[i2rs] Comments on draft-ietf-i2rs-ephemeral-state-21

[Joe Clarke <[e-mail]>]

I have read through the latest -21 rev of 
draft-ietf-i2rs-ephemeral-state, and I'd like to raise some additional 
(mostly minor) comments:

Section 1:

OLD:

The I2RS Architecture
    document [RFC7921] abstractly documents a number of requirements for
    implementing the I2RS requirements

NEW:

The I2RS Architecture
    document [RFC7921] abstractly documents a number of requirements for
    implementing the I2RS.

===

You talk about YANG as a data modeling language and NETCONF/RESTCONF as 
protocols.  Good.  However, in section 1, you go on to say:

What does re-use of a protocol mean?  Re-use means that while YANG,
    NETCONF and RESTCONF are a good starting basis for the I2RS protocol

YANG is not a protocol.  This is a nit, but perhaps YANG can be dropped 
from this sentence since it speaks to _protocol_ re-use.  Either that, 
or say something like:

What does re-use mean in this case?  Re-use means that while YANG, 
NETCONF and RESTCONF are good starting points for defining the I2RS data 
model and protocol...

===

Section 2:

The requirements are technically distilled from RFC7921 AND RFC7920 now.

===

Section 2, bullet 1:

OLD:

The I2RS protocol SHOULD support an interface asynchronous
        programmatic interface interface with properties of described in
        section 5 of [RFC7920] (e.g. high throughput) with support for
        target information streams, filtered evens, and thresholded
        events (real-time events) sent by an I2RS agent to an I2RS Client
        (Key points from section 1.1 of [RFC7921]).

NEW:

The I2RS protocol SHOULD support an asynchronous
        programmatic interface interface with properties of described in
        section 5 of [RFC7920] (e.g. high throughput) with support for
        target information streams, filtered evens, and thresholded
        events (real-time events) sent by an I2RS agent to an I2RS Client
        (Key points from section 1.1 of [RFC7921]).

===

Do a pass through this doc to lowercase the 'c' in client.  You have a 
number of cases that read "I2RS Client" where we previously standardized 
on "I2RS client."

===

Section 7:

OLD:

MUST BE

NEW:

MUST be

(This has been called out by others.)

Joe