[cap-talk] How desirable / feasible is a persistent OCAP language?

Bill Frantz frantz at pwpconsult.com
Sun Jul 27 21:09:39 CDT 2008


The last time I discussed the upgrade of persistent objects,
<http://www.eros-os.org/pipermail/cap-talk/2008-April/010727.html>,
it was in the context of the KeyKOS/EROS/CapROS space bank. (The
comments may also apply to Coyotos.) I picked this example, because
I think it is the most difficult upgrade problem in the design of
these systems.

One observation is that when the long-term state is kept in a well
defined format, upgrade becomes easy. Examples include the KeyKOS
kernel (but not the objects it implements), Unix file system
implementations (those that only depend on pages and inodes), and
file editors (which depend on the format of files).

Alan Karp was discussing last Friday the point that the most
important thing that needs to be persisted in a capability system
is changes in the long-term capability relationships, and that
these changes can conceptionally be persisted by rewriting the ACLs
for the capabilities active entities receive when they authenticate
themselves to the system.

Cheers - Bill

-------------------------------------------------------------------------
Bill Frantz        | The first thing you need when  | Periwinkle
(408)356-8506      | using a perimeter defense is a | 16345 Englewood Ave
www.pwpconsult.com | perimeter.                     | Los Gatos, CA 95032


More information about the cap-talk mailing list