It’s a good idea on the face of it, allowing people to see where they’ll get the best broadband speeds, if you follow the link to the map, or look at the detailed data, it’s actually not very useful at all.
Nor, sadly, is the associated table of data, which you can download, or browse through on the website.
What the data tells you
The data tells you the overall performance, which is based on a combination of the other factors, the average broadband take-up, percentage of people receiving speeds below 2Mbits/second – which is the point at which it’s not much of a broadband service, and you’ll struggle with things like iPlayer – the availability of ‘superfast’ broadband, and the average modem sync speed.
The sync speed is sort of related to the throughput you’ll get; it’s the speed at which the broadband modem talks to the local exchange, and you’ll never get a faster connection than that.
You might, of course, get a much slower one – for example, if your ISP doesn’t have enough onward capacity, either from the exchange, or from elsewhere in their network, then you will be paying for a speed you will never get, or seldom in busy periods. But the data published by Ofcom doesn’t tell you that.
More missing information
In fact, this data is pretty basic; sure, it’s good to know if superfast broadband is available – and as the figures in each area rises, it’ll be trumpeted as proof that Ofcom, the government, or the industry is really doing wonderful lovely things for consumers.
But what would be more interesting is a detailed breakdown. For example, on many exchanges a system called ‘LLU’ or Local Loop Unbundling is in place, which allows competitors to BT to install their equipment in the telephone exchange and connect it to the wires that lead to the subscribers’ homes.
Some of this equipment is better than others, allowing some companies to provide faster access. Some exchanges have a wider range of providers in them, which may mean faster or – just as important to some – cheaper services are available. But the Ofcom data tells us nothing about the penetration of LLU, or how many exchanges have cheap as well as fast services available.
One of the biggest flaws, however, in my view, is that the data is broken down by ‘Adminstrative authority.’ The intention is probably that this should be useful to local government, so they can see the lie of the land in their area. But whether through laziness, sparseness of raw data, or incompetence, the actual presentation doesn’t help an awful lot.
Look at Hampshire, for example. This is a fairly rural county, with a few sizeable towns and cities. Both Southampton and Portsmouth have ‘unitary authorities’ which is term derived from how the local government is structured. And so they count as separate administrative regions, and each has its own entry in Ofcom’s table. Neither Basingstoke nor Winchester does, and so the data for both of them is simply lost amongst the ‘Hampshire’ entry.
The same is true in Cambridgeshire, and at it’s most extreme take a look at the top of the Ofcom map, and click the large area there, labelled ‘Highland.’ It’s a mostly rural area, and pretty sparsely populated. It does, however, contain the city of Inverness, which has a population of around 60,000 (2008 figure).
A report from the Highland council puts the population of their whole area at around 215,000 (2005 figure). Allowing for changes since those two figures, we can estimate that around a quarter of the population of this area is found in Inverness.
It doesn’t take much common sense to realise that broadband is often much better in rural areas, with more services available and faster speeds, thanks to proximity to the phone exchange. So just how much are those figures for the Highlands skewed by the fact that 25% of the population are in one place, with better connectivity that the rest of the huge area?
Want to know where to get fast broadband?
If one of the intentions of this data is to tell people where they might get fast broadband, it’s next to useless. Not all the data will be as skewed as it’s likely to be for the Highlands, but it’s still not much use. Edinburgh has the fastest broadband, the headline says. But which parts of Edinburgh? Will you get the same speed in a desirable townhouse in the New Town as you would in an industrial unit close to the airport?
There’s no way of knowing. Just as there’s no way of knowing anything about broadband in London, other than one figure which covers all the boroughs – making it next to useless for anyone who wants to know what experience they might get if they were to move.
And there are other reasons to want to know some of this data, too. For instance, high speed broadband is seen as essential to providing access to government services in future. One of the aims is that most interaction will be done online, allowing post offices and local council offices to be closed to save money.
But many services are used by those with lower incomes, so it would be useful to know if those services are equally available or if, for instance, the companies investing in providing faster broadband are concentrating on the more affluent areas, where they can make a better profit, and ignoring those poorer areas.
Knowing that sort of information could be vital for public policy; it could reveal a need to invest in infrastructure in some areas, or to offer subsidies to help make sure a digital divide doesn’t worsen, for example.
But Ofcom’s map tells you none of this. You’ll never know, from these simplistic figures, if the provision of broadband to the people of Hackney is as good as it is to those of Westminster or the City of London. You can’t find out if some cities have a good service, because they’re lumped with a huge rural area.
Focussing on the administrative boundaries chosen might have made sense to whoever put the data together, but it doesn’t make it very useful for anything other than grabbing a quick headline.