W3C home > Mailing lists > Public > w3c-ietf-xmldsig@w3.org > July to September 1999

RE: Multiple signatures in a sig_block

From: Richard D. Brown <rdbrown@Globeset.com>
Date: Wed, 21 Jul 1999 13:41:45 -0500
To: "'Chris Smithies'" <Chris_Smithies@penop.com>, <w3c-ietf-xmldsig@w3.org>
Message-ID: <009701bed3a8$afe24280$0bc0010a@artemis.globeset.com>
Chris,


If deemed necessary, grouping of signature elements can still be specified
by the application DTD. I would rather keep the XMLDSIG specification
(syntax and procedures) and its associated data model as simple as possible
as long as it does not preclude any option at the application level.

On the other hand, I do not share your point of view WRT the benefit from an
evidentiary standpoint. Quite systematically, attachment of multiple
signatures to a document raises the problem of validity if at least one
signature is deemed valid while some cannot be verified or are deemed
invalid. I think that there are two scenario where multiple signatures would
be applied by a same person, for a same intent, and a same content:

1- plurality of recipients having different trust or key management
requirements: In such circumstances, it is probable that a given recipient
will not be able to verify the signatures intented for others. Therefore,
there is no evidentiary benefits since the recipient cannot assess the
validity of the signature stack.

2- application/regulation requires that multiple 'marks' be attached to the
document: A real world example would be applying a manual signature on top
of the organization stamp (though one could argue that in this case there
are two different signers). Such requirement could make sense only if all
the marks can be verified by the recipients. In such circumstances, grouping
certainly helps evidence that all the marks have been attached. But such a
requirement seems very much application specific and I would rather let the
application deal with it. To some extents, this reminds me about
counter-signature, which is already left to the application layer (mostly an
authorization issue).

Sincerely,

Richard D. Brown
Software Architect - R&D
Globeset, Inc. Austin, TX - U.S.


> -----Original Message-----
> From: w3c-ietf-xmldsig-request@w3.org
> [mailto:w3c-ietf-xmldsig-request@w3.org]On Behalf Of Chris Smithies
> Sent: Wednesday, July 21, 1999 8:59 AM
> To: w3c-ietf-xmldsig@w3.org
> Subject: Multiple signatures in a sig_block
>
>
>
>
> The value of allowing multiple Signature elements in a
> SigBlock (sig_block,
> signatureBlock, whatever...) is that in the future, multiple signing
> technologies may well be used in the course of the same
> transaction, with
> the same intent, by the same person, governing the same
> content. It may
> also be that different signature standards are required by different
> recipients of the document. It would clearly be efficient and
> useful from a
> procedural and evidentiary point of view for these signatures to be
> combined together into a single entity.
>
Received on Wednesday, 21 July 1999 14:42:10 GMT

This archive was generated by hypermail 2.2.0 + w3c-0.29 : Thursday, 13 January 2005 12:10:07 GMT