If you live in the UK and order a new BT phone line, from January 2006 it won't be a BT engineer driving up in a BT van to install it for you. And if you're a customer of one of the many other phone companies, you won't have that slightly disconcerting experience of finding a representative of the incumbent doing the installation work.
Well, ultimately it will still be BT, but as from mid-January the vans and the engineers and indeed the local network itself will all be part of a new BT subsidiary, Openreach.
It's a curious and bold creation, offered by BT to the UK regulator, Ofcom, as a solution to competitors' fears that BT's local loop business was too closely integrated with the rest of BT, and that this unfairly favoured BT's business over those of its competitors, which still rely on the incumbent to connect its customers.
So BT came up with the proposal, accepted by Ofcom, to split off the whole of its local loop operation, from the distribution frame in the exchanges down to the connection into end users' premises, into Openreach.
Steve Robertson (above), CEO of Openreach, recognises he has a challenge. Openreach has been carved out of BT Wholesale, which at the same time — see previous pages — is embarking on the bold 21st Century Network plan. "We're down the bottom of the value chain looking up with great interest, and it's all those folks who have got the smart sexy stuff to do that are going to drive this."
But it's obvious that 21CN and all the other developments in the industry are driving demand for new services and for bandwidth, and Robertson's new operation will be responsible for delivering all that to the end users, on behalf of BT and other operators. "Openreach is a child of the broader market dynamic" which is transforming the industry, he says.
Equivalence for operators
His first task is to provide "a level playing field" for his customers — which are not the end users but the operators. "As an industry we get wedded to jargon so we say we're going to create equivalence of input systems. What the heck does that mean? Everybody who buys products from Openreach does so in the same way."
So when BT Retail orders connections from Openreach from an exchange to its customers it will not get priority over, say, Cable & Wireless. "Coming from an environment where we've been part of a vertically integrated telco that does provide challenges," he admits.
"Everybody gets the same provisioning systems, the same repair systems, the same access to our engineers, and so on. It is a big change for us. It's also a big change for all our service providers."
But one of the questions that's not clear — probably still is not to Robertson and his colleagues at Openreach — is how exactly the company will be driven. It has inherited an infrastructure that includes optical fibre into businesses and copper pairs under footpaths and streets. Some of it is new, but much of it will need renewing at the demand for bandwidth increases.
"The challenge that's facing us inside Openreach is that this infrastructure needs to be fit for purpose for all those wonderful new services," says Robertson.
"Our first priority has to be delivering service and a foundation stone of that has to be improving the quality of our infrastructure. What does that mean? We need to look at can we have fibre to the cabinet or not. There are all kinds of engineering decisions that have to be made that will generate capital programmes."
He recognises that this will be a challenge. "The prioritisation of how we spend money and where things are to be done has to be done in conjunction with our customers, with the business. There's a lot more work we need to do in that respect. How do we productise our services?"
Meanwhile there's an immediate priority: "We need to minimise loss because the demand for greater and greater speeds will be with us for a long time," says Robertson. In addition "we need to improve reliability, we need to improve our repair performance, we need to improve our provisioning performance, we need to improve the way that we bond electronically with our customers and our suppliers , and there's a lot of stuff that needs to be done that's right at the top of the agenda."
Openreach is at the very early stages, he readily admits. "It feels like we've come a long way, but when you look at the future landscape we've got a huge amount of work to do."
What happens in Openreach's access network has to relate to the core networks and to content demands and the needs of the end users. "If that chain gets broken at a time when we're all moving very quickly and there's lots of new stuff happening then it can be dangerous for the industry."
Openreach will need to have privileged access to the plans of all communications providers. "There are going to be lots of people investing lots of money to create new businesses and we are supplying almost all of them." They will need to trust Openreach.
The company will not have a public face. "We're going to have to work with communications providers to make sure that every person knows that when you see an Openreach van and there's somebody knocking on the door they know who it is. We've got to contribute to this but I do not want Openreach to be the name at the front of people's minds. I want people to know they are an AOL customer, a Wanadoo customer, a Cable & Wireless customer, a BT customer."
The demarcation question
The demarcation between Openreach's network and the networks of BT and its competitors appears fairly clear at the moment. "In terms of the infrastructure it's the first optical distribution frame or the main distribution frame in the local exchange. The ownership of all parts of the infrastructure from there out to the end user is Openreach, including the frame itself," says Robertson.
But it's less clear what happens as broadband advances. If fibre moves out into the access network, with DSLAMs on the kerb delivering many megabits to customers down just a few metres of copper, where will the boundary be then?
"I honestly don't know the answer to that," says Robertson. "It's a fantastic question, because clearly it's a very difficult question to answer because I don't think there have been any conclusions drawn about that."
Ofcom's strategic review in 2005, which led to the creation of Openreach, "is deliberately fuzzy around that, because it's a complex question to answer", he says. "There isn't an answer to that yet and it's something we'll be working on with our customers and with Ofcom and with the rest of BT. Clearly it's important."
And if operators decided to use WiMax, or another wireless technology, for the local loop, would that be Openreach's? Robertson pauses before saying: "That's an interesting question. At the moment we are focussing very much on the copper infrastructure and the fibre infrastructure because our plate is full in terms of stuff to do. We've looked at this and we don't think that it's going to be an issue for us in the next few years."
But "in practical terms" if there were a move towards wireless distribution, "it's almost inconceivable that anything like that could happen without Openreach being central to it in some way", he says.
New access technologies
"Would we own the infrastructure? I don't know. We've got the engineering capability and the weight of engineering presence that you would need in that sort of environment. I think it's unlikely that anything radical in the access space that was going to be a mass application of new access technology wouldn't somehow touch Openreach. The question is in what way? Do you own the asset or do you not own the asset?" It's not an immediate problem, fortunately, but he's aware that it might become one.
In the immediate future, Openreach is responsible for ensuring that broadband signals get through. Sometimes that "gets down to little things, such as the sorts of connectors we've used historically in the network". They introduce losses, "and there are millions of them". So "we need to be continually looking at how we can improve the quality of the infrastructure".
And Openreach is going to have to develop a suite of products to cater for a range of communications providers, from those that want to deliver several channels of high definition TV over 24 megabits of DSL to those that want a simple, straightforward product. "Over time we need to be able to differentiate service," he says.
"This is a market where differentiation and segmentation will happen, probably in ways we haven't thought of yet. We need to be agile, we need to be very customer connected and we need to make sure we deliver what is needed for their businesses. If we don't there won't be a place at the table for Openreach."
Above all Robertson recognises that "we're in the middle of a massive industrial transformation", and "we will go with that transformation or we will be part of telecoms history", he says. "Whether we will be part of the future or not is very much in our own hands." GTB
See also interview with Paul Reynolds of BT Wholesale
The state of the access network
What precisely is the state of the infrastructure that Steve Robertson and his colleagues at Openreach are now responsible for?
The answer, according to Eric Gries of Evolved Networks, is that probably they don't know. "When someone orders DSL, whatever they tell you, almost 50% of the time they're guessing," says Gries, a former Amdocs executive who has recently taken over as CEO of Evolved.
He's not specifically talking about BT and Openreach's access network. "The problem is universal," says Gries. But Evolved Networks is closer to BT than to most incumbents: it is, indeed, a spin-off from BT's own business, now set up as an independent company to market systems to determine as accurately as possible the state of the cables under the roads.
Sometimes the problem is that operators aren't quite sure where the cables run, he says. There are maps, yes, but "a single exchange can have 240 maps associated with it", and records aren't always scrupulously kept up to date.
It is "telcos' dirty little secret", says Gries. "The maps are terrible." And as customers demand higher and higher bandwidth, for applications such as IPTV, the problem will potentially get worse.
And how do you define the access network, he asks. The quality of cabling inside customers' homes will also be important — but that's privately owned by the resident. "You won't just be able to check that there's dialtone," says Gries.
Evolved Networks is busy marketing software which digitises information about the access network and then validates it using a logical database. "We get 98% accuracy," he says. And then it enables engineers to keep it up to date.
"A lot of companies just don't know what's going on in their networks," Gries warns. GTB