Document: draft-groves-eccsi-00.txt
Reviewer: Francis Dupont
Review Date: 20110113
IETF LC End Date: 20110118
IESG Telechat date: unknown

Summary: Ready

Major issues: Almost Ready

Minor issues:
 - The signature length in 3.3 is 2N+4 when computation and examples
  give 4N+1?
 - There is no justification that if s' is too big to fit within
  and N-octet integer then q - s' is always small enough.
  (BTW I know the justification so it works but currently the
   specification is a bit distressing to implement :-)

Nits/editorial comments:
 - 1 page 3: e.g. -> e.g.,

 - 1 page 3: minimised -> minimized

 - 2 page 4: KPA -> KMS (in "that there is one KPA, one user" ...)

 - 3.3 page 6: possible typo: 2N+4 -> 4N+1

 - 4.1 pages 6 and 7: 'B' should be listed here

 - 4.1 page 7: standardised -> standardized

 - 6 page 11 (twice): authorised -> authorized

 - 7.2 page 13: KPAK second line needs re-indent

 - 7.2 page 15: I can check Sig length is N+N+(2N+1)=4N+1

 - Acknowledgement -> Acknowledgment (i.e., American spelling please :-)

Regards

Francis.Dupont@fdupont.fr

PS: my dictionary doesn't like 'affine' which seems correct for me
(BTW draft-mcgrew-fundamental-ecc-04.txt uses the word 'affine' too).