begin quoting David Brown as of Sat, Aug 30, 2008 at 04:40:57PM -0700:
> On Sat, Aug 30, 2008 at 03:54:04PM -0700, SJS wrote:
>
> >Ah, that's an artifact of your notation. If you use a *reasonable*
> >notation, then the displayed ordering will match the actual ordering:
>
> Depends on your notion of "reasonable". Most people would expect to
> enter the passwords in the order they are listed on the sheet, in
Which is NOT the order they are computed in, which is where I made a
stupid assumption. My apologies.
> which case each password is the hash of the one following it on the
> list, which is exactly what I said in the beginning.
Please don't use "following" or "preceding" in your explanations without
providing concrete examples.
Diagrams would not go amiss.
> The important distinction is that the hashes are entered in the
> opposite order they are computed in. That was the novelty of the
> invention when S/Key was created.
Hm...
Okay. I see, I think.
We compute a sequence of hashes, and print them out in reverse order,
except for the last one, which we put on the server.
The user then starts on the top of the (reversed) list, and enters the hash
there. The system hashes that value once, and checks against the stored
value; if it matches, the old stored value is discarded, and the user's
input is made the new stored value.
User's list Stored value
H(H(H(s))) H(H(H(H(s))))
H(H(s)) H(H(H(s)))
H(s) H(H(s))
Even looking at that, I have a hard time saying "the following hash was
computed first" -- I keep applying "following" to the act of computation.
--
Next time I'll look up an algorithm in two different places.
Stewart Stremler
--
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
