Re: HTTP-NG: status report

Simon E Spero (ses@tipper.oit.unc.edu)
Tue, 22 Nov 94 14:04:48 -0500


This needs a more detailed response, but here's a few instant sound bites :-)

1) Working group
Should be one set up after the BOF at the IETF in San Jose next month

2) What is needed- I outlined a few of these in the architectural overview
more will undoubtedly be presented at the BOF and WGs

3) Why change - I also outlined the rationale for changing in the overview -
this is expanded in the HTML version (to be made available when
Dave gets back from London tommorow.

4) When - part of the design criteria was the need to have something that
could work well in a phased transition - the proxy approach works really
well for this - in theory there only need be two machines running the new
protocol either side of a long pipe, and all old clients could proxy through
them to see a huge speedup (need to be pretty hefty machines though :-)

5) Servers parsing documents - This is fine, providing the results are cached.
One of the problems with server-side includes is that because they aren't
referential transparent, you can't in general cache the results.

Simon
-----
Contract with America - Explained! |Phone: +44-81-500-3000
Contract: verb |Mail: ses@unc.edu
1) To shrink or reduce in size - the economy contracted +-----------------------
2) To become infected -My baby contracted pneumonia when they stopped my welfare