Threat Model for BGP Path Security
draft-ietf-sidr-bgpsec-threats-09

Section 8. Really? No-one provided useful or notable individual input? 
Sigh.

Thanks for addressing my comments.

I support the publication of this document and I only have two quick points...

1. I agree with Barry's DISCUSS point on providing a definition of PATHSEC in the document rather than referencing the WG charter.

2. It may be worthwhile to mention in section 3 that "inaccurate" is a subjective term when discussing network operators' views of a BGP Update.  Given the flexibility noted in section 4 about local policy, it is difficult to externally judge whether the result of a network operator's action is inaccurate or simply a change in local policy.

The distinctions in the threat characterized seems somewhat arbitrary, and appear to have  conflated motivation with methodology. e.g. hacker crimnal isp nation-state and so on seem somewhat arbirary categories, nation states might easily for example employ the methodologoies of the other(s) and vice-versa.

Thanks for the discussion. I've made my previous discuss
point a comment that you can handle as you choose. I do
still think that editing the draft to reduce the number of
references to the charter (and adding a URL for the 
version you mean if you keep some) would be a fine 
improvement. 

-- old discuss

I'm not sure that arguing that some residual threat ought
not be addressed because its not defined in an RFC or
because of the current charter is really ok. Shouldn't the
underlying technical reasons be what's presented in this
document?

-- old comments

- section 3: nations might want to censor (or get others
to do that on their behalf)

- section 4, intro: shouldn't we now also consider passive
wiretapping, at least to the extent that visible BGP
traffic might assist in setting that up more efficiently?
While that might not translate into a requirement for
confidentiality, I think it may be worth considering as
part of the attack.