Andrew Martin <[email protected]> writes:

> I can appreciate the security-based argument against relying on IP, but
> at the same time I think there are good arguments for the off-campus /
> on-campus configuration.

Oh, I do too, don't get me wrong.  I just don't want to change our
advertised *level of assurance* based on that data point.  The LoA is
there for applications that need a specific LoA for compliance purposes,
and the LoA numbers that Stanford is advertising are defined by a NIST
standard.  I don't think that the IP location is sufficiently secure data
to amount to a second factor for LoA purposes.

I certainly understand the more general use case of only requiring
multifactor for off-campus accesses.  I think it's a good way to ease into
using multifactor.

> I can empathize with Russ that the configuration options get complicated
> when you try to make webauth easily support all of these options and
> wish the team good luck with that :-) However - we'll be waiting for an
> elegant solution. In the meantime, in case you're interested, with
> Fletcher's help we put together a rewrite hack which seems to be working
> as follows:

Yup, that looks like the right idea to me.

-- 
Russ Allbery <[email protected]>
Technical Lead, ITS Infrastructure Delivery Group, Stanford University

Reply via email to