[babel] Shepherd's review of draft-ietf-babel-hmac-02

[Donald Eastlake <[e-mail]>]

Section 1.2, Bottom of page 3.
"robust random number generator" -> "robust random number [RFC4086]
generator". Could also add the RFC 4086 reference at other occurrences
of "random" but reference should be included at least once.

Section 1.2, Top of page 4, third bullet item.
This is a little confusing to read. I assume that the point about the
packet being accepted by node A is that it's a valid packet (has a
good HMAC TLV). But I think the average reading will be wondering
"What is this node A and why isn't it mentioned again later?" Not sure
if there is a reasonable wording change that improves this...

Section 5.2:
The diagram is odd. I would recommend sliding the Index field and
preceding vertical bar character to the left so it starts aligned
with the Type. Also, the PC is fixed length, which is not shown in the
figure - I think you really have to at least mention the length of the
PC in this section. Perhaps best to change the figure to something
like the following:
    0                   1                   2                   3
    0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |     Type      |    Length     |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |                             PC                                |
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   |            Index...
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-
or
   +-----------+
   |    Type   |         1 byte
   +-----------+
   |   Length  |         1 byte
   +-----------+-----------+-----------+-----------+
   |                      PC                       |  4 bytes
   +-----------+-----------+-----------+-----------+
   |            Index...                              variable
   +-----------+-----------+-----------+-----...

While it is reasonable, for implementation convenience, that there is
a maximum size for Index, perhaps for cryptographic strength, there
should be a minimum length, maybe 12 bytes?

Thinking some more about PC, seems like you either (1) need to say
that circular arithmetic is used [rfc1982] (bad idea in my opinion),
or (2) say what happens when PC hits all 1's (presumably adopt a new
Index, probably a better idea). It might be useful to also state that
PC is an "unsigned integer" or the like.

Section 7.
Should say IANA is "requested" rather than "instructed". (Or, after
the number are allocated, should be re-worded to indicated that the
IANA actions have been completed.)

Appendix B.2.
Heading should end with "-hmac-01", not "-hman-00".

Thanks,
Donald
===============================
 Donald E. Eastlake 3rd   [phone] (cell)
 1424 Pro Shop Court, Davenport, FL 33896 USA
 [e-mail]