[2834] in cryptography@c2.net mail archive
RE: "damn the bitmaps..."
daemon@ATHENA.MIT.EDU (Trei, Peter)
Wed Jun 24 16:23:14 1998
From: "Trei, Peter" <ptrei@securitydynamics.com>
To: "'perry@piermont.com'" <perry@piermont.com>,
"'cryptography@c2.net'"
<cryptography@c2.net>
Date: Wed, 24 Jun 1998 15:55:13 -0400
Perhaps NIST could send you a printed copy. The
problems appear to be due to aliasing problems
while scanning. I've been able to figure out most
of them by comparing e's and c's on the same row.
While they are individually ambiguous, when there
are two more-or-less distinct populations of
ambiguous blobs, you can make intelligent guesses
as to which is which.
Peter
Of course, what we really need is a reference
implementation, which would be sadly unexportable.
> From: Perry E. Metzger [SMTP:perry@piermont.com]
>
> I'm currently attempting to type in the F permutation table from the
> Skipjack spec. Unfortunately, thanks to the High Quality Bitmapped
> Document, it is horribly difficult to distinguish the "e"s from "c"s,
> and unfortunately the authors used lower case!
>
> Does anyone know how to get better copy of the F table?
>
> Perry
