The problem with that approach is that it tends to result in
solutions which only address one problem, while a more
forward-looking design could transform the same basic capability into
a general enabling mechanism. Standardization of overspecialized
solutions will just hold back the Web.
> Those that aren't useful in the context of the Web just
> won't get implemented in a mainstream browser... Does
> anyone think that Telnet will migrate into the browsers?
Exactly...and look at WAIS for an even more poignant example of a
useful, but complex and overspecialized protocol which is gradually
being de-integrated, and will probably be used mainly through HTTP
gateways in the long run.
The lesson I draw from this is that getting _any_ new protocol into
browsers is going to be very difficult, unless it has a truly broad
and diverse constituency. What is the likelihood of a helper
application being designed with a protocol that's more generally
useful than Telnet or WAIS? My guess, very low.
Competition may choose superior solutions, but it doesn't create
them. For that you need foresighted people.
--------------------------------------------------------------------
Paul Burchard <burchard@math.utah.edu>
``I'm still learning how to count backwards from infinity...''
--------------------------------------------------------------------