Re: [jose] Proposed resolution of header criticality issue

I’ll add some cheers — this does look like substantial progress.

I assume the fields such as “epk”, “apu” etc that sometimes must be understood, and at other times must be ignored (depending on “alg” or “enc” value) would NOT be listed in the “crit” field as they are defined in the “base specs”.

Being in the “base specs” is the right criteria for whether a field should be listed in “crit” as long as “base specs” means: “base specifications for the particular “alg”/”enc” values”. It shouldn’t mean (and doesn’t have to mean) the base spec for the whole JOSE system.

P.S. “meta” might be a nicer label than “asd”.

--
James Manger

From: jose-bounces@ietf.org [mailto:jose-bounces@ietf.org] On Behalf Of Tim Bray
Sent: Tuesday, 12 March 2013 12:43 PM
To: Karen ODonoghue
Cc: jose
Subject: Re: [jose] Proposed resolution of header criticality issue

Cue wild cheers from the peanut gallery where non-cryptographers sit.  MustIgnore is infinitely more robust and open-ended than MustUnderstand.  -T

On Mon, Mar 11, 2013 at 5:31 PM, Karen O'Donoghue <odonoghue@isoc.org<mailto:odonoghue@isoc.org>> wrote:

Folks,

A side meeting was held Sunday with a number of jose working group participants to try to resolve the open issue related to header criticality. The following are the proposed resolutions from the meeting. Point 5 of the proposed resolution below is actually independent of the other 4 points, and could be considered separately. This will all be discussed in Wednesday's meeting.

In addition to the text below, there was some agreement to replace the "understand" text with something a bit more explicit like "must process". However, that text has not been rolled into the summary text below yet.

Thank you to Jim Schaad, Mike Jones, John Bradley, Nat Sakimura, Martin Thomas, Eric Rescorla, Matt Miller, and Richard Barnes for your efforts (and my apologies if I missed someone).

Regards,
Karen

1:  Change the language “Additional members MAY be present in the JWK.  If present, they MUST be understood by implementations using them.” to “Additional members MAY be present in the JWK.  If not understood by implementations encountering them, they MUST be ignored.”  (And make the same change for JWK Set as well.)

2:  Characterize all existing JWS and JWE header fields as either must be understood or may be ignored.  “alg”, “enc”, and “zip” must be understood.  “kid”, “x5u”, “x5c”, “x5t”, “jwk”, “jku”, “typ”, and “cty” can be ignored because while not using them may result in the inability to process some signatures or encrypted content, this will not result in a security violation – just degraded functionality.  Other fields such as “epk”, “apu”, “apv”, “epu”, and “epv” must be understood and used when “alg” or “enc” values requiring them are used, and otherwise they may be ignored.

3.  Define a new header field that lists which additional fields not defined in the base specifications must be understood and acted upon when present.  For instance, an expiration-time extension field could be marked as must-be-understood-and-acted-upon.  One possible name for this would be “crit” (critical).  An example use, along with a hypothetical “exp” (expiration-time) field is:

  {"alg":"ES256",
   "crit":["exp"],
   "exp”:1363284000
  }

4.  All additional header fields not defined in the base specifications and not contained in the “crit” list MUST be ignored if not understood.

5.  Define a new header field “asd” (application-specific data) whose value is a JSON structure whose contents are opaque to and ignored by JWS and JWE implementations but for which its contents MUST be provided to applications using JWS or JWE, provided that the signature/MAC validation or decryption operation succeeds.  The intended use of this field is to have a standard place to provide application-specific metadata about the payload or plaintext.

_______________________________________________
jose mailing list
jose@ietf.org<mailto:jose@ietf.org>
https://www.ietf.org/mailman/listinfo/jose