Re: Paper & pencil password algorithm

WTShaw <lurens1@xxxxxxxxx> wrote:

The problem with bits is that there are only two of them. Using groups
of them does not answer the fact that binary is fundamentally a weak
number system or you would not be needing to be so absorbed in
patchwork fixes.

It "fine-tuning" not patchwork fixes. :-)

Base 10 digits are also weak. Depending, certain
higher bases are much stronger and can be rated from very good
to very bad.

Please can you give an example of a very good base and a very bad one
and explain why one is good and the other bad in the context of this

In stream ciphers with binary lineage, the same thing goes. With
higher sets, the assumption that characters will be in some sort of
usual order is a prejudice or that all inputs will have the same set
and order. There are many more possibilities and saying "Oh, that
would be easy," does not mean that it is. It's just politics to
disregard the value of deranged sets of characters of selected
qualifying sizes used in mechanisms that are strange to those with a
limited and/or one-sided of crypto.

I'm sorry to say that I find your writing poetic but very hard to
understand. There could be something of great value in that paragraph,
but I am unable to see it. Would you clarify your meaning please?

It's time for all here to admit that cryptography is a huge field and
we should encourage its growth is all avenues.

Absolutely right! Long live crypto!

James Taylor