Re: "Hits" pragma

James Pitkow (pitkow@cc.gatech.edu)
Thu, 17 Aug 1995 14:42:55 -0400 (EDT)


Hello,

<95Aug16.095730pdt.2763@golden.parc.xerox.com> from "Larry Masinter" at Aug 16, 95 09:57:19 am
>
> There's other kinds of sampling, too. For example, you could
> selectively turn off caching for randomly assigned host addresses
> e.g., hash the host name, and day k of a n-day cycle, turn off caching
> for hosts whose hashed IP address mod n == k.
>

Cool idea, except for the non-unifrom distribution of IPs and the number
of users behind each IP. Furthermore, this distribution does indeed vary
from site to site, so analysis tools would have to generalize the distribution
(which is most likely a member of the gama family) and then paramtertize this
for each site to achieve a robust random sampling methodology - tricky stuff.

I liked you first idea more.

Jim.