W3C home > Mailing lists > Public > www-talk@w3.org > November to December 1995

www & objects - was OLE/COM - Microsoft's Strategy For Once Again Dominating The Software Industry

From: Sankar Virdhagriswaran, Crystaliz Inc. <sankar@fcrao1.phast.umass.edu>
Date: Sat, 09 Dec 1995 13:39:07 +0000
To: "Daniel W. Connolly" <connolly@beach.w3.org>
Cc: www-talk@w3.org
Message-Id: <9512091813.AA17936@fcrao1.phast.umass.edu>

>"The Future of Object Technology in the WWW"
>Submitted by Mark Madsen, Information Services Framework, APM Ltd 
>http://www.w3.org/pub/Conferences/WWW4/Panels/obj3.html

IMHO, Mark asks the right questions. The first and most important question
to answer is what *new* applications will be enabled by making Web object
ready/object capable/etc.

Web is about collaboration between people. In the act of collaboration, they
might use some applications/resources. So, sometimes it is collaboration
between people and applications all in one bag. The most popular type of
collaborative applications is a well known list. Web, today, at best covers
one of a handful of collaborative applications. Therfore, the question to
answer is what of the other applications can be developed better with object
enabled web.

This question is *not* answered by blaming Microsoft or comparing
Microsoft's OLE with CORBA. The current HTTP infrastructure is more to blame
w.r.to not enabling some of the collaborative applications. It has very
little to do with whether WWW is object enabled or not.

The most popular collaborative applications are (you can find this in any
Lotus Notes brouchure and if you looked carefuuly in some other places, you
can find the rest of this list):

1. Publishing - Web today is a publishing medium. Most of the efforts in
integrating some form of an object broker with the Web are aimed at
addressing the need to publish some database information to the world.

The following applications have an object component. But, fundamentally,
objects perse won't help.

2. Tracking - progress tracking, project management type tracking, etc. Web
does not support it today and until it becomes async. and supports
meaningful notification capability, doing this is hard. 

3. Workflow - Ooops, there goes the async. word again (objects, hmmm...,
useful but having it is not really going to solve the async. problem). 

4. Conferencing - Multicast (where are the objects ?)

5. Discussions/Annotations - Notifications, very little objects

6. Concurrent applications (version management, transactional workflow): Web
has a problem with its naming scheme for versions. Of course doing workflow
is not possible, so why worry about transactional workflow.

7. Collaborative authoring: Now, this requires objects. But, not the CORBA
or COM objects. The objects here are about the document being
collaboratively authored (be it word processing documents or military
logistics plans). CORBA/COM objects are at too low level for people to
people interaction. It also requires asynchrony.

So, capitans, are we not focussing on the wrong end of the problem. Let us

If you folks are going to be at the W4C conference stop and see our poster.
We present our solutions to these issues.

Let us make HTTP into a truly asynchronous (reference and time)
infrastructure, think about objects that people care about, and then argue
about whether OLE/CORBA/Java/what-have-you will help.



  
Sankar Virdhagriswaran                         Phone: (508) 287 4511
Crystaliz Inc.                                 Fax:   (508) 287 4512
Received on Saturday, 9 December 1995 13:26:52 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 27 October 2010 18:14:18 GMT