[JOS-UI] Compatibility might be off-topic

[Gilbert Carl Herschberger II]

Some of you might believe that my concern about compatibility is off-topic
for the user-interface mailing list. You might not see the a connection.
You might not be concerned right now with the run-time compatibility of the
user-interface three to five years down the road. You may be more
interested in these other things.

Yes, we need a place to play. It is just experimental, right? So, it
doesn't have to be compatible with anything. It is just a learning
experience, right? So, there's no reason to be overly concerned at this
point. If we're going to build JADE-UI (or just-another dead-end
user-interface), if we're not going to build on the strenght of the Java
platform, why bother?

How do you maintain your enthusiasm for something you know--for sure--isn't
going to last? Have your given all this, both the user-interface and all
programs that depend on that user-interface, a death sentence?

But I can still see a connection. With Java, it is possible to build a
user-interface subsystem that never becomes obsolete. It is possible to
build a subsystem that looks forward to innovation, embraces change, and
keeps on running all the old code, too. It is possible to give this a
chance to grow, multiply, and live.

Java, the programming language, is a fundamentally different programming
language. It is the first of its kind. There is nothing like it. It is
fundamentally different for these three reasons.

1. It is based exclusively on dynamic linking.
2. It has separate, self-contained interfaces.
3. It has a very large namespace.

It is new. It does not have an architecture like C or C++ at all. It sets a
new standard. You can do things with Java that have never been done before.
With this new standard, we must have new expectations. Before there was
Java, we couldn't build a future-oriented operating system with a
future-oriented user-interface.

People have reached a concensus that you can't build for the future with
COBOL, Visual Basic, Fortran, Pascal, and RPG. It's true. It is not
feasible to build a future-oriented platform with these old languages. Even
Linux is reaching backward into the past.

Sun Microsystems built a new and incompatible abstract windowing toolkit
(called the AWT). They used Java; but, they did not build on the strength
of Java. What then? They built yet another new and incompatible abstract
windowing toolkit (called Swing). They used Java again...and almost found
out what Java can do when you put your mind to it.

What is the point of building yet-another-toolkit? There is untapped
potential here to create a user-interface above and beyond anything the
world has ever seen. We can build a masterpiece.

Imagine this: you can build two, three or four user-interfaces that run
simultaneously on a single machine. You're free to experiment, without
getting yourself into trouble. You can build bridges between the old and
the new. You can run old applications in the latest and greatest
user-interface, without changing a line of code (if that's what you want).

The rules have changed. Java changed the rules. What sounds radical and
impossible on all of the other operating systems may become routine and
taken for granted on JOS.

Give compatibility a chance.


_______________________________________________
UI maillist  -  [EMAIL PROTECTED]
http://jos.org/mailman/listinfo/ui