[Cryptography] New White Paper: GhostLine - Information-Theoretically Secure Multi-Party Chat

[Tony Patti]

>  Kent Borg wrote Monday, September 8, 2025 4:15 PM:
>
> Because I can't put a number on how much xor-ing in any one RNG helps the result, some would say means that RNG should not be xor-ed in. 

Hi Kent,

As part of my five volumes of _Cryptosystems_Journal_ 
I designed and built and published electronics which I named "RANGER Devices" (RANGER = RANdom number GEneratoR)
which were composed of 16 crystal oscillators (in a 4x4 matrix) 
running asynchronously from the PC, and connected via a parallel port cable.

The interesting thing about crystal oscillators is that they have 60/40 waveform symmetry,
so you would assume they would be bad as a source of random bits.

BUT the key (pun intended) is that the crystal oscillators operated asynchronously from the PC.
AND (TO KENT's EXCELLENT QUESTION): 
XORing 4 of the crystal oscillators (in hardware) was VERY GOOD, 
XORing 16 of the crystal oscillators was EXCELLENT.

Tony Patti
crypto at glassblower.info
(ARPAnet NIC IDENT "TP4")