[cap-talk] password capabilities & object capability model?

Mark Miller erights at gmail.com
Thu Aug 21 21:18:18 CDT 2008


On Thu, Aug 21, 2008 at 5:37 PM, Karp, Alan H <alan.karp at hp.com> wrote:
> MarkM wrote:
>> I consider an unguessable swiss number to be a simple form of crypto --
>> perhaps the simplest form.
>
> The only requirement for a swiss number is that it be unguessable.
> Cryptography is one way to generate such numbers, but there are others.  For
> example, there's a guy at PARC who uses a Lava lamp.  The numbers he
> generates are unguessable, but there's no cryptography involved.  I think
> the name should reflect some essential property, not the means of obtaining
> that property.

I don't care how the number are generated. A Lava lamp is silly but
fine. But the first part of the protocol of "tell me the secret number
and I'll give you the money" is comparison of large unguessable
numbers. To me, this is the simplest degenerate case of crypto. But
since everyone else finds this use of the term confusing, I'll stop
using it that way.


-- 
Text by me above is hereby placed in the public domain

 Cheers,
 --MarkM


More information about the cap-talk mailing list