Thursday, December 16, 2010

Britain's superfast broadband disconnect?


While it was encouraging to read the following words about public sector network opportunities in the UK's new broadband strategy, published last week and greeted with an entertaining little slip-up on BBC Radio 4's Today programme (don't worry James; those of us that share the Culture Secretary's surname have heard 'em all before, more times than we care to remember in fact), it seems a few doubts have crept in along the way as well:
"5.19 Public sector networks like those that connect our schools, hospitals and other public building to the Internet are often mooted as the answer to improved connectivity for locations where broadband connections are slow and it is easy to understand the frustration of those that know that a fibre passes within close proximity of their residence. It is correct that considerable public investment has been made in these networks, but it is not unfortunately correct that they provide an instant solution to slow connectivity. 
5.20 Whilst there is definite scope for the re-use of public sector networks, particularly to address the absence of adequate backhaul, those that exist today may require extensive technical alteration to make them fit for purpose, which might render them impractical or too expensive for re-use, or may be subject to commercial contracts or procurement constraints that do not permit their reuse for the household consumer market. 
5.21 Neither of these are absolute barriers, although in practice budget may be the defining factor, but where the re-use of and existing public sector networks provides an efficient means of improving household connectivity it will be part of the solution. Indeed there are some examples where the model has been applied, for example the North Yorkshire network, which was built and specified with the purpose of aiding and improving household connectivity in mind and further examination of the commercial arrangements will be explored during the BDUK superfast broadband pilots. 
5.22 It does need to be highlighted that each service delivery to a school or public building using fibre means that route has been upgraded and has been readied making it easier and cheaper to upgrade. This is a potential contribution to upgrading routes to rural areas, and where suitable, we will look to make this re-use a reality."
As far as I am aware, no-one has ever suggested that such networks would provide an instant solution to slow connectivity, so it's a bit disingenuous to suggest that they have. And it's always been the case that such networks would require development and most likely bandwidth upgrades to provide additional service in this way. But, given that upgrading them to support additional use would also improve services for their original users, particularly rural schools that stand to benefit hugely from improved broadband access, surely it's worth putting the effort in to address these hurdles? Worryingly, the language used above looks (to me anyway) more like an attempt to find reasons not to pursue this opportunity, rather than an encouragement to embrace it.

This undercurrent of doubt also doesn't sit very well with points made earlier in the paper. In particular, the section on the important role of local authorities and regions, in partnership with communities, in stimulating and coordinating broadband developments:
"4.2 What Governments have seldom done – but which we are proposing for the first time – is enable communities to influence or take part in extending access networks...
4.3 We are taking a new approach to delivering connectivity in rural and hard to reach areas that the market will not provide for. Where local authorities have superfast broadband as a development priority, BDUK will work with them to source an upgrade to the data transport infrastructure. This will be the foundation for the Government’s £530m investment commitment over the lifetime of this Parliament.
4.4 BDUK will also explore the viability of a broadband community hub at a local level – which could provide the means of extending networks where the community will either take responsibility for the actual civil engineering of the network or take greater control over managing network elements. Networks can then be extended over time to provide enhanced access to broadband for individual premises in a variety of ways. For example, an operator’s cabinet can be equipped to support the splicing of fibre builds into the access network. Interfaces can be made available such that wireless networks or indeed community managed femtocells can be added to the network."
Given that local authorities and regions have invested heavily in broadband access over the last 10+ years, surely it makes much more sense to utilise this existing infrastructure than to deploy duplicate over-builds? And what better "broadband community hub" could there be than a school? Doesn't the line on "extending access networks" rather contradict the previous extract, which suggests that this might be just too difficult?

My guess is that ever-helpful industry folks have been having a quiet word on the back of re-use proposals, to suggest that "you really don't want to be bothering with these fiddling little school networks you know". This is principally because industry will want to get its hands on the cash with as few strings attached as possible, to build new private assets it can then sell back to the public sector for the next 20 or 30 years. A market-led approach to broadband, in fact, as is being deployed in Cornwall.

I think consolidating and extending access to existing publicly-managed networks is a far more appropriate use of public monies. This would further develop publicly-owned and managed assets to benefit communities first and companies second. Of course the private sector has a huge role to play, as such networks are commissioned from commercial providers, but, most importantly, the critical mass of management and development resides in the public rather than the private sector under such an approach.

Here's the bottom line: you're much more likely to get an infrastructure that meets both your current and future requirements if your relationship with your providers involves "commissioning from" rather than "being sold to". The former benefits end users first and shareholders second, while in the latter, it's the other way around.

Thursday, December 02, 2010

Net neutrality, Comcast & the FCC...it's just like old times


Following a lengthy period of discussion and debate, which brought suggestions from a wide range of parties, FCC Chairman Julius Genachowski yesterday announced how he intends to preserve “internet freedom and openness” – or, to put it more succinctly, he unveiled his latest proposals on net neutrality.

He began by revisiting previous misbehaviour:
“...we have seen clear deviations from the Internet’s openness – instances when broadband providers have prevented consumers from using the applications of their choice without disclosing what they were doing.”
Absolutely we have – have a look at this previous post to see what Comcast got up to in this regard. Yesterday’s announcement reiterates consumers’ right to transparency and to access the lawful content of their choice, as well as their right  to a level playing field which precludes “unreasonable discrimination in transmitting lawful network traffic”. These consumer protections are balanced against recognition of ISPs’ needs:
“...broadband providers need meaningful flexibility to manage their networks – for example, to deal with traffic that’s harmful to the network or unwanted by users, and to address the effects of congestion. Reasonable network management is an important part of the proposal, recognizing that what is reasonable will take account of the network technology and architecture involved.”
And in relation to the legal basis for this, given that that the ruling made in April 2010 went in Comcast’s rather than the FCC’s favour, it seems there has been a change of heart:
“Informed by the staff’s additional legal analysis and the extensive comments on this issue over the past year, the proposal is grounded in a variety of provisions of the communications laws, but would not reclassify broadband as a Title II telecommunications service. I am satisfied that we have a sound legal basis for this approach.”
So this spells the end of the third way proposed in May 2010, where broadband services would be reclassified as Title II services, albeit with only a handful of Title II provisions applying to broadband: only the transmission component of a broadband access service would be recognised as a telecommunications service, so providers wouldn’t be regulated in relation to, for example, web-based services and applications, e-commerce sites and online content. More on this from the BBC. The FCC's board of commissioners will vote on the proposals at a meeting on December 21 2010.

It’s interesting (deliberate?) that the FCC’s announcement coincides with Comcast making further headlines in relation to net neutrality, but for a different reason this time. From the FT (Comcast at the centre of ‘net neutrality’ row):
“The US cable giant was accused late on Monday of demanding fees for the first time in return for carrying internet movies and other traffic for Level 3, an internet backbone network operator. Level 3 accused Comcast of “effectively putting up a toll booth” on its broadband networks so that it could unilaterally set the price for online content that competes with its own services The Colorado-based company recently won a contract to carry video on behalf of Netflix, whose inroads into streaming movies pose a long-term challenge to Comcast’s own cable television business...Comcast said that there had been a doubling in the amount of traffic it was being asked to handle on behalf of Level 3, with the Netflix deal believed to account for the jump. It added that it was being asked to carry five times as much traffic for Level 3 as it sent in the other direction, which it said justified the imposition of fees.”
A Comcast blog post claimed that that it already charged other internet backbone companies the same fees that it was seeking to apply to Level 3. Level 3’s original complaint is here and their response to Comcast’s reply is here. Clearly an additional complication here is that the Netflix service could be argued to be a direct competitor to Comcast’s own services. In a word? Messy.

This is a good example of what Ofcom referred to as a two-sided market in their June 2010 consultation: ISPs serve both content providers (such as the BBC, in relation to iPlayer content – perhaps not the best example, given recent history!) and downstream customers accessing their content (you and me). Philosophical arguments about whether such fees undermine the so-called principles of Internet openness and freedom have to be balanced against the reality that broadband traffic volumes are continuing to increase rapidly whilst ISPs’ revenues remain largely static. Surely this can’t be sustained if we all want to continue to benefit from new, innovative services that can exploit the full potential of next generation networks? This graph from Ofcom’s 2010 Communications Market Report (page 283) shows the proof of this:


I find it frustrating that complex issues like these all get lumped together under the banner of net neutrality when I think there are some important distinctions to be drawn, as I’ve suggested in this previous post. There is one good thing about this continuing debate though –at least it gives people like me plenty to write about!