application/pkix-attr-cert

[Russ Housley]

Bjoern:

>* Russ Housley wrote:
> >I suggest that we need a MIME type for attribute certificates that is
> >similar to the one already defined for public key certificates
> >(application/pkix-cert) and CRLs (application/pkix-crl).
>
>(I take it this is draft-ietf-pkix-attr-cert-mime-type-00.txt).
>
> >    To: ietf-types at iana.org
> >    Subject: Registration of MIME media type application/pkix-attr-cert
> >
> >    MIME media type name: application
> >
> >    MIME subtype name: pkix-attr-cert
> >
> >    Required parameters: None
> >
> >    Optional parameters: version (default value is "1")
>
>This needs a definition for the non-default values (syntax, semantics,
>perhaps implementation considerations, e.g., for unsupported versions).

I'm tempted to drop this.  The types defined in RFC 2585 use this 
construction, and we have never needed it.  Is there another type 
that does this and has good text to steal?

> >    Encoding considerations: will be none for 8-bit transports and most
> >    likely Base64 for SMTP or other 7-bit transports
>
>This needs to be one of the values defined in RFC 4288.

I'll do some homework.

> >    Security considerations: Carries an attribute certificate
>
>In the draft, a pointer to the security considerations section of the
>draft should probably be included here.

Will do.

> >    Interoperability considerations: None
> >
> >    Published specification: RFC 3281
> >
> >    Applications which use this media type: Any MIME-complaint transport
>
>This could be more specific. One or two additional sentences as to the
>nature of "attribute certificates" would be useful. Also note that the
>Abstract should have at least three lines.

Will do.

> >    Additional information:
> >      Magic number(s): None
> >      File extension(s): .AC
> >      Macintosh File Type Code(s): none
> >
> >    Person & email address to contact for further information:
> >    Russ Housley <housley at vigilsec.com>
> >
> >    Intended usage: COMMON
> >
> >    Author/Change controller:
> >    Russ Housley <housley at vigilsec.com>
>
>I believe the change controller needs to be a standards organisation;
>I believe the preference for RFC-defined types is "The IESG".

No problem.

Thanks for the prompt feedback.

Russ