Re: Proposed solution to the "predetermined states" problem
Peter McGoron 23 May 2026 16:59 UTC
On 5/23/26 12:28, Wolfgang Corcoran-Mathe wrote:
> You're right that the solution I proposed would give you only 255
> predetermined states, one for each non-zero constant binary port.
> In practice, I doubt anyone will need that many states. (Remember,
> this is*only* relevant if you can't / won't save port states to
> disk. In the absence of that restriction, you can have as many saved
> states as your storage will allow.) Still, it's an ugly fact that
> makes the constant-port approach look a bit hacky.
Another way would to have a constant bytevector port. That allows more
than 255 predetermined states and one could fill it with, say, 256 bytes
of random data from /dev/random and just keep it in the source file.
-- Peter McGoron