For the WAN work I'm doing right now, the first bottleneck in practice
is bandwidth. When that's fully addressed, the next will be latency. I
think the Squeak performance will only matter later.
We do use plugins. The direct need (e.g., for video) is as an
abstraction for platform-specific libraries. But speed is a nice
side-effect. (Of course, this hybrid approach has worked well for Python.)
I imagine that Squeak performance would matter very quickly if I tried
to do platform-independent in-world physics.
But the important thing, I think, is to build stuff that can be done
now, rather than waiting for other development teams and other technologies.
David P. Reed wrote:
> Well, Dan's $5,000 wasn't picked up before it expired on 12/31/2006.
> I think the whole world is waiting for Coke from Ian Piumarta.
> Strongtalk looks interesting only if you don't think Coke will pan out.
University of Wisconsin - Madison
Division of Information Technology
mailto:[hidden email] jabber:[hidden email] voice:+1-608-262-3724