Ben Laurie <ben(_at_)algroup(_dot_)co(_dot_)uk> writes:
Peter Gutmann wrote:
Ian G <iang(_at_)xxxxxxxxxxxxx> writes:
Ben Laurie wrote:
Obviously the I-D should be updated to reflect this (and clearly no-one
has ever implemented v3 keys from it).
(I'm not qualified to rewrite the draft on that issue.)
Why not just include the pseudocode (or more usefully straight C code) for
the operation? That's the ultimate definitive description, any programmer
can understand it, and you don't run into the problems you do with the
english description.
(Even if an implementer doesn't understand it, they can always just cut &
paste the code).
Isn't there some dumb IETF rule against that?
I doubt it, many RFCs containing nontrivial algorithms (specifically ones
where the easiest way to explain them is with code) already include full
implementations, e.g. the MDx RFCs and the Punycode RFC.
Peter.
(Completely off-topic, but while I'm mentioning Punycode does anyone know what
the *real* story behind this was? It seems to be the Intercal of character-
set representations... does the background explanation by any chance begin
with "Some mathematicians got drunk and made a bet..."? :-).