March 17, 1997
Bringing order to the chaos of Internet push
By Bob O'Donnell
Something is usually better than nothing, the old saying goes, and when it comes to the
wilds of Internet-based push technologies and the complete lack of standards in that
world, the old maxim still rings true.
The something in this case is the Channel
Definition Format specification proposed by Microsoft at last week's Internet World
Expo. (See "Microsoft
proposes uniform interface for receiving 'push' content".) The proposed
specification, which is backed by numerous Internet push and content vendors, would permit
properly enabled browsers to receive content from any Web site using any push mechanism
that supported the format.
In theory, this proposal could overcome the one huge hurdle facing push/pull mechanisms
on the Web: The lack of standards. As appealing as the concept of Internet-based push/pull
technologies may be, the absolute chaos that exists in that world now forces individuals
to select content based on technological limitations rather than personal interests.
Instead of freely choosing which type of information they want to receive, users often
select content based on what their push/pull client supports. Or, they get placed in the
ridiculous position of having to download, install, configure, and learn several different
push/pull mechanisms to get the mix of content they really want.
In addition to the hassles this entails for end-users, multiple clients on multiple
desktops can cause big headaches for IS managers trying to support a standard desktop
configuration. By implementing products that support Channel Definition Format (CDF), IS
can provide users with a service they want, but avoid the problem of locking themselves
into unstable proprietary formats. In addition, I can imagine a centralized server product
that coordinates all requests for CDF-based information updates (a la PointCast's
I-Server) that will let network managers control the amount of traffic on their corporate
LANs.
So, it seems as if CDF could be an important tool in solving some of these very real
problems, right?
Well, yes, in theory, but there are a few potential obstacles. First, archrival
Netscape is not on Microsoft's list of supporters for the proposed standard. Apparently
Microsoft chose not to include Netscape in its discussions regarding development of the
standard, which seems like a childish, political move that's undoubtedly justified as a
competitive strike. Frankly, leaving out the market-leading vendor of Web browsers on a
discussion and proposal that could have a profound impact on how the Web is used is just
plain dumb.
Second, the fact that the standard comes from Microsoft immediately puts it into
question by certain Internet-related companies and individuals. I personally don't buy
this, but it seems that unlike the desktop PC market, where Microsoft-proposed standards
are instantly embraced, Microsoft-proposed Internet standards are greeted with unusually
high degrees of skepticism and concern. Already, several vendors left out of the
specification-building process have expressed concern that the specification is too
simple. (Of course, HTML 1.0 was pretty simple too, but look where we are today.)
In spite of all these issues, I think the proposal is a very positive step forward, if
for no other reason than it places a stake in the ground and offers a starting point for
further development. Push/pull mechanisms are talked about as the next coming of the
Internet and, if properly implemented, I think they could have an amazing impact on how we
all access and receive information. But until some kind of order arises from the chaos
that the market is faced with, push/pull technologies offer nothing more than an empty
promise of better things yet to come.
©
Copyright 1997, by InfoWorld Publishing Corp., a
subsidiary of IDG Communications, Inc. Reprinted from InfoWorld,
155 Bovet Road, San Mateo, CA 94402. Further reproduction is prohibited.
