ietf-dkim
[Top] [All Lists]

Re: [ietf-dkim] Splitting the DKIM base doc

2006-03-25 12:11:04


Barry Leiba wrote:
I think the work of splitting it will be small, and won't affect the schedule,


Given what Eric said he planned to do, with respect to the base document's discussion and reference to the key service, I agree.

If I understood correctly the change he is planning will a) contain only and exactly what is essential to specify the base document's functional requirement, and therefore b) de-couple a key service specification from the critical path of the base document.

Here is a revision to the architecture diagram that went into the threats document. It attempts to identify the standardized components and their inputs and output. I think it is sufficient for the current discussion, to decide which parts need to be in separate documents.

For reducing dependencies in the critical path, a specification of interface parameters, rather than for a specific document, ought to suffice.

The diagram is still quite rough and I suspect it contains errors. So, please send me corrections and enhancement.



                                           |
                                           | - RFC2822 Message
                                           V
                         +=======================================+
                         |      ORIGINATING OR RELAYING ADMD     |
                         |                                       |
   ......................+  Canonicalize  (Header, Body, c-alg)  |
   .         ............+  Hash (c-Header, c-Body, h-alg)       |
   .         .        ...+  Sign (h-Header, h-body,              |
   .         .        .  |       Domain, Selector, Priv-Key)     |
   .         .        .  |                                       |
   .         .        .  +=================+=====================+
   .         .        V                    | - Message
   .         .   +------------------+      |   (Domain, Selector, Key)
   .         .   | Canonicalization |      |
   .         .   |    Algorithm     |      |
   .         .   +----+-------------+      |
   .         V        .                    |
   .  +-----------+   .                    |
   .  |   Hash    |   .                    |
   .  | Algorithm |   .                [Internet]
   .  +------+----+   .                    |
   V         .        .                    |
+---------+  .        .                    |
|   Key   |  .        .                    |
| Service |  .        .                    |
+--+------+  .        .                    |
   .         .        .                    V
   .         .        .  +=======================================+
   .         .        .  |      RELAYING OR DELIVERING ADMD      |
   .         .        .  |                                       |
   .         .        ..>|  Canonicalize (Header, Body, c-alg)   |
   .         ...........>|  Hash (c-Header, c-Body, h-alg)       |
   .....................>|  Verify (h-Header, h-body,            |
                         |          Domain, Selector, Pub-Key)   |
                         |                                       |
                         +=======================================+
                                           |  - Verified Domain
                                           V  - [Report]
+-----------+            +=======================================+
|  Signer   |            |                                       |
| Practises +...........>|         Signer Evaluation             |
|  Query    |            |                                       |
+-----------+            +=======================================+


--

Dave Crocker
Brandenburg InternetWorking
<http://bbiw.net>
_______________________________________________
NOTE WELL: This list operates according to http://mipassoc.org/dkim/ietf-list-rules.html