Projects like this make me hopeful for the world, but fretful for America. Not because of the increased competition or outsourcing or anything like that, but because by and large we've created a political climate where large infrastructure developments are nearly impossible to implement. We eschew any sort of tax-funded large digital infrastructure projects while at the same time we do our best to keep regional ISP monopolies in place, killing any meaningful competition.
The fact that Kenya now has hundreds of miles of fiber optic cable is awesome. The fact that most of the internet I use in the USA ends up last-mile-ing over coaxial lines laid in the 1970s is terrible and unlikely to change in the next 10 years.
(Of course, our non-digital infrastructure is crumbling for largely the same reasons, so that's even more stuff to worry about).
Second, are coaxial lines really that bad? You do need literally bundles of fiber for backbone activity, but last-mile? DOCSIS 3.0 is 42Mbit/channel, with as many channels as the endpoints can support. Comcast has already offered 305Mbit. We will need fiber last-mile one day, but not nearly any time soon.
I suspect the real limitation is a grab-bag of things like:
- Increasing speeds highlights damaged coax, which would then require repair
- Slowing the growth in speed means they can continue to charge a lot for service that is easy to provide
- The true bottleneck is probably somewhere in the ISP, for example routing infrastructure, which would need to be upgraded
Eh, Fiber is expensive. It's getting better, though; a big part of the expense is the skill required to splice fiber, and they are developing better and easier ways to splice all the time. I betcha the last mile is going to suck for most people for a long time. And you know? I don't think it matters as much as we think. If it does matter for you (It matters for me, certainly) you will move to the places where fiber is available. Yeah, google fiber gets all the press, but surewest has been doing fiber to the home in Sacramento for close to a decade now. (I mean, fiber matters a lot to me, but not enough for me to move to Kansas... or Sacramento, for that matter. Ugh.)
Here in silicon valley, both the city of Palo Alto and the city of Santa Clara (see http://svpfiber.com or email me if you want to get in on an in-progress project) have municipal fiber rings. It's aimed at businesses, and /only/ does the last mile (which is to say, once you get to the data centre you will still have to pay someone for transit to the internet. And you will have to pay to trench to your house, so you are probably looking at $500/month minimum and maybe $10K one time to trench in to your house. Still, worth it to not live in Sacramento. Ugh.)
I mean, what I'm trying to say is that there are really good tubes, here in America. I, myself, have 12 gigabits of burst capacity (a 6.2 gigabit commit) Yes, our last mile for most people is shit. But the last mile for everyone, everywhere is shit. Just like in Africa, I personally predict that it's going to be wireless, for most of us. And for most of you? wireless is just fine. I mean, yeah, cheap fiber for everyone like they are said to have in Japan, or in Kansas or in Sacramento would be super great, but I don't think that only having that in a few places will cripple us. I mean, when it really matters, even here in America, if fiber matters to you, you can get it by paying a small (compared to developer salaries) fee, or by moving somewhere unpleasant.
The fact that Kenya now has hundreds of miles of fiber optic cable is awesome. The fact that most of the internet I use in the USA ends up last-mile-ing over coaxial lines laid in the 1970s is terrible and unlikely to change in the next 10 years.
(Of course, our non-digital infrastructure is crumbling for largely the same reasons, so that's even more stuff to worry about).