W3C home > Mailing lists > Public > ietf-http-wg@w3.org > April to June 2021

Re: Genart last call review of draft-ietf-httpbis-cache-16

From: Mark Nottingham <mnot@mnot.net>
Date: Mon, 31 May 2021 12:07:13 +1000
Cc: gen-art@ietf.org, draft-ietf-httpbis-cache.all@ietf.org, ietf-http-wg@w3.org, last-call@ietf.org
Message-Id: <5FF6DC9E-1A66-4F75-92FD-0305E15342C3@mnot.net>
To: Mohit Sethi <mohit.m.sethi@ericsson.com>
Hi Mohit,

Thanks for the review. Responses below.

> On 31 May 2021, at 12:14 am, Mohit Sethi via Datatracker <noreply@ietf.org> wrote:

> - In the HTML version of the draft, the reference to [Semantics] does not work
> properly. I looked at the xml source which looks fine. I suspect it is
> something to do with the tooling.

Julian has covered this, I think.

> - It is not clear to me which draft is creating the "Hypertext Transfer
> Protocol (HTTP) Field Name Registry". It seems both this draft and
> draft-ietf-httpbis-semantics are creating it? Perhaps you could remove the text
> in this draft saying "introduce the new" and just ask IANA to update the
> registry with fields in Table 1 of this draft.

That text includes a reference to the Semantics text creating the registry, so I think it's clear which is doing the work. It's important to mention this because it's a fairly substantial change from IANA's standpoint.

> Nits/editorial comments:
> - Section 1: When does a client or server act as "tunnel"? I don't know if it
> is absolutely necessary to explain the term. You can decide.

This is defined in Semantics; I've added a reference in https://github.com/httpwg/http-core/commit/ee8c9306972

> - Section 1: HTTP caching's goal is significantly improving performance -> HTTP
> caching's goal is to significantly improve performance?

That introduces a split infinitive.

> - Section 1.3: Maybe it is obvious to many readers, but I was not sure what is
> meant by a "canned string"?

Fixed in https://github.com/httpwg/http-core/commit/51bfc119585

> - Section 3 vs Section 4: "A cache MUST NOT store a response to a request
> unless:" does not have a comma before unless while "When presented with a
> request, a cache MUST NOT reuse a stored response, unless:" has a comma before
> unless?


> - Some of the bullets in section 3 and 4 were hard to parse. Take for example:
> "When presented with a request, a cache MUST NOT reuse a stored response,
> unless: the stored response does not contain the no-cache cache directive
> (Section, unless it is successfully validated (Section 4.3), and". I
> am not sure how to simplify the text on all these requirements.

We've struggled with how to clearly state these requirements for some time now; the feedback we've received during and since RFC7234 indicates that this format is preferred. That said, the specific phrasing is undoubtedly not perfect, and we're open to suggestions for improvements.

Cheers and thanks again,

Mark Nottingham   https://www.mnot.net/
Received on Monday, 31 May 2021 02:07:47 UTC

This archive was generated by hypermail 2.4.0 : Monday, 31 May 2021 02:07:54 UTC