What does Python fix?

Courageous jkraska at san.rr.com
Tue Jan 22 21:49:23 EST 2002


>Swing and things like Swing demonstrate cracks in the "write once
>run every concept," basically illustrating that the overall approach
>has its flaws.

>Actually, we've had very good luck with Swing.

I can believe that you do; however, this won't change the fact that
Swing is horribly bloated, slow, and responsible for most of the memory
leaks encountered by your typical Java programmer. Swing is in many
way the Achilles' heel of Java. If someone says "Java is slow," they're
usually quite wrong. They have gotten a perception of Java's slowness
from the mollasses like speed with which Swing runs. In fairness, this
disctinction is in some ways academic. Let's just say, however, that
the Java _language_ isn't slow at all, not particularly in the 1.4b3
releases where even I/O is pretty damn quick.

>Not sure what problems you've run into.

Oh, nothing like you're thinking. I'm the author of a rapid prototyping
toolkit built using Swing. Swing has good API appeal. It's slow, however,
and having dealt with Swing for some time, I decided that it's slowness
is in part due to its internal design approach. I'm looking forward to
seeing IBM making progress on SWT. On Windows, it rages. They have a lot
more work to do, however.

C//




More information about the Python-list mailing list