The Royal Society of Edinburgh has taken a long hard look at the issue of how the BT project to improve broadband speeds across Scotland will fare. The conclusion is that in the rural areas of Scotland some 450,000 premises will not get superfast broadband, which is at odds with the Draft Budget Plan that has an aim of 95% superfast broadband by the end of 2017-2018.
Ignoring any calculations it is not 100% clear as to what the targets are for Scotland, when the contract with BT was signed in July 2013 the approved BT press release quoted "the new project will ensure that 85 per cent of Scottish properties have access to fibre broadband by the end of 2015 and around 95 per cent by the end of 2017" and in a very recent quote Nicola Sturgeon did not mention superfast in relation to the 95% figure. This makes judging any of the projects results so early as difficult, particularly as while the report is getting almost universal coverage saying the roll-out will be a disaster its optimistic estimate using a pure FTTC roll-out is that 85% will get superfast speeds.
In terms of the calculations we are possibly uniquely placed outside of people like Point Topic to know about some of the caveats and the dangers there in. Postcode data for the United Kingdom is fairly easy to source, but things start to fall apart when estimating line lengths from cabinets and the big unknown over what will happen to Exchange Only lines. Even the choice of 1.2km as a line length that will support 30 Mbps is open to question, as this will vary according to the gauge of the copper used in the local loop and the level of take-up and corresponding cross-talk. Plus it would appear the cabinet to postcode data is not fool proof, as where a postcode is served by two cabinets they split the premises in half, where some previous cabinet data leaked out actually gave the proportions.
|Upper limits for copper-based superfast||FTTC||FTTC+EO||Households left behind|
|Large Urban Areas||87%||92%||84,000|
|Other Urban Areas||89%||92%||60,000|
|Accessible small Towns||84%||92%||17,000|
|Remote Small Towns||80%||93%||4,600|
|Very Remote Small Towns||81%||90%||3,200|
|Very Remote Rural||18%||31%||62,400|
Link shows 366,500, but on checking figures we got the higher figure)
The big unknown is how much FTTP BT may deploy in Scotland, potentially FTTP used appropriately can solve the long line issue, the issue is the cost and hence why the FTTC areas are almost always addressed first. It should be remembered that any FTTP deployed as part of a BDUK project, includes the same wholesale install costs and line rental as the FTTC product, i.e. GEA-FTTP 80/20 product is no more expensive than the up to 80 Mbps FTTC service. Only the fibre on demand service carries the high install and monthly fee.
The study has ignored Virgin Media coverage which is available in Glasgow, Edinburgh, Livingston, Kirkcaldy, Dundee, Airdrie, Denny and Perth, Ofcom reports 37% of Scottish homes have the option of Virgin Media and this can add a crucial few percent to the availability of superfast broadband. Add in the unknown factors over Exchange Only lines and the flawed assumption that EO lines will be served from the exchange (reality is that clusters close to an exchange usually have the cabinet installed close to the exchange, but long distance EO clusters will most likely have the cabinet inserted deep into the network) and one can start to see the level of uncertainty climb.
So where do we stand, well the reality is such that even if 95% superfast coverage is reached, for those in the final 5% they are still missing out. The key thing is that people need to remember that when the politicians showcase 95% coverage they are not talking about 95% of rural areas or a postcode group like Kirkwall, but 95% of the whole of Scotland.
In short we would love to have published some of our figures on speed estimates, but the caveats and gaps in the information that meant large assumptions had to be made has meant while we will talk about the foot print of specific cabinets where we have verified its actual location we are not able to publish a UK figure with a high degree of confidence beyond the estimate based on cabinet line length data given to Ofcom many years ago, but where only the summary was available rather than the full 30 million or so line lengths.