Kizuna West

A bit more on the Big Society. I mentioned that Rory Stewart wants faster broadband for his constituents in rural Cumbria. Now Rory is a decent guy – I knew him at school – but I don’t think he’s on to anything much with his play to relate the issue of rural broadband provision to Cameron’s Big Society. We might eventually end up with better broadband in Cumbria, but we’ll have gotten there by the hard road: the plan doesn’t do much for the idea of little platoons. The Guardian has a bit more about how Stewart describes his project. Apparently there are three components to it:

1) The government part. The government is going to open up its public infrastructure. It is going to allow us into the fibre optics thick pipes that run to the schools. It is going to put pressure on Network Rail to let us into the thick pipes that run along the Carlisle Settle line.

2) We, as communities, will get a very small government subsidy equivalent to what they would have given us in terms of their universal service commitment. You roll out a parish pump which is to say you go into that thick pipe at your school or on the railway and you bring out a little fibre optic cabinet. Then, and this is the key point, the parish comes and puts together its own plan to get the stuff from the parish pump into their home.
Many of our communities will want to go for fibre optics to the home so they can have super fast stuff. Others will be content to put a wireless hub on top of the pump that will give them two megabytes.

3) The final government support for the community is to provide a loan. If it costs a £1,000 to put broadband into your house, if you have a soft loan over 15 to 20 years that is only costing you £50 a year.

Now there’s not quite enough technical detail here to comment on viability. What we can do, though, is make a quick comparison with the way Japan has set about solving the same problem. In 2008, JAXA launched a satellite – Kizuna – which allows any rural Japanese household to connect to the internet at 155 Mbps download and 6 Mbps upload. That’s the domestic transmission rate. Small businesses get 1.2 Gbps download. Compare this with the default 2 Mbps rate mentioned in the Cumbrian plan. Note, that’s megabits. Rural Japanese already get between 75 to 600 times the data rate planned for Cumbria, once the parish meetings are held, and the thousand small disagreements about what to do have gotten thrashed out.

In terms of subscriber requirements, the Japanese subscribers only need to install a satellite dish (45 cm at the lower data rate) to get connected. The Cumbrians are expected to raise loans to get cables laid to their houses (note: wouldn’t line of sight microwave be better in some cases?).

In terms of scope and timescale: Kizuna gave coverage to the whole country from launch day onwards. The Eden Valley Big Society plan, once implemented, will cover part of one county of England.

The talk of parish pumps and railway lines is charming, but I think it’s a shame to be literally parochial about something like this. There are situations where a society needs to amass all of its resources to be effective. Communications infrastructure is one of those situations. What’s more, investment in satellites, specifically, is consistent with fiscal stimulus as generally understood (there are British satellite manufacturers). The Big Society talk is surely better saved for the human-to-human stuff, if we’re going to hear about it at all.

Anway, what do Fistful readers suggest vis-a-vis rural broadband. How has this been solved elsewhere?

10 thoughts on “Kizuna West

  1. Satellite broadband is extremely economical, at most $500mm for the two satellites (one on orbit spare) plus perhaps $200 per terminal.
    Downsides:
    Poor reception in rain, bigger (perhaps 36 inch) dish needed in northern areas, must have a south facing antenna site, much higher latencies make interactive internet games and voice communications difficult.

    As a quick and effective nationwide demonstration of the capabilities of broadband, satellite is a superb solution, even if it is a transition stage to a fully fiber linked community.

  2. I think it’s unethical.

    If I want to live in a remote village in Scotland, on what basis do I demand others pay in any way for my services?

  3. Several communities in British Columbia have experimented with radio transmission, which has worked well in urban areas (skyscraper to skyscraper). Does anyone have any evidence that this has worked outside an urban area? (I would like to see positive results but…)

  4. Cable installations in the US are estimated at $2000 on average, so satellite is roughly one quarter the cost or less, depending on the number of users.
    The satellites ab initio nationwide coverage eliminates the need to discriminate against more remote locations on grounds of cost or ease of access.
    Small communities in unusually difficult sites may benefit from a master antenna with a local net, wired or wireless, much like as if it were an apartment building with multiple dwellings.
    RF links with enough band width to carry video tend to be line of sight and there are exposure limits, so routing them should be carefully done.

  5. I would point out that KIZUNA is, at the moment, a technology-demonstrator. Not a solution. For example, it may provide 155Mbps (OC-3) service, but it provides a total of three channels. 30 15Mbps downlink customers at a contention ratio of 10:1.

    Really, nothing is a substitute for fibre. You can fit 10 KIZUNAs into one strand at OC-192, and provide similar speeds in both directions.

  6. Well yes, but if it works as they hope, the Japanese will be pretty far along the road to something everyone can officially call a ‘solution’.

    And are you sure that ‘three channels’ and ‘155 Mbps’ gives you license to do the arithmetic you’re doing there? I’m sure contention ratios come into it somewhere: JAXA doesn’t say.

  7. The JAXA data sheet specifically mentions 3 channels up/down with 6/155Mbps each, and suggests that each channel can be used for separate geographical and functional tasks using beamforming to aim and on-board IP routing. That suggests that each channel represents one radio transponder. So you could support three clusters of terminals/three local exchanges at a 10:1 contention ratio.

    Now, a production satellite using the same technology would likely cram in a lot more transponders in an airframe the size of a bus and need Ariane-5 ECA to launch.

  8. Kizuna ought to be the Japanese for ‘jam tomorrow’. Inmarsat are offering 50Mb satellite connections worldwide, but they’re not economical for individual households, only for ships and aeroplanes.