Details of Optimisation Table VM

Bill House bhouse@dazsi.com
Tue, 6 May 1997 15:27:02 -0700


John,

You mentioned that the underlying concept of the OTVM is patented. How are we
to base a freeware VM on patented technology?  I'm afraid that this issue would
have to be cleared up before we could move forward. I'd hate to get 90% down
the road and start receiving love letters from MS lawyers. Such issues are
easier to get around in ANSI/ISO agreements than they are under freeware
standards like the GPL (which specifically prohibits patents, if I'm not
mistaken).

Setting that aside for the moment, the basic idea sounds intriguing. We already
have a RISC-like instruction set (the AgentBase VM has 36 regular instructions)
that mimics a chip architecture pretty well. ABVM could form the basis of your
FIS. Although I don't think we really need an OT to be multilingual (we already
do that using ABVM alone), the extra layer may provide infrastructure for
things like secured modes of operations, etc., that ABVM does not currently
take into account. Hmmm...

This is interesting, as the new Samsung "system-on-a-chip" could probably be
burned with ABVM in hardware. That might speed things up a bit. <g>

Bill House
--
http://www.dazsi.com
The views expressed are mine alone,
unless you agree with me.