Re: [6lo] Comments on draft-hou-6lo-plc-04

Dear Anand,

Thank you very much for your review. Please find my responses below.

Best regards, 
Remy

> -----Original Message-----
> From: 6lo [mailto:6lo-bounces@ietf.org] On Behalf Of S.V.R.Anand
> Sent: Thursday, October 18, 2018 5:54 PM
> To: 6lo@ietf.org
> Cc: 6lo@ietf.org
> Subject: [6lo] Comments on draft-hou-6lo-plc-04
> 
> Dear Authors,
> 
> I went through the draft and I think it is useful. Wish to share my initial
> thoughts on the text.
> 
> - According to my understanding, the draft covers IEEE 1901.1, IEEE 1901.2
> and
>   ITU-T G.9903 PLC standards together. After glancing through ITU-T G.9903
>   specification it appears that the 6LoWPAN adaptation is already defined
> there.
>   It would be nice if the draft covers those aspects that highlight the major
>   differences that are not covered by ITU-T G.9903. If the difference is not
>   substantial, the draft can just focus on IEEE 1901.1, IEEE 1901.2.
[Remy]It is true that ITU-T G.9903 has included the selections from RFC4944 and RFC6282 in the body text. Thus the related description is normative. In our draft, the G.9903 related contents, such as stateless address autoconfiguration, LL address, unicast address mapping and header compression, are well aligned with ITU-T standard. What is in our draft but not in ITU-T G.9903: neighbor discovery protocols in RFC6775 and RFC6775-update, the new outcome of 6lo. As for the two IEEE standards, the description on IPv6 adaptation does not cover all the aspects, and is included as annexes, thus not formal standards. In this draft, we are aiming to provide an overall design of PLC IPv6 adaptation layer based on IETF standards and their modifications.

> - Since routing aspects are outside the scope of 6Lo charter, wonder how
>   important is the routing protocol section for the draft. Of course, LOADng is
>   deeply embedded into ITU-T G.9903 in the form of mesh under routing. Let
> me know
>   what you think.
>  [Remy] The description on routing protocols is just informative as a part of the PLC technology overview. Thus we put it in section 3 instead of section 4.

> - Informative reference to LOADng can be included.
>
> - Refer to the following text in Section 3.
> 
>   "A routing protocol (e.g., RPL [RFC6550] or AODV-RPL
>   [I-D.ietf-roll-aodv-rpl]) at the Network layer is optional according to the
>   IEEE 1901.1 and IEEE 1901.2 PLC standards mentioned in this document."
> 
>   Depending on the way the reader reads it, the above sentence can be
>   interpreted as "IEEE 1901.1 and IEEE 1901.2 PLC standards are specifically
>   saying RPL and AODV-RPL are optional". Further, "according to the IEEE
> 1901.1
>   and IEEE 1901.2 PLC standards mentioned in this document" can be read as
> the
>   document is defining its own version of IEEE 1901.1 and IEEE 1901.2 PLC
>   standards. Can you rephrase it ?
>  [Remy] Thanks for pointing it out. We need to rephrase it: A routing protocol MAY be required in tree and mesh topologies.  The routes can be built in mesh-under mode at layer 2 or in route-over mode at layer 3.

> - Refer to the text in Section 4.
> 
>   "However, due to the different characteristics of the PLC media, the
> 6LoWPAN
>    adaptation layer cannot perfectly fulfill the requirements."
> 
>    At this point of the draft, it is not clear what the requirements are. It is
>    better to substantiate with few unfulfilled requirements that provide
>    motivatation for the draft.
>  [Remy] Good point! For example, the stateless address autoconfiguration and unicast address mapping defined in RFC4944 can't be used for IEEE 1901.1 due to the different L2 address format. Thus modifications are needed to adapt IEEE1901.1.

> Regards
> Anand
> 
> _______________________________________________
> 6lo mailing list
> 6lo@ietf.org
> https://www.ietf.org/mailman/listinfo/6lo