- From: Daniel DuBois <ddubois@spyglass.com>
- Date: Tue, 02 Apr 1996 16:18:46 -0600
- To: mcmanus@nysernet.org, http-caching@pa.dec.com
At 09:17 AM 3/14/96 -0500, Patrick McManus wrote:
>- -> Specifically, I'd propose using:
>- -> Cache-control: max-uses=NNNN
>- -> on responses, where NNNN is a positive integer.
>
>To summarize, we're talking about a token system, where the cache is
>given N tokens to use however it sees fit (1 token to serve a page to
>a client, or it can redistribute its tokens to other caches).
>[...]
>I am not a demographics expert, but it would seem to me that a sample
>size of 1 in 8 or 1 in 12 is more than sufficient for statistical
>purposes, and the bandwidth and server load savings are tremendous.
It is true that this token system does not give the origin server the abilty
to know actual hit counts though, correct? I mean, just because a server
received a subsequent request from Proxy A on a URI Proxy A had previously
been given 10 tokens/uses for, doesn't mean that proxy has served 10 copies
of that URI response. Maybe it just served one, and then the 2nd request
came after the max-age.
I don't really care about demographics, I'm just curious if this fact is a
concern for those who do. In other words, will this solution keep people
who care about hit counts from using cache-busting or 'validate-every-time'
techniques. In yet other words, are there
1a) people who care about hit counts, and
1b) people that don't, or are there
2a) people who care about hit counts,
2b) people who care a little bit but not enough to break this scheme, and
2c) people that don't.
If the world represents #1, this scheme might not have enough/any value.
-----
the Programmer formerly known as Dan
http://www.spyglass.com/~ddubois/
New direct dial phone: 708-245-6577
Received on Tuesday, 2 April 1996 23:02:42 UTC