- From: Scott Lawrence <lawrence@agranat.com>
- Date: Mon, 22 Sep 1997 14:51:07 -0400
- To: HTTP Working Group <http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com>
So far, the response to my survey questions has been underwhelming.
I am hoping that this is because there was not a sufficiently
concrete proposal. If it was because you just haven't gotten to it,
feel free to do so. In any case, I'd like to put forward a somewhat
more structured idea in hopes that it will generate some more
interest. I don't consider any part of the following to be cast in
stone; please suggest improvements.
================
Each Thursday from 16 October 1997 through 20 November 1997
inclusive will be a Test Day; anyone who plans to participate must
send mail by 23:55 GMT on the preceding Tuesday to a central
registrar. The registrar will forward the information by email on
Wednesday to all who have signed up. Each participant provides:
Organization:
User-Agent or Server string: (may be approximate)
HTTP Role: [origin, proxy, tunnel, client, robot, ...]
HTTP Version:
Address: (DNS host names and/or IP addresses for the HTTP
implementation)
Time: (GMT times the system will be active or available)
Contact Name: (person to contact with an issue)
Contact Email:
Contact Phone:
Contact Hours: (GMT times the contact is available, should overlap
the span in Time, but may be a subset)
Notes: other relevant information, possibly including URLs for
information on where to find background material.
A single organization may register as many participating systems as
they need, but should register separately for different roles or
substantially different configurations so that other participants
can distinguish them easily (for example, my company will probably
make available one origin server that is using a clock and one that
is not).
Note that HTTP Version is included; it will be useful to have 1.0
(and even 0.9) implementations participating in all roles,
especially if they are instrumented to provide traces.
Some ground rules:
- Participating systems may be configured to allow access only by
announced participants for that week.
- Participants will, on request, make a reasonable effort to
provide whatever relevant log or trace data they have to other
participants to resolve problems.
- If a problem or possible issue with another participant
implementation is found, that issue will be communicated to the
contact for that implementation promptly. Only if the
involved participants disagree on the correct behavior or if
the involved participants agree to do so will the issue be
brought to the working group mailing list.
- Any other disclosure of problems found with other
implementations during these tests is poor form.
- Communicating positive results to other participants is
strongly encouraged.
--
Scott Lawrence EmWeb Embedded Server <lawrence@agranat.com>
Agranat Systems, Inc. Engineering http://www.agranat.com/
Received on Monday, 22 September 1997 12:29:45 UTC