Pros and Cons of @Jeremy_Hunt Superfast Broadband Strategy Document #digitalbritain

DCMS Minister Jeremy Hunt has finally announced the government strategy for providing “superfast broadband” to the final third. I’ve read the speech, the press release and the 64 page strategy document and this is my interpretation of where it is all at.

The government has the laudable aim for the UK of having “the best superfast broadband network in Europe by 2015. Moreover gov is not letting the grass grow under its feet. We have already seen work progressing on the 4 Big Society projects (initially three but apparently none included BT so a fourth was added).

Another positive is that the Universal Service Commitment of 2Megs is being rolled into the “superfast” activity. The investment in an infrastructure to just provide 2 Megs is a waste of money.

The announcement talks of a ‘digital hub’ in every community by the end of this Parliament. This is great. You do however have to read between the lines to see what is going on.
The idea of a hub stems from the concept of the Digital Village Pump as is now installed in Ashby de la Launde and is being looked at for the Cumbrian Big Society project. This concept brings a high speed fibre connection into a community and allows for that connection to be used to connect to a variety of means of terminating to local end users.

There are however some worrying indicators. In today’s announcement there are constant references to BT together with “cabinets” and “fibre connectivity to the nearest exchange”. DCMS has also now confirmed that in saying digital hub they do indeed mean FTTC. BT has said that it intends to tender for each project covered by the £830m of funding made available for this activity and that it will match any government funding. On the face of it this might not sound like a bad thing. BT has said that such an arrangement would allow it to extend superfast broadband reach to 90% or more of the population.

The real issue is something that Jeremy Hunt alluded to unwittingly in his speech in saying

  • “…unless you take extraordinary risks, you won’t survive in the digital world. I want our broadband infrastructure to make it possible for our entrepreneurs and investors to take those risks.”

It would appear that the government is taking a safe, non-risk based option here. The signs are that it is lining up BT to provide the digital hubs into these communities. Superfast broadband to 90% of the population would get the UK a long way towards Jeremy Hunt’s stated objective.

So is this a bad thing we have to ask ourselves? The problem is that BT is not a company that is going to take risks. BT is also too big to be able to innovate. Everything BT does has to scale, which is one of the reasons that the government will inevitably want to partner with it. In this case however scale = inflexibility and lack of innovation.

If, as reading between the lines suggests, we are going to see FTTC as the solution for the final third this has the following issues:

  • Once FTTC is in that is it. The end user will be stuck with a copper based solution for a long time to come. BT has said that it won’t be upgrading users to FTTP if they already have FTTC. Note that the residents of Ashby de la Launde already enjoy 100Mbps symmetrical FTTP with an upgrade path if necessary. My own view is that 100Mbps symmetrical is the minimum standard we should be aiming for. This is supposed to be a long term investment.
  • BT does not currently allow competitors access to its cabinets to connect their own services. This will prevent innovative communities and service providers from providing cost effective solutions to that last 10% that still wouldn’t be getting FTTC. BT’s preferred solution for this 10% is a copper based BET technology that facilitates the government’s 2Meg USC.
  • Even if competitors were allowed cabinet access, the backhaul for FTTC is expensive – on a wholesale basis up to 3 x the cost per Megabit as putting in your own fibre backhaul.
  • The government would effectively be extending to BT a monopoly status in these areas – something that successive governments have been working hard to erode – to the great benefit of UK plc it might be added.

It seems fairly clear to me that BT will probably win the majority of tenders. For one thing today’s strategy document effectively hands it to them because the government has said that it does not see any reason to change the way fibre rates are calculated.

  • “First, that the decisions of the Valuation Office Agency are made independently of ministers. It is not our role to decide who is liable for what under the business rates regime. Second, that the existing rates regime has been tested in court numerous times and no ruling has required any change to the regime. Third, that while in general we favour a low tax environment for new investment; it is right that non-domestic property should continue to be taxed to provide the essential public services we all rely on.”

This means that only BT is likely to be able to submit a competitive bid – all other network operators will be required to pay rates on their connectivity.

There are also other issues that weigh the scales in BT’s favour. Third party access to BTs poles and ducts has been mandated by Ofcom and we await a proposal from BT in January telling us how they are going to do this. BT’s most recent offer to NextGenUs (Ashby’s network operator) required them to use BT engineers (and consequential high labour rates & uncertain availability ) to do all the work. NextGenUs were also being quoted 21 days repair time for any problems. This is not a viable business situation. They would almost certainly repair their own problems within hours. It is very important that Ofcom negotiates hard with BT re this. Ofcom’s reputation in the industry for being another department of BT does not augur well.

If, as it appears, that BT is being lined up to take most of the cash available for NGA I can understand why the government is taking this approach. Let us not however delude ourselves into thinking that this is the best long term strategy for UK plc. This strategy is not an example of innovation and risk taking. It is anti competitive and is likely to be a step backwards from the progress of recent years. FTTP and true open access are the only sensible long term solutions.

Enjoy this article? Please share it with your friends.

share on Facebook share on LinkedIn share on Twitter share on Google59share on Comments
  1. agree totally tref. You got IT in one.
    Shame the councils and RDAs aren’t as savvy and will shortly hand our digital future over to the copper cabal.
    Great disappointment today for next generation access. ie we aren’t getting it. for decades. All this work will be to do again in another few years. You would have thought government would learn from the 2003 fiasco when BT took public funding and promised ADSL to everyone. Well we didn’t get it then, and the digital divide will grow even wider if they are allowed to do FTTC everywhere. Its a scandal just waiting to happen…
    chris

  2. Paul Browning

    One other problem with a FTTC based solution, what happens to customers who don’t connect via a cabinet ?

    There are plenty of people who either live so close to the exchange that they connect directly or live in older areas where cabs were never installed that BT seems to be deliberately ignoring.

  3. Neil

    The real issues are profiting from a dominant market position to castrate competition (trample on the small cost-effective suppliers), imposition of a limited technology (FTTC) and restraint of trade (backhaul costs, pole and cabinet access) – all well-known matters which OfCom should have dealt with long ago and must now, quickly.
    But, customer requirements are likely to be satisfied for the foreseeable future if everyone gets broadband in the 8 – 40 Mbps range at a price little different from at present. Is that unacceptable?
    What precisely has to be done to alleviate the worst constraints? Fixing the legal issues, enabling open access to cabinets and poles, and enabling competitive backhaul?

  4. Fastnow-slowlater

    “Note that the residents of Ashby de la Launde already enjoy 100Mbps symmetrical FTTP with an upgrade path if necessary. My own view is that 100Mbps symmetrical is the minimum standard we should be aiming for.”

    But that’s shared not per household, its not a fair comparison to FTTC. In fact FTTC would offer better speeds per household than the 100Mbps shared.

  5. Fastnow-slowlater

    And why would this be a BT monopoly, it will offer wholesale access. Something smaller telco’s can’t do, THAT would be a monopoly.

    If you know of another Telco that can put match the governments funding and rollout access to as many places as quoted and wholesale it then I hope they get a crack at some of this cash.

  6. Liam Crozier (Legolash2o)

    I agree with Fastnow-slowlater, it’s a shame there is not an option to install a fiber cable from your house to the FTTC cab.

    Wouldn’t it be possible to have an fiber cable going into a splitter within the FTTC cab, where if a home user wants to have FTTP they would effectively put a cable between the cab and the home.

    Is that possible?

  7. PhilT

    “BT does not currently allow competitors access to its cabinets to connect their own services.”

    what about sub loop unbundling, which does precisely that ??

  8. PhilT

    Liam – FTTC is a short term decision, it does nothing for subsequent FTTH it just puts an amount of connectivity (4 or 6 fibres ?) into a cab with a DSLAM in it and the terminations for the sub loop tie cable. Having roadside electronics and van rolls to connect up users is a bit poor for the 21C with our labour rates.

    The cabs are shipped pre-assembled from China, there’s no room to add anything (ie they are designed for their intended purpose). If you’re going to do FTTH economically on a large scale (as opposed to the odd isolated proof of concept) something like long range GPON from a greatly reduced number of phone exchanges would be nearer the mark.

  9. Somerset

    cyberdoyle – still going with the copper story? Please listen to professionals with experience in engineering and business.

    There is still no justification for 1G to a home. It sounds good and exciting but the requirements are not there. Fibre for new builds and long distances make sense but are you proposing the government funds FTTP everywhere? I’d be happy to get it but would I use it?

  10. There’s nothing hugely wrong with FTTC itself, it’s a fair technology and could be improved upon in the future as DSL tech’s adjusts to focus exclsuively on shorter copper runs from the cabinet rather than the exchange.

    However I agree with the other points regarding competition and believe that Ofcom is not doing enough to level the playing field and prevent anti-competitive practices. This also goes beyond the superfast debate and also stems into other areas, like SFI’s.

  11. Liam Crozier (Legolash2o)

    I think FTTC for villages would be good, in an ideal world FTTP everywhere would be ideal. Maybe BT would be able to invest more if they could be allowed to sell there old copper wires…

    Make telephones use VOIP and have maybe 2 AA rechargeable batteries in the modem so they can always make telephone calls in power cuts.

    The money should only be spent on fiber if they truly want a “future-proof” infrastructure. Maybe BT can make mini-exchange building in a village like the DVP for fiber connections to the homes, also with FTTP, BT could close exchanges which saves them money!

  12. Somerset

    Mark, I don’t understand the competition issue. Gojng with any company than a BT Openreach solution means you don’t have a choice of ISP.

    Liam – BT is looking at a FTTC/P solution with battery power in the router to keep it going during a power cut. This means copper from exchange to cabinet can be taken out. BT need to get Ofcom to agree to a solution.

    Surely the big issue for FTTP is digging up the roads and pavements in built up areas, if only some would understand.

  13. Liam Crozier (Legolash2o)

    Somerset i agree with what you said to Mark, BT don’t have to pay as much fibre tax AND they can provide it on a wholesale basis, providing choice.

    “BT is looking at a FTTC/P solution with battery power in the router to keep it going during a power cut”, do you have a link?

    Yeah digging up roads is a big issue, they could run the fibre through the existing ducts where a cab currently exists and then use poles? Our town has a lot of telephone poles, and i mean alot. They could put a splitter on the same pole as the old copper box (don’t know its technical name lol)

  14. Ted King

    Someone needs to tell your MPs that if they go along with this scheme HMG could be found in violation of EU human rights and competition laws. BT is currently receiving a subsidy from HMG in the form of reduced fiber taxes – which should have been ramped up over time as non-BT rates were ramped down (going for revenue neutrality).

    @ Somerset – “There is still no justification for 1G to a home.”
    I’ve used various data communication speeds from 1200 bps ear-muffs (late 1970s) thru T-1 (late 1990s) and on up to today’s various fat pipes. SPEED MATTERS ! Whether you are a student, a knowledge worker, or just having fun the speed of your connection makes a HUGE difference.

    There may not be a need for that fat a pipe right this instant but the planning for the backhaul to support something close to those speeds should be on the shelf with funding being pursued. I see the lack of widespread DekaMega (ten megabit/sec symmetric) at a reasonable price (under $50 US) in Europe and North America as something amounting to criminal neglect and economic sabotage by the major telcos. How can a decent education and cost effective health care be delivered without fat pipes ? Do we want to warehouse ourselves next door to the hospital ?

    P.S. I was recently offered $15/mo. USC-class DSL by AT+T. I nearly died laughing while reading the offer because at my current spending level (~ $50/mo.) for POTS and dial-up (56Kbps) ISP I should be getting flat-rate ISDN (128 Kbps) and either Wi-Fi access while roaming or USC-class DSL. I also suspect that the price will go way up after the one year term is over. (Fool me once [admin. fee on un-used long-distance service] – shame on you, …)

  15. Somerset

    Ted – I have no issue with fibre for all new builds and serving distant areas. People here are questioning the cost of effectively replacing the complete copper local end.

    What bandwidth is needed for education and health care?

  16. Ted King

    I would say DekaMega is the minimum that should be rolling out RIGHT NOW on a national basis. This allows for two-way streaming video, basic voice, remote telemetry, and leaves some slack for unforeseen needs. Part of the problem is that both the incumbent telcos and the cablecos only want to offer asymmetric products to residential customers. If you need a faster upload speed they want you to buy a business product at flesh-kilo prices (Shylock updated). This mind-set hammers those people who either wish to work from home (both home business and remote staff) or are temporarily home-bound due to a health problem (broken leg, pneumonia, etc.).

    By the way, I’m not saying that the old copper plant should be jettisoned in a hurry. Instead, we need to wean ourselves off of the copper teat. There needs to be a bridge option where the fiber link to a BT cabinet can be pooled with other fiber backhaul so that if one link dies then the other link(s) can provide some amount of bandwidth. [Remember those news articles that describe the havoc caused by a broken cable or exchange (backhoe, fire, flood, etc.) ?]

    Things like a “Digital Village Pump” that offer both Wi-Fi NOW and a fan-out point for FTTP as funds become available are essential. Unfortunately, most of BT’s competitors are too small to have enough crews and enough financial leverage to get the job done in a reasonable time frame. So I suspect that the best thing to do for now is to give a dinosaur like BT one small (HMG should reserve slots for competitors) bite at the DVP / backhaul apple CONDITIONALLY. The conditions would be :
    1) An FTTC cabinet is NOT the same as a DVP; and
    2) If they drag their feet then disqualify them from future projects.
    I personally do not want a fiber repeat of ISDN’s strangulation by over-charging. That’s part of why I remain on POTS and dial-up – the prices on better service do not make economic sense for me, the customer (not enough bits for the bucks).

    NB – I don’t work for a telco / cableco / ISP nor an advocacy group. I’m just a long-time techie with data communications experience (sysop, field tech., and sysadmin).

  17. Somerset

    What’s DekaMega???!

  18. Trefor Davies tref

    Thanks Ted that is very insightful. At the very minimum whoever lays fibre must make it available to others to use and not just allow others to resell FTTC.

  19. this post is about the most interesting ont’net at the moment, well done Ted, wake those dinosaurs up and well done Tref for giving us a platform.
    We really need to get this show on the road, and break this copper stranglehold which is stifling innovation.

    As has been said before, get the fibre out to the rurals and market forces will deliver in cities.
    chris

  20. Somerset

    Fibre is just a connection from A to B with nothing on the end. We must define exactly what is meant by ‘laid fibre’ and ‘available to others’. It’s the detail that is important and I think it’s currently missing.

  21. Somerset

    cyberdoyle – what innovation is limited by existing broadband connections – forget the transmission medium for a moment, if you can.

  22. Trefor Davies tref

    Somerset – that’s a fair point. If you look at the Ashby de la Launde roll out NextGenUs was able to terminate onto an existing fibre run in Digby. If this was available for each POP under the government’s strategy then that could sort many of the concerns. This doesn’t sound like it could happen from an FTTC cabinet so it might be necessary either to build a small brick building aka Ashby or have a second cabinet there. It would also be necessary to be able to negotiate backhaul costs as the WBC per meg bandwidth costs would be high.

  23. Liam Crozier (Legolash2o)

    The problem with building a ‘small brick building’ is that BT would have to make it quite big so….

    1) Other providers can provide LLU and need space for their equipment
    2) BT would need to fit big batteries in the building so the homes can use phones during power cuts

    BT need to be able to remove old copper if the property has fiber and ISPs should not be obsessed to full LLU even though full LLU makes it cheaper for customers.

    The copper they sell, can be used to go towards more fiber since copper these days is worth a lot more. In fact lots of people are now stealing BT copper cables lol lol.

  24. Somerset

    Tref – what was this ‘fiber run in Digby’? Are you suggesting that something like the BT ethernet POPs are provided (by someone and at some not inconsiderable cost) everywhere?

    Initially, maybe, these are only required in no and slow spots as it would be difficult to justify where FTTC/P is/will be happening. These areas need to be identified.

    How does an FTTC cabinet fit into this?

  25. Somerset

    Liam, assuming you can remove a complete cable as it may have ISDN or alarms etc, running over it.

    What’s the scrap value of 20m of dropwire?

  26. Trefor Davies tref

    No there was an existing pop in Digby to service a private business. This isn’t likely to be the same everywhere. However if someone is running a (publicly funded) fibre to an FTTC cab (the notional digital hub in the strategy doc) then the same cable will have multiple strands that can be used to service others. If the plan is to put a digital hub into every community it is a very small effort to turn this into the foundation of a Digital Village Pump.

    There is a slight caveat here in that if someone has already invested in an FTTC cab then the viability of a FTTP proposition will be reduced. Doesn’t mean that he option shouldn’t be there though.

  27. Somerset

    tref – we need a detailed diagram! Where would this (publicly funded) fibre go to and what would it connect to?

  28. Liam Crozier (Legolash2o)

    I agree, a detailed diagram would be nice.

    Somerset, ‘around £2500-£3000/ton for clean copper’, one site somewhere said £4,800 a tonne. Depends where you sell it.

  29. I think the difference in emphasis on “digital hubs” in the press release and in the strategy paper “Britain’s superfast broadband future” is rather strange. It’s the lead item and main focus in the press release, but only merits one paragraph in the strategy document, which does seem to suggest that we’re only talking about being able to make different types of connection to a streetside cabinet. It sounds like a kind of mad, fragmented version of SLU, with the wireless operator taking a connection, as well as communities digging their own fibre and asking to be connected up. I understand that the viability of this idea is still being explored, but a bit more detail would have been nice.

  30. Somerset

    Liam – calculation for 3km of 0.5mm diameter please!

  31. Somerset

    Liam – possibly £42, but I may be totally wrong. The answers 42!

  32. Liam Crozier (Legolash2o)

    I wouldn’t have a clue, it’s done by weight….

  33. Trefor Davies tref

    Somerset/Liam
    The government strategy is to fund/part fund digital hubs in communities. We are told that these are effectively FTTC cabinets which, by the nature of the beast, are on the end of fibre runs. There are thousands of these communities and I doubt the map/diag is there yet. Check out the thinkbroadband notspot map http://www.broadband-notspot.org.uk/coverage-map.html#5,55.354,-4.8,all,2

  34. Somerset

    Where are the fibres from the hubs going to? Do we assume the government funding pays for installation/build and the ongoing rental/maintenance/support is from the users?

  35. Trefor Davies tref

    Yes Somerset – that is the model

  36. Somerset

    tref – Where are the fibres from the hubs going to? Where is the connection to ‘the internet’ and where is the management?

    ps – the non spots map needs cleaning, some are next to exchanges!

  37. Ted King

    Re : Location of DVPs / fiber hubs –
    Why build a new building ? A typical village would have at least one of three prime candidates for a hub – a town hall, a church, or a pub. While access to the equipment might get a bit complicated the sense of local ownership and the use of a site that is a natural center of gravity increases the comfort level.

    “Where are the fibres from the hubs going to?”
    Basically to paying customers near the hub initially with possible discounts or early installation for investors (cash or sweat equity). Those people farther away than a few hundred meters may have to settle for either Wi-Fi or line-of-sight wireless as a stopgap. Keep in mind that Wi-Fi service near the hub should be offered as soon as the hub is hot. That gets people used to a somewhat fatter pipe and generates some revenue through either the “Internet Cafe” model or some sort of co-op service.

    “Where is the connection to ‘the internet’ and where is the management?”
    As I understand it there are three sorts of players : BT, new-age telcos (e.g. Rutland), and co-ops. If the hub is operated by BT then it is their headache. If the hub is operated by one of the other two then the backhaul has to find its way to a peering point and then be relayed on out into the internet. Keep in mind that this is TCP/IP – a fault-tolerant, fully automatic (after the cabling is in place and power turned on) way of connecting diverse networks. Once the peering agreements are in place and the port charges are paid then the system justs works. The management issue breaks down into a setup phase (peering, port assignment, initial connections) and a maintenance phase (add / drop customers, repairs). In my book the real bitch kitties are pricing and billing. Those would be handled by the hub operator (who may also own the hub).

    Management elements = traffic control (part contention, part account type) + customer (add / drop / repair) + outside world (peering, routing, regulatory, etc.)

    Caveat – The above are the opinions of an observer who’s been through various copper installs (POTS, analog leased line, DDS, T-1) from the customer side. I’ve been reading various articles about fiber installs in Europe and the United States but haven’t gotten hands-on experience yet. Corrections to any of the above are welcome.

  38. Somerset

    Town or village hall is ideal, council owned.

    I’m really interested to see how to get from the village hall to all the properties in the village…

  39. Trefor Davies tref

    THe above link is to a blog post about the Ashby Digital Village Pump. It has a photo of the DVP (behind indeed a civic building), a photo of the line of a trench dug down hte side of a field carrying fibre, and a photo of a fibre connection to the house. They laid 30kms of fibre. There is also wireless connectivity temporarily in place to reach some places.

    http://www.trefor.net/2010/11/13/the-digital-village-pump/

    I did do a video of a ride in a truck along the line of the fibre which if i can find will put on youtube and link to it.

  40. Samuel

    Oh man, what would we need “1G” for? For starters what is a “G”? I think the present way in which internet speeds are advertised is very confusing, what is Mbps? I bet you anything that in an office or a classroom if passing a file from one computer to another via the internet or USB people would ask how big the file was and be replied to “its 2MB” or “1.7GB” so whats all this bits nonsense?

    Internet speeds should be given in MB or GB, the latter will only come with FTTH/FTTP – even that acronym is started to get confusing! Fibre To The Home/Premisses.

    Now, what would I need it for: well 1080p video streaming right now can work, but it has to buffer. Youtube would take about seven minutes to buffer a 1080p, when FTTH would have it up and ready instantly, as if it were locally stored – which is much better! YouTube has also tried a proof of concept Super Hi-Def 4096×3072 (http://www.techexplorer.in/2010/07/youtube-now-supports-4096p-super-hd-videos.html) which won’t even play on my computer! but that could be a limit of my graphics card?

    Then there’s the weather, recent snow had many stuck at home where not only could they not get to work, but even out to the shops (or couldn’t face it), children also, a day off school is one thing, but a week! You see, people wan’t good jobs and why leave the house to commute and be stuck in a stuffy office when you can work from home, or run your own business. People can communicate with VoIP or, more likely in the near future, with 3D life size hD holograms, that’s going to be some strain on the connection!

    Then there’s the issue of a single connection, in our house we live as separate entities: its a multi-share and just today housemate got BT to install a separate line for him so he can play X-Box! The thing about that console is that it was introduced as raw power, so HD this and real-time that! only going to get worse – erm, I mean more demanding! Now imagine a household of two to three kids all pulling at the same time.

    Then there’s the connected home with HD cameras all monitoring your home, surveying everything from your Turkey’s from theft (big problem at Christmas!) to new born children or pets… are you going to want that in SD, no its going to all be HD and S-HD like 4096p! What about the connected home, everything talking to each other, devices that is.

    Then of course people forget about upload speeds, they always come as an afterthought. Well with Fire Optic connections the downs as good at the up, which you’ll need with the above scenarios.

    So you see there is already a requirement for high speeds, if you give more, people take more, services will crop up. Why do Flash websites still take time to load? because designers just pack more in, yet it feels like flash site always took time to load.

  41. Trefor Davies tref

    Here’s the video of me riding around the route of the fibre dig in Ashby. It isn’t always visible though for some stretches you can see the brown line where the grass has not yet grown back. It was a bit of a roller coaster ride and I make no apologies for the amateur nature of the film. You can imagine that provided you have hte cooperation of the landowners it is much quicker and easier to lay fibre in these rural areas.

    http://www.youtube.com/watch?v=SqBJDQ2J9Sw

    enjoy 🙂

  42. Fastnow-slowlater

    Ted , I’m not sure what you mean by this:-

    “Once the peering agreements are in place and the port charges are paid then the system justs works.”

    Connecting a backhaul to the Internet isn’t cheap and this is the bit I’m not getting with these “Digital pumps”. People keep throwing around great headline speeds for DVP’s like 100Mb up/down 1Gb up/down but that is shared between many many houses… not per household

    Such an idea will bring the internet to people who can’t currently get it I guess, but it won’t bring the next gen speeds “per household” as its simply not affordable.

  43. Trefor Davies tref

    Fast now
    You need to understand how contention works. Not everyone uses all their bandwidth all the time. Every adsl or fttc (or fttp for that matter) connection is contended.ie the backhaul bandwidth is shared.

    a gig backhaul would server 10k or more customers – you wouldn’t turn that much bandwidth on if you only had 500 or 1k customers – in that case it would be uneconomical and unnecessary.

  44. Fastnow-slowlater

    I know how contention works tref, I’m comparing this to FTTC. With BT Infinity I believe the backhaul to the exchange guarantees you 15Mb down. I don’t believe these pumps have the same guarantees. I mean its impossible to do so with a 100Mb backhaul to 180 homes for example.

    That’s what I’m getting at, I know bandwidth is shared and not everyone maxes out their connections.

    You say a gig backhaul would serve 10,000 customers or more. On FTTC it serves 266 or 288 (can’t remember which) the amount of FTTC users the cab can serve and if that minimum 15Mb cannot be guaranteed due to numbers they increase the backhaul to cope.

    Obviously handover to the ISP in question and their connection is another matter but I’m talking about the backhaul from the DVP or FTTC to the exchange/POP.

    The DVP version doesn’t seem to offer what FTTC does.

  45. Ted King

    I believe the difference is in the number of pairs that can be lit. FTTC only has room for a couple of pairs. A DVP / hub has more space and can light up more pairs or more easily upgrade the transceiver for a higher bit-rate. In some respects an FTTC cabinet is a sort of straight-jacket – the designers go through a lot trouble stuffing in the electronics while leaving room for the cables, air-flow, and the repairman’s tools.

    Also, speaking as someone who’s had to work in confined, dirty, and cold spaces on his employer’s
    data cables (co-ax, RS-232, bus+tag) I’m all in favor of a decent sized chamber that has elbow room and lots of mounting points (plywood and / or racks).

    On the point of bits vs. bytes I suggest switching to ISOs (disk images).
    Given : DekaMega (10 megabit/sec. symmetric, /10 to allow for overhead)
    Then : 60MB / minute
    __CD ISO (700MB = an OS disk for techies, lots of vacation snaps for regular people)
    ____under 12 mins.
    __DVD ISO (4.7GB = a basic movie)
    ____under 80 mins.

  46. Fastnow-slowlater

    I’m not sure how many pairs can be used on FTTC, even if it were only two (not so sure myself) that’s 20Gb of backhaul (if they wanted) for a single cabinet of 288 users. Sounds ok to me.

  47. Ted King

    On the surface ten (or twenty with two pairs) Gbps seems like plenty. But how big of a needle’s eye does that camel have to pass through back at the exchange ? This is something a network tech. checks for in any network design – where are the choke points ?

    PC – 100 Mbps / 1 Gbps
    Firewall – 100 Mbps / 1 Gbps
    CSU/DSU* – up varies; down 100 Mbps typical
    FTTC cabinet – up 10 Gbps; down capacity 28 Mbps avg. (35 Mbps raw less 20% for overhead)
    Exchange – ???????

    *Cust. Svc. Unit / Data Svc. Unit

    P.S. I’m in the habit from my teletype / ASCII async. days of dividing raw bit rates by ten (10) instead of by eight (8) to get bytes. This works pretty well even with today’s fat pipes. Also, when dealing with LAN’s be prepared for the software to take a chunk out of the capacity. My overhead on 100 Mb Ethernet with MSWindows in the loop was around thirty-five percent (35%).

  48. Somerset

    As tref says – 1G would serve 10k customers. This will be a quote from experience, despite what some may think, not everyone continually downloads data at their maximum connection time all day.

  49. Somerset

    ‘maximum connection speed all day’.

  50. Fastnow-slowlater

    Sure Somerset I’m not saying they would but I’m saying its a difference between FTTC and DVP. FTTC gives a guaranteed minimum throughput to the exchange (leaving aside the bandwidth that goes on from there) a DVP does not.

    Earlier comments on other articles suggested that heavy users actually get less of a priority.

  51. Somerset

    And what happens at the exchange? Why, oh why, don’t people know the whole story…

  52. Ted King

    “As tref says – 1G would serve 10k customers.”
    If you do the math the average capacity in that case is 100Kbps (hectakilo) – that’s basically one and a half voice calls (an ISDN B-channel is 64 Kbps). In other words Tref is being traditional and taking advantage of a statistical reality – some will be idle, some will be talking, and some will be surfing with few at a high level. But if there is a popular streaming video from the BBC or a sporting event then the gigabit port will be choked (unless somebody turns on multi-cast).

    FTTC vs. DVP
    There is no way that an FTTC cabinet can beat a DVP setup once you get away from a pure-voice model. The DVP is more flexible : multiple backhauls, sidehauls, and local processing (caching server, hosting services, etc.). What is a sidehaul ? What if several neighboring villages decide to connect village-to-village (DVP-to-DVP direct) ? This would be done to provide redundancy and cut costs on backhaul. This is part of the TCP/IP architecture which supports both hierarchical and ring style networks. And for those who think this is something new the standard exchange configuration for the old AT+T had load balancing logic that would route calls up one or more layers, sideways if needed, and then back down to the destination number.

  53. Somerset

    You can do all that in the exchange the cabinet connects to – if you want to. It’s all down to costs and benefits.

    Until there are detailed designs for a particular area we don’t know what best.

  54. Ted King

    Local control and competition. If Village-A is served by BT and Village-B is served by Virgin then doing it at the exchange would be a nightmare. Or if both are served by BT and it doesn’t want to offer the desired service at that exchange then the villages are out in the cold. But if the two villages form a co-op and get technical assistance from a small player like Rutland Telecom or NextGenUS then they, the co-op, control the features, costs, and timetable.

    Keep in mind that this is NOT rocket science. The only time the kit gets complicated is when it has to be shoved into one or two boxes and left out in the weather. A very small DVP might just use a SOHO eight port Ethernet switch that is gigabit capable. The setup could be a quarter sheet of plywood on the wall with eight boxes mounted – backhaul interface, Ethernet switch (one port open for testing), and six customer terminals. Wi-Fi would be a combination of customer anchors and mesh fill-in a la Meraki. It may look and sound complicated but it’s really a checklist driven by the features you want. A network planner earns his salt by knowing what to add for robustness and growth (battery back-up, a second sheet of plywood, etc.).

    So when a telco of any size tries to fob you off with “It’s too technical for you to understand.” that’s grounds for disqualifying them. Why ? Because either they don’t know what they are doing or they are in the process of overcharging you. Yes, the electronics are a horrendous mystery. But those black boxes are implementing some very simple, yet powerful, concepts. Getting one’s messages from A to F relies on packet switching principles worked out decades ago. [The British illustrator and cartoonist Thelwell would be a good source of illustrations (e.g. the impending meeting of two shepherds and their flocks as an example of collision and contention).]

    To those who are lurking –
    Ask questions and demand answers that make sense. Between Google and Wikipedia one can assemble a basic understanding quickly. You don’t need to spend big bucks on classes like the ones I’ve taken (Pacific Bell seminar on data comm., Zilog seminar on the Z80 family, college courses on programming, etc.). And don’t limit yourself to non-fiction – some of the books and films out there are very illustrative of the ideas behind modern networking. Seek and ye shall find.

  55. PhilT

    Openreach FTTC doesn’t have any *guaranteed” data rates, but the PIR is stated as 20Mbits/s downstream or the sync speed if lower. There are several fibres into a BT FTTC cabinet – either 6 or 8 – and they have 288 local loops of VDSL capacity.

  56. Somerset

    See this – http://tv.theiet.org/technology/communications/index.cfm – Fibre to Britain 2010, interesting discussions.

  57. Pingback: Read all about it – wheel invented – or a call upon the people of Lincoln to vote #deappg #FTTC | trefor.net

  58. Somerset

    PhilT – are these numbers of fibres installed or blown in as required?

  59. Pingback: @edvaizey @jeremy_hunt please read this #deappg #digitalbritain #fttc #fttp | trefor.net

Leave a Reply

Your email address will not be published. Required fields are marked *

*