Re: [Sidrops] Adam Roach's No Objection on draft-ietf-sidrops-lta-use-cases-05: (with COMMENT)

[Randy Bush <[e-mail]>]

>> Please consider adding the boilerplate specified in RFC 8174.
> 
> Or, alternatively (and my preference), re-word that brief paragraph in
> the Security Considerations so that it doesn't use "MUST".  I find
> "Hence they MUST be implemented to assure the local constraint." hard
> to understand anyway, so re-wording might help.

as mirja kühlewind pointed out, that paragraph was a bungle.  how about

6.  Security Considerations

   Though the above use cases are all constrained to local contexts,
   they violate the model of a single Global RPKI, albeit to meet real
   operational needs.  Hence the result must be able to be validated as
   if the changed data were part of the validatable Global RPKI while
   including the local context, perhaps with the addition of trust
   anchors or authenticatable patching of trust.

   Modification 'recipes' may lack authentication.  E.g., if
   modifications to the tree are passed around a la SLURM files, see
   [RFC8416], what was object security becomes, at best, transport
   security, or authentication by other trust domains such as PGP.

randy