Re: Happy Valentines Day

Sundar Narasimhan (sundar@ascent.com)
Sat, 14 Feb 1998 22:01:06 -0500


Okay, I am going to take this as a cue to get my two cents and two licks
in. These views represent my own and may or may not be closely held
nor even believed by others.
Greg: Thanks for taking a shot. Would the following be fair?

- You seem to be saying that XML can represent all kinds of state
including those needed by most computations extant today.

- The behaviors (methods in your description) access and manipulate said
state primarily through using the meta-data facilities provided within
XML.

- Someone is working on making all of the services layer in CORBA
(such as naming, persistence, versioning, transactions, events etc.)
or DCOM -- to name two widely supported/available distributed object
models work thru XML? and that it will be widely available soon.

If you read relational history -- it took at least two decades to
develop -- i.e. for those that believed a separation of data model and
application logic can hold. Even now it can be argued that this view
has not been fully realized. I find your faith that universal
application and data integration can occur thru XML fascinating, but
somewhat naive. Of course I'm willing to be convinced by examples or
other work.

ps. I am not trying to take issue with those that believe that XML
will be useful for Web pages, or desktop publishing. What I'm still
curious about is the hype behind the ties between OO, distributed
objects and xml. (Of course, more than one practioner has privately
admitted to me that the real reason might be that the OO community has
gotten caught flat-footed by the Web, and that many in the CORBA and
(D)COM communities want quite desperately to have something
intelligible to say about the Web and its future course. Look at all
the saliva running across the OMG in response to the Enterprise
JavaBeans spec ;)