From: IAB/IESG For Response to: ITU-T Q1/SG15 Response requested by: 2011/3/15 In response to: Liaisons LS-243 and LS-244 from ITU-T Q1/SG15 Title: Regarding ITU-T standards for IP home networking and "Protocol for identifying home network topology", G.phnt Submission date: 2011/1/28 Contact: Patrik Faltstrom Monique Morrow Thank you for your liaisons LS-243, "Notification of ITU-T standards for IP home networking" and LS-244, 'Draft Recommendation G.phnt, "Protocol for identifying home network topology"', posted 2010/12/15. We appreciate Q1/SG15's interest in collaboration with the IETF in areas of mutual interest regarding home networking. At present, the IETF does not have a "HomeNet" working group. While the formation of such a working group was considered in the past, no working group was formed. The following IETF working groups may have an interest in work undertaken by Q1/SG15, as listed in LS-243 and LS-244: v6ops, opsawg (for G.phnt), 6man, intarea and tsvwg. For reference, the current list of active IETF working groups is always available at http://datatracker.ietf.org/wg/ We have some comments based on an informal review of the attachment to LS-244 titled "Initial draft of G.phnt". The protocol, as we understand it from the attachment, is related to IETF Internet protocols but does not appear to modify or adversely affect the operation of any IETF Internet protocols. If the protocol is, indeed, limited to discovering devices and device configuration, which is collected and reported by an agent, there should be no issues with the work being undertaken by ITU-T Q1/SG15. If, on the other hand, the protocol is intended to provide a means for service discovery or device configuration, it may be in conflict with existing IETF Internet protocols. In this case, the IETF and ITU-T should manage the development of the protocol under the processes in RFC 4775. We were unclear about some aspects of the protocol specification, which we hope will provide constructive feedback. It is unclear whether internet layer addresses would be carried in LLDP, which would constitute a layering violation because LLDP is a data link layer protocol. The security considerations in the specification were lacking. As specified, the protocol would represent an easy way for an attacker to gather information about devices on a home network, or may represent a mechanism through which an entity such as a service provider might gather information about a home network that the owner of the network might not want to reveal. Finally, related to the concerns expressed above about the nature and intent for the protocol, the specification is unclear about future extensions: for example, will the specification be extended to allow the protocol to act as a topology discovery mechanism across multiple links? In this regard, the title of the protocol may be a bit misleading, as, in its present form, there is really no network topology discovery, per se; rather, devices on a single link are discovered and information about those devices is reported. We request that ITU-T Q1/SG15 consider and respond to our technical comments and questions, and we request Q1/SG15 liaise back to us any further developments in this work.