Re: [jose] #103: Rename "alg" to avoid future collision problems

["jose issue tracker" <[e-mail]>]

#103: Rename "alg" to avoid future collision problems

Changes (by [e-mail]):

 * status:  new => closed
 * resolution:   => Postponed


Comment:

 While I still think it is a mistake not to do the rename, I will bow to
 the pressure of the group in terms of the apparent problems that they have
 with respect to the backwards compatibility problems of doing this.  It
 would have made things much cleaner in the future and for enforcement of
 what algorithms can go into what fields.  This has been a problem that has
 been recurrent for CMS when dealing with signature algorithms as the
 restrictions were not cleanly stated using 1988 ASN.1

 It will be possible to do the necessary logic of doing key management for
 MAC in a future document by the following:

 A new element in the signature headers is defined called recipients.  This
 element will contain a JSON array which in turn will contain a JSON object
 that corresponds to a recipient element from the JWE specification.  This
 will break stupid parsers that look for the alg string without doing the
 JSON parse first in order to determine if it is a JWS or JWE element as it
 might encounter the JWE alg element before the JWS alg element.  That will
 be considered to be an acceptable failure mode for this change.

-- 
-------------------------+-------------------------------------------------
 Reporter:               |       Owner:  draft-ietf-jose-json-web-
  [e-mail] |  [e-mail]
     Type:  defect       |      Status:  closed
 Priority:  major        |   Milestone:
Component:  json-web-    |     Version:
  signature              |  Resolution:  Postponed
 Severity:  -            |
 Keywords:               |
-------------------------+-------------------------------------------------

Ticket URL: <http://trac.tools.ietf.org/wg/jose/trac/ticket/103#comment:2>
jose <http://tools.ietf.org/jose/>