[coyotos-dev] IDL code point selection

[Jonathan S. Shapiro]

On Wed, 2007-07-11 at 09:14 -0700, Charles Landau wrote:
> At 10:57 AM -0400 7/11/07, Jonathan S. Shapiro wrote:
> >3. I am not at all clear what you are trying to achieve here. Is there a
> >flaw in the "cryptographic hash of fully qualified name" scheme that you
> >can articulate?
> 
> You said it:
> >  >  > At 10:46 AM -0400 6/19/07, Jonathan S. Shapiro wrote:
> >  >  > >Cryptographic code points are regrettable because they cannot be
> >  >>  >efficiently demultiplexed.

The only code space that can be demultiplexed really efficiently is a
sequential assignment of some form. The problem with this is that there
is no automatic assignment strategy that is both sequential **and stable
over edits**.

While addition/deletion of methods definitely does break compatibility,
we don't want to completely break binaries that may be slow to update.
This requires a stable assignment scheme. Unfortunately, alphabetic
sorts don't qualify.

shap
-- 
Jonathan S. Shapiro, Ph.D.
Managing Director
The EROS Group, LLC