To tsbsg13@itu.int Georges Sebek ghassem@nortel.com CC IETF PWE3 WG Internet Area AD Internet Area AD PWE3 Co-chair PWE3 Co-chair statements@ietf.org Yaakov Stein For Action Please bring this liaison to the attention of Q7/13. Thank you for your liaison entitled Use of term "pseudowire" and definition of stitching function https://datatracker.ietf.org/liaison/329/ We would like to thank you for the information on your draft new Recommendation. We agree that what you call a Generic Adaptation Layer (GAL) is either a single segment PW, or a multisegment PW. We would appreciate your using the term "pseudowire" (PW), and believe that standardizing terminology is best for the industry. We understand that you have included normative references to the necessary IETF RFCs. We request that you include a clear statement that there is no intention to change the behavior defined in the normative IETF RFCs, and in the event conflicting definitions, the IETF RFC is the authoritative source. In addition, although the focus of this document is the multi-segment case, we advise adding (for reference) figures prior to Figures 6-1 and 6-2 depicting the single segment case. Furthermore, we feel it would be useful to specifically state that the subnetwork connection inside the S-IWF can potentially modify the PW label. Although this is implicit in the switching capability at the PW layer, this is the new feature of the MS-PW. Regarding your question about the use of a single S-PE or a pair of S-PEs connected by a simple link, we feel that both cases are possible. In the MS-PW architecture draft (draft-ietf-pwe3-ms-pw-arch-03.txt) we differentiate between the intra-provider and inter-provider cases. The inter-provider reference model of figure 5 of this draft maps to your Figure 6-3 when "PW segment 2" is a link. On the other hand, for the intra-provider case, or when two providers have a relationship of trust, the single S-PE case is feasible. Regards Stewart Bryant & Danny McPherson IETF PWE3 Co-chairs