Note: This ballot was opened for revision 09 and is now closed.

(Ralph Droms) Yes

I have one comment with my independent reader hat on.  I was a little
surprised, after several potential uses of SDNVs were described earlier
in the document, to read:

  In general, it seems like the most promising use of SDNVs may be to
  define the Length field of a TLV structure to be an SDNV whose value
  is the length of the TLV's Value field.  As previously discussed,
  this leverages the strengths of the SDNV format and limits the
  effects of its weaknesses.

I was left wishing for a little more explanation of why the authors make
this statement and wondering why the other potential uses for SDNVs
were judged less promising.

(Jari Arkko) No Objection

(Ron Bonica) No Objection

(Gonzalo Camarillo) No Objection

(Russ Housley) No Objection

(Tim Polk) No Objection

(Dan Romascanu) No Objection

(Peter Saint-Andre) No Objection

The introduction states:

   One example of SDNVs being used outside of the 
   DTN protocols is in Hixie's Web Socket protocol
   [I-D.hixie-thewebsocketprotocol], which includes 
   a binary frame length indicator field identical to an 
   SDNV.

Referencing draft-hixie-thewebsocketprotocol-76 seems problematic, given that this individual I-D was superseded by draft-ietf-hybi-thewebsocketprotocol and the work of the IETF's HYBI WG. If this text is not really needed here, I would urge its removal to prevent confusion regarding the canonical documentation of the WebSocket protocol.

(Robert Sparks) No Objection

(Sean Turner) No Objection