[secdir] review of draft-ietf-karp-threats-reqs-05

[Stephen Kent <[e-mail]>]

This is a re-review of this document.  I reviewed version -03 in August 
of 2011. I provided an extensive
set of comments and edits in  an effort to improve the readability of 
this doc.  Some of the edits were
accepted, but many others have been ignored. New text (several pages 
worth) has been added, which has not improved the overall quality of the 
document.

This document is very, very badly written. It includes made-up names for 
attacks, bad definitions, a messed-up terminology section, an 
inconsistent discussion of threats and attacks, and a set of 
"requirements" that are a mix of useful, vague, and silly statements. 
(One of my favorite examples is the definition of INTERFERENCE attacks, 
which begins by saying that "ADDING NOISE" is a type of INTERFERENCE 
attack. Since this does not appear to be a discussion taking  place in 
the RF context, this is not a helpful bullet! The extensive use of 
uppercase words
is also not much of an aid to readability.)

The threat/attack discussion is a hodgepodge; it gives the reader the 
sense that the topics that have been included are arbitrary, with no 
sense of a taxonomy or a comprehensive, consistent treatment of threats 
and attacks.

This document requires significant work to become an RFC that will be a 
useful guide for the KARP WG,
and not an embarrassment to the IETF.

An annotated, edited version of the doc is attached.

Steve