Pierre Pfister <[email protected]> wrote:
    > I agree with Markus and Steven on most points, but let me clarify a few
    > aspects.

    > - First, I think this is really cool if another area wants to use
    > hncp. So I don’t see any reason we couldn’t adapt the documents to
    > allow it. We should be careful *not* to make homenet specifications too
    > complex thought. Having too many abstraction layers may not be
    > desirable.

Copy and paste is an acceptable way to develop specifications.
The IETF owns the text, if someone wants to use it, let them write a
document.  If the documents can be converged later on, great.

    > hncp-transport, and another for hncp-homenet. I’m a bit more concerned
    > about removing trickle algorithm behavior. We can abstract the TLV

If you remove trickle, then you might as go back to OSPFv3.

    > - Homenet-hncp intends to be incremental in terms of TLVs. Other TLVs
    > will come, and, if hncp is widely used in homes, vendor-specific
    > options will come as well. So I think ‘ignore TLVs you don’t know’ is
    > the correct behavior.  Nevertheless, I understand hncp could be used

We actually need a "critical" bit on the Type value then.

--
Michael Richardson <[email protected]>, Sandelman Software Works
 -= IPv6 IoT consulting for hire =-

Attachment: pgp7eJvHM66sD.pgp
Description: PGP signature

_______________________________________________
homenet mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/homenet

Reply via email to