Pay-as-you-grow pragmatism ruled the fiber agenda at BBWF 2012

“We tend to overestimate the effect of a technology in the short run and underestimate the effect in the long run”. Amara’s law will probably apply to superfast-broadband services, but it’s little comfort to those operators faced with investing in the multi-billion dollar networks that will support them. While most have one very solid short-term reason for investing in next-generation access – competition from cable – what they might gain in the longer-term remains unclear.

So, it was no surprise that strategies that enable operators to hedge their bets took center stage at Broadband World Forum in Amsterdam last week – quite literally. On the first day, the cavernous main room was given over to the DSL Evolution stream about how copper networks were evolving to support faster speeds over longer distances.

Discussions I had with operators and vendors both on- and off-stage throughout the three-day event also reflected a growing consensus that “future-proof” fiber-to-the-home technologies are not right for many operators in most cases in the here and now.

Pay-as-you-grow based on what you know

A number of telcos told me that their NGA roadmaps were likely to be much more fluid over the coming years. Moves to extend fiber from the street cabinet to the home might happen within five, ten, fifteen or even thirty years’ time.

While emerging DSL acceleration technologies would no doubt influence these decisions, demand for speeds will play a greater part. Almost all operators I have spoken to acknowledge that few consumers are willing to pay a significant premium for superfast speeds. At the same time, there is a growing confidence among telcos that offering value-for-money bundles of moderately-fast fixed broadband with other telecoms, TV and over-the-top services will be enough to stave off the threat from cable.

Indeed, engagements that we’ve had with several European cable operators suggest that they are no clearer about how to capitalise on their superfast-speed advantage than the rest of the industry.

This skepticism about speed is reflected in demand for NGA technologies. Despite the fact that 100Mbps and 1Gbps services first became available to consumers in 2004, two vendors at BBWF 2012 confirmed that they have not had a single order for 10GPON equipment, launched about 18 months ago with the aim of helping operators to take broadband speeds to the next level.

This revelation plus the fact that many of the world’s most advanced services have been stuck at 100Mbps for years, adds further credence to our theory that Nielsen’s law – which states that Internet connection speeds for high-end home users will increase by 50% every year – has stalled.

The power of the prickly premises problem

Operators’ interest in pay-as-you-grow strategies is also motivated by a dawning realisation of quite how hard – and expensive – rolling fiber right to the home is. In particular, the challenge of installing fiber in front gardens, buildings and individual homes has been “vastly underestimated”, to paraphrase a number of operators and vendors my colleagues and I have spoken to.

This part of process can account for 60% of the cost of connecting a home with fiber – and sometimes more, according to one vendor. This is partly because it can be subject to all manner of obstacles that are almost impossible to plan for. Working with local authorities, utilities and other local infrastructure owners to deploy fiber to cabinets or streets is hard enough; dealing with thousands of building owners and millions of home-owners is massively more complex.

Indeed, it is a cause of some regret for those operators which have FTTH embedded in their technical or political ideologies. An executive of a state-sponsored national broadband network, for example, told how suburban sprawl in his country meant that some homes would cost the equivalent of over US$30,000 to wire up because of their distance from the public main road.

There is also the problem of the consumer. An operator in an advanced Asia Pacific market described how ADSL customers interested in moving to its cheaper FTTH tariffs often lose interest when they discover they will have to take time off or spend their weekend days waiting for an engineer to call. Realising that engineers will also have to dig up their immaculate gardens and driveways or drill through their pristine walls is also a turn-off for many home owners.

All these problems mean that operator interest in G.fast – variously branded as Omega DSL or GigaDSL by competing vendors – is growing. This technology promises to sidestep the problems of FTTH/B by enabling operators to provide 500Mbps speeds over 100m of copper as early as 2015-2016. Tellingly, one vendor claimed that an operator had not been deterred from pursuing G.fast despite estimates that some deployments could cost 10-20% more per home than rolling out in-building/home fiber.

Thinking short-term to succeed now

All these developments mean that telcos and vendors should focus on three key areas where they can reap rewards now rather than invest speculatively in technologies which might only yield benefits in several years or even decades:

  • Think multiple, tactical evolutions: Uncertainty over demand for NGA speeds and services means many operators have already committed to DSL acceleration as the most cost-effective means to compete with cable in large areas of their footprints. The next step is to refine plans and technologies to ensure they can quickly and cost-effectively extend fiber closer to the home in multiple ways if and when the threat from superfast cable grows.
  • Rollouts need to be fault-tolerant, not customers: With debates over future fiber architectures more or less decided – and little operator interest in such technologies today – vendors need to focus on systems that enable the provisioning of services with as little trouble and costs as possible – to users as well as operators. Broadband is a vital utility, but one which customers want delivered with minimal disruption.
  • Focus on the bundle rather than the bitrate: Few telcos can afford to fight a speed war with cable – upgrading to full-blown FTTH everywhere would cost too much. But for now, that doesn’t matter. Broadband speeds still rate below TV, price and increasingly OTT services when consumers come to choose bundles. Putting together the right package at the right price means telcos can still compete, whatever “inferior” NGA architecture they choose.
Comments
  • FTTN rubbish November 6, 2012 at 5:19 am

    What a pathetic article. Faster broadband over copper by having to put a node every 100 metres only a complete idiot would think this is feasible. Given that Malcolm “Dial Up” Turnbull wants his nodes at 1km to give people the broadband lottery of anywhere between 25-80Mbps requires 50000 – 70000 nodes then at 100m we would need an extra 200000 – 280000 powered air conditioned fridge size nodes cluttering up our nature strips.

    Let’s face it we will need fibre to the home FTTH if not now then in the next few years. So do it right first time by getting the government to install it (without costing the taxpayers a cent. 7% return) and the telcos access to it. The current NBN roll out has had a 44% uptake for 100/40Mbps so the faster speeds are required now.

    People peddling FTTN as a solution are definitely in the minority and are basically proposing it for purely ideological reasons. They need to get out of the way and let the majority enjoy the digital revolution the NBN is bringing to Australia

  • Avatar
    Rob Gallagher November 6, 2012 at 10:06 am

    Thanks for the comment – this is obviously a subject you feel very passionately about!

    I’d just like to clarify a few points:

    This article is about telecoms operator-funded next-generation access rollouts, not government-backed ones, the Australian NBN plan or Malcolm Turnbull’s views. Granted, I mention one state-backed project, but largely to illustrate some of the challenges any fiber-to-the-home rollout will face.

    When it comes to such operators looking to fund their own investments in NGA, those backing or leaning towards FTTN are not in the minority – in fact, the number of those committed has grown greatly over the last two years or so.

    This is partly because of their own experience With FTTH/B or that of their peers.

    Sadly, many FTTH/B rollouts around the world have not seen anywhere near the kind of rapid takeup levels that you describe in Australia – or at least, not until they have started pricing it at about the same price or less than current-generation broadband, a strategy which was not in line with their original business plans.

    In addition, the widespread takeup of 100Mbps+ FTTH/B in advanced markets has yet to result in a significantly higher or different Internet usage than in markets dominated by current-generation broadband services.

    True, a tipping point could occur when new applications and usage emerges, similar to when video became more widespread on current-generation broadband. But this is small comfort to operators, particularly given that third-parties riding “over the top” over their new networks might ultimately yield the majority of the benefits.

    Clearly, governments have much more of an interest in making more speculative investments in fiber in order to drive benefits for industry and society as a whole. But for a growing number of telecoms operators which have to invest in networks with their own/shareholders’ money, going straight to FTTH across the board is looking increasingly unwise.

  • Avatar
    Rob Gallagher November 7, 2012 at 9:01 am

    And to further clarify…

    The operators we have spoken to are looking at G.fast as an alternative to rolling out fiber in front gardens and driveways or within buildings and homes. The logic is that if they feel the need to provide speeds of up to 500Mbps, they can use G.fast to avoid any problems related to digging up driveways or gardens or entering the home.

    It is correct that there are issues related to where such equipment resides, how it is powered and potential interference, which operators and vendors are currently working through. But at this stage, these do not appear to be so insurmountable to deter operator interest in G.fast. Our views on likely deployment strategies are set out here: http://blogs.informatandm.com/5787/g-fast-a-question-of-commercial-radio-manholes-prison-sentences-and-indoor-versus-outdoor-engineers/

    But demand for G.fast – and again, I stress, from operators funding their own NGA rollouts – will very much depend on when operators feel the need to offer speeds only FTTH/B can provide.

    At present, many are comfortable that various DSL acceleration technologies (vectoring, line-bonding, phantom mode – all, of course, where relevant and feasible) combined with tactical moves to bring fiber closer to the home will ensure they can remain competitive with cable, come what may.

  • Post a comment