New document on "Simple hit-metering for HTTP"

At the Montreal IETF, Paul Leach and I volunteered to write up a draft
of our design for a simple hit-metering mechanism for HTTP.  We said
we'd have it done by the end of July; Larry Masinter said "does that
mean August 1?", and I said "no, it means July 31."

So (with a few hours to spare), we managed to get something done that
we think should work fairly well.

Our goal was NOT to solve the general problem of collecting demographic
information; it was to reduce the incentive for origin servers to
defeat caching merely so that they could collect simple hit-counts, of
a sort that the caches could just as easily collect for them.  However,
we believe that our design solves a large chunk of the problem, without
significant complexity in the protocol or the implementations.

I believe that the design we described originated with me; however,
Paul was the one who had to explain to me how and why my design
actually worked.  He also did all of the word-processing and a lot of
the drafting.  The two of us have generated at least seven drafts
before subjecting the rest of you to it.

You can find a copy at

    http://ftp.digital.com/~mogul/draft-ietf-http-hit-metering-00.txt

We invite constructive criticism of the draft; we don't expect it to be
flawless.  But we'd appreciate that critics recognize that we are not
trying to solve every aspect of the demographics problem, but are
proposing a design that does a decent job with minimal complexity.

-Jeff

Received on Wednesday, 31 July 1996 17:17:54 UTC