Br0kenTeleph0n3

Following the broadband money

Lies, damn lies, and broadband data

with 27 comments

Is someone fiddling the broadband numbers, and if so, why?

INTUG, the international telecom users group, believes that officials in charge of Europe’s broadband roll-out are being misled by “vested interests”. It has written to Europe’s Digital Agenda boss Neelie Kroes, asking for the source of the claim that the UK has full coverage for basic broadband.

The claim was made in a Digital Agenda working document that said, “Denmark, Finland, France, Luxembourg, Latvia, Malta, Netherlands and the United Kingdom) have already achieved full coverage for basic broadband services”.

INTUG told Kroes, “Our UK Member, CMA (Communications Management Association), and the UK Federation of Small Businesses (FSB) met recently with OfCom and UK government representatives …to discuss broadband roll out. It was clear from the discussion that full coverage for basic broadband is still far from being achieved in the UK.

“This misrepresentation of a reality has been a consistent concern of users in the UK since BT’s often quoted claim of 99.6% coverage of broadband, which was similarly misleading. We would be grateful if you would share with us your source for this information. We can then assist the NRA (national regulatory agency i.e. Ofcom) in ensuring that future statistics on this sensitive topic are reliable and not unduly skewed by vested interests.”

This is not the first time questions have been raised about broadband claims. Announcing a 4G mobile technology trial in Cumbria this week, Everything Everywhere CEO Olaf Swantree said, “New independent research shows that one in five households in Britain could depend on 4G for superfast broadband in the coming years.”

Br0kenTeleph0n3 revealed in March that BDUK’s coverage maps exclude wireless coverage, even wireless networks that local councils are using today.

Sources say that BDUK is also worried about the accuracy of the maps of physical infrastructure such as street cabinets. Apparently BDUK’s maps show cabinets that are often 1km out and one is known to be 5km out. This, of course, drastically affect the ‘reach’ of BT’s Infinity broadband service, which one source put it, “goes 1300m with a following wind, and broadband itself which dies at around 5km”.

A further problem is that only Openreach engineers can tell the actual line length between a street cabinet and a subscriber’s house, and the information in BT’s asset database does not always reflect the actual situation.

These inaccuracies mean local councils may be making investment decisions using information that is wrong and potentially misleading.

We understand that BDUK is at last starting to worry about these issues as it prepares to release almost £400m to local councils to spend upgrading the networks in their “final third” areas.

Advertisements

Written by Br0kenTeleph0n3

2012/05/04 at 08:02

27 Responses

Subscribe to comments with RSS.

  1. Even though the ageing PSTN infrastructure must be near the end of of its economic life, it is understandable that the incumbents are approaching this from a strictly commercial viewpoint, rather than from a due diligence aspect of the Governemnt’s and BDUK’s view as a vital national infrastructure.

    The BT Press release for Surrey DC12-048 dated 30 March 2012 is clearly perpetuating their confusing statements:-

    http://www.btplc.com/news/articles/showarticle.cfm?articleid=%7b908d390e-2896-4daa-86ab-669e7488634e%7d

    “…premises in Surrey with access to the high-speed technology to more than 230,000. And further major investment is under way; BT has announced plans for its super-fast fibre network to pass another 95,000 premises in the county by Spring 2013.”

    Perhaps BT could confirm the maximum number of services they are able to provide with the current deployments? To assist in this calculation we note from the public domain release of the BT Database dated 15 December 2011, covering their installation phases up to 9a / 9b, there are a total of 1,112 PCP green cabinets included in the FTTC deployment. This includes quite a number that are not to be provided with SuperFast equipment at all. Local observations around Guildford, Godalming and Cranleigh demonstrate that almost all of the provisioned cabinets are only equipped with 100 pair link cables and just a few with 200 pair link cables which appear to be the maximum capacity of those units. (Without very costly re-works.) This yields figures of 111,200 (or even 222,400 using the 200 figure everywhere) services in stark contrast to the 325,000 addition from the above press statement.

    Possibly the sting is in the tail of the press release as it states:-

    “BT’s deployment plans are subject to an acceptable environment for investment.”

    and, much more concerning, that the design is not fit for purpose, nor will it be available, everywhere. Some presumably will have to wait until after 2014 possibly using a lower performance solution:-

    “Due to the current network topography, and the economics of deployment, it is likely that some premises within the selected exchange areas will not initially be able to access fibre-based broadband. However, Openreach is actively looking at alternative solutions for these locations.”

    Unsurprisingly there is no detail of what solutions are being considered, when they might be available, what costs are involved nor the expected performance. Surely any alternative solution must exceed the sometimes very poor performances observed with the current deployments ?

    Surrey Hills

    2012/05/04 at 08:57

    • The devil is in the detail, and regional press invariably blindly reproduce the data.

      Openreach FTTC rarely enables all cabinets on an exchange, something we have highlighted on thinkbroadband previously and usually mention.

      To actually sit down and prove the BT figures would be a very long process.

      As for number of tie pairs between the two cabinets, Openreach has said that something like a 50% takeup is planned for phone lines connected to a cabinet (remember the upgrades are not automatic), and some will not to pay more, or their provider does not offer the service yet.

      In areas with high demand, there has been public reports of a second FTTC cabinet.

      Andrew Ferguson

      2012/05/04 at 09:26

      • Thanks for the comment, Andrew. Do you think the situation you describe is satisfactory if we are about to spend hundreds of millions of pounds upgrading the exchanges an dcabinets in “uneconomic” areas? If not, what should be done about it, and who should do it?

        Ian Grant

        2012/05/04 at 09:32

      • To answer “Thanks for the comment, Andrew. Do you think the situation you describe is satisfactory if we are about to spend hundreds of millions of pounds upgrading the exchanges an dcabinets in “uneconomic” areas? If not, what should be done about it, and who should do it?”

        Look at Lancashire, half the money is BT’s own, and will be same around UK so its reducing risk, not removing risk. So company is still left to allow some commercial realities to play.

        The only way to guarantee is to match 1:1 lines to VDSL port, which would be expensive, and not popular as each green cabinet may need two companion cabinets (some areas have people objecting to the extra cabs already) or going for a full FTTH deployment. Which will be a paid for option in 2013.

        This also means that the 10% who are NEVER going to upgrade or get internet access will have wasted ports.

        I would add that you can bet other countries in EU have the same issue, but not speaking Spanish, German or myriad of other languages hard to analyse the extent. Some people appear to think the physics of broadband and economics are a uniquely UK problem.

        As for the alternate solutions, I have asked and recently got a FTTC/P will be used, when asking about Exchange Only lines, i.e. a cabinet may be deployed for the long EO lines, and those close with no cabinet may get FTTP.

        The situation from Openreach is evolving as demand rises, and the ROI becomes clearer. In short if the 2/3rds shows good ROI, then in 2016/17 Openreach may have naturally extended coverage to 80%. The BDUK system is a way of shortening the timescale that is all.

        Andrew Ferguson

        2012/05/04 at 11:04

    • You have been told before about the connection between the number of customers and the link cables. Please don’t repeat it, it’s based on take up.

      Somerset

      2012/05/04 at 13:29

    • The reduced number of links has nothing to do with the cpacity of the network. All of the fibre equipped cabinets are fed by 4 fibre blown fibre bundles and the initial phase of installation only one or maybe two fibres are active, leaving two or more spare for future capacity (ie initial order take up). Each fibre is capable of supporting around 100 connections.

      G Talbot

      2013/11/06 at 13:59

      • Perhaps your pearls of wisdom emanate from Martlesham Heath but I can assure you that your description does not match the actual deployment in Ewhurst. The final distribution node (i.e. the aggregation point) terminates with an ongoing 7 tube outer as far as the first PCP joint pit where indeed a quad tube supplies the FTTC. The 7th tube hasn’t even been capped to keep it in good order for further use.

        However the first FTTC can only have two tubes available as another quad tube continues to the central village where again it splits into two quad tubes going north and south. These quad tubes can only have two tubes each available. When the fibre was blown it was in one tube only for each FTTC containing a single quad bundle. Those are terminated within the FTTC into a quad termination box with four pigtails but only one is used at present to connect to the MSAN even though in this ECI case there are seven ports on the card. Furthermore the design does not allow for any further future fibre distribution from the FTTC going elsewhere.

        I.e. although there are ample fibres within the distribution node, it will be quite impossible to distribute more without new road ducts (as the existing salt-glazed ducts are already in a very poor state and jammed full of other cables and tree roots. A quite different active cabinet for a fibre switch and patch panel would be necessary to provide P2P FTTP as others are already doing.

        Merrow Drover

        2013/11/07 at 01:09

  2. Are the teleco’s telling porkies?

    I don’t think so, they firmly believe the figures, and no-one has found a way of checking the data to the level some would like without it costing a fortune.

    The BT 99.6% figure, is based on all but 21 out of 5500 exchanges offering ADSL, and the number of lines were people have applied for broadband, but it failed to work. (Some 16 or is 18 of these have an alternative wireless service and are in the Highlands and Islands)

    Now as not everyone has yet ordered ADSL, and those with poor speed estimates may not even bother, the statistic is open to question. If you add the caveats I am sure people would understand it better.

    Also what is a basic broadband service? If the two way satellite is allowed in the definition, then majority of the EU is covered. If it goes down to 256 Kbps, for DSL then figure wont be that bad in the UK.

    If I had £5 for every poster on the thinkbroadband forums, who said they had slow broadband, and it turned out that you could get it to run faster with some tweaks to wiring, or they were misunderstanding the units involved, or they had bought a pile it high sell it cheap heavily congested service then I would be living in a mansion.

    There is a real case for going around to every home in rural areas and surveying them, and then going back to improve speeds for those that are slow. We see this on our maps, where people close to an exchange get slow speeds, but a person with an optimised setup can be km’s away and get better.

    Andrew Ferguson

    2012/05/04 at 09:33

    • Thanks, Andrew. You raise points about speed and contention. Should local authorities include in their requests for tender for rural upgrades target figures such as the EU’s 30Mbps/100Mbps targets?
      Also, would you be happy to provide here a list of things end users might do to optimise their service?

      Ian Grant

      2012/05/04 at 10:43

      • Not being privy to the confidential bid documents hard to say what they’ve told local authorities.

        http://www.thinkbroadband.com/guide/fibre-broadband.html has some guidance figures for distance to the cabinet, based on Ofcom line length research, and then cross referencing profile 17a VDSL2 graphs.

        The line lengths will vary from cabinet to cabinet, so beyond a chart like above, you are asking someone to sit down and work out the line lengths, which is not always correct anyway. In short adding to the costs of rollout.

        In terms of improving service, at the simplest is
        http://www.thinkbroadband.com/videos/iplate-installation.html

        http://www.thinkbroadband.com/faq/sections/radsl.html has a pleothra of information e.g. ring wire removal, IP profiles etc

        Most people just need someone to visit for 30 – 60 minutes to check speeds and sort out wiring. Of those complaining of poor speeds would bet a good third can be improved.

        Andrew Ferguson

        2012/05/04 at 10:55

      • Thanks Andrew. I’m sure some wil find that useful.
        I accept that broadband is a technical product masquerading as a consumer product, but still – do we really think people would pay to have someone optimise their broadband?
        I doubt it, which is why the comms industry needs to get busy creating a true (happy) consumer experience for broadband. Think iPod, plug & play simplicity, “it just works”, and similar nostrums.
        Just for now, let’s leave out any debate about the effect of traffic shaping or blocking access to Torrent servers etc 😉

        Ian Grant

        2012/05/04 at 11:17

      • In terms of experience and industry provide help, yes could do better would be teachers report, but if that meant 25p per month extra on the bill you can bet people would move to another cheapest provider.

        Now full fibre of course removes all the RFI problems, but costs a fortune to install.

        Andrew Ferguson

        2012/05/04 at 11:21

      • And yet the Swedes say FTTH can earn a 1.5 times return on investment after five years, even in rural areas. I believe the FTTH Council has also produced attractive numbers, at least for the usual timeframes we think of in the telco industry.
        Is it perhaps that the time-frames have shortened, that even infrastructure outsourcing is becoming so fashionable?

        Ian Grant

        2012/05/04 at 11:31

      • ROI on full fibre, well I pay Sky £7.50 a month for broadband, so over five years they’d get £450 to support me, connect me to local node and run their network etc. So unless someone is installing fibre for £50 hard to see that working out well.

        Am sure ROI can be demonstrated, but not so sure UK consumers will pay the £25 to £35 a month for it, plus the £100 to £250 setup fee. Yes I would, but would the millions chasing the lowest cost telephone and broadband?

        Remember UK fibre (apart from Virgin) is a wholesale network, so where profit from bundles can help ROI this is not the case for Openreach.

        Andrew Ferguson

        2012/05/04 at 11:39

      • Fair point. However, Openreach is by far BT’s most profitable division despite Ofcom’s regulation of its prices, and has been increasingly so in recent years.
        It looks to me as if BT is making its money “at the factory gate”, which suggests it is probably overcharging for wholesale products. Either that or the market is just incredibly competitive at the wholesale and retail level.
        Given that VM and Sky both report accelerating uptake of high speed bundles it would seem there’s still consumer money on the table. B4RN being able to raise cash for its 1Gbps FTTH dig suggests there is an appetite for FTTH. Perhaps it’s not being addressed correctly.

        Ian Grant

        2012/05/04 at 11:56

      • Sky has only offered fibre for a few weeks, so not sure what data you have seen.

        ADSL2+ demand is high, but that is price and unlimited usage allowance driven.

        Virgin Media, the doubling upgrades help to boost speeds, and while 30Meg is slowest it sells. Still many left of 10 or 20Meg at the moment, and even after current round of doubling most will be on 20 Meg.

        Again be wary of even statements to investors.

        Andrew Ferguson

        2012/05/04 at 12:12

    • I moved a router for someone from an extension socket to the master socket and the speed went from 600k to 3M! TBB maps are very revealing for the potential for speed increases.

      Somerset

      2012/05/04 at 13:37

  3. It is a worrying situation for the ‘bidders’ inside the BDUK scheme indeed, who have to work on possibly incorrect data, and it will impact directly on costs for these L Authorities when they try to implement the stunning ‘2mb bottom line’ so many will see in rural areas.

    In my area we have several folk, all ‘promised up to 4.5mb’ by the BT sales algorhythm, who are stuck on dial-up. Some at 8.5-9km line length with corresponding high line losses. To be fair to BT (!) Ian Livingstone did react to my complaint about this and since Christmas these connections are now being offered ‘up to 2.5mb’ (hoorah!) but are still not THAT impressed. If this sort of ‘data’ has gone into BDUK (from Analysis Mason’s survey?) ……………………The 2008 SEEDA survey paints a different picture. It puts a potential 92 of the 142 postcodes no my exchange alone into the ‘possibly sub 2mb’ arena. Cost that out?

    Add in the 70% wireless coverage at 40mb on the Isle of Wight that ‘Click4Internet’ provide, that BDUK do not ‘know about’ (or the IoW council ?apparently?), the bits of West Sussex where Kijoma provide wireless b/b and probably many, many other examples and …………….what is Jeremy’ H’s phone number? [perhaps a mobile would be a good idea :-))) ]

    Mike Phillips

    2012/05/04 at 10:42

    • The errors affect Openreach/BT just as much.

      Database errors on top of the engineer log reports being wrong will have built up over the years, talk to any long serving Openreach line engineer.

      What amazes me is that back of fag packet estimates we did in 2006 or so, surface in various consulting reports. At the end of the day no matter how expensive the consultant unless you survey every single premise and telephone line there will be exceptions, and errors.

      Are the errors within a tolerable level? Obviously they are not for those affected, but when providing county or country wide statistics that non-maths geeks can understand one ends up having to drop a myriad of caveats.

      Andrew Ferguson

      2012/05/04 at 11:09

      • “talk to any long serving Openreach line engineer.” – yes, Andrew, I have and I understand the issues there. We are faced, however, with a significant and limited amount of people’s money being spent on what could be incorrect data. What I am struggling with is to see how anyone is going to be able to ‘trust’ the promises of BT’s NGA roll-out and the quotes.that come in for the gap-filling at 2mb+ and 24mb+ in the BDUK plan. Will it, with this amount of money at stake, be acceptable for ‘E&OE’ to be the caveats on these returns with a shrug of the shoulders and a ‘that’s life’ attitude and will it then just be blamed on the ‘previous’ government in 2016?

        By the way, I am seeking an accurate definition of the BT ‘homes passed’ phrase. Can you define how a home is ‘passed’? Is it merely that it is in an NGA exchange area, or does 24mb+ need to be actually available at the front door ie – ‘today please, Mr Murphy’?

        Mike Phillips

        2012/05/04 at 12:04

      • Homes passed and NGA.

        Next Generation Access, break that done, and there is a BT diagram confirming this that NGA includes ADSL2+, since 20% of UK households (on most recent BT Wholesale figures) are still connected to a ADSL only/20 CN exchange. ADSL2+ is available at 83% from Sky and 90% from TalkTalk.

        So you need to talk technologies to avoid getting messy figures.

        A home is passed for ADSL2+ or ADSL if it is connected to a telephone exchange that offers those services. It becomes a red (part of 0.4%) if they order service and it fails to work, or line at a later date has ADSL removed due to it not being economic to repair. BT removed all distance limits in 2006, so will now try to run service, and record we have seen is around 11km.

        A home is passed by FTTC if its cabinet is on the deployment plans for FTTC. No distance limits apply, though at order the checker will give estimates and some providers do reject order, helps to keep advertised speed data higher.

        A home is passed by FTTP if the plan is to get fibre to within 25metre (check figure) and then do final bit when customer orders service.

        Don’t have source, but am pretty good on memory for figures, but the average is that 40 to 85% of homes on an exchange will get FTTC under the current 2/3rds roll-out.

        Virgin Media coverage is similar, i.e. if their cable passes your house you are in their 48% coverage, but on ordering if job will cost > £300 to connect they say no.

        Andrew Ferguson

        2012/05/04 at 12:21

  4. Thanks, Andrew, that’s a help. So it is not, as some folks hint, actually ‘capable of connection at 24mb+’.

    Mike Phillips

    2012/05/04 at 12:23

  5. I think we need to recognise we have a two-fold problem here. I do not think we should necessarily confuse the ‘commercial’ roll-out by BT and the problems facing LAs in the framework although they are inextricably interlinked, of course.

    Whatever we all think of the dreadfully antiquated FTTC system BT are installing and selling, one cannot level any blame at them for choosing a commercial appraisal on what and where and how.

    The problem from where I sit is that Local Authorities, local and national Politicians, and indeed the media are bedazzled by the words from BTOR about ‘we will do xxx%’ and we are ‘passing yyy homes’ and cannot see their way past the froth to what it will actually mean. BDUK have set many traps, willingly or unwillingly. The fact that EU/state funding CANNOT be applied in NGA black areas calls for absolute clarity in what makes a postcode or area ‘NGA Black’ (yes, I know what the EU say). I have yet to receive an answer from my local MP, Nick Herbert, raised some months ago, as to what ‘NGA Black’ means in real terms. As far as the BDUK plan is concerned, we need to know the actual number of premises that will be ABLE to receive 24mb+, the BDUK target after the BT deployment, not just which postcodes are ‘NGA Black or Grey’ because they have perhaps one cabinet with perhaps 100 pairs on an enabled exchange – and even then probably not able to deliver 24mb+ to all those 100 due to line length and quality. I have yet to find any ‘guaranteed’ bi-directional speed definitions associated with ‘NGA’ and indeed Andrew tells us that ADSL2+ qualifies as NGA. These are the drivers that will affect the success or otherwise of County and National plans when targetting the UK definition of HSB, which is where I am primarily focussed, given the limited budget of £530mill. In the case of many rural areas, just getting a minimum of 2mb to 100% of premises wit POT will take a lot out of the money, involving new cabinets (with new-dig ducting – hopefully not with just copper cables ….!).

    There are, to me, too many ‘black holes’ in the whole NGA Black scenario and how Counties are expected to produce a realistic budget if the data is not accurate……………………….? I fear many will (as I understand some have already) realise they cannot achieve the targets on the existing funding, and will either have to accept a possibly embarrassing political ‘defeat’ or stump up more cash – which will not be easy.

    Mike Phillips

    2012/05/06 at 10:40

  6. […] more over at ISP review and BrokenTeleph0n3. […]

  7. Latest TBB article http://www.thinkbroadband.com/news/5204-lse-calls-for-government-to-put-more-into-broadband-pot.html suggests that the ‘state investment’ may be about 50% of what is needed to bridge the gap between BT and HMG targets. It puts the shortfall at around £1100mill.

    Mike Phillips

    2012/05/07 at 14:13

  8. […] Ferguson, whose comments on our “Lies, damn lies and broadband data” story  have made an extensive contribution to our understanding of some of the more arcane […]


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: