2008-08-15»
bandwidth and storage and europe and america»After doing my reading into fiber in San Francisco, I’d learnt a couple of things: firstly, there’s a lot of fiber around, actually, and secondly, a lot of the fiber under me was owned by Astound, who bought it from RCN, San Francisco’s previously weak-as-tea competition to Comcast.
As it happened, he next day I got some flyers through the letter box from Astound offering 10Mbps for $60 a month. As I’ve been tottering along with 4Mbps/1Mbps for $55 with Comcast, I thought I should look into that.
Competition is a marvellous thing. Wherever in the US Comcast has been facing it, I discovered, they have been magically upping their rates to 16Mbps. Simply calling Comcast support and hinting I was going to shift caused them to mention this fact, and five minutes later, I’m running at around 20Mbps/3Mbps for $65 a month. Add to that the $170 terabyte turned up today, and I feel like I’ve just leapt up about twenty countries in some OECD chart. I guess what I should do now is call Astound and see if they will offer to move PAIX into my bedroom cupboard for $65.99.
Up until now, I’ve always assumed that the UK’s consumer bandwidth situation has been rather better than most of the US — a tidbit gleaned from smug Brit slashdotters, and envy-enducing reports from my friends about their DSL deal-shopping. The received opinion is that the US dropped the ball almost immediately after rolling out broadband, and was promptly outgunned by most of the rest of Europe, something that begging outside telcos for ISDN-level speeds in most of Silicon Valley confirmed for me.
Now, after spending a few minutes on Speedtest’s worldwide self-selected statistics, I’m not so sure. I was originally assuming there was some American-bias to the stats, but digging deeper that doesn’t seem to be the case. The US actually does pretty well compared to Europe (except for those bastards in Sweden, etc) these days.
One thing I’ve learnt is that nation-spanning preconceptions like this are often temporarily true, but not for half as long as they hang around. Pleasantly schadenfreuderish viewpoints have a lot of lag to them. Take mobile phone adoption in the US. When I first arrived here, the difference between US cellphone culture and the UK was stark, and I, like many foreign-media journalists, would frequently dine out on the gap. In 2000, you couldn’t actually consistently text people on other networks; nor would it be reasonable to expect a stranger to have a mobile phone at all.
Then, in I think about 2003, I was crunching some stats about the crappy cellphone penetration in the US for my European friends to gawp at. Instead of doing an us-and-them comparison, I did a time-based one. How far behind was the US chronologically from the UK? It turned out that the US had just crossed 50% of households owning a cellphone. Laughably small compared to 2003 Britain, where it was close to 80% (I am surely misremembering these stats, but bear with me). But nonetheless, pretty much exactly the same as 2000 Britain, my original basis for smugly lording it over the Yanks. America’s primitive phone culture was, it turned out, only as primitive as the futuristic super-advanced one I’d left three years ago.
Sure enough, when my family came to stay that year, my usual prattle about how Americans don’t have mobile phones like we do was swiftly undermined. “What are you talking about, Uncle?” said my annoyingly smart niece, “Look around you. They’ve all got mobiles.”. And they had, and my anecdotes were like drinking from yesterday’s half-empty, cigarette-filled party beers.
I think the same thing is happening with broadband. From 2000-2008 there was a much bigger consumer rollout in Europe than in the US, partly because of government-compelled competition in European telecom (and aren’t I a bad libertarian for even suggesting that), and some really terrible decisions both by business and regulators in the United States. But those differences are slowly closing out as both continents start reaching the limits of DSL and the current infrastructure.
I imagine folks will disagree, which is fine. I’m not entirely sure of the position myself. The real question is: how could we test this? Are the Speedtest stats enough on their own?