Frank,

I am not sure I understand all you said. Of course, I am not talking about
*concurrent* access to the card from the two hypothetical applications. What
I believe would be most likely is that at any time the user is playing with
either application or the other, not both. So there would not be a
concurrency/sharing/consistency/arbitration problem. On the other hand,
having one API controlling access to the card seems too much of a
limitation. I do not think having a pkcs#11 DLL below OCF is a viable
option, at least today. By the way, while it is very useful that OCF
supports PC/SC (via the bridge DLL) it is instead a pity that the average
pkcs#11 library is not based on PC/SC too (if I am not mistaken: please
correct me).

Adriano

> -----Messaggio originale-----
> Da: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
> Inviato: mercoled� 29 dicembre 1999 13.18
> A: Santoni Adriano
> Cc: [EMAIL PROTECTED]
> Oggetto: Re: R: [OCF] RE: �OCFA OCF-based and PKCS11-based access to
> same data on the same card: possible?
> 
> 
> 
> 
> Adriano,
> the scenario that you described does make sense to me and is 
> something that we
> have discussed many times in our group. What you pose as a 
> more general question
> of interoperability does in deed not have an easy solution.  
> But it is easy to
> understand that there must be one single point in the system, 
> that finally
> controls and arbitrates the concurrent access to the same 
> card in the scenario
> that you describe.  So, it can be either ...
> ... OCF, if all applications are based directly or indirectly on OCF,
> ... a PKCS#11 access layer, if all applications are based on it,
> ... PC/SC if that particular implementation (probaby Windoze) 
> is using PKCS#11
> and OCF on PC/SC,
> ... your own self written access layer that takes over this 
> "traffic cop"
> function and which is used by all applications.  And if the 
> applications are
> using the same data, like in your example cryptographic keys, 
> they better use
> the same view of and access mechanism to them.
> 
> OCF can not do anything about card accesses that go around 
> it, as can not
> PKCS#11, as can not PC/SC as can not your or mine self 
> written access layer.
> All must go through one common channel in order to guarantee 
> concurrent access
> without interference.  In my view OCF and PKCS#11 are both 
> good candidates of
> being that one access controlling instance of choice, but 
> only one of both for
> the same card (in the case of PKCS#11 implemented based on 
> OCF this instance is
> OCF).
> 
> We have written a very sophisticated PKCS#11 access layer for 
> a filesystem card
> an for a JavaCard.  We know what optimizations we have done 
> internally to store
> the data safely and securely and performantly on the card.  
> We would not advise
> anybody to try to access the data on the card directly 
> without the control of
> the PKCS#11 access layer.  We would not do so if it were our 
> development project
> :-)
>          Frank Seliger
> IBM Pervasive Computing Division
> Schoenaicher Str. 220,    71032 Boeblingen,   Germany
> [EMAIL PROTECTED]                                            
>             Tel.
> +49-7031-16-3142
> 
> 


---
> Visit the OpenCard web site at http://www.opencard.org/ for more
> information on OpenCard---binaries, source code, documents.
> This list is being archived at http://www.opencard.org/archive/opencard/

! To unsubscribe from the [EMAIL PROTECTED] mailing list send an email
! to
!                           [EMAIL PROTECTED]
! containing the word
!                           unsubscribe 
! in the body.

Reply via email to