>My tastes (obviously) run to a more evolutionary approach for HTTP. I'm
>unconvinced that the performance problems require a flash cut to a binary
>protocol. Spero has shown that doing multiple transactions over one
>connection achieves signficant performance improvements. His response is
>to change HTTP drastically. Mine is to do so within the current overall
>design.
Agree. It certainly does seem that most of the performance improvements
made by HTTP-NG could also be made in HTTP-TOS, with a simple modification
to allow multiple objects to be retrieved/posted over a single TCP/IP
connection.
Are we really willing to switch to something totally different and
incompatible, just to eke that last 10% speed improvement? I'm skeptical.
--
Eric W. Sink, Senior Software Engineer -- eric@spyglass.com
I don't speak for Spyglass.
"Can I get a direct flight back to reality, or do I have to change planes
in Denver?" - The Santa Clause