Meadhbh,
A major use case that's lacking in the standards you mention is support for
security domain separation. In high assurance environments, a crypto device
typically separates two security domains (protected & unprotected), performing
cryptographic operations at the boundary. The existing standards operate within
a single domain-- they enable a crypto device to accept data to be
cryptographically transformed and return the results directly to the caller who
initiated the operation.
This type of high assurance use case is common with diplomatic and military
communications and shows opportunity in sensitive commercial enterprises like
banking, as well. Specialized cryptographic hardware designed specifically for
these types of environments is available. So, we're not talking about tokens
or smart cards, although potential solutions (like the one we've proposed) can
accommodate those use cases as well. Nonetheless, you'd probably be better off
using something like Cryptoki for those situations.
For an introduction to this problem domain, see section 1.4 in our
Internet-Draft:
http://tools.ietf.org/html/draft-lanz-cicm-00#section-1.4
If you're interested, but can't attend the bar BoF, you can still participate
via the mailing list:
https://www.ietf.org/mailman/listinfo/cicm
Regards,
Lev
From: Meadhbh Hamrick [mailto:ohmeadhbh(_at_)gmail(_dot_)com]
Sent: Wednesday, March 10, 2010 11:59 AM
To: Novikov, Lev
Cc: IETF-Announce; IETF-77
Subject: Re: [77attendees] High Assurance Cryptographic API Bar BoF at IETF 77
in Anaheim, CA
It's going to be tough for me to attend, but I would love to hear why cryptoki,
pc/sc or cdsa don't work for your use cases. Also, are you talking about a
specific kind of token? Soft tokens? 4758s? NCipher boxen? Smart cards?
Is there a resource on the web describing the problem domain, or is this an
informal meeting to discuss problems with existing solutions or new use cases?
-cheers
-meadhbh
On Mar 10, 2010 7:28 AM, "Novikov, Lev"
<lnovikov(_at_)mitre(_dot_)org<mailto:lnovikov(_at_)mitre(_dot_)org>> wrote:
There will be a bar BoF on high assurance cryptographic APIs during IETF 77.
When: Tue, March 23, 2010 from 17:30 - 19:00 (tentatively)
Where: Mix Restaurant (in the Hilton Anaheim)
Menu available at:
http://www1.hilton.com/en_US/hi/hotel/SNAAHHH-Hilton-Anaheim-California/dining.do#1
Commercial cryptographic APIs (e.g., PKCS#11) are not suitable for high
assurance cryptographic devices that have special considerations (e.g.,
security domain separation, rich module & key management). We'd like to discuss
the problems surrounding creating a logical model for high assurance
cryptographic devices and a potential solution we've published as an
Internet-Draft:
https://datatracker.ietf.org/drafts/draft-lanz-cicm/
Interested parties are invited to join our IETF mailing list for potential
updates on the bar BoF and/or other matters relating to high assurance
cryptographic APIs:
https://www.ietf.org/mailman/listinfo/cicm
Thanks,
Lev Novikov
The MITRE Corporation
_______________________________________________
77attendees mailing list
77attendees(_at_)ietf(_dot_)org<mailto:77attendees(_at_)ietf(_dot_)org>
https://www.ietf.org/mailman/listinfo/77attendees
________________________________
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf