I guess I should be concrete. Suppose the instruction set is 6502.
Below, if we start at the first position, A9, the A register would be loaded with the value A0, then there would be relative branch. However, if we start executing at the second, A0, the Y register would be loaded with 10 and then there
would be subroutine call. One could imagine modeling the effects with a constraint solver to embed two distinct programs in the same byte sequence. Neither would be encrypted, but one of them wouldn’t be visible without changing the readers reference frame.
I don’t think a reverse engineer would spot it just from a disassembly.
A9 A0 10 20 10
From: Friam <[hidden email]> on behalf of Jon Zingale <[hidden email]>
Reply-To: The Friday Morning Applied Complexity Coffee Group <[hidden email]>
Date: Friday, May 29, 2020 at 9:23 PM
To: "[hidden email]" <[hidden email]>
Subject: Re: [FRIAM] Optimizing for maximal serendipity or how Alan Turing misdirected ALife
Marcus,
you write:
The experience of being out-of-phase with a conversation has the same gist.
You summarized much of my experience with Friam. Can you say more about
how it is
like homomorphic encryption, but in plain sight? There is a sense that homomorphic encryption (relative to the privacy discussion) is in plain sight
(public key), so I am guessing you have something different in mind.
Jon
| Free forum by Nabble | Edit this page |