W3C home > Mailing lists > Public > public-awwsw@w3.org > June 2009

Re: Back to HTTP semantics

From: Jonathan Rees <jar@creativecommons.org>
Date: Fri, 12 Jun 2009 14:47:46 -0400
Message-ID: <760bcb2a0906121147v43e8dbc0yf1a5e5acc6b0da87@mail.gmail.com>
To: Pat Hayes <phayes@ihmc.us>
Cc: AWWSW TF <public-awwsw@w3.org>
On Fri, Jun 12, 2009 at 2:00 PM, Pat Hayes<phayes@ihmc.us> wrote:
> Apologies for ignorance, but are HTTP responses timestamped in any way? Can
> one tell by looking that something has been cached for a long time since it
> was created? Because if not, I don't see how this can help much.

Yes. The information you get from a GET/200 response is that the
entity continuously corresponds to the resource in an interval. One
endpoint is in the past (Last-modified:), the other is in the future
(Expires:), and the present is in between (Date:). Neat, huh? And if
you know the server's clock is off by 3 minutes, you can correct for
that...

But even without that, it might be helpful to record whatever you know
about correspondences through time. You could make plots, do diffs,
and so on. Text mining on the entities corresponding to the resource
identified by http://news.google.com/ might be pretty interesting.

Jonathan
Received on Friday, 12 June 2009 18:48:21 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 12 June 2009 18:48:21 GMT