Re: [ietf-types] Request for review: application/evd+xml media type

Bjoern Hoehrmann <derhoermi@gmx.net> Thu, 07 July 2011 15:13 UTC

From: Bjoern Hoehrmann <derhoermi@gmx.net>
To: Yves Lafon <ylafon@w3.org>
Cc: ietf-types@ietf.org, public-ws-resource-access-comments@w3.org, Bob Freund <bob.freund@hitachisoftware.com>, ietf-xml-mime@imc.org
Subject: Re: [ietf-types] Request for review: application/evd+xml media type
Date: Thu, 07 Jul 2011 17:13:32 +0200
Message-ID: <r7ib17pq6s3iji57q4fnqpj11dnmit7hqm@hive.bjoern.hoehrmann.de>
References: <alpine.DEB.1.10.1107070940400.4139@wnl.j3.bet>
In-Reply-To: <alpine.DEB.1.10.1107070940400.4139@wnl.j3.bet>
MIME-Version: 1.0
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 8bit
Sender: owner-ietf-xml-mime@mail.imc.org
Precedence: bulk

* Yves Lafon wrote:
>Here is the registration document for the application/evd+xml media 
>type [1] by the W3C Web Services Resource Access Working Group [2].

Has this been submitted before? It seems to me that the document has had
Candidate Recommendation status for a couple of months, and W3C policy
is to submit on entering Last Call, but there is nothing in my archives.

><<<
>This appendix defines the 'application/evd+xml' media type which can be 
>used to describe EventDescription documents serialized as XML.
>
>MIME media type name:

You are using an outdated template, the current template is in RFC 4288.
The current templates structures and names some fields differently.

>Security considerations:
>
>     none

Well, at the very least you should point out that the ones in RFC 3023
are likely to apply to implementations of the specific type here aswell.

>Interoperability considerations:
>
>     There are non known interoperability issues.

This looks like a typo.

>     Fragment identifiers:
>     An EventDescriptions fragment identifier references a particular Event 
>Type within an EventDescriptions document. The Event Type referenced is 
>the Event Type with the @id attribute whos value equals that of the 
>fragment identifier component.

In addition to the typo, this seems insufficient and I am not sure how
compatible this is other types and what some people would like to change
the existing specifications to. I note in particular that you don't say
whether for instance "#x" would match id=' x ' (note the spaces); I also
note that you don't define the `id` attribute as ID attribute and don't
say how this works for actual ID attributes or magic ones like xml:id.

>     Base URI:
>     As specified in [RFC 3023] section 6.

The only reason to include this that I can think of is to mislead people
into thinking RFC 3023 has anything to say on that even though it does
not. If you don't want to say anything about Base URIs, then don't.
-- 
Björn Höhrmann · mailto:bjoern@hoehrmann.de · http://bjoern.hoehrmann.de
Am Badedeich 7 · Telefon: +49(0)160/4415681 · http://www.bjoernsworld.de
25899 Dagebüll · PGP Pub. KeyID: 0xA4357E78 · http://www.websitedev.de/

Re: [ietf-types] Request for review: application/… Bjoern Hoehrmann