21st Century connectivity: lessons for the UK

10 Nov

I’m on my way to London to speak at an event entitled 21st Century connectivity: lessons for the UK next Thursday. The event is organized by Reform and Sky and is billed as follows:

Thursday 12 November 2015
09.00 – 12.30
Altitude, Millbank Tower, 30 Millbank  

How the UK achieves world-leading digital infrastructure is hotly debated.

Ofcom has launched a once in a decade review of digital communications. Government is helping to deliver super-fast coverage (with an ambition for ultra-fast). Internet providers are rolling out the latest technologies to consumers. The Treasury’s productivity plan and Spending Review framework highlight digital technology as one of the key themes of public service reform in this Parliament.

But how ambitious should we be? Is the UK attractive to international investors and innovative companies with new technologies? Is our communications infrastructure up to global standards? What can we learn from other markets, from institutional investors and from those who have delivered world-leading connectivity elsewhere?

Reform, in partnership with Sky, invites you to a unique policy summit to discuss how we ensure the UK has a world-leading digital economy for the next decade and beyond.

Through three panels and a keynote speech, domestic and international experts will set out how we can deliver the digital infrastructure needed for efficient public services, a productive economy and a stronger society.

Lunch will be provided after the summit. Please register your interest at events@reform.co.uk. Attendance will be confirmed by email.

The list of speakers (myself excluded of course) is frankly amazing:

  • Matt Houlihan, Head of Government Affairs, UK&I, Cisco
  • Dr Peter Cochrane OBE, Cochrane Associates
  • Dr Adam Marshall, Director of Policy, British Chambers of Commerce
  • Dana Tobak, Managing Director, Hyperoptic
  • Vanessa Oakley, General Counsel, Chorus, New Zealand
  • Per-Olof Gustafsson, Deputy CEO, Stokab, Sweden
  • Benoit Felten, Diffraction Analysis
  • Jonathan Ford, Chief Leader Writer at the Financial Times
  • Gerard Lyons, Chief Economic Adviser to the Mayor of London
  • Robert Wall, Senior Principal, Canada Pension Plan Investment Board
  • Nick Delfas, Redburn Analysts
  • Andrew Griffith, Group CFO, Sky

The event is free to attend, so if the topic is of interest, don’t hesitate!

EC Confusion and Prejudice around the OTT terminology

6 Nov


I have said in the past that the OTT (Over the Top) acronym is toxic and should not be used, except possibly in a very specific context of describing how a digital service is delivered to end users. The European Commission recently issued a ‘public consultation on the evaluation and the review of the regulatory framework for electronic communications networks and services‘ in which the acronym OTT is used repeatedly. This is an important document, designed to flesh out major evolutions in telecom regulation.

I never imagined that the European Commission would use the term OTT to describe a type of player in the industry.  Not only that, but that they would do it without any consistent definition and in the context of an official consultation: the commission can’t seem to agree if the term designates market players or services.

Consider for a moment how cautiously BEREC handles the term ‘OTT’ in its recent report:

“OTT is a term frequently used but often not clearly defined. Some use the term to define a group of actors; others use the term to qualify a category of service… BEREC acknowledges that the term OTT might be regarded as incorrect or having a negative connotation.”

Compare this to the way in which the EC use the term in their consultation. In the document’s introduction (p3) we read:

“Since the last review in 2009, electronic communications networks and services have been undergoing significant structural changes characterised by [...] more complex competition with the convergence of fixed and mobile networks and rise of retail bundles as well as emergence of new online players (so called OTTs) along the value chains which challenge the traditional role of Telcos and Cablecos [...]“

In this usage, OTT designates market players. It’s not describing types of services but suggesting that it’s intention is actually to target specific companies without defining them. Worse still, it does so by using the terminology coined by network operator lobbyists to designate American Internet Giants. Sure, it’s not explicit, but it’s there.

Later in the consultation though (p55), things get a little more specific, but no less confusing. “Over-the-top (OTT) services”, the commission writes, “are increasingly seen by end-users as substitutes for traditional Electronic Communication Services used for interpersonal communications, such as voice telephony and SMS.”

So we’ve shifted from using the term to describe market players to using it to describe services. The definition seems relatively narrow, although it’s also very subjective. How does the fact that users see a service as a substitute for telephony or SMS make it OTT? And who decides which service end-users see as substitutes? Is the commission proposing to do European-wide end-user surveys to determine which online services they see as substitutes?

To me this is a non-sensical definition. Defining the exact scope through the subjective eyes of end-users is absurd and ineffective. Regulatory decisions need to rely on precise definitions, and this isn’t one.

But it doesn’t stop there! 10 pages further down (p65), the Commission asks about substitutability, and frames the question as follows:

“Do you think that traditional electronic communications services [...] can be functionally substituted by OTT services or platforms with communication elements (e.g. internet telephony services, web messaging services, webmail services, social media platforms, other)?”

One way to rephrase this would be: do you think that the services that end-users think are substitutable are actually substitutable? Confused yet?

Or, alternatively, this is a third, different, definition. And one, incidentally, that includes virtually every online service you can think of, since most online services include communication elements. Are we talking about AirBnB (which has embedded messaging), Call of Duty online (which has live audio communications), or are we talking about Skype and Whatsapp?

I have met my share of policy people who are completely disconnected from reality. I have also met a bunch of smart people at DG Connect over the years. I simply cannot understand how something so sloppy could make it into probably one of the most important EC consultations in the telecoms field in a long time.

It’s hard not to think that the incumbent lobbyist terminology wormed its way into the consultation, alternatively describing OTT as services and as a substitute for other acronyms like GAFA (Google, Amazon, Facebook, Apple.) This in itself suggests an inherent bias in the way the consultation was written, and therefore in the way the results will be analysed.

I was tempted to respond, despite the tediousness of it all, to expose that bias, but it turns out I can’t do that as Diffraction Analysis because I’m not registered with the commission in Brussels. I can only do it as a  nameless individual which seems kind of pointless. So much for wanting to hear the voice of the Industry. In fact, thinking about it, the whole thing seems biased: not only are the respondents restricted, but it’s excessively long, the interface is clunky and the questions are all closed. It’s as if it was designed for those who have an army of lawyers and policy people whose jobs it is to write such responses. Who could that possibly be ?

Photo: ‘In Over (the Top of) my Head’ by Benoît Felten

Plucking the low-hanging broadband policy fruit

5 Nov


Ever since moving to Asia last year I have been looking around for what was happening broadband policy wise and gradually realized how much work needs to be done in this area. Earlier this year I wrote an article for FibreSystems entitled The Many Paths of Asian Fiber. This was the beginning of a thought process that matured in a speech I made at CommunicAsia in June (Telecom Asia did a write up about the speech with a somewhat more spectacular title than I felt my speech was).

To cut a long story short (and I of course encourage you to read the two articles above), there is no model deployment for fixed broadband in Asia, and policy makers are by and large not addressing the core issues that stifle private initiative. There is a prevailing view that no infrastructure investment will happen without public subsidies, and I think that view is just plain wrong.

Earlier this week, I spoke at the FTTH Council Asia Pacific’s Gimme Fiber Day in Kuala Lumpur. Telekom Malaysia is (in many ways) rightly held as one of the enlightened incumbents in the region (deploying fiber, embracing open access, etc.) but something struck me while I was there: if you look at the amazing commercial success that HSBB1 (the first phase of their fiber deployment) has been, with low deployment costs and high take-rates, you wonder why it needed public subsidies in the first place. This would have been a perfectly viable infrastructure project, with a breakeven (according to my back of the napkin calculations) well below 10 years. And remember that 10 years for infrastructure is already a very good breakeven.

All this to say that before thinking of publicly financing wireline infrastructure deployment, policy makers should look at the low-hanging broadband policy fruit that would eliminate many of the pain points of deployment and make private initiative viable on its own terms. Here are two of these low-hanging fruit:

  • Fibering new properties up front: many countries in Asia are building new accommodation at a very fast rate. The emerging middle-class aspires to live in apartments that have all the modern amenities. Yet few if any of these countries have grasped the connectivity opportunities these new apartment blocks offer: most of them are still connected with copper, even in countries where copper was never much deployed in the first place. I have had a hard time tracking consolidated numbers for new builds in APAC, but I know a number of markets where the growth of new accommodation is a big thing (India, Indonesia, Malaysia). In all of these countries, a law mandating that fiber ducting and/or fiber be deployed when the property is build, with a proper mechanism for patching that fiber to whichever service provider’s network decides to serve the property would make a massive difference in wireline broadband adoption very fast. And the cost would be exactly zero since deploying fiber to a new home is no more expensive than deploying copper (in fact, it’s likely to be slightly cheaper.) A few years ago, Diffraction Analysis did a study on real-estate benefits of FTTP deployment in Europe. One of the aspects we examined was legal obligations. If you’re interested in this topic, you can check the webinar we did at the time or get in touch with us.
  • Laying fiber duct alongside all other infrastructure work: digging to lay fiber is the primary cost driver for deployment and therefore the primary challenge. By mandating that fiber ducts be laid alongside all public infrastructure work (water, electricity, roads, etc.) and handing the management of those duct assets to a public entity, governments can drive the cost of deployment down dramatically. A number of attempts have been made in the US to pass just such a law and one is under discussion right now. They call it ‘Dig Once’. For emerging markets, this could be a golden opportunity at no or very little cost.

Both of these low hanging fruit have limited cost. And even if they require a little public money to cover the cost differential, the amounts would be an order of magnitude lower than broadband subsidies and no doubt well spent.

Photo: Low-Hanging Fruit (c) Benoit Felten

NBN Co should NOT open its network to retail customers

9 Sep


It won’t come as a surprise that I follow what’s happening with Australia’s NBN Co with interest. After all, even if I’ve been highly critical of the plan’s implementation and it’s political weaknesses, it still is a national broadband network and as such worth looking into. But some things slip by me, and so it was with no amount of surprise that I read Paul Budde’s post on Linked In this morning entitled NBN Co Should Open Up Its Network to Others. I’ve met Paul and he’s generally been fighting the good fight in Australia on this issue, but I think it’s fair to say that in this instance I could not disagree more with the stance he is taking.

In essence, what Paul argues for is the ability for NBN Co to sell directly to retail customers. This is currently not easily feasible, he says, because NBN Co has 121 points of interconnection nationally and no wholesale product to connect these points of interconnection. It is also not allowed by the regulator ACCC.

To me this entirely undermines the point of structurally separating NBN Co in the first place. In the new regime NBN Co will have no retail contact with any customers, business or otherwise (apart from technical maintenance). They are a pure wholesale player. This is exactly as it should be. Any other regime recreates a publicly owned incumbent, one that has every incentive to act nefariously towards other market players and screw customers. There’s a reason retail monopolies were broken down, and NBN Co was designed in part to address the fact that Telstra post-liberalization was still too much of a retail monopoly.

I have often argued (unlike Paul) that the sin of the original NBN Co design was not that it had too many POIs but that it had too little. In order for competition to be healthy it needs to be able to optimise its costs, and the best way of doing that is to invest to get closer and closer to the customer. This substitutes a CAPEX-intensive model to an OPEX-intensive model and benefits the market. NBNCo is structurally designed to only offer an OPEX-intensive model and I think that’s a core issue.

I don’t have enough context on what’s going on here, and I need to dig deeper, but it seems to me that allowing NBNCo to offer products on the retail market would be detrimental on many levels: it would undermine the structural separation, if not deal a deadly blow to it; it would create distrust in the market between NBNCo and its customers. And ultimately it would require for NBNCo to be more than an infrastructure player. Considering how hard it’s been for the organisation to get to where it is today, I’m not sure that’s even sensible…

 Photo Credits: (CC) Geee Kay

Is Structural Separation in the UK really that risky?

21 Jul


I rarely disagree with the excellent Martyn Warwick, but sometimes there are exceptions. In an article in Telecom TV entitled BT Threatens Decades of Litigation Over Forced Sale of Openreach, Martin concludes that:

An ill-considered and badly executed sale of Openreach with no guarantees as to its future performance by a new owner would be a recipe for disaster with the buyer breaking promises and failing to make the investments necessary to make Broadband Britain a reality outside the major cities. Meanwhile, Virgin Media goes from strength-to-strength.

This misses, in my opinion, two major points that I wanted to reiterate here:

  • first, a sale is definitely not the way to go for structural separation. A spin-off, as implemented by Telecom NZ is the right approach for a number of reasons: first, it means that the shareholder structure only changes if the shareholders themselves want to sell their shares in the new vessel, so it’s not really a story of a single (and potentially evil according to Martin) buyer acquiring the crown jewels.
  • second, a structurally separated entity lives only on wholesale revenues. As a consequence it has every incentive to deploy better infrastructure, especially in a competitive environment where someone like Virgin is doing the same.

As Thomas Langer and myself discovered when we released our report late last year entitled Can Structural Separation via Spin-Offs Help Europe Achieve its Broadband Ambitions, there’s a lot of half-truths and outright lies that have been patiently spun by incumbent lobbyists about Structural Separation. In fact, Martin quotes BT’s CEO Gavin Patterson in his article in a way that is very telling:

“This is a commercial enterprise and if there’s uncertainty we will defend the rights of our shareholders, undoubtedly. It puts that investment very much at risk. At the end of it, and if we’re meant to be looking at the next ten years, what do you want to look back on? Do you want to look back at 10 years of litigation and arguments?”

The threat here is clear: a US style ‘litigate ’til they give up’ approach. But what about the shareholders?

As we demonstrate clearly in our report, there is every chance that a well executed structural separation via spin-off will result in two major achievements: increase value for shareholders and allow Openreach to extract enough cash-flow to envisage long-term infrastructure investment instead of the (at best) mid-term solutions currently put in place with VDSL and its upcoming life support.

The fact that this would be done by an entity with no financial ties to any market player and therefore would put every market player on a level playing field is, in many ways, the cherry on the cake. Indeed, it’s been seen to work very well for New Zealand, and the Czech Republic’s O2 is currently undergoing the same process. This is neither new nor rocket science.

The fact is that, if Gavin Patterson truly works for the shareholders, he should be considering Structural Separation very seriously instead of threatening the UK government.

Image Credit: 3D Judges Gavel (cc) Chris Potter

Not exactly Uber-innovative

22 Jun


Uber is, or at least was, until recently, presented as a paragon of innovation. The company is, we have read many times, revolutionizing the transportation for hire industry through radical innovation.

I’ve had an instinctive distate for Uber from the start, failing to see how the various players in the ecosystem they built could make decent money. There is no doubt that Uber’s model ensured Uber was making money, but I had the feeling that the drivers certainly couldn’t, long term.

When their bully PR tactics emerged last year, my distate turned into frank dislike, but then what does that matter to anyone? The fact that I don’t use them doesn’t make an iota of difference. I did however get increasingly annoyed at them being used as an example of radical innovation. Let’s look into that for an instant:

Is the innovation in their app allowing you to hire a driver using the app on your phone? Sure that’s neat, but every large taxi company in the world matched that within months of Uber launching in their market. Could they have done it earlier? Sure. But the fact that it was so easily matched shows that that’s clearly not where the innovation lies.

Is it in the dynamic pricing? Beyond the fact that dynamic pricing is hardly a new thing in general (even if it hadn’t been applied to that industry until Uber did it), it doesn’t seem to be such a big selling point for users. Furthermore, at least according to one economist quoted in this fascinating article on Daily Finance, it’s ripe for abuse and will likely backfire as a consequence.

Uber is often presented as a great example of the “sharing economy”, is that where the innovation lies? Actually, Uber drivers are usually taxi drivers who moved to Uber, so there’s no “sharing” there, not more than there used to be. UberPop is a proper sharing platform, but it’s undermining the main Uber driver’s income, possibly undermining the whole business model. So how is that smart innovation? I’d say it rather gives sharing a bad name.

No, the innovation is in one simple, and as yet unadressed (at least until last week) regulatory issue: Uber has designed a model that allows it to completely avoid the labor and fiscal burden of employed drivers. Uber’s innovation, in a nutshell, is in fiscal evasion.

And the most amazing thing this about it is that everybody knows it: when the California Labor Commission last week decided to treat Uber drivers as employees every analyst and tech journalist on the planet said Uber was screwed.

Labor laws exist for a reason. They may seem frustrating to entrepreneurs (I should know) but they are, for the most part, individual rights earned after long labor battles. They protect workers, and they help finance the help needed by those who lose their jobs. Circumventing labor laws isn’t innovation, it’s just morally despicable.

So let’s forget about Uber as an innovator and focus on companies that truly are innovating, for the benefit of all rather than simply for the benefit of their own wallets.

Portuguese FTTH Webinar Replay

22 Jun

Video Webinar 26 May 2015 Why Consumers Love FTTH in Portugal – The FTTH Consumer Experience Study from paftthcouncileu on Vimeo.

Exetel’s Magical Cost Accounting

16 Jun


Towards the tail-end of last week, tech journalist Juha Saarinen reported how Australian ISP Exetel was going to terminate the contracts of 400 heavy using customers. This surprised me, because I don’t think that’s legal in most western markets, but apparently it is in Australia. I engaged in conversation with Juha on twitter, and others joined. As part of that conversation, Information Age editor Ry Crozier pointed me to another article in PC World that has a bit more about the reasons for the decision. One passage in that article quoting Exetel struck me:

This has allegedly resulted in the ISP losing money on a number of accounts, including “one single user costing Exetel a loss of over $600.00 in a single month and more that 300 users costing us between $30 and well over $200.00 each in the Month of April and then repeating that sort of usage in May”.

I immediately asked myself how it was even possible for a single customer to cost AUD600 a month. I did one of the back of the napkin calculations I’m now infamous for, and this is what I came up with. Let’s assume for a minute that we’re only talking about traffic costs, not the whole wholesale access + infrastructure + overhead shebang (which should be no more than AUD30 anyway).

Transit in Sydney is very expensive (for reasons I won’t get into here, but might be worth examining in the future). Ballpark, in the upper end, it’s around USD15, AUD19. This is likely right for Exetel who is a small ISP.

If you attribute transit costs to a single customer, you have to assume that that customer is solely responsible for an increase in your peak transit (when you’re not peaking, the marginal cost is zero since your bill is based on peak capacity). That is already stretching believability, but let’s assume for a second that that’s even possible.

If that customer was responsible for an increase in peak traffic all by himself, he would be using 600/19, ie. 32Mbps of traffic all by himself. Throughout all peak hours of a whole month.

Since most of Exetel’s customers are ADSL customers, it’s safe to assume that this one is too. So at best, said customer has 15Mbps download capacity and 1Mbps upload capacity (if he lives next door to the central office). So 16Mbps is the maximum capacity this customer can possibly use on the network.

And yet cost accounting (if done with an understanding of the cost structure of transit) attributes him twice that in capacity used full time throughout the peak hours of the month.

This leads me to the unavoidable conclusion that Exetel’s accounting is terribly wrong. What they have most likely done is look at the overall cost of their transit and divided that by the amount of MB downloaded by each customer.

This is not about picking on Exetel specifically. There are still dozens, probably hundreds of ISPs worldwide who do just that. When we published our report on datacaps three years ago (Do Data Caps Punish the Wrong Users?) we hoped it would help the non-network people within ISPs to understand where the costs of transit really were. It seems that hasn’t worked as well as we hoped despite coverage in Techcrunch, Ars Technica, This Week in Tech, Wired and many others.

So we have decided to heavily discount the report, down to a measly €250. The data may be outdated, but the logic behind the cost accounting isn’t and, clearly, needs to be spread wider!

Free Webinar on French Broadband Usage and Satisfaction

10 Jun

On Tuesday June 16th 11 AM CET, Diffraction Analysis will present the high-level results of a study undertaken with the FTTH Council Europe on the French Market. This is a Free Webinar which you can join by registering here.

The results of the study will be presented by Benoît Felten, CEO of Diffraction Analysis and the session moderated by Joeri Van Bogart of the FTTH Council Europe. In it, we will answer questions such as:

  • Are FTTH users in France more satisfied with their broadband than DSL users ?
  • How do broadband users in general perceive the different broadband platforms ?
  • Do FTTH users do more things or different things online compared to DSL users ?

Join the discussion!

What’s faster than superfast ?

2 Jun

This morning I watched the above speech by Göran Marby of Swedish Regulator PTS. It meanders a bit, and (anecdotally) I was shocked by the apparent confusion between IP and Internet, but it makes some good points, and in particular it raises an important question about the future of networks. The assumption seems to be that the millions of devices that we predict to be connected will do so on the basis of a mobile (cellular) network framework. But most of these devices won’t move. In Fiber rich Sweden, the alternative (that they be connected on the basis of a fixed network paradigm) is appealing indeed.

This is something that interests me a lot, but I don’t think my thought process is quite mature enough to really address yet.

Instead I want to (slightly humorously) latch onto a joke Marby made in his speech about the UK’s “superfast” broadband. If 30Mbps is “superfast” he says, the standard 100 Mbps in Sweden should be called “superduperfast”.

We’re currently delving back into a project that’s been on the backburner for too long, with the aim of analyzing advertising messaging and strategies for FTTP.

If I was Virgin or Hyperoptic (or Sky / TalkTalk, assuming their fiber plans in York are actually moving forward), I would latch onto this concept and launch a campaign arguing that if BT is delivering superfast, then either of these other players are delivering superduperfast. Done well it would be funny, and would degrade the “superfast” tag, which frankly means nothing. If copper champions can devalue the advertising power of “fiber”, surely devaluing “superfast” is fair game?

(Note: I notice that Cityfibre used “ultra-fast” in their announcement, but I still think superduperfast would work better)