Re: Minor fixes to draft 05
Roy T. Fielding (fielding@liege.ICS.UCI.EDU)
Wed, 03 Jul 1996 15:14:21 -0700
> I have problems with a couple of these, since they appear to change
> the specification in ways that either might be inadvertent on your
> part, or haven't been discussed sufficiently.
Not inadvertent, but wrong is always a possibility.
> *** 3786,3793 **** [13.1.1 Cache Correctness, leftover from prior change]
> origin server is violated (see section 13.1.5 and 14.45).
> 4. It is an appropriate 304 (Not Modified), 305 (Proxy Redirect), or
> error (4xx or 5xx) response message.
> - and it is the most up-to-date response appropriate to the request the
> - cache has (see section 13.2.5, 13.2.6, and 13.12).
>
> If the cache can not communicate with the origin server, then a correct
> cache SHOULD respond as above if the response can be correctly served
>
> I'm not sure why you propose deleting those lines.
Because, when I rewrote that section after draft 04, I put those lines
in the first paragragh (i.e., look above it if you have the draft handy).
It was just a paste error when Jim applied the changes (no worries).
> *** 4022,4029 **** [13.2.3 Age Calculations, as requested by Ben Laurie]
> age of a response or cache entry.
>
> In this discussion, we use the term "now" to mean "the current value of
> ! the clock at the host performing the calculation." All HTTP
> ! implementations, but especially origin servers and caches, should use
> NTP [28] or some similar protocol to synchronize their clocks to a
> globally accurate time standard.
>
> --- 4019,4026 ----
> age of a response or cache entry.
>
> In this discussion, we use the term "now" to mean "the current value of
> ! the clock at the host performing the calculation." Internet hosts that
> ! use HTTP, particularly those hosting origin servers and caches, should use
> NTP [28] or some similar protocol to synchronize their clocks to a
> globally accurate time standard.
>
> If I understand your use of the word "Internet" (as a replacement for
> the word "all"), this change now means that intranet hosts (i.e., those
> not directly reachable from the Internet) are not expected to
> have synchronized clocks.
>
> I have no idea why anyone would want to run an intranet Web any
> differently than the Internet Web. But in any case, this is a
> "should", not a MUST, and so I see no reason to change the suggestion
> from the draft-05 version.
No. As Ben mentioned several times, it is ludicrous to suggest that
HTTP *implementations* should use NTP -- instead, the hosts running
HTTP should also run NTP. If you can rephrase that in a better way, fine,
but the existing text is just plain bogus.
> ***************
> --- 5929,5941 ----
>
> 14.16 Content-MD5
>
> ! The Content-MD5 entity-header field, as defined in RFC 1864 [23], is a
> ! MD5 digest of the entity-body for the purpose of providing an end-to-
>
> Minor nit: shouldn't this be "an MD5 digest" rather than "a MD5 digest"?
Fine by me (I only changed the commas in that one).
.....Roy