Re: IANA Media Types

Hi,

From:  "Dournaee, Blake" <bdournaee@rsasecurity.com>
Message-ID:  <E7B6CB80230AD31185AD0008C7EBC4D202A1B612@exrsa01.rsa.com>
To:  "'reagle@w3.org'" <reagle@w3.org>
Cc:  XML Encryption WG <xml-encryption@w3.org>
Date:  Mon, 29 Oct 2001 15:52:25 -0800

>Hello All,
>
>I am pondering the choice of the media-type tree referent in [1], Section
>4.3
>
>Does anyone know why the media type tree is hosted at www.isi.edu (or if
>this isn't "the" tree, why isn't it at www.iana.org, or something similar).
>I am just wondering about the history of this. Further, these media type
>designations are the same ones referred to in the DSig spec as used in the
>"MimeType" attribute for the <Object> element. The values suggested here are
>the bare media types names without the fully qualified URI shown in the XML
>Encryption specification.

IANA and media types predate the dominance of the web. Jon Postel who
was IANA was at ISI, so when web pages were first created, they were
put there. It makes no sense to try to nail this stuff down at
isi.edu. They are also refereneable via iana.org but, in fact, I'm not
aware of any commitment by IANA to maintain the current accidental
structure of these web pages.

This sort of problem, where a field is being set to URI (or Content-Type)
syntax and it is necessary to shoe horn the other syntax (a Content-Type
or URI respectively) into it is EXACTLY why I wrote the draft whose
current version is at

<ftp://ftp.ietf.org/internet-drafts/draft-eastlake-cturi-02.txt>

>This seems odd, is there a compelling reason to have these be different
>between the specifications? With anticipated use of both of these
>technologies together, this just seems weird.

Actually, URIs and content types can be syntactically distinguished,
so the field could be their union, but that has its own complexities.
Or you could have two different attributes, MIMEtype and URItype or
something, but that also has its own complexities.  Or MIME types
could be registered for Element and Content, although I guess since
this is in the W3C it just has to be URI syntax, and that does provide
consistency and a lot of flexibility...

>[1] http://www.isi.edu/in-notes/iana/assignments/media-types/media-types

>Blake Dournaee
>Toolkit Applications Engineer
>RSA Security
> 
>"The only thing I know is that I know nothing" - Socrates

Thanks,
Donald

Received on Tuesday, 30 October 2001 12:20:41 UTC