[Cryptography] Possible reason why password usage rules are such a mess

[Phillip Hallam-Baker]

On Thu, Mar 5, 2020 at 2:50 PM Radia Perlman <radiaperlman at gmail.com> wrote:

>
>
> On Wed, Mar 4, 2020 at 9:58 AM Peter Gutmann <pgut001 at cs.auckland.ac.nz>
> wrote:
>
>> There has been some speculation in the past over why we have so many
>> cargo-
>> cult password security rules that make no sense in any modern context, the
>> prime example being the need to change passwords periodically.
>>
>
> I've never heard a good technical explanation for requiring periodic
> password changes, but wouldn't all the arguments about why it's silly to
> require frequent password changes apply to requiring certificate renewals?
> (and while we're at it, though I don't want to distract from the "why must
> certificates be periodically renewed" question...why does my driver's
> license, which proves who I am, not work for getting on an airplane if the
> license is expired...I can understand if they won't let me fly the plane
> with an expired driver's license, but I'm just planning on being a
> passenger.)
>

Yes and no.

The WebPKI is not just an authentication scheme, it conflates
authentication and authorization. CAs can and do revoke certificates for
breaches of faith. This is particularly common for code signing. Domain
Validation and free certs have pretty much eliminated the accountability
the WebPKI was designed to provide.

So the reason for expiry is to cap the window during which certificate
status is reported so that it is possible to revoke a cert for a host whose
key was compromised or has been used in violation of TOS.

Now also remember that in 1995 when we were putting this together, we were
using RSA1024 and RC4. We were not at all confident in the work factor of
any of our cryptographic algorithms. Today we are very confident in the
security of X.448 and AES256.

In the Mesh I have taken a totally different approach. I don't have
automatic expiry of assertions. I separate authentication and authorization
and I use a blockchain-ish approach to deal with changes of authorization
status.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://www.metzdowd.com/pipermail/cryptography/attachments/20200305/f53675b8/attachment.htm>