#175: Section 5.2. Message Decryption

 A.  the MUST in the last sentence of paragraph one ignores multiple
 recipients.

 B.  Step 1 - remove the parsing on compact serialization - belongs in
 section 7.1

 C.  Step 2 - MUST is not really required.  Also what happens if somebody
 puts in the padding characters?  Is this really a killer?

 D.  Step 4 - This step should be discussed in the data model and not here.

 F.  Step 5 - need a positive action in the event that the MUST NOT is
 violated.  Re-write to MUST and be checked and then die

 G.  Step 5 - the second sentence duplicates what should be in the data
 model

 H.  Step 6 - this step can be deleted.  It amounts to a NOP

 I.  Step 8 - problem with this and the multi-recipient case.  'If the key
 for this recipient item is not known, then continue processing the next
 recipient.  If there are no recipients remaining then fail processing of
 the JWE message."

 J. Step 9 - if you are doing this, it is better to keep things separate
 and not combine direct and non-direct methods in a single step.  It is
 confusing.  Additionally, all of these algorithms produce a CEK when you
 have finished processing according to the algorithm definition - it is
 just a question of how the algorithm itself works.  Algorithm details
 should not be included here unless they are critical and very unique.
 Thus direct makes sense to treat separately.

 K.  Step 10 - The note on multiple recipients is buried to deep to be
 found.  It should be deleted from here and make at a high level.

 L.  Step 10 - I don't understand what you are trying to accomplish with
 the 'To mitigate ...' statement.  Are you really worried about timing
 attacks at this point?  if so then you need to make sure that you have
 discussed all of them (you haven't) and we are now talking about errors
 returned to the application protocol which has never been discussed
 before.  This should be a separate section

 K.  Step 11 - don't start with otherwise

 L.  Step 11 - format of encrypted key value is algorithm specific not JWE
 specific.

 M.  Step 13 - should be moved up as it affects error processing

 N.  Step 14 - Remove ASCII

 O.  Step 14 - Talk about the aad value in terms of the data model and not
 the serialization

-- 
-------------------------+-------------------------------------------------
 Reporter:               |      Owner:  draft-ietf-jose-json-web-
  [email protected] |  [email protected]
     Type:  defect       |     Status:  new
 Priority:  Editorial    |  Milestone:
Component:  json-web-    |    Version:
  encryption             |   Keywords:
 Severity:  -            |
-------------------------+-------------------------------------------------

Ticket URL: <http://trac.tools.ietf.org/wg/jose/trac/ticket/175>
jose <http://tools.ietf.org/jose/>

_______________________________________________
jose mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/jose

Reply via email to