When two (almost) become one

Briefing: As Voice over IP becomes a reality, what are the likely business benefits, asks Nick Booth

Briefing: As Voice over IP becomes a reality, what are the likely business benefits, asks Nick Booth

Has there ever been a period in the history of IT when the voice and data networks weren't threatening to converge? This sequence of events must be one of the most prolonged the industry has ever seen.

It began, arguably, with the invention of the modem. Once someone came up with a machine for sending data down a phone line, it was only a matter of days before the industry visionaries logically concluded that voice and data networks would merge into one. Decades later, we are still waiting for the merger to be finalised.

There have been a lot of false dawns on the way to the rationalisation of our networks. Just under a decade ago, when the commercial possibilities of the Internet were being explored in earnest, a number of suppliers had the same idea for the medium's first killer application: sending voice communications over the Internet.

The business case was strong enough to convince a whole range of venture capitalists to fund a variety of competing product developers. Once companies realised that you could make international calls across the Internet, at local rates, they thought there would logically be an explosion in demand for these products. One supplier chose its product launch to predict the death of the telecoms industry, since the phone companies' revenue would disappear.

Two years later BT was still reporting record profits. But all the fledgling Voice over the Internet (VoIP) companies have almost disappeared.

Sending voice calls over the Internet is still a desirable goal, given the massive savings it will bring companies. "Voice over IP is a lot harder to do than most people realise," says Steve Dunwoody, chief technical officer of Equinox Solutions, which claims to be the UK's first convergence service provider. "In many cases you can't use either of your existing networks, but build a separate one."

Nevertheless, there is a lot more to convergence than simply sending voice calls over a cheaper network. Desirable though this cost- cutting application may be, it is not immediately deliverable. Convergence is also capable of delivering a range of applications that will increase productivity.

Messaging is a case in point. So many systems of communication have sprung up in the last 10 years that they are starting to become counter productive. You can spend hours answering all your phone, fax and e-mail messages, many being duplicates from people who have decided to reach you on as many systems as possible. Rationalising these systems into one would save time, especially if the new system provided better management of messages.

The Internet Protocol (IP) is succeeding here where it is struggling in carrying voice. Ian Short, marketing manager of the IP business solutions division of Getronics UK, says IP is enabling a long-overdue reassessment of our working lives. "We now all have PCs, desk and mobile phones, access to faxes, pagers, voice and e-mail. But we still have problems contacting each other and end up wasting half the day leaving messages for other people or deleting those from other people," he says.

Unified messaging, Short promises, will offer you the chance to route calls and messages to the appropriate place depending on how you wish to configure the system. This sounds hellishly complicated. "The communications flow will adjust to a person's personality, presence and the time of day," Short insists.

Could it be that instead of spending all day answering messages, we'll spend all day configuring messaging systems? Not so, argues Ian Morris, co-founder of a new distribution company, Equip, which specialises in convergence products. Products from suppliers such as Praxon have been capable of unifying all messaging systems into one medium for some time. They've just been too expensive.

Morris expects companies to adopt the single-port-of-call method for communications in a big way over the next year, since the technology has now reached maturity. "Everything will be done from a single phone number for DDI. All your e-mails and faxes can be read out to you by a machine over the phone," he says. "It will be picked up by companies now because it's easy to use and it's come down to the right price point."

A mass take-up of unified messaging is unlikely, however. Small to medium-sized businesses have some way to go before they get to the stage where they can start converting their voice calls into e-mails and vice versa. Many have yet to come to grips with the Internet. Consultant Doug Twining, of PA Consulting Group, says there are still technical barriers to be overcome, despite what the suppliers may claim. "The first challenge is the conversion of existing voice technologies to IP-based systems. Then there are the data format standards limitations to the existing voice profile for instant messaging standard. Not to mention the simple Internet fax profile. Mobility will be hampered by wireless bandwidth and memory limitations."

The biggest breakthrough in convergence, predicts Morris, will come in video streaming. However, it is difficult not to be sceptical about the promises of video services. Video conferencing has been mooted as a business tool for as long as most people can remember.

It's not that long ago that BT was urging people to spend £400 on a telephone handset so they could see a passport picture-sized blur on a screen that vaguely resembled the person they would be talking to but whose lip movements seemed to be 10 seconds behind their speech. This, said BT's publicists at the time, was going to "revolutionise business" and drastically reduce the amount of business travel involved for anyone who has to attend lots of meetings.

Those same claims are being made today. The difference is that the roll-out of various types of digital subscriber line services will bring about a massive upsurge in available bandwidth. This, coupled with technical advancements in cache technology, means that video and audio can at least be streamed from a server to all points of a network. Presentations can be broadcast across the network without affecting the movement of the rest of business traffic and it will be possible to watch training videos at the desktop.

The technology is there, it is up to companies to find practical uses, says Morris. "Let's say we were looking for a new PR company. We could watch all the pitches on video before deciding who we would see," he says. "You can extend that principle to anyone who has to meet and greet lots of new companies."

One of the major advances in business communication has been in customer care. The integration of computers and telephony systems (which isn't strictly convergence because the two systems don't run across the same network) demonstrated how much more efficient companies could be if they brought the two systems together. Computer telephony integration improved customer services vastly. The next generation of customer care systems aims to take the improvement a lot further.

Here convergence has a part to play too. "Screen popping", a process where a customer's records are retrieved automatically by the system and brought up onto the terminal of the operator who is taking the call, brought about a marked increase in productivity. More importantly, it kept the customer from losing patience with their supplier and going elsewhere.

Online commerce needs something similar. While it's bad enough to phone a company and get passed from operator to operator, none of whom are familiar with your case, at least you are already a client. Visitors to online shops come up against problems before they have entered into a contract, which makes it easier to take their business elsewhere. Online buyers frequently want to question a human operator before they place an order. There are two main reasons why they don't. "They may already be using their only line to get access to the Net," says network consultant Peter Williams at 3Com. "The only way they could call the number is on their mobile phone, which illustrates other reason why they might not call. They don't like paying charges."

'Call me' buttons are increasingly popular on e-commerce sites, especially those that sell to consumers. The buyer enters their phone number into the template, then waits for a call from a salesperson at their leisure.

There's an argument that the most successful example of convergence will be that data will migrate to the mobile voice network. By 2002 there will be more mobile phone subscribers that fixed-line subscribers, argues Martin Cassidy, vice-president of Ericsson's worldwide datacom channels. By 2003 more mobile users will be using phone and palmtops to access the Internet than users of desktop PCs.

"Convergence will take place between the mobile comms infrastructure as well as the fixed-line network. One is growing and the other isn't," says Cassidy, "so it's unrealistic to expect the fixed-line voice and data networks to gel completely, because a lot of business communication is gradually being moved onto the mobile comms network."

Ultimately, all networks are moving away from the old model, where systems were monolithic, closed and proprietary to one supplier. All the equipment, be it a PBX on a customer's premises or a switch on the carrier network, contained all the intelligence. Devices, such as handsets, were dumb terminals, in the same way that mainframes had centralised intelligence and lots of dumb terminals hanging off them. In the Internet world, the network is dumb and the intelligence resides with all the computers (from servers to PCs, palmtops to handsets) that attach to it. All applications, voice included, will ride across this data network infrastructure. This is less convergence than submergence of the voice network into the Internet.

"Network convergence really is just around the corner, but in order to take maximum advantage, today's network managers need to understand current telecoms spend and network inventory, prior to jumping onto a new bandwagon," says William Green of Compass Management Consulting.

In the meantime, it's a question of making sure that the new technologies actually deliver recognised business benefits, rather than having the latest applications for the sake of it.

Beware! Convergence could cost you money

Even people who don't believe that voice and data will work well together over a single network seem to accept that convergence will save them money.

But will two disparate systems work together? Not everyone thinks so.

It means too many different things to different people, complains Stephen Hope, a technical consultant at telecoms provider Energis. "To manufacturers it is the latest bandwagon to push new kit. To resellers, it is the latest set of buzzwords."

There are issues with having a single architecture, warns Mark Stancombe, European regional manager for network performance expert Quallaby. "Take asynchronous transfer mode (ATM). The voice community wanted the ATM cell size to be 32K because voice only needs small packets. But the data community wanted 64K packets because bigger packets are better for carrying data.

"So the ATM Forum chose 48K as the packet size, which ended up being no good to anyone," Stancombe added

The assumptions about costs only work if you think that today's transport costs are going to stay as they are forever. Of course they won't, says Hope. Broadband is coming down in price, there are more players and more price competition for all kinds of services, especially when the local loop is opened up. Then optical networks will eventually bypass the need for today's expensive, and relatively slow, electronic networks.

One of the costs of converged services will be management. But there is an alternative school of thought: who will need expensive management when bandwidth becomes so cheap?

Read more on Voice networking and VoIP