RE: HTTP 1.1 issue 02: 3.11 Entity Tags

>  Constructive comments are encouraged.

Do you think it would be useful to separate the proposal for
allowing an entity tags with a greater uniqueness scope?

For other purposes, tying the entity tag uniqueness scope to
the delta coding header wouldn't be as useful.

 - Allowing naively simplistic implementations of delta encoding.

  I suggest you remove "naively"; those who might value
  simplicity of implementation over some of your
  other goals may not be "naive" but just operating
  under other constraints.

# based on an entirely false analogy between MIME and HTTP...
# It is too late to fix the terminological failure in the HTTP/1.1
# specification

I don't think that the dictionary definition of "entity"
is any more relevant than the dictionary definition of "protocol".
Terminology always fails to meet our expectations for it.

I don't think that the relationship between MIME and HTTP is
an 'analogy', nor is it 'entirely false'. Your grumpy complaint of
the word "entity" here doesn't really advance your cause.
And it wasn't "too late" to fix a "terminological failure"
in January 1998 (the date of your Internet Draft.)
Rather, there was rough consensus that it was not a "failure".

I suggest that you continue to explain how the concepts of
"entity" in HTTP and MIME differ, as it is a useful discussion,
but keep the introduction of "instance" neutral:
"to avoid confusion, we define a new term in this memo:"

Now that we're about to issue the HTTP/1.1 DS as an RFC, you might
revise the reference [9] and the tense of "has been proposed".

Do you mean this to be 'on the table' as a proposed standard?

Larry

Received on Tuesday, 10 November 1998 11:24:14 UTC