For people who didn't read my post on the "introductions" forum, I'm a politician (boo! hiss!) for the Pirate Party UK (hooray?) and I'm trying to learn enough about the issues involved with rolling out broadband and the way it's been handled under BDUK (and other subsidy schemes) to write some articles about it, and maybe make some useful FOI requests.
I'm going to try and keep he politics out of this post, because I'm well aware that I *probably* have the wrong end of the stick about *some* of this stuff so I thought I'd check with you folks before making any dumb 'series of tubes'-type comments.
So, this is what I
think I have found out,
please dive in and tell me where I've got my facts wrong:
There are (or were) 3 different sets of government targets.
The Government set up BDUK to deliver
1) 2Mbps to 100% of the population, and 30Mbps to 90% of the population, by 2015. This has been dropped in favour of;
2) 2Mbps to 100% of the population, and 30Mbps to 95% of the population, by 2017.
and the EU wants
3) 30Mbps to 100% of the population and over 100Mbps to 50% by 2020.
These goals are arguably not all that useful because they don't specify any measure of upload speed, contention ratios or throttling, and none of them deliver what the majority of experts actually want, which is FTTH.
From a technical point of view, the things that need to be done to achieve the first two targets can roughly be split into 5 parts.
1) Getting '2Mbps for all' is mostly about dealing with long copper/aluminium lines that run from green cabinets to people's houses.
2) 30Mbps to 90% needs a combination of exchange upgrades and;
3) Installing new big green cabinets, generally called DSLAMs, next to the existing small ones called PCPs where exchanges have been upgraded and;
4) Running fibre to those cabinets through existing ducts, some of which my be blocked, and there's also the issue of;
5) Exchange only lines, for which there doesn't seem to be a solution at the moment.
If 2Mbps for all is done with copper not fibre, then some (all?) of that's going to have to be replaced in 3 years with at least FTTC in order to deliver on the 30Mbps for all EU target.
BT has pretty much finished announcing where it will be doing it's commercial FTTC roll-out, and BDUK funding goes to areas where there is no commercial solution. This isn't just as simple as going for the bits BT missed, because Virgin owns cables in some areas, and there are rural schemes (using a variety of technologies )in some places too.
Here's where is all gets a bit opaque, because there's a chinese wall between the public-facing bit of BT and BT's Openreach division (BTOR), who both keep a lot of information needed to know if BDUK are getting value for money secret because it's commercially sensitive, and because the contracts between BT and the 44 councils are confidential (although there seems to be some mechanism by which BDUK compare the deals?).
Random bits of technical knowledge that I've got from various forums that may be right or wrong:
a) DSLAM cabinets used by BT come in 2 sizes, 288 connections (of which sometimes only 144 are installed) and more rarely 96 connections?
b) 'not spots' in areas with upgraded exchanges happen because BT have decided it's not commercially viable to install a DSLAM cabinet in places with too few subscribers?
c) BT won't say exactly what number 'too few subscribers' is, but we might be able to work it out through anecdotal evidence - for example the row in Kensington and Chelsea was over 108 cabinets supplying 24,000 homes and businesses, so we know that 222 lines is economically viable because that's the minimum each cabined could have… assuming BT isn't using social profiling (disclaimer: I'm in a 'not spots', but a much less posh one. For this reason I've looked into not-spots a lot more than other issues, but I would like to understand the other stuff better too)?
d) Even if BT have 'done' an area and put a DSLAM in, it might not be possible to get FTTC there if the DSLAM is full and BT don't think it's viable to put another one next to it?
e) If you're in a 'not-spot', it's impossible to pay BTOR to put a DSLAM in?
f) BT are using a 20% take-up rate to judge viability?
g) The BDUK contracts are secret, but a well argued FOI request ought to be able to reveal them after a bit of a battle, because of the precedent set here:
https://www.whatdotheyknow.com/request/liverpool_direct_bt_contracts#comment-14142h) The number of lines per exchange is public knowledge, but the number of lines per cabinet is not?
Questions that I therefore think I ought to be asking (but I'd like to know if these really
are the right questions)
1) Is 100% 2Mbps going to be possible, given that speed is affected by building's internal wiring?
2) What is the magic number of lines that would makes a not-spot cabinet viable, if the exchange has been done and fibre already goes there?
3) Why is there disparity between per-cabinet costs for different councils? Why do BTOR not quote per metre of cable and per cabinet?
4) Is BDUK's 90% availability really satisfied if the infrastructure can only cope with 20% take-up?
5) Is anything BDUK are funding in to hit the 2017 targets going to need to be ripped out again and replaced to hit the 2020 targets?
6) Why is there a 6-9 month gap between a BDUK contract being signed and work actually starting? This is being put down to planning and agreeing costs between BT and councils, but surely BT must already know the costs to have deemed things not commercially viable, and the Councils must know them to have agreed the price?
7) is BDUK gap-funding or paying for everything? I.e. if something costs £1m, but would only make BT £900,000, are BDUK giving BT £100,000 or £1m to do it?
8 ) Shouldn't BTOR be required to quote for and accept upgrade orders from customers other than BT? This seems like a no-brainer to me, and I'd be interested to know why it wouldn't work… BDUK will have a list of the costs it was quoted but could't afford, so why not publish them and let not-spot villages have a whip-round, if they want to? This sounds obvious to me, it's not commercially sensitive info as it's not commercially viable work, it will get subsidies down for the next round, and BTOR will get more money because it's not going to be chipping it's portion of the BDUK money.
Sorry for the long post, there's a lot for a newbie to get their heads round when it comes to understanding this stuff!