Re: WG Review: Source Address Validation in Intra-domain and Inter-domain Networks (savnet)

[Stephen Farrell <stephen.farrell@xxxxxxxxx>]

Hi Fred,

On 03/06/2022 21:44, Fred Baker wrote:
> I guess I would like to understand what privacy-promoting mechanisms
> you would find acceptable. 

My concern is that such a thing might not exist, so I can't
give you a positive answer I'm afraid. (Where "thing" is a
privacy-friendly SAVI-like Internet-scale something - if I'm
reading the charter so badly that nothing SAVI-like is
envisaged then perhaps that indicates other problems with the
charter text.)

> For example, I could imagine some form of
> nonce or address-hiding mechanism known only by the communicating
> parties, perhaps exchanged during some form of encrypted call setup
> protocol.

That's why I asked about research. If it could be done I'd
bet some academic has published on the topic. (It's not my
field so I've not looked, hence just asking.)

Cheers,
S.

> Sent using a machine that autocorrects in interesting ways...
>
> > On Jun 3, 2022, at 1:26 PM, Adrian Farrel <adrian@xxxxxxxxxxxx>
> > wrote:
> >
> > Hey Alvaro and Stephen,
> >
> > > > I oppose the creation of this working group on the basis that
> > > > it makes no mention of privacy. Extending the kind of
> > > > privacy-unfriendly source address validation mechanisms 
> > > > (unwisely IMO) used, to something deployed at Internet-scale, 
> > > > could be a major error.
> > >
> > > The WG won't be chartered to extend existing mechanisms.
> > >
> > > If there's text that gives that impression we should fix it.
> >
> > Weeeell, I read...
> >
> > | The "Source Address Validation in Intra-domain and Inter-domain
> > Networks | (SAVNET)" working group will define routing
> > protocol-independent architectures | and procedures to accurately
> > determine the valid incoming router interfaces | for specific
> > source prefixes.  The accuracy of the enhancements is expected | to
> > improve upon current SAV mechanisms.
> >
> > ...to mean that procedures and enhancements would be defined.
> >
> > Actually, I interpreted the whole charter as "examine existing
> > approaches and develop new techniques" and read it in that light.
> > Maybe the charter could be clearer up front that no new mechanisms
> > or extensions to existing mechanisms will be defined.
> >
> > What am I missing?
> >
> > Cheers, Adrian

Attachment: OpenPGP_0x5AB2FAF17B172BEA.asc
Description: OpenPGP public key

Attachment: OpenPGP_signature
Description: OpenPGP digital signature