Re: [apps-discuss] Last Call: <draft-ietf-appsawg-http-forwarded-06.txt

On Tue, 10 Jul 2012 08:43:43 -0700
SM <sm(_at_)resistor(_dot_)net> wrote:
> > > In Section 6.3:
> > >
> > >    'To distinguish the obfuscated identifier from other identifiers,
> > >     it MUST have a leading underscore "_".'
> > >
> > > I suggest removing the requirement and using "can".  The implementer 
> > > can decide what to put in that field.  
> At 04:28 10-07-2012, Andreas Petersson wrote:
> > I think that will make parsing harder, and give no benefit at all.
> It allows for more random bits of information.
(adding context again)


Hi,

Can you substantiate a bit on this statement?
How is it "random bits of information" when the specifications says
that it MUST be underscore?

As far as I can think of, the only thing that it will tell is that the
implementation is following this specification.
So, on the contrary; the more "degrees of freedom" that is given to the
implementation, the easier it would be to do fingerprinting.


Cheers,
 Andreas

signature.asc
Description: PGP signature