ietf-smime
[Top] [All Lists]

RE: Countersignature Attribute

1998-09-15 14:35:46
I don't know about other products, but Outlook and OE tooke the approach of
ignoring countersignatures.
 
jim
 

-----Original Message-----
From: Russ Housley [mailto:housley(_at_)spyrus(_dot_)com]
Sent: Tuesday, September 15, 1998 1:40 PM
To: jsp(_at_)jgvandyke(_dot_)com
Cc: pgut001(_at_)cs(_dot_)aucKland(_dot_)ac(_dot_)nz; 
ietf-smime(_at_)imc(_dot_)org
Subject: Re: Countersignature Attribute


John:

Prior to getting this note, I took a stab at revisions to the same section
of text.  I do not know how S/MIME v2 implementations handle
countersignatures, but I took the opposite approach that you did.  I was
trying to ensure that current implemntations would all conform.

Here is what I wrote:


A countersignature attribute can have multiple attribute values.  The syntax
is defined as a SET OF AttributeValue, and there must be one or more
instances of AttributeValue present.



The UnsignedAttributes syntax is defined as a SET OF Attributes.  The
UnsignedAttributes in a signerInfo may include multiple instances of the
countersignature attribute.



The fact that a countersignature is computed on a signature value means that
the countersigning process need not know the original content input to the
signing process.  This might have efficiency advantages, but it also has
security disadvantages.  Therefore, countersigners must validate the
signature value prior to signing it.  This validation requires processing of
the original content.



A countersignature, since it has type SignerInfo, can itself contain a
countersignature attribute.  Thus it is possible to construct arbitrarily
long series of countersignatures.





Russ


At 11:56 AM 9/15/98 -0400, John Pawling wrote:
Peter (and friends),

I agree with your recommendation.  This results in a change to my comment
to
CMS-06 to read as follows:

5) Sec 11.4, Countersignature: Please change as follows:

OLD: "A countersignature attribute can have multiple attribute values."

NEW: "The UnsignedAttributes syntax is defined as a SET OF Attribute.  
The UnsignedAttributes in a signerInfo MUST NOT include multiple 
instances of the countersignature attribute.  The Attribute syntax defines 
attrValues as a SET OF AttributeValue.  A countersignature attribute 
MAY include one or more instances of AttributeValue.  There MUST NOT 
be zero instances of AttributeValue present in the attrValues SET
OF AttributeValue."

Does anybody disagree with this recommendation???

- John Pawling




<Prev in Thread] Current Thread [Next in Thread>