Issue 321, proposed resolution

[<Frederick.Hirsch@nokia.com>]

Dana and SwA/MIME experts:

Issue 321:  clarity of the cipher text encoding for attachments 

Draft 9 of the SwA profile [1] has changes intended to address and close
this issue, using approach #3 in the mail below. The cipherdata is
explicitly indicated to be base64 encoded. Changes are at lines 376,
382, 413, 432, 497 in the pdf diff.

The earlier text treated the cipher data as binary octet stream, relying
on Content-Transfer-Encoding and attempting to reduce processing steps
for encoding - an alternative resolution might be to make that more
explicit and obvious in the wording. 

Can you please review draft 9 and indicate whether the changes are
appropriate (base64 encoding), whether instead a binary octet stream
should be specified, and whether use of Content-Transfer-Encoding
requires discussion. (Specific text proposals would be helpful.)

I would like to be sure I correctly captured the intent of earlier
feedback and close this issue properly.

Thanks

Regards, Frederick 

Frederick Hirsch
Nokia

[1]
http://www.oasis-open.org/apps/org/workgroup/wss/download.php/9038/wss-s
wa-profile-1.0-draft-09-diff.pdf

-----Original Message-----
From: ext Dana Kaufman [mailto:dkaufman@forumsys.com] 
Sent: Monday, August 30, 2004 3:39 PM
To: Hirsch Frederick (Nokia-TP/Boston); wss@lists.oasis-open.org
Subject: Comment on wss-swa-profile 1.0 draft .8

Here is some feedback based on draft .8:

It is not clear as to how to interpret/transform the encrypted contents
of the attachment.  There are three incompatible possibilities that
different implementers might conclude from the specification:

1) Possibly the encoding of the encrypted data is inferred from the
Content-Transfer-Encoding, i.e. binary means raw bytes and base64 means
base64 encoded bytes.

2) The encoding of the encrypted data is inferred from the SwA transform
specified in the specification.  The specification does not seem to
indicate that the SwA transform does base64 decoding, but the example
implies that it does.

3) The encoding is explicitly specified though transforms in the
CipherReference, as is the norm in XML Encryption.  base64 decoding
would be done with an explicit transform following the SwA transform.

Look like the approach to use needs to be clarified.

Dana S. Kaufman
VP of Product Management
Forum Systems, Inc.
Tel: (781) 788-4232
E-Mail: dkaufman@forumsys.com
Visit http://www.forumsys.com