Live from Stavänger - Part 1

The first session of the Stavänger OECD Workshop on Fibre investment and policy challenges is just wrapping up. A session called Technological developments geared to policy makers, featuring Herman Wagter (Amsterdam Citynet), Marvin Sirbu (Carnegie Mellon), and Karl Bauer (Leoni).

Herman Wagter's presentation was very interesting, but of course knowing some things about the Amsterdam project, it wasn't all new to me. The focus of the presentation though, was making an economic case for Open Access. Very interesting. One of the challenging points Herman made was that topology and technology should not be confused. Herman argued very convincingly that while common sense and future-proofness dictate that homerun should be the topology, that doesn’t preclude from using PON in the CO to optimize usage costs.

Marvin Sirbu's presentation first explained three theoretical competition models: infrastructure based competition, unbundling and bitstream. Marvin explained that there were three different schools of regulatory thought as to where the natural monopoly laid: conduit, dark fiber, or bandwidth. He then described some specifics of the different topologies. One of the arguments that comes again is that deep splitted PON doesn’t make much economic sense because a single customer forces you to provision equipment that covers 32-64 customers. An aggregation point infrastructure allows for a better scalability and also allows for a form of unbundling at the aggregation point (provided you have enough space available). I guess Herman's earlier point could be said to be that putting the aggregation point at the CO is the best way to keep all your options open and maybe even serve different customers (consumers and businesses, for example) with different technologies.

Another interesting point Marvin made was that soaring VOD usage may have impact on backhaul costs, and regulation of the metroplolitan and regional backbones may be required too in some areas.

Another very interesting aspect of Marvin's presentation is an analysis of breakeven on the network roll-out. If you can generate 50$ per subscriber to amortise the cost of the network you need 20% penetration. If you only get 30$ you need 33%. That hints that at best you might be able to support 2 fiber to the home networks. But this overlooks customer acquisition costs and assumes 90%+ penetration. Facilities based competition therefore simply cannot work. Key message to regulators: do not wave away open access requirements in exchange for fiber investment.

Karl Bauer's presentation was a lot more focused on technology and operational aspects of roll-out. Very interesting, but a little low in the service layers for me to grasp! A number of interesting points though: the speed of the roll-out has significant incidence on breakeven, so you want to optimise all you can before you start rolling out. Similarly, authorisations to wire homes should be acquired before the roll-out phase to optimise costs.

A good start with a few challenging points ! Login later today for more !