[mpls] Review of draft-ietf-mpls-1stnibble-06

[Donald Eastlake <[e-mail]>]

Hi,

I tried to look at this draft with fresh eyes. I think it is technically
sound and a good idea.

Probably my strongest comment is that, while this draft has a Security
Considerations section, it seems quite weak. How about something more like
the following two paragraphs:
      This document creates a new IANA registry for and specifies changes
to the treatment in the data plane of packets based on the first nibble of
data beyond the MPLS label stack. One intent of this is to reduce or
eliminate errors in determining whether a packet being transported by MPLS
is IP or not. While such errors have primarily caused unbalanced and thus
inefficient multi-pathing, they have a potential to cause more serious
security problems.
      For general MPLS label stack security considerations, see [RFC3032].


More minor comments:

Last line before the 1.1 heading: "determined on the PFN" -> "determined
based on the PFN"

Section 2.1.1, point 2, unbalanced parentheses.

Section 2.2, "depend" => "dependent"

Section 2.2: The quoted paragraph from RFCC 4928, its replacement
paragraph, and the OLD TEXT and NEW TEXT below that should all be indented.
I looked at the XML and there do seem to be indent="..." attributes which
has worked for me in the past so I'm not sure what the problem is. Suggest
trying <blockquote>...</blockquote>

Section 2.2: also, it seems like the two substitutions of text in RFC 4928
should either both use the NEW/OLD TEXT convention or neither of them
should.

Section 2.2: The new text for RFC 4928 says "A new document is to be
published". It always seems dangerous to predict the future and, if this
text is adopted and is true, then it won't be true after this "new
document" is published and there will then be an incorrect statement in RFC
4928 as updated. I'm not sure what would be best here. You could,
presumably, say something like "At the time of this document, it was
planned to ..."

Section 2.3: I guess it's nice to thank IANA but the registry creation that
this document would cause is a routine action for IANA. For drafts where I
am an author, if any individual in IANA makes a suggestion that improves a
draft, I try to add them to the acknowledgements section. But I don't think
it is appropriate to have explicit thanks to the IANA function for
something routine that is done in hundreds or thousands of RFCs,

Section 3.1.1: I don't think this needs to be an extra subsection and it
certainly doesn't need an all capital letters MUST. Suggest just deleting
3.1.1 and adding the following to the text at the top ov 3,1: "The
assignment policy is Standards Action."

Thanks,
Donald
===============================
 Donald E. Eastlake 3rd   [phone] (cell)
 2386 Panoramic Circle, Apopka, FL 32703 USA
 [e-mail]