A better approach to security

[Ian G <iang AT cacert.org>]

On 6/09/10 1:34 PM, Mario Lipinski wrote:
> Hi,
>
> Am 25.05.2010 02:38, schrieb Ian G:
> > And, it is a BirdShack decision to not provide geoloc in any future
> > rewrite of the code.
>
> Imho this applies to the business logic. For the website this should be
> discussed again. But will the website be critical?


Hmm, ok.  I am guessing the website will at least be sensitive because 
the BL has to take its word that it is authenticating the user.

Well, ok, that assumes client cert access, which the website handles. 
If we stuck with passwords, then the website could prove some form of 
authentication across to BL.

(I was assuming that client cert would be the major access, and 
passwords would only be used for account recovery.  But it does rather 
bring in a gap in authentication.)

iang

Attachment: smime.p7s
Description: S/MIME Cryptographic Signature