,
Cambridge CB3 0RD, U.K.  <URL:>;  <>
Voice: +44-1223-568934; Reception: +44-1223-515010; Fax: +44-1223-359779

Received on Wednesday, 12 July 1995 06:59:25 UTC