Who maintains the infrastructure that provides 'universal service and access for all'?

While the scramble for broadband licences, spectrum, and target markets is well and truly on, very few people are stopping to think about the nuts and bolts of building and then operating and maintaining the physical infrastructure on which broadband will be delivered, says Howard Earley, Plessey chief operating officer.

In the words of Communications Minister Roy Padayachie, at the New Age business briefing, in Sandton, on 9 September 2011, the South African government is unambiguous in its intention that, for instance, the reallocation of spectrum in the migration of broadcasters to digital transmission will “be directed towards the attainment of the goals of universal service and access for all”. Other African countries are on the same road.

Why? Because it is now universally acknowledged that the information communications technology (ICT) industry is the enabler of other industries. Even in emerging economies, the private and public sectors are now heavily dependent on affordable, converged communications technologies that enable them to engage with the global economy on competitive terms. As has happened elsewhere in the world, African economies will eventually become utterly dependent on being able to communicate by voice, video and data easily, quickly, reliably and cost-effectively.

And, of course, there's always the prospect of technology evolving. Some years ago, we were kept very busy building 8 000 mobile network base transceiver stations and wireless local loop systems sites across many African countries. Demand for those kinds of sites has tailed off, but soon there's going to be demand for 4G sites. They'll be smaller, but there'll need to be many more of them.

Many countries will also have to confront the local loop unbundling issue South Africa is facing now, because most countries started out with a single national telecoms operator that laid most of the cable to homes and businesses. There, as in South Africa, the question will arise of who lays and then maintains the last mile of cable that replaces that of the incumbent or, indeed, who manages and maintains the existing last mile on behalf of the multiple service providers who want to share access to that last mile.

In the early days of telecoms, it was taken for granted that the telco would build, operate and maintain its own infrastructure. Now, while many telcos do still operate their own infrastructure, fewer actually build it, and many outsource the maintenance of it.

In the new world of converged communications, where telcos and service providers have their hands full creating, marketing and delivering products and services to their customers, building, operating and maintaining infrastructure is even less core to their business.

For one thing, it can be extremely hard, physical work. Think of the people who have to risk being out at sea in all weathers, searching with remote-controlled robots on the ocean floor for - and repairing - faults in the undersea cables carrying voice and the Internet. That's not just hard work, it's highly specialised.

In terrestrial terms, the specialisation is just as demanding. In the United States, for instance, there are dozens of Internet exchanges, located mostly on the coasts, which gather the undersea cables and disperse them over land across the country. The cables enter the exchanges through the floor and splay out along metal trellises to the hundreds of clients' servers in the building. The cables from individual carriers are then patched into land-based cables that radiate from the exchange, connecting service providers with other exchange points in urban areas and, from there, to homes and businesses.

Technicians are in the exchange building at all times, taking care of routine maintenance - tightening a loose connection there, rewiring a patch panel here. Even in a hurricane, the building is staffed. Twenty-four hours before a storm hits, all essential personnel are already inside and they don't leave until the storm has passed. And everything is monitored from a NASA-like network operations centre elsewhere in the building.

The walls are made of seven-inch, steel-reinforced concrete that can withstand hurricanes. Environmental control is fanatically precise, keeping condensation off the circuits, and the servers cool. There are dozens of back-up diesel generators.

So, ok. We don't yet have that kind of Internet exchange in Africa. But we will. And the point is that the operation and management of infrastructure requires its own specialised focus and expertise.

The really interesting thing about infrastructure operations and management (O&M) is that the skills and experience needed can be repeated and replicated at many different sites, while being monitored and managed from a centralised control facility. In its way, O&M is an off-the-shelf offering.

Yes, of course, minor adjustments have to be made to fit specific geographies and particular types of service. The O&M for a base station will differ from that of a fibre optical regeneration site. But the O&M for base stations in general will, in all critical aspects, be the same. As it will for fibre regeneration sites.

It makes no sense, therefore, for each telco or Internet service provider to have its own in-house infrastructure O&M team when it can buy the service on an outsourced basis from a specialist - and gain all the benefits of economies of scale and highly experienced management talent that only an outsourcer can provide.

Also, if your infrastructure O&M outsourcer has already built your facilities, then why waste the insight gained in that process by bringing O&M in-house? Why spend the time, money and executive effort in reinventing the O&M wheel when your infrastructure is a hygiene factor and not a differentiator? Why invest in achieving a world-class O&M capability in-house when what you really should be doing is designing world-class new content services for your end-users?