[E-Lang] A more critical question (was: an example impatience policy )

Karp, Alan alan_karp@hp.com
Thu, 22 Mar 2001 12:06:49 -0800


Zooko wrote:
> 
> However what we, and everyone else, need is a distributed, 
> attack-resistant,
> scalable mutable namespace, which seems like a difficult 
> problem and perhaps
> not even feasible.
> 

We think we built one with e-speak.  E-speak Beta 2.2 is built on the
concept that all names are local to a specific e-speak client (process) with
name mappings held in the core (TCB) address space.  There is no name
guessing attack, because the name is only meaningful to a single party, and
there are no name mapping attacks because the core requires explicit
authority to remap.  The e-speak product has relaxed this approach a bit by
introducing global references to these mappings, but the mappings themselves
are always used for resource access.

_________________________
Alan Karp
Principal Scientist
Decision Technology Department
Hewlett-Packard Laboratories MS 1U-2
1501 Page Mill Road
Palo Alto, CA 94304
(650) 857-3967, fax (650) 857-6278
https://ecardfile.com/id/Alan_Karp
http://www.hpl.hp.com/personal/Alan_Karp/