This is an interesting idea Chris. Start from the least served areas then build out.

This has a number of things going for it. Firstly such a programme would qualify for state aid which would be an incentive. Secondly it would establish rural regions not just as niche player but also as key players. Thirdly it would emphasise links to the urban core because providers would have to find commercial wins - city regions would become true city regions.

While not being at the forefront of people's minds this suggestion does also create a potential for a more democratised view of NGA because it creates the opportunity for those rural communites to create their own solutions and then to sell on rather than wait for the opportunity to connect to already established, large scale, centralised, commercial opportunities owned by the big four. This suggestion has sustainablilty build in.

The researchers say "Using emulation technology combined with real-world infrastructure, the tests showed that the network is able to serve between 1,000 and 4,000 users within 20 kilometres of the main ring with symmetric internet connections at speeds of around 300 Mbps. Separately, the researchers also demonstrated that the technology could be used to transmit optical signals up to 100 kilometres from the central office in order to provide up to 250 homes with asymmetrical 10 Gbps downstream and 2.5 Gbps upstream connections."
Great, but they also say it uses existing fibre. That's a big problem right there, which would be resolved by building the network locally and joining the dots, as it were.
I'm not sure how much the existing duct networks would support the ring topology suggested by the researchers. You can bet your bottom dollar the incumbents are not going to play ball willingly.
That said, the lab results look interesting enough for someone to start exploring the commercial models. Perhaps Mrs Kroes can find a few euros for that.

I would rather go for the credo 'Broadband for everyone'. It is perfectly possible to deploy networks for inner area's and less served area's together. An economy of scale and scope and sharing costs and benefits would contribute to fulfill this credo for a start.

Looking at the Dutch situation: 2 million citizens are living in area's with non-sufficient connections. Normally the costs of deploying and services would be too expensive for investors. With an honoust non-revenu-driven businessmodel it is possible to:

- get the connection (people, companies)
- get ROI (investors, whether they are public of private)
- get to an open market situation for ISP's, which leads to competitive prices and better quality

The key is act locally, but share knowledge, investments and services with other parts of Europe. Look at the Swedish situation, where the OpenNet model is a good and working way to deploy nextgen networks without 'cherry picking'.

It is about creating no exceptions, everybody needs a connection, then everybody should get it.

Chris, your comment "transit charges are too high" is spot on. I like to say there is an ocean of capacity being sold one drop at a time. it is not a technology issue, rather a pricing, software and marketing issue. All investment has to be approached from a horizontal perspective. For instance the cost of 196 bundle fiber running down the street is high if there are 1-2 buyers, but low if there are 100 buyers buying 1-3 strands. 3 strands, btw, supports ANY conceivable business model for the foreseeable future. Elsewhere I will introduce the concept of digitization, 3 prior digital waves, and how they pave the way for a 4th and final wave of digitization. The key in all this is modeling marginal cost under an array of demand contexts. Only then can we efficiently sell the ocean of capacity.