World Wide Web.  Where are we heading next?


Submitted By Amine Mekkaoui
Cambridge Technology Partners
304 Vassar Street, Cambridge Mass, 02139
Tel  617-374-8419			


It seems that most of us at this point have caught up with the fact
that HTML/HTTP technology integrated with a database in the back end
makes things a little bit more interesting than merely retrieving an
HTML based document.  HTML authoring tools, particularly those with
links to databases via the Common Gateway Interface have been
available for a while now.

So what's next?

As corporate America sees the benefits of making legacy and existing
client/server applications available to the World Wide Web community,
the developer community will be busy resolving issues on how to
integrate a static HTML-based client into existing applications and
maintain the state of an application throughout a session; how to
enable server-based applications to handle millions of users arriving
via the Internet; and, how to manage the distribution of data and
processes without rewriting existing business functions.  So far this
has caused frustration.

While HTML is a well-defined technology, it does have its
shortcomings, specifically in performance, richness of the user
interface, interoperability with existing application and data, and
scalability.  Performance suffers because of the extreme level of
overhead, due to the requirement for connection re-establishment on
every transaction and the excess formatting content specified by HTML.
Interface facilities are limited to those that can be implemented in
the context of a "page". Integration is complex because HTML is not
designed to integrate with current server-based applications.
Scalability is lacking because of the difficulty inherent in
distributing processing using the HTTP server model. All these factors
limit the degree to which current Web applications can meet changing
business requirements.

Sun's Java language has emerged as an industry-recognized language for
the Internet.  The Java language holds significant promise.  The
language solves problems on the client-side by improving performance,
enabling the creation of dynamic real-time Web applications, and
providing the ability to create a wide variety of user interface
components. However Java is a recent development.  As Sun’s Java
matures our current HTML problems will begin to disappear.

>From the perspective of the server, the lack of robustness and the
architectural complexity in integrating a distributed architecture
with today’s web clients suggests that the OMG model promises a
bright future for the World Wide Web community. By applying the OMG
model with JAVA we will be able to distribute data and processes, be
platform independent, integrate with existing computing applications,
and create a dynamic and richer client environment.

As it stands today, there are large and growing number of Web-related
tools on the market that address distributed processing, hardware
independence, and provide common methods for disparate objects to
interoperate.  Some of the vendors that are leading the pack in this
area are NeXT software with their WebObjects, ParkPlace-Digital with
their Visual Wave, and SUN’s Microsystems NEO products combined
with IONA’s Orbix which is Corba 2 compliant.