Slashdot Mirror
Can Mobile Broadband Solve the UK Digital Divide?
MJackson writes "Lord Carter's interim Digital Britain report recently proposed a new Universal Service Obligation (USO), which would effectively make it mandatory for every household in the UK to have access to a broadband service capable of 2Mbps by 2012. Since then there has been much talk about Mobile Broadband (3G, 4G) services being used to bridge the UK Digital Divide, but is that realistic? The technology has all sorts of problems from slow speeds and high latency to blocking VoIP, MSN Instant Messaging and aggressive image compression ... not to mention connection stability."
but how many people can it support on a tower at a time before it slows to a crawl?
Yes, it may be a little socialist in some respects, it really forces a good thing onto the people with very little downside except short term funding issues.
If you think that short term funding issues should take precedence over long term societal growth, then by all means reject this proposal. But it should be noted that that sort of short term thinking is what led to the collapse of the American auto industry and the subsequent begging for bailouts.
It is forward looking policies that brought Korea and Japan to the forefront of broadband technology. With every new home wired for fiber and existing lines being replaced at a rate of 3 miles per hour, these Asian countries have already made investments that Western countries should have been making 10 years ago when the DotCom boom was in full effect and money was plentiful.
No, in the sense that the technology is severely limited compared to a hardline.
Yes, in the sense that, with a little strategic gaming, cell derived wireless technology is almost certainly the cheapest way to minimally satisfy whatever universal service obligations end up being imposed. Unlike landline buildout, where you'll actually need to spend real, verifiable money building real, verifiable connections to every lower-income hovel that you can't be bothered to bother with; a wireless "universal" system could simply involve tacking a horrifically crippled lowest tier option onto the infrastructure you are already building to sell to cost-insensitive business types.
It is fairly likely that, unless astonishingly carefully drafted by public spirited experts, the USO will underspecifiy what is actually required to access the internet pleasantly. You'll be able to satisfy the requirements by demonstrating the availability of an X megabit connection from at least one top floor flat per postcode, while saving money and/or upselling hard, by blocking like crazy anything that isn't vanilla port 80, and not really bothering about latency, packet loss, and spotty connections among your less preferred customers.
Don't get me wrong, the mobile stuff has its place, since you can't really trail a fiber line around behind you when you move about. As a means of "universal access", though, I strongly suspect that it is a good solution only in that it will be the cheapest way to offer something nominally resembling an internet connection, not by virtue of actually being any good.
In particular, my concern would be the effect on the development of the internet. Available bandwidth spurs development of new uses for the internet, which spurs greater demand for bandwidth, which spurs improvement of bandwidth supply, and so forth. Reliance on extremely expensive or crippled internet access guts that. If the internet access is costly or lousy, interesting uses of it will stagnate or shrivel. If they do that, the stagnant status quo is under no pressure to upgrade, and there things stay.
What?!?
I use a 3G HSDPA service regularly with two different laptops that have built-in HSDPA modems from Sierra Wireless and Ericsson. I also use Nokia and LG phones over Bluetooth tethering (since I'm in Australia and have sensible carriers that don't lock that down).
I get a public IP address. No NAT. No filtering, either. Full use of VoIP (SIP or *ick* Skype), etc. No dodgy proxy hacks with image compression or other nasties. It's just a regularly IP service.
It's fast. Not ADSL2+-over-wifi fast, but quite fast enough for everything I need to do, including VNC/RDP remote control of machines at work, SSH, etc. Latency is occasionally a wee bit high, but nothing too bad.
It's pretty stable - it only goes a bit flakey when going through (eg) a train tunnel where it completely loses reception. Even then, it often just transparently recovers without apps or the OS ever really noticing. Sitting in one place, it's rock solid.
I use VoIP via my 3G service in my laptop regularly, via both SIP and (when forced, reluctantly) Skype. It's pretty darn solid; the only issues are VERY occasional quality drops due to latency spikes.
With a 1GB per month data allowance (for a wallet-smashing $15 per month ... so, about the price of a decent lunch) I can get a lot done. My carrier, Three (Hutchison), is the best priced data carrier in Australia, but Vodafone and Optus aren't too much worse and they have much better coverage, so this is hardly unique.
So ... if your 3G service sucks, it's because your carrier sucks, not because the technology does. Unfortunately, it looks like carriers DO suck in the US and the UK, though for different reasons.
In the US, you get hardware you've bought and paid for but is locked down so hard you can barely breathe next to it. Want to install your own apps? Better pay to unlock that feature. Want to use bluetooth/wifi tethering? Better get the "Internet" plan to unlock that feature. Want to use another provider's SIM with *YOUR* hardware, even after your contract has expired? Tough luck.
In the UK, it doesn't seem to be so much locked down as crap. Blocked and filtered up the wazoo, WAP-like transparent proxying and HTML/image reprocessing, private IPs handed out with all traffic through proxies or NAT, etc. Ick.
This will have to change ... but it's a carrier problem not a technology one.