[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <200408201852.i7KIqIRM011759@turing-police.cc.vt.edu>
From: Valdis.Kletnieks at vt.edu (Valdis.Kletnieks@...edu)
Subject: RE: MS should re-write code with security in mind
On Fri, 20 Aug 2004 12:23:35 EDT, Barry Fitzgerald said:
> An interesting cost benefit analysis of this would be to take the amount
> of bandwidth increase if people used encrypted/authenticated
> p.s. I'm not sure where to start to get valid numbers on this. Every
> scenario I've been able to think of in the time it took to write this
> e-mail has major methodological flaws.
Estimating the overhead is pretty easy.
As a straw man, take the size of the PGP signature on this message as a fairly
fixed overhead for one style of "authenticated". For "encrypted", there's a
good chance the message will actually end up *smaller*, because most crypto
packages compress before signing, in order to (a) make it faster (as the
compression is probably quicker than signing) and (b) more secure (the reason
English text compresses so well is because there's lots of entropy -
compressing first gets rid of a lot of the entropy).
For another datapoint, consider the following DomainKey header I saw in a mail
from an early adopter:
DomainKey-Signature: a=rsa-sha1; s=mail; d=resistor.net; c=simple; q=; b=Hm PldyQdhZsXA12xUJ0oHscqlYYfF/E/H2T1MowOryfJnLfIIZGGUjYvSGMo2rFo
That will also probably be a "usually fairly fixed size overhead" tag.
The hard part is coming up with a good estimate of what % of spam will
be properly authenticated once spammers start using the cached credentials
already present on the PC in order to send out their spam....
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 226 bytes
Desc: not available
Url : http://lists.grok.org.uk/pipermail/full-disclosure/attachments/20040820/ad161962/attachment.bin
Powered by blists - more mailing lists