I have attended a good number of meetings about broadband where a mixture of providers, politicians and lobbyists are present and by and large always find that they are lacking in actual technical content and some in the past have been told to dumb down responses when getting into the technical aspects (where technical can simply mean discussing upstream speeds). So to read the write up via an anonymous source in Computer Weekly for a meeting with alternate network providers to discuss the final 10% problem for rural broadband has me screaming at my screen, both for what appears to have done at the meeting and the subsequent coverage.

The DCMS, BT, politicians and campaigners are it appears stuck in a loop where there is a mystical final 10% for broadband across the UK that can be identified with one or two magical database queries. Alas given the widely varying targets that the local authorities have implemented the situation is far from simple, e.g. Surrey if it hits its target might only have 2% falling below the lowest accepted definition of superfast (a connection speed of 24 Mbps or faster), but at the other end of the spectrum Scotland has a target of 85% with option of fibre broadband by the end of 2015, rising to 95% in 2017 and Northamptonshire already has a 100% superfast target set for 2017. Now of course we cannot say for certain what speeds the people will get and there in lies a major problem, as people will enjoy speculating based on their local experience of broadband.

The Computer Weekly article appears to have been written from the viewpoint that the BDUK process has failed to deliver anything or will not deliver what is expected, i.e. the usual UK habit of everything is a disaster and or a botch job. My view is that it is far from perfect, but at least lets look at the data over the next couple of years before we lynch a politician over it.

We suspect that councils should be able to identify areas that will not be covered by the main superfast component of their project, but if they do this now before the full planning and roll-out it would mean that any claw back clauses to improve coverage (i.e. money handed back by BT due to commercial levels of take-up) would be effectively useless and would simply sit around until the date triggers where they are handed back to the council. The change in June 2013 when the £300m we already knew about was paraded as a new £250m investment by the Government and a push to get superfast coverage to 95% at some point in 2017 was also a moving of the goalposts and that has ramifications for any smaller alt-net type project, especially given there is talk of going even further.

I have voiced my view before that the only way to have avoided a lot of the current mess would have been to wait for the commercial roll-outs to complete and then assess what was left and only then consider spending public money. Market failure for the BDUK final third project has only ever been a paper tiger, hence the delays in the EU granting State Aid Approval. A delay to the main £530m BDUK project would have allowed alt-nets time to gain the foothold they want, and have made issues like the Open Market Review much easier to judge. It may have even meant that Openreach identified more areas where a commercial roll-out was viable, saving the public purse money.

If the BDUK process and EU State Aid Rules are boiled down to the ultimate conclusion, you have the situation where you are creating mini monopolies part funded with public money, this charge is commonly thrown at BT but it is just as correct for other firms using public money. BT is cast as the predatory vampire squid, but I suspect that any other operator of similar size would behave the same way.

So while we can voice moral outrage now at the idea that councils may want to wait till 2015 or later to be sure where their projects will be when they finish, this actually sounds like a reasonable solution if you want to avoid the risk of legal objections to various projects. Just maybe the DCMS should have franchised the likely most rural areas and back in 2011 said that these areas are outside the main project, but only available for RCBF type projects.

There are some people looking at the speeds of 15 Mbps being mentioned but this is where things get a bit more technical and a read of what the EU State Aid rules say is needed:

In assessing projects for State aid approval, the NCC requires that NGA technologies that are used in NGA white intervention areas must provide the same outputs as those defined for other established NGA network deployments. Specifically, the NCC will expect to see that the technical solution:

  • is capable of providing access speeds in excess of 30Mbps download, not only by reference to theory and technical standards, but also by evidence of calibrated performance measurements of an existing deployment within the area of interest or an demonstrably equivalent deployment in a similar geographical environment;
  • typically provides at least a doubling of average access speeds in the target NGA intervention area;
  • must be designed in anticipation of providing at least ~15Mbps download speed to end-users for 90% of the time during peak times in the target intervention area, as demonstrated by industry-standardised or reliable independent measurements;
  • must show how the solution would adapt to maintain capability and end-user experience in changes to key parameters such as increased take-up and increased demand for capacity, and be able to show using clear calculations that this is both technically and commercially viable;
  • must have characteristics (e.g. latency, jitter) that enable advanced services to be delivered e.g. video-conferencing and High Definition video streaming to be provided to end users as evidenced by trials results not necessary obtained within the area of interest; and
  • have longevity such that one might reasonably expect increases in performance within the next 7 years.
Guidance notes for the role of Next Generation Access technologies in addressing superfast broadband market failure under the UK’s State aid scheme

 

This seems pretty clear, in that the 15 Mbps 90% of the time, is referring to peak time speeds, or in layman’s terms a measure of congestion. Thus it avoids a bidder for a project delivering a connection speed of 140 Mbps to every home but once 1,000 homes are connected at peak times people can only get dial-up type speeds because they only installed 250 Mbps of backhaul.

The underlying need for connections to provide speeds in excess of 30 Mbps to qualify as superfast intervention still applies. Though there are still some projects operating to a 24 Mbps definition for superfast, but this is generally those where no ERDF funding is used.

The problem now is that a good number of people will latch onto this 15 Mbps figure and claim this is all that the project will deliver, the only way this would happen is if every cabinet rolled out by Openreach as part of the BDUK contracts was in a field on its one with the nearest property 1km away.

I do not think the Openreach roll-out is perfect but for the money the public purse is spending it is difficult to see what we could have realistically got better. The Fujitsu bid was if to be believed FTTH to a coverage level of 80%, which means just 7% of premises given current commercial levels, with the remaining 20% using a mixture of alternative technologies.

At the end of the day we have councils that they may be struggling to cope with the data onslaught that mapping the projects involves, this is not as an insult to the councils simply a fact that it takes time and resources to handle mapping tasks - this is something that the BDUK could have helped with rather than providing a near useless website that still really is based around a couple of spread sheets shared via Google.

The current stand-off was solvable but only if the BDUK had insisted on contract bidders at the time of tendering giving an absolute commitment on what speeds individual postcodes or parish councils would receive and was contractually obliged to hit those targets. Now with the contracts signed and locked away, a change like this will have BT (and it would have any other provider) asking for time or more money to actually take on the work. Given that at the Public Accounts Committee we heard that other potential bidders were dropping out due to other conditions already imposed.

The technology that would be the perfect solution and we all know what it is – Fibre to the Premises suffers from the cost of delivery and as the politicians wanted something they could parade in the 2015 General Election there was never the money or time for millions of homes to be connected using FTTP.

 

 

9 Responses


  1. PhilT on 08 Oct 2013

    “BDUK’s National Competency Centre (“NCC”) requires that …” in case anyone else wondered who NCC was.

    Worth noting that Altnet’s had about 10 years to fill in BT ADSL notspots and in general did not do so.

    The 15Mbits/s 90% of the time should be a useful protection against willy-waving wireless schemes with a total capacity well below 100 Mbits/s due to spectrum and bit rate limitations.

  2. JohnS on 08 Oct 2013

    Good article but what a response from PhilT! Protection against wireless schemes? I’ve been working with wireless altnets for 15 years and they’ve connected tens of thousands of users in some of the hardest to reach areas of the UK, including Shetland, Western Isles, Lake District etc. They’ve been delivering NGA speeds since before the term was even dreamt up. As the DCMS identified yesterday at the F10 event the key is backhaul and WISPs can beam it in where it’s needed and are not restricted to the design of an aged PSTN network. To be so dismissive displays a lack of understanding of what WISPs can and do actually achieve now, without any public money. 15Mbps for 90% of the time would be no issue for most WISPs.

  3. gerarda on 09 Oct 2013

    “Worth noting that Altnet’s had about 10 years to fill in BT ADSL notspots and in general did not do so”

    Many attempted to do so – but in the absence of a universal service obligation BT were allowed to cherry pick leaving the areas to be covered unviable even for community funded projects.

    • PhilT on 14 Oct 2013

      BT enabled virtually every exchange with ADSL and did nothing about the notspots, just like the Altnets didn;t for 10 years.

      I don;t see what BT cherrypicked – where there lines/exchanges were ?

      • gerarda on 14 Oct 2013

        If you spend a few minutes googling you will find dozens of local schemes that were rendered unviable by BT enabling exchanges without any commitment to give 100% coverage in that area. A USO would have meant either BT would have to address the notspots before enabling or not enable allowing an altnet access to sufficient potential subscribers to be potentially viable

  4. gah789 on 09 Oct 2013

    Both the article and comments largely miss the major constraint on the development of community networks, which is the availability and cost of backhaul. Altnets could not fill in ADSL notspots because you had to have 500+ customers to cover the cost of a leased line and the risks involved in contracting capacity in advance of proven demand were too great.

    The costs of backhaul have fallen but current use patterns mean that systems with less than 200 customers will struggle to be viable if they rely on a BT leased line. I am familiar with many small communities with less than 100 potential customers and broadband speeds of less than 0.5 Mbps because they are so dispersed. It doesn’t matter what the wireless system can deliver if the cost of backhaul is going to absorb all of your revenues – or monthly charges are out of reach of the majority of your customers.

    This is not the final 10% – more like the final 2% – but it is still more than 500,000 premises over the UK. You can’t put all of them on satellite as there isn’t sufficient capacity. Anyone who thinks that 3G/4G is a viable alternative has simply never looked at the economics of building and operating the masts required – and has forgotten that masts require backhaul too.

    So, whatever happens any serious attempt to honour the USC has got to address the backhaul issue as the number 1 barrier to any solution.

    • PhilT on 14 Oct 2013

      Personally I see spectrum as a bigger constraint on wireless than backhaul.

      I don’t see traditional “leased lines” as viable backhaul these days, Ashby-de-laLaunde uses an Openreach fibre ethernet circuit (BES / WES type thing) from an OLO who sound “inexperienced”.

      Are we saying they are too expensive too ?

  5. Dragon1945 on 14 Oct 2013

    I was getting 2.1 MB back in April. Now it is 1.79 MB. I need 4MB to access a lot of services, but I am too far from the Exchange. Instead of putting fibre in to outlying areas, the houses local to the Exchange, with good BB speed already, were treated to Fibre. My Talktalk contract says “up to 8MB”. The houses near the Exchange were well in excess of that before Fibre, so when am I getting the same service that we all pay for when I can’t even get 2MB? When I complained I was first told I had to pay £50 for an engineer to make a visit. Then told in the circumstances it would be free. However , if he wasn’t coming to install Fibre from the Exchange to replace 50 year old copper cable there wasn’t much point. Imagine my amazement when Talktalk rang me recently asking if I would like Fibre for an extra £10 a month. According to their maps we have fibre installed outside the garden. Underground cable. Funny that. None of us has seen any work done here since the BT Engineers found Cave Spiders and refused to work on the cables, and none of the residents here have Fibre from any of the companies we use. Maybe I should have acted ignorant, paid for the non existant Fibre, then sued them.

    • andrew on 14 Oct 2013

      Without knowing any details on your location hard to say, but can give a general overview.

      Generally the fibre services are installed to a green street cabinet, which is within 100m to 1km of your property. Nothing to do with fibre passing your home in the street. For 0.5% of the UK it is the other way around, i.e. you actually get a fibre from the street all the way into your home.

      TalkTalk currently Only sells the cabinet version at present, so the question is whether your street cabinet has or has not been upgraded and putting your FULL address into the checker at https://www.btwholesale.com/includes/adsl/main.html should tell you what services are available at the wholesale level.

      On those near the exchange getting the upgrade, it is more complex, as it all depends on where the cabinets are located for your exchange. Some exchanges have those on faster ADSL2+ speeds not seeing anything better, so it is down to personal experience.

      On the current speeds, it is always worth checking your current line speeds to see if they are under performing. Do this by taking router stats from the modem i.e. http://www.coolwebhome.co.uk/stats/routers.html#hg532 and using a calculator like http://www.coolwebhome.co.uk/calc/ to see if they are roughly right.


Leave your comment