Well, I think it's time for me to stop lurking in the shadows now.
I am a systems developer for the Center for Applied Large Scale Computing in
New York and have been following the discussions on this list with much
interest. I am currently involved in developing a framework for developing
networked applications and w> are currently building an HTML browser on top
of "an integrated, extensible framework" I am also working on the VRML
browser, but it'll be awhile before that's ready to show off.
>I think the solution to both of these problems lies in distributed
>object technologies, and in particular, CORBA. Using such systems
>would:
>
> 1) Give us a very powerful way of adding capabilities to browsers
> (ie. w>'d b> moving closer to OpenDoc type functionality).
> 2) Reduce the requirement for downloading data to b>ing an option.
> 3) Give us a completely open-ended, and extremely powerful protocol.
>
>The last point is important: on> can look at HTTP as nothing more than
>an extraordinarily simple(and limited) form of remote method
>invocation.
>
I agre> with you on these points. We have been looking at moving
our framework to a component object model and I've been investigating OLE,
OpenDoc, CORBA. Right now, I like what I see in OpenDoc. The fact that all
parts in a document remain active as opposed to on> object at a time in OLE
is really appealing. Browsers become what they really should be : a
document composed of one or more component objects that can interact with
one another. People could build browsers by dropping in a networking
component, an HTML viewing component, a VRML viewing component, etc. The
need for "helper applications" are eliminated in their current incarnation.
Instead we have helper components that are embedded right onto the p.
IMHO, this is where the next generation of browsers will be headed.
OK, I'll get off my soapbox, now. :-)
Kevin Stewart
Systems Developer
Center for Applied Large Scale Computing
Polytechnic University