Persistence: a proposal

[Scott L. Burson]

   From: Patrick Logan <patrickl@servio.gemstone.com>
   Date: Thu, 22 May 97 13:07 PDT

   >>>>> "Scott" == Scott L Burson <gyro@zeta-soft.com> writes:

       Scott> I have some AI research I want to do which requires a
       Scott> high-performance persistent Lisp with firstclass
       Scott> continuations.

   Have you considered RScheme from Paul Wilson's OOPS group at Texas?

I didn't know about it.  I'll take a look.

   They are just starting to address performance issues, but you'd be
   taking advantage of their continuing efforts. They already have the
   continuations and the transparent persistence. Most likely they have
   better performance than ZetaBase for day-to-day Lisp programming.

Uhh, maybe.  I have designed ZetaBase from the beginning to support the kind
of work I want to do, although I think it has wide applicability.  E.g.,
although small commits are relatively expensive in ZetaBase, it makes
committing tens of megabytes of stuff at a time quite cheap, compared to many
of the designs I have seen.

-- Scott