Dear Olivier and Job, Thanks for your feedback, and sorry for delayed answer (I thought I was already subscribed to the mailing list, but I was not). I was sure the question about PIC and ADD-PATH was going to appear, and I am going to address it in the presentation. The answer is that we mainly target scenarios where you can't use PiC either because the operator is using next hop self (which we understand is fairly common) or because there is no IGP to update the pointer.
Going back to Job's questions: 1) Would it perhaps make sense to rank updates by BGP next-hop (using a metric like traffic volume), rather than destination prefixes themselves? Each BGP next-hop in effect represents a group of prefixes. We are not pushing any ranking algorithm. We used the traffic because it was the most relevant concern at that moment, but we can also consider other algorithms. For instance, we received a comment about considering critical services like the DNSs at the top of the list. For sure I agree we can consider ranking traffic by next hop, but as we are mainly targeting the next hop self situation this can be tricky. 3) Are you looking to publish your work in context of IETF or IRTF? We are still to decide how and where to present this. IETF is definitely one option. Any suggestions about how to go about resenting this in the IETF? do you suggest to just have an informative presentation in something like GROW WG or do you think it would make sense to write a draft describing the mechanism and see if any WG is willing to take this on? BR, Juan Brenes -- JUAN BRENES BARANZANO Universidad Carlos III de Madrid
