[Cryptography] Montana: A Post-Quantum Blockchain with Time as Scarcity

[Renato Schiavinato Lopez]

On Wed, May 20, 2026 at 4:28 PM Ron Garret <ron at flownet.com> wrote:

> With cryptocurrencies, optimists assume that large blocks which haven’t
> moved in a long time are associated with lost keys, which will not be found
> or brute-forced.  I imagine optimistic assessments which have held true for
> a long time could, likewise, go out the window rather quickly if the
> status-quo changes.
>
> There is one big difference: the only way to actually destroy fiat
> currency is to destroy physical tokens, and those scale *physically* with
> the amount of money they represent.  It is actually quite hard to lose a
> million dollars.  I'd be surprised if it has ever happened.
>
> By way of very stark contrast, crypto keys do not scale physically with
> the amount of money they represent.  Keys representing a significant
> fraction of the world economy can be stored in a device literally the size
> of your little finger.  It is really easy to lose a Ledger Nano S PLus.
>
> Also, in practice, the requirement to keep keys secret mitigates against
> backing them up unencrypted, and so losing access to keys can happen if
> someone simply forgets the pass phrase used to encrypt them, which can also
> happen very easily.
>
> rg
>

Ron,

You hit the nail on the head regarding the physical asymmetry of crypto
security.

This exact friction is what motivated a design I’ve been working on, which
I shared on this list last week: Schiavinato Sharing (BIP39-Native
Threshold Backup over GF(2053)).

The goal is to solve the exact problem you described: how to split a BIP39
mnemonic into physical, unencrypted threshold shares (so there is no
passphrase to forget), while ensuring that recovery does not depend on the
survival of any specific software stack, hardware vendor, or operating
system.

By instantiating Shamir directly over GF(2053) - the smallest prime field
containing the 2048 BIP39 word indices - the arithmetic remains
human-executable. If the original software tool disappears in 20 years, an
auditor, lawyer, or heir can realistically reconstruct the seed using
durable paper artifacts, a basic calculator, and modular arithmetic, in
just 30-60 minutes.

It bridges the gap between the physical scaling security of traditional
assets and the cryptographic sovereignty of digital keys, without forcing
the user to choose between an exposed plaintext seed and a fragile digital
encryption layer.

For those interested in how the linear consistency layer and fault
detection are handled under this model, the specification is open:
https://github.com/GRIFORTIS/schiavinato-sharing

Best regards,
Renato Schiavinato Lopez
GRIFORTIS
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://www.metzdowd.com/pipermail/cryptography/attachments/20260520/8d3743b2/attachment.htm>