Or maybe it's just sour grapes. 8-/
> MB> I realized that fairly late in the game - 1994 - but I got the
> MB> hell out of Unix/C/RDBMS as soon as I could.
> Okay, now you're scaring me. I'm not an RDBMS person, but I consider
> myself to fall into the Unix/C camp... Well, Unix/C and the usual host
> of scripting languages. I'm with David where "objects" are concerned...
> I've never quite understood the lure. So if I don't jump on the
> "object" bandwagon (at some level), my only alternative, long-term, is
> to write apps for Win95/NT-wielding machines? Ouch.
The good news is that the Web world is morphing into the object world
(at the same time the reverse is also happening - though much slower, and
with lots of "Not Invented Here!" problems). Feel free to start from
either side, so long as you're looking ahead to convergence.
Picture serializing a Java Bean into XML (using URLs for classes that
provide the implementation, like the Smalltalk object model) and then back
again. The Javasoft view is that that Bean isn't useful until it's been
deserialized back onto the JVM heap - yet the Web manages to do a whole
lot with serialized-to-HTML objects (and encapsulated scripts) that we call
"documents". Go figure.
What's the difference between a Bean exporting a property "Name" that has
the value "Mark Baker", and the following XML document?
Not a whole lot I'd say. Each is just a serialized version of the other.
I posted some of my thoughts on this topic to FoRK and dist-obj back in
Note that a lot of this is old news to many FoRKers. I'd even hazard a
guess that more FoRKers than dist-obj-ers *really* get this stuff.
-- Mark Baker, Ottawa Ontario CANADA. Java, CORBA, XML, Beans http://www.iosphere.net/~markb email@example.com ICQ:5100069
Will distribute business objects for food.