Hello,

If you look at the Musclecard Protocol definition - there is a figure 
which shows the 35,000 foot picture of the architecture:

Applications
CDSA / PKCS-11
MuscleCard framework
PC/SC
Drivers
Cryptographic Device

in that order.  Like the network stack it is necessary not to have too 
large of a plug-in stack.  If each card had to have it's own PKCS-11 
module, it would be difficult to ensure interoperability and quite 
cumbersome to write CSP's for cards.  Consider MuscleCard framework for 
smartcard specific applications, and one layer below PKCS-11 or CDSA.

Applications need the PKCS-11 or CDSA architecture to 
manage things like CA's, certificate handling, etc.  The MuscleCard API is 
well capable of handling cards like the Finnish ID PKCS-15 card with the 
right plug-in.  Even if it wasn't it could still fit at the PKCS-11 layer.  
At the Musclecard layer you are more worried about containers than 
certificates,  Certs are stored as objects, keys are unwrapped and stored 
as keys, pins as pins, etc.

Dave

***************************************************************
Unix Smart Card Developers - M.U.S.C.L.E.
(Movement for the Use of Smart Cards in a Linux Environment)
http://www.linuxnet.com/
To unsubscribe send an email to [EMAIL PROTECTED] with
unsubscribe sclinux
***************************************************************

Reply via email to