Archive | Access RSS feed for this section

So much for mobile-only network strategies…

16 May

I had a “wow!” moment this morning reviewing a presentation given by WiK’s Scott Marcus entitled State-of-the-Art Mobile Internet connectivity (thanks for Chris Marsden for pointing this out).

In particular, there’s one graph in there that blew my mind, I’m reproducing it here.

 

Source: Mobidia / Informa (2013)

This shows the proportion of mobile data traffic that is offloaded to private wifi networks (in blue) or public wifi networks (in light blue). Basically, only a quarter to a third of the data traffic consumed by mobile devices is actually delivered over mobile networks (except in Japan and India where it’s half).

Wow.

Take that, “we’ll only need mobile networks in the future” posse…

Hilarious NBN commentary…

3 May

I already published a humorous video on the Australian Coalition’s NBN plan last week, but this one is even better. Why don’t we get this kind of humour in France ?

Interview on Australian Telecom Podcast Crosstalk

22 Apr

Last week I was interviewed by Phil Dobbie on the Australian Telecom Podcast Crosstalk about the viability of on-demand FTTP. The whole show addresses issues of copper pull-through with an AAAC representative and on-demand FTTP as seen from the UK perspective as well as my own comments on the issue.

Don’t put your trust in on-demand FTTP…

12 Apr

On Tap by Benoît Felten (benfelten)) on 500px.com

One of the most interesting features of this week’s alternative NBN plan published by the Australian opposition (see my article Australia’s NBN becomes a political football on Telecom TV) is this notion of “on-demand FTTP”.

On paper, there’s a lot to be said for this concept. Basically, the operator (public or private) deploys Fiber to the Node, but a consumer can pay to get Fiber to the Premises if they really want it. This feature is not unique to the alternative NBN proposal, it’s also being trialled by BT in the UK as highlighted in this article. What’s interesting about the concept is that in theory it addresses the demand issue. In a nutshell, we’ll deliver “better broadband” to you for free, but if you really want “kick-ass broadband”, you can pay extra to get it.

Except it’s a technically absurd concept, for two reasons:

  • firstly, because the costs of FTTP deployment that are commonly considered (say around €1000 per home in dense urban areas) are mass deployment costs. These costs can be met when you send teams on the field to connect every home. If you send teams on the field to connect one home, the costs are much much higher. Just look at how much large businesses are paying to connect their premises: it’s in the tens of thousands of euros. Presumably, there’s a comfortable margin built into that price, but it still gives a sense of what the real cost would be.
  • secondly, and very ironically, both BT and the Australian NBN chose point to multipoint architectures. That means that if a customer was to ask for an FTTP connection in a given area, the whole active equipment chain from the customer’s ONT to the splitter to the OLT would have to be activated for a single customer. The costs of that are absolutely prohibitive.

And I won’t even mention the network management absurdity of maintaining two parallel infrastructures, one of which only serves a very small number of customers.

On-demand FTTP is a fallacy. It will either never materialize or be priced in such a way as to convince customers not to take it. The reason it’s being put forward is purely as a marketing argument for BT and the Australian opposition to be able to say “those that really want it will be able to get it”.

Not unless they’re filthy rich, they won’t…

 

Photo Credit: On Tap by Benoît Felten

Debunking the Free-Rider Myth

4 Apr

The debate on net neutrality has been heated these last few months, with various initiatives both at policy levels and ISP levels highlighting the need to clarify how internet traffic is carried so that journalists and members of the public can form an opinion on these issues for themselves. A while ago, I was asked by Google to write a paper aiming to do just that. It was released yesterday under the title There’s no economic imperative to reconsider an open internet.

I’m quite proud about this. Working on this paper gave me the opportunity to really explore the net discrimination arguments and examine their worth. My conclusions are expressed in the title: there is no big issue related to the cost of traffic management, even as the traffic itself increases. I hope you find the paper interesting, and feel free to spread it around.

Ventura’s Stefan Stanislawski gives the low-down on FTTH Financing in Europe

26 Feb

Excellent short summary of Ventura’s excellent report on funding.

Net Neutrality Debate à la Française

16 Jan

Over the Seine

Since the recent (and thankfully short lived) decision by Free to block Google’s advertising content for users of their set-top box, the French net (from government to activists) has been abuzz. Even the international tech press has commented on what looks from the outside like an emblematic struggle between an ISP and an OSP (See FastCompany’s French ISP Free Blocks All Web Advertising and the thoroughly researched pre-kerfuffle article on GigaOm YouTube sucks on French ISP Free, and French regulators want to know why).

Government condemnation of Free’s patent net neutrality violation has been mild, to say the least. Fleur Pellerin (French delegate minister in charge of small and medium businesses, innovation and digital economy) did state that Free’s “approach was not acceptable” (Free: Fleur Pellerin demande de mettre fin au blocage de la pub) but that “it is not illegal under French law” (Pellerin : «Free et les éditeurs vont trouver un compromis»). Over the last week-end she even stated that Xavier Niel (Free’s CEO) “asked the right question, a question that needed to be asked” (Blocage de la publicité: Niel a “posé la bonne question”, selon Pellerin).

So it was with some trepidation that I attended a round table at the ministry on Tuesday morning entitled ‘Neutralité de l’Internet’ (Internet Neutrality). A lot of people were in attendance, an interesting mix of greybeards (and future hairy greybeards) and suits. The round table was organised in two sessions with different speakers although to be fair there wasn’t really a clear underlying theme to each. I tweeted extensively, and you can find my tweets here and here. I’m going to try to pull the many threads of the discussions into a coherent whole, using the things I quoted in tweets as anchor points. Forgive me if I don’t fully succeed. One last thing: moderation was excellently handled by tech and economy journalists Solveig Godeluck and Guillaume de Calignon from Les Echos who managed to keep things moving despite too many speakers with too much to say!

The first session started with a somewhat technical explanation by Kave Salamatian of why Net Neutrality was a new issue, the thrust of the argument being essentially that it stems from the Internet’s nature as the first separated delivery network ie. one that isn’t medium specific. Questioned on the academic’s views on what was really happening in the networks, he made an elegant analogy to Plato’s Cave:

Agustin Diaz-Pines from OECD and Vesa Terava from the Electronic Communications Policy Directorate at the European Commission both presented some slides. Diaz-Pines insisted on the inherent efficiency of the current financial (or, as the case may be non-financial) mechanisms underlying internet connectivity while Terava showed the results of a recent Berec study highlighting the extend of throttling on ISP networks in Europe. Neither really offered solutions though (more was expected there from the EC, and the general feeling in the room was that they were yet again stating the obvious: that there are net neutrality violations in Europe…)

Things started heating up a little when French MPs started talking. Laure de la Raudière and Corinne Erhel spoke about the main results of a parliamentary investigation they published on the topic nearly two years ago. It was quite clear from Laure de la Raudière’s arguments that at a political level there’s a mix of very different threads that should (in my opinion) be unraveled for the debate to move forward. This is what I said in the heat of the moment:

I thought MP Christian Paul‘s statements were much sharper in that respect. He started by stating that the dichotomy between managed services and “the rest” was dangerous:

He then addressed the often heard argument of ISPs needing a contribution from OSPs (in France, the polite way of saying it is “a better share of the value accross the value chain”) by asking if we really believed that more revenues in today’s networks would mean more investment in tomorrow’s network:

He didn’t lay it thick on that argument, but considering the French operators’ track record in FTTH so far, it’s certainly a potent argument. Finally he ended his short adlib intervention by stating that the emergency was to legislate on non-discrimination now:

And indeed, this ties back to the minister using the argument that it’s not illegal to absolve Free in one of the above interviews. On that note the first panel came to a close with Corinne Erhel stating the important notion that relationships between players on the internet should not be reduced to the (often contentious) relations between the big guys:

Consensus from the first panel was that non-discrimination was an issue everyone was in agreement on and the sense from the MPs was that the government should legislate on that. The twitter feed (mostly in French), especially during the EC presentation was very much all “enough commissions and studies and reports and hearings, we all agree on this, let’s act”. The second panel was a little more fact oriented, if only for the participation of Benjamin Bayart, the head of FDN a small not for profit French ISP. Bayart is vocal, articulate and seems afraid of nothing. On the same panel was Octava Klaba, the CEO of OVH, the largest French hosting company (and one of the largest in the world). In a sense, while their participation was very welcome, it also highlighted the complete absence of representatives from large ISPs or the large OSPs on either panel. The first real numbers mentioned all morning came from Klaba (whose name I originally misspelt):

One of the things he stated also that I was waiting for someone to express what the piddly little amounts that this whole debate revolves around:

When Yves Le Mouël started talking, we were hoping for some action at last: as a representative of ISPs in France, he’s known for being pretty adamant in defending their positions. I don’t know if he felt cornered (which, to be fair, he probably was, both on the panel and in the audience) but he wasn’t very combative. He tried to explain that the whole discussion was not really about the money

That in itself is interesting since that argument is repeatedly put forward by his members’ representatives. In fact he himself used that same argument a few minutes later stating that ISPs (more specifically network operators) represented 90% of the investment in the sector (and conveniently ommiting to mention how much of the revenue they represented) and as such deserved some solidarity. I wasn’t very convinced by that argument, to say the least:

The way he framed his argument though, probably paves the way for the way the government is framing theirs since a lot of his rhetoric was covered in the Minister’s speech a little later (sorry if I’m getting ahead of myself here). The core concept is that of an ‘economic signal’. When asked to explain exactly what that meant, how that signal was to be sent, to whom and what the results would be, Le Mouël was less than clear. What I could gather is essentially that because most of the peering does not lead to an exchange of actual money, it doesn’t send an inherent economic signal to the parties involved. And that’s wrong. As nebulous as that sounds, it seems to be the latest track that the steam train of legislative reform is on. Or rather regulatory oversight. Indeed, Le Mouël insisted that his members (ISPs) did not want a law, they just wanted to have the ability to appeal to the regulator (which, as I understand it they already can). Of course, he could not end by not stressing (quite subtly) that there’s an “us” vs “them” thing going on here. For full transparency, in the following tweetquote, the stuff in brackets is mine, not his:

In fact, Laure de la Raudière mentioned something about that recourse to the regulator a little earlier, she said:

ARCEP was on the second panel and seemed to take a relatively hands free approach, stating that they believed peering should be the result of commercial negociations and not regulation of legislative action. In a sense, that comment which came towards the end of the panel was a good summary of the panelists position. I’ll come back to what I think that means later, but the contrast with the minister’s speech which followed was striking. In that speech, she outlined again most of the arguments that had already been mentioned, strongly reaffirmed attachment to non-discrimination but argued for a more balanced share of revenues accross the value chain and expressed a (non-specific) need for an “economic signal”. In other words, she built the case for intervention… only to conclude that she would ask a commission to examine the issue. After that, there was a bit of a Q&A which I won’t bore you with, I’ll just quote a couple of statements from the panelists which I thought were both spot on and provocatively framed:

 

 

At the Q&A I raised two questions, one about asymetry and one about paid transit. The first one was aimed at Yves Le Mouël. I asked him if his members would be willing to pay the OSPs if the traffic flows reversed in a few years time when symmetrical access became the norm, as seen already in some Nordic countries. He hesitated and answered that his members would address that issue when the time came if it ever came.

My second question was to the minister who has been using the example in several interviews that Dailymotion paid for peering when Youtube didn’t and that wasn’t fair. I asked her if the fact that Dailymotion was the property of Orange might not make their example a little biased. She answered that the fact they were owned by Orange didn’t mean they didn’t peer with Free, SFR and other French players and therefore was neither here nor there. But then Octave Klaba took the microphone and made a statement that really stunned me:

That tells me I need to investigate this, which I’ll do in the next few days. I don’t know which payment the minister is referring to (since none of this is public), or who is right, but I’d sure like to know.

My feeling at the end of the conference was as follows:

I don’t think the ISPs want to push for a law, or even any more interference from the government or the regulator in this matter. I think they’ve been painstakingly building the arguments for government to tolerate known breaches of net neutrality as long as they’re not too visible. Since the OSPs don’t want government intervention (but would support a non-discrimination law) it’s a deadlock. What’s completely polluting the debate is the tax issue. The French government, like other governments in Europe, is mightily pissed off at Facebook, Google, Amazon and others for not paying proper taxes in France through tax evasion. I think they’re right to be pissed off about this. But the ISPs are using that as an argument to justify their misbehaviour: it’s the nice French guys who dutifully pay their taxes vs. the evil American guys who don’t.

That political ire is tainting the debate, and that’s why the government wants to be seen to take it seriously while not acting really (at least that’s my interpretation). The consensus on non-discrimination is there, from all political sides and both greybeards and suits. But there won’t be a law anytime soon. Because in the eyes of the government, the quid pro quo has become: be dutiful tax-paying companies, OSPs, and then we’ll look into it. Of course they’d never admit that that’s how it’s now framed, but I strongly believe it is.

All the rest, “economic signals”, “traffic asymetry”, “network investment”, it’s all window-dressing, aimed at suggesting that there is a legitimate reason to examine all of these issues when in fact it weighs so little in the cost structure of the ISPs that it shouldn’t even be a point of interest for government.

That leaves me a little concerned. On the one hand, I think (I hope) it’s unlikely that the economy of traffic exchange will be affected by any government decision anytime soon. On the other hand, the evident sympathy that Free has garnered with the government by dropping their little bomb a fortnight ago is worrying. That my government – or in fact any government – would think that such an action need not be strongly condemned is a big concern. That they might further think that it’s a good way of framing the argument is even more preoccupying.

More than ever I believe that network neutrality needs to be protected. If it won’t be protected by law, it must be protected by citizens who understand what there is to lose if breaches are tolerated any further.

 

Photo Credit: Sylvain Courant photographies

A Plea for a Radical Change in European FTTH Policy

14 Dec

 

European FTTH policy isn’t working. It’s not in any way controversial to say so, and I have done so repeatedly in the past. When it comes to offering comprehensive solutions and alternate routes of course, things get trickier. When Diffraction Analysis released Can the New Zealand NGA Model Be Replicated? a few months ago, our goal was to suggest that the example of New Zealand could constitute an inspiration for European policy makers.

A couple of weeks back, the FTTH Council published a study they commissioned entitled Financing Stimulus for FTTH. The study was undertaken by Ventura Team and Portland Advisers, and it’s frankly a stunning piece of work. The headline in the Council’s PR was that “at the present rate it will take 92 years to fiber up Europe”. It’s a strong message, and a clear incentive for change, but it only scratches the surface of what is in this study. In fact the reason it took me so long to write about it is that I had to read it and understand it first. It’s a very detailed piece of work and while the language and concepts are mostly clear, you still need a lot of focus and attention to get through it and grasp the messages.

I really encourage you to read this piece of work, even if you’re not from the EU, because there are many things in there that I haven’t seen elsewhere. In particular, the study makes the connection between the players who are likely to deploy and their incentive to do so (or not), and the players who are likely to fund the deployment and their incentive to do so (or not). That analysis puts in sharp relief the disconnect between the strategies of telecom incumbents in Europe and the expectations of infrastructure funds. It examines the nature of the financial players involved and the amount of funding they could make available for such deployment, if the conditions were right for them. There’s a table on p34 which is the clearest transverse view of funding of telecom infrastructure I’ve ever seen. Very clear and to the point.

The study also delves in-depth on one of the most important aspects of a fiber switchover, ie. the relative price of wholesale copper vs. wholesale fiber. We’ve heard all kinds of things on these issues in recent months, with both incumbents and competitive operators arguing that copper wholesale prices should go up to free up more cash to invest in fiber. One would have to be very naïve indeed to believe that either would actually invest more in infrastructure were that to be the case. The study clearly demonstrates that this approach is nonsensical if the policy goals are to be met, and that the only workable scenario is actually to have copper wholesale prices at least 15% higher than fiber wholesale prices. Not only does this make sense going forward, but it’s also part of the social contract of regulated prices, designed to ensure a re-investment in network assets that didn’t effectively happen for copper.

Finally, the study examines a framework that would enable telcos to remain masters of their own infrastructure fate in black areas but would structure the grey and white markets through institutionalised public-private partnerships that lead to the creation of NetCos in a way reminiscent of the New Zealand scenario. Needless to say that aligns with my views, but is way better argued in the study than I could ever have done.

Long story short, if you care about this issue and want to examine a really clever way to move forward, this is a must read. I can only commend the FTTH Council Europe for commissioning and publishing such a study. It’s the most comprehensive piece of work in this area that I have seen in a long while, and hopefully the fact that it’s widely distributed will win hearts and minds where it matters.

Telecom Italia goes full-blown FTTC

10 Dec

A few years ago, it looked like Italy might be a place where interesting broadband deployment models might emerge. In particular, a consortium of alternative operators all working together pushed a radical proposal to pool resources for the deployment of a unique national FTTH point-to-point network. The key to making that work of course was the willing participation of incumbent Telecom Italia. Having rejected FTTC as a solution, and being cash-strapped, it looking for a while like Telecom Italia might agree.

Alas, things have changed and the Italian model now looks like so many others: Telecom Italia is embracing FTTC with FTTH deployment being a reality only in Milan (using Metroweb’s infrastructure). The plans for FTTC deployment target 6.1 million homes passed end 2014 in 100 cities. Somewhere in there is the Milan FTTH and possibly one or two other cities, but that sounds like wishful thinking to me.

Meanwhile, the ambitious Lombardia FTTH project seems to have ground to a halt as well. I haven’t heard anything about it in ages, and I don’t know if that’s for financial, technical or political reasons (and probably a mix of the three).

So there you have it. For a long time, Italy was a leader in the European rankings because of the legacy of Fastweb’s deployment in Milan. But these numbers have stayed flat for years, and it’s not looking like they will be growing anytime soon.

City Fibre Holdings in York

6 Dec

Ever since the official announcement earlier this year, I have been keeping my eyes peeled for more information on developments in York. An article in Thinkbroadband earlier this week summarized what we know about the plans so far: Fibre to within 200 metres of 80% of York businesses by March 2013. The title really says it all.

There have been some interrogations about City Fiber Holdings’ plans going forward considering the history of the company they purchased the assets from (i3) and the relative silence that followed the purchase. Now with services having been launched in Bournemouth and the York project going forward the dynamic is become apparent again.

Considering CFH is currently the largest FTTH network in the UK despite their relatively small size, their development is going to be interesting to follow.