Re: [P2PSIP] WGLC for draft-ietf-p2psip-share-04

Hi Alissa, Hi WG,

I have reviewed draft-ietf-p2psip-share-05, and here are my comments.

General comments:
in my opinion the draft is in good shape and reads well. I have a few nits
and editorial suggestions detailed below. I believe these can be addressed
quite easily with a quick resubmission and my impression is the doc is
ready to go.

Detailed comments:

in Section 1: refer to RFC6940 (and which section, if applicable) the first
time specific terms are used such as "RELOAD Usage" or "RELOAD security
model". Spoiler: I will have a lot of such comments below ;)

in Section 2: for reader convenience, I suggest listing the key terms
(without recalling their definitions) imported from RFC6940, and the
p2psip-concepts
draft in the paragraph right after the 2119 boilerplate.

in Section 3.1: in step 3, I suggest being explicit that the 8bit part is a
suffix (least significant bits)

in Section 4.1:
- "...Alice is also granted (limited) write access..."
Either explain what "limited" means here, or remove this adjective.

- "Note that overwriting existing items in an Access Control List that
reference a    different Kind-ID..."
Clarify: different from what? I suppose you mean that the overwrite results
in changing the Kind-ID

- "The Resource Owner is allowed to overwrite any existing ACL item, but
should be aware of its consequences."
Either quickly explain / give examples of consequences or remove this
sentence.

in Section 5.1:  "The specifications in this document scheme adhere to this
paradigm...".
add reference to RFC6940 (and the exact section). It will help readers
grasp quicker what draft-ietf-p2psip-share specification adds here.

in Section 6.1:
- first sentence "Write access ... solely be issued by the Resource Owner."
rephrase needed (confusing as readers already know that delegation is
possible).

- "... stored in the numerical order... starting with the index of the root
item...".
I have a (stupid) question: What if the Node-ID of the an authorized peer
with ad=1 has a node-ID that is numerically smaller that that of the owner?
I suggest rephrasing in order to clarify this corner case, just to make
sure no one is confused?

in Section 6.5: Step 1. reference "as per RFC 6940 Section X.Y."

in Section 6.6: Because it is possible here, I would have preferred to see
the last 2 paragraphs written in steps + pseudo-code style if...else..else.
But that's a matter of taste.

Best,

Emmanuel

On Tue, Apr 21, 2015 at 9:08 PM, Alissa Cooper <alissa@cooperw.in> wrote:

> Yes, that’s fine, thanks.
> Alissa
>
> On Apr 21, 2015, at 1:40 AM, Emmanuel Baccelli <Emmanuel.Baccelli@inria.fr>
> wrote:
>
> >
> > Hi Alissa,
> >
> > if it is not too late: I am currently reviewing the document. ETA early
> next week.
> > Sorry for the delay. Is that alright with you?
> >
> > Best,
> >
> > Emmanuel
> >
> >
> >
> >
> >
> > _______________________________________________
> > P2PSIP mailing list
> > P2PSIP@ietf.org
> > https://www.ietf.org/mailman/listinfo/p2psip
>
>