HTTP/1.1 revision 6 concerns/issues

Jim Gettys told me I should address these items to this mailing list for
possible additiion to your errata list.  There are five items I noticed,
of varying significance to the spec.

1. (technical-ish)

The "TE" header should be added to the list of hop-by-hop headers in
section 13.5.1.  This is implied by the Connection-header protection
mandated in 14.39 (the TE definition), but for completeness it should be
present in both places.

2. (editorial)

Section 8.2.3, first sentance: "an client" should be "a client"

3. (editorial, significant)

Section 8.2.4 refers to section 8.2.3 for guidelines on when a client
should retry a request.  The section 8.2.3 it refers to was deleted in
revision 6; 8.2.3 in Rev6 discusses the 100 (Continue) Status.

4. (editorial)

Section 8.2.3, the second bullet under "Requirements for HTTP/1.1 
proxies", I assume this requirement is subject to the same conditions as
the first bullet, but this could be made clear.

5. (technical?)

Section 12.2, paragraph 1 refers to the specification reserving the header
name "Alternates", but this header only appears in RFC2068 and in the
backward compatibility section (19.6.3) where it is mentioned in passing.
The suggestion that "Alternates" is a viable way to select the best
representation when Alternates is (presumably) deprecated and (clearly)
not defined in the current specification may be out-of-place.  (If nothing
else, a reference to RFC2068 would be appropriate since the only other
mention of Alternates in this draft is in 19.6.3 which says "we removed

Thanks for your attention,
Your lives aren't small, but    \\ Adam Davenport Bradley,  Grad Student
you're living them in a small    \\ Boston University   Computer Science
way. Live openly and expansively! \\  353-8921/MCS211
II Cor 6:12-13 (The Message)  <><  \\

Received on Friday, 29 January 1999 09:49:11 UTC