In general, the draft technically is in a good shape, but I have a couple of
"nits" to be fixed before concluding anything.

---

This text is kinda ambiguous, I think we should use "none" instead of "any",
right?

Old:

If the target BGP Identifier in any of the Node Target extended community
does not match with the local BGP Identifier, this node is not the target
node of Update, and the information in the Update is not eligible to be
installed on this node.

New:

If none of the Node Target extended communities match with the local BGP identifier,
this node is not the target node of Update, and the information in the Update
is not eligible to be installed on this node.

---

>Target BGP Identifier (4 octets): The BGP Identifier of a target node. 
It is a 4-octet, unsigned, non-zero integer as defined in [RFC6286].

The document doesn't say anything what if we receive BGP-ID as 0.0.0.0?
Is this malformed, should be ignored, installed anyway or what? Technically,
0.0.0.0 is a malformed BGP-ID, but you know - hackers do not follow RFCs.

Another question: which BGP-ID should be used? The one that is received/sent
in the BGP OPEN message, and what if the confederation is used? We should explain
this in the draft too.

---

>One or more Node Target extended communities MAY be carried in an Update
message to designate a group of target BGP nodes.

So, identical (duplicate) is fine? Can we say that?

---

>In this section, the mechanism for intra-domain scenario is described, the
mechanism for inter-domain scenario is for further study. The domain here
refers to an administrative domain, which may consists of one or multiple
ASes managed by a single operator.

This is technically fine, but according to RFC 6286, identical ASNs can be used
in the same "domain". Maybe we should clearly say that.

---

>The sub-type of the Node Target Extended Community is TBA.

TBA is already assigned => 0x09.

---

>It can be a transitive extended community with the high-order octet of the
type set to 0x01, or a non-transitive extended community with the high-order
octet type set to 0x41.

Should we write down that this new community MUST be stripped when going beyond
the administrative domain?

I think also that draft-uttaro-idr-bgp-oad might be mentioned too regarding
what is "domain".

---

Internet-Draft is empty, should it be "Standards Track"?

---

>The mechanism of using the Node Target extended community to control the BGP
route distribution only to particular BGP nodes is out of the scope and will
be specified in a separate document.

This text is kinda "redundant", right? It's in the 1 and the 3 paragraphs.