As best I can tell, the technical substance of sidr-repos-struct-03 is
in fine shape. That said, I found the document difficult to review,
largely due to the style of its writing. While I recognize that this
may be merely a matter of stylistic preference, I find the writing
unnecessarily verbose and unapproachable.
I notice that the only two other WGLC reviews of this doc came from
native English speakers, one of whom has been deeply involved in the
work for years. I think the doc has not gotten adequate review, I
think that is in large part because of the writing.
I think this document would benefit greatly from an editorial pass,
probably from a different editor. Assuming someone can be found to do
that, I support delaying the publication in the interest of producing
a more readable document.
And, as I noted in my WGLC review of the architecture draft, I have
some reservations about pushing these docs out without having the
validation algorithm nailed down (including the handling of
incremental deployment). In the case of this document, noting that it
has normative references to the architecture and manifests docs, which
are more likely to see changes from changing the validation scheme, I
don't see any harm in sending it forward.
Specific suggestions
Danny McPherson's comments on ths draft from 3 December 2008 don't
appear to have been incorporated nor responded to on the list. (e.g.
"complete set" is still a strange phrasing, and the spelling nits
haven't been fixed)
Rsync is not mentioned in the intro. Instead of just referring to
"publication points", adding the concreteness of "rsync" at that stage
would help make the doc clearer.
Section 1: "A Resource Certificate describes an action by an Issuer
that binds a list of IP address blocks and AS numbers to the Subject
of a certificate, identified by the unique association of the
Subject's private key with the public key contained in the Resource
Certificate." "describes an action"? Surely there's a better way to
say that.
I'm not sure that Section 2.2 paragraph on naming certificates is
consistent with section 4's guidance on persistent naming. Which may
be more about how hard it was to understand section 2.2 than about any
real differences.
Nits
The rsync reference appears to be truncated.
Spelling:
s/performss/performs/
s/Becuase/Because/
s/Whn/When/
s/traveral/traversal/
(and there are surely others, including some that Danny pointed out)
-- Sam
_______________________________________________
sidr mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/sidr
