----- Original Message -----
From: "Peter Gutmann" <[email protected]>
To: <[email protected]>
Sent: Thursday, October 09, 2014 3:03 AM
> Ralph Holz <[email protected]> writes:
>
> >As an addendum to my last mail, I would like to add that TLS itself
does not
> >have a threat model.
>
> Very few crypto/security standards do.  DNSSEC has one, but that was
tacked on
> well after the RFCs were written.
>
> The other way of looking at it is that crypto/security standards all
(well,
> almost all) have the same threat model, which I refer to in my book as
the
> Inside-Out Threat Model: Whatever this standard happens to defend
against is
> defined to be the threat.

or look at security standards as being the defense against some threat.
I have been involved several times in adding security to an existing
application protocol and the Security AD has started by asking what
threats the application wishes to defend against.  After which it is
possible to evaluate TLS, e.g., as suitable - or not.  So here it should
be a question of what threats face IMAP, SMTP, HTTP and so on and, at
least in the case of the last, it then depends on what is happening
about HTTP, buying a theatre ticket with a credit card v transferring
the funds to buy a house.

Tom Petch







>
> Peter.
> _______________________________________________
> Uta mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/uta
>

_______________________________________________
Uta mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/uta

Reply via email to