Re: Very best practice for RSA key generation

Phillip Hallam-Baker <[email protected]> Mon, 28 Oct 2019 12:52:26 -0400
Newsgroups gmane.comp.encryption.general
Message-ID <CAMm+LwgQK9FO3n4ofLXN0-1jWMsEZ=2Gepv7deap8dFsZokesQ@mail.gmail.com>
On Sun, Oct 27, 2019 at 6:35 PM Bill Stewart <[email protected]> wrote:

> On 10/17/2019 12:23 PM, Phillip Hallam-Baker wrote:
> > The basic idea is that a user has a key which expressed in Base32 looks
> > like this:
> >
> > ZAAA-UJUY-H7TF-SFLK-CWAW-TKC4-O5HQ
>
> This is a UI suggestion, not a cryptographic one, but in addition to
> supporting keys like that, it might be useful to also allow keys
> to be input in "Correct Horse Battery Staple" dictionary-word form.
>

That is already a suggestion in the Internet Draft but not one that has
been implemented yet.

The basic idea is to take dictionaries of 32,768 or so words for common
languages, form a Merkle tree out of them and print the apexes out. These
can then be used in combination with a simple Web Service that simply
extracts the proof chain for specific words.

With 65K words (feasible for English), that means 16bits per word. A 2^128
work factor requires ten words including the overhead.

_______________________________________________
The cryptography mailing list
[email protected]
https://www.metzdowd.com/mailman/listinfo/cryptography