Re: [media-types] [Rats] registration for +cwt

[Darrel Miller <[e-mail]>]

Michael,

Could you share what scenarios you envision for clients processing an application/eat+cwt as a cwt without understanding what an eat is?

It is my understanding that structured suffixes are not intended to convey instructions on how to parse a media type, but to provide a clients that are not familiar with the specific media type the ability to fall back to the more generic semantics of the suffix.  https://www.rfc-editor.org/rfc/rfc6839.html#section-2

I get the impression that primary goal of registering the set of media types https://www.ietf.org/archive/id/draft-ietf-rats-eat-media-type-01.txt is to provide decoding instructions rather than conveying additional semantics.  I wonder whether registering just "application/eat" and a "format" parameter might be a simpler solution.

Darrel

________________________________
From: media-types <[e-mail]> on behalf of Michael Richardson <[e-mail]>
Sent: Friday, April 7, 2023 9:17 PM
To: Roy Williams (COSINE) <[e-mail]>
Cc: Ned Smith <[e-mail]>; [e-mail] <[e-mail]>; [e-mail] <[e-mail]>; [e-mail] <[e-mail]>
Subject: Re: [media-types] [Rats] registration for +cwt


Roy Williams (COSINE) <[e-mail]> wrote:
    > The question I have is if CBOR and COSE are simply
    > content-transfer-encodings and not content-type.  This gets into a
    > nebulous area when discussing reasoning over volumes of data.

    > If all the content is signed it would show as homogenous data
    > application/cose and thus any filtering would require tunnelling into
    > the payload which could be expensive.

Don't forget Accept: headers.

CBOR/COSE are not CTE.
(Base64 is CTE)


--
Michael Richardson <[e-mail]>   . o O ( IPv6 IøT consulting )
           Sandelman Software Works Inc, Ottawa and Worldwide