Fiberevolution

Punitive charges for rights of way threaten Indian broadband expansion

2012-01-25T14:00:00+01:00

If you read me on a regular basis, you will know that I'm certainly not amongst those that condemn regulation by default, far from it. I think in many countries, and at least as far as the telecom sector (and the finance sector, but that's not the topic here) are concerned the issue is not too much regulation it's too little.

However, there are clear examples, especially in emerging markets of regulation and policy pulling in opposite directions. Case in point: India. The Indian government in the last couple of years has been gung ho about broadband deployment and adoption, especially wireline broadband (to a large extent, wireless broadband deploys itself in India). India is a very odd market in that respect: with a population upwards of 1.2bn inhabitants, there are only 10m homes with wireline broadband in the whole of the country, a penetration rate over population of 0,8%, one of the lowest in the world.

The Indian government plan is therefore necessary and timely (if not a little late). The most visible aspect of the plan is the intent to fiber up 250 000 panchayat. These aren't planned as access networks but rather aggregation networks which can then be extended to customers by wireline or wireless access networks. And in a market where copper deployment is minimal, there is a clear opportunity to leapfrog copper altogether.

Or there would be, if prices for rights of way weren't unbelievable punitive. This recent article from the India Express entitled BSNL complains of higher digging charges by civic bodies details what we're talking about. BSNL (the state telecom operator operating over the whole of India except Delhi and Mumbai) is complaining that the rates charged by certain states for rights of ways are three times higher than the rates paid by electricity companies for no apparent reason. Said states respond that that's the "standard" rate for telecom players, so there's no reason to complain.

Here's the hitch though: the rate that the eletricty company is paying is equivalent to USD 30 000 per km. And the rate BSNL is charged is therefore USD 90 000 per km. In a country where existing "unlimited" 512kbps broadband offers cost roughly USD 15 a month (750 Rs), a price that is definitely not affordable by the majority of the population, the business case for widespread deployment of any wireline network simply collapses on the basis of costs of rights of way alone.

It's a shame when one thinks of the potential in a country with an emerging middle class, low labor costs and an appetite for content... My understanding though is that these costs are under the jurisdiction of state governments or even municipal bodies below state governments. While that may explain in part why the national government policies are at odds with the punitive rates for rights of ways, it suggests that there is opportunity for an enlightened local or state government to dramatically lower these costs and thus open up the infrastructure market in a radical way... and reap the economic benefits in added growth, business attractiveness and other income taxes... Anurag Bhatia, a local blogger and tech brain who alerted me to these rates tells me that the state of Gujarat may be implementing such a policy. I need to investigate that.

Google Fiber Trouble

2012-01-24T14:00:00+01:00

For a number of weeks now, we've been hearing that Google has entered some troubled waters when it comes to its Kansas City Fiber project. The first customers were supposed to be connected in early 2012, and that's clearly not the case. Furthermore, local newspapers report that the fiber network deployed by the Mountain View giant is far from being ready for commercial operations. The delay, it seems, is due to Google and local stakeholders disagreeing on the conditions and rate at which Google should be able to access the poles to attach its fiber to.

Reading between the lines, it seems that Google thought, or hoped, that because of their brand name and because of the visibility of the project for Kansas City, they would be able to speed up normal procedures and get preferential rates. Reality, alas, seems to have reasserted itself. The Google brand cannot steamroll over everything, apparently.

The latest rumor is that Google would consider buying Surewest, a Sacramento based FTTH and cable operator who has been operating its cable plant in Kansas City for a number of years. Fierce IPTV reports this as a rumor, but reports it nonetheless. The rationale being that by using Surewest's existing authorizations, Google could speed up its deployment and meet its targets.

For what it's worth, I don't buy it. The original lofty goals of the Google Fiber initiative have already been considerably watered down, and the "blueprint" nature of the project is already seriously in question, but if Google needs to buy an existing and successful business to achieve its goals, not only is it admitting failure, it's throwing out the window any pretense that their model is replicable and that other municipals could do what they're doing.

Still, there's no doubt Google is in rocky terrain right now. The company has been known in the past to pull the plug on projects that didn't achieve their targets. Will Google Fiber be the latest victim?

Thinking Long-Term

2012-01-23T11:56:11+01:00

I was reading the november issue of Wired on the plane this week (the beauty of digital copies on tablets is that you can catch up on past issues!) and something struck me while reading the article on Amazon and Jeff Bezos entitled Jeff Bezos Owns The Web In More Ways Than You Think. I wanted to share that short quote (and strongly recommend reading the interview; it's a clear illustration that the Apple model is not the only option...) because it's relevant to infrastructure discussions:

Our first shareholder letter, in 1997, was entitled, “It’s all about the long term.” If everything you do needs to work on a three-year time horizon, then you’re competing against a lot of people. But if you’re willing to invest on a seven-year time horizon, you’re now competing against a fraction of those people, because very few companies are willing to do that. Just by lengthening the time horizon, you can engage in endeavors that you could never otherwise pursue.

The revolution is not where you might think...

2012-01-13T15:52:08+01:00

The French media have been abuzz since Tuesday following the press conference by Iliad's Xavier Niel on the launch of Free's Mobile offering. Mobile may not be the core topic for this blog but since I talked a lot about Free in the past and certain aspects of their business model and since I will soon be writing about their fiber fiasco, it made sense to lay the groundwork for a broader reflexion on disruptive business models while it's hot.

I won't go in any great detail on the announcement itself and Free's offers, you can find context about it in English easily if you google it up. Om Malik has a general context article (he met with Niel before the release, so it's not specific, but it's a good primer) and The Guardian has an article that describes in broad terms the offering they have launched.

What interests me is the answer to two questions, essentially:

how much is Iliad moving outside of its proven model with this launch?
how much of a revolution is this really?

The first question might sound odd, but in my opinion it's really important for two reasons. First because Iliad has made its mark in France and beyond with a very specific and counter-intuitive model that has proven it works. Second because they have shown in the past, especially with their FTTH initiative that they can fail when they move too far out of their model.

The second question is a broader reflection on the French market and how much disruption this launch is likely to generate.

A partial extension of the Free model

If I had to summarize in three bullet points the specifics of Free's offerings so far and why they were successful, I would say this:

simple offers with few options to maximise legibility for customers
aggressive prices that appeal to people on a budget or casual users
rich bundles with innovative services that appeal to geeks and push new usages amongst the general population

With these new offers, Iliad checks the first two marks, but completely misses out on the third, which raises an interesting question: is price sufficient to sustain the model?

I'm not convinced it is.

No doubt that Free will attract customers who are price sensitive, but if they only attract such customers, they won't get the market share they hope for (Niel mentioned a 25% market share goal, though didn't say what the business model breakeven was) and they will have attracted low-profitability customers which ultimately might not be such a bad deal for the other players in the market.

The real make-or-break question is, without any innovation in their offering, without any specific services or usage, will they attract the kinds of customers who are currently paying upwards of 50€/m to the other operators?

Of course right now everybody is up in arms, Orange, SFR and Bouygues have slashed severly in the prices of their "cheapo" brands (thus confirming the not so subliminal message Free is pushing that they were "screwing their customers" until now) and it looks like 50% of the market will subscribe to Free. When the hulubaloo dies down, many customers are just going to do what most customers always do: stay where they are because the hassle of change is just too much to deal with.

Without a cool/geeky element to the offer, Free is effectively missing out on what made their success in DSL: low price combined with innovative service components.

Fans of the offer are saying that these things will come, and maybe they will but I wonder if Free hasn't missed the best possible window to introduce disruptive usage into the mix (and by the way, in case anyone wonders, things that could have made a difference: femtocells in every box, fully integrated ott apps, 4G/LTE, 3 screen approach, etc.)

A revolution in perception, maybe

Iliad entering the market was presented as a revolution by the company itself and most of the press, but is it really? It's unlikely to be a revolution in usage unless one thinks that 3GB of data in a 20€/m plan will boost mobile data usage exponentially (both in volume and adoption). I doubt that somehow, but you never know.

If there's no usage revolution, the only revolutionary aspect left will be the bloodbath. That's extremely likely, even if Free fails to reach it's penetration targets. Within a few days of them announcing their tariffs, competitors had slashed some of their prices in half or more (with, as mentioned above, a likely lasting negative effect in perception) and that's a path that there's no coming back from. It's understandable that they would run around like headless chicken, and it's not like they had no time to anticipate, but it's probably fair to say that they did not expect such price aggressiveness.

Customers are whooping with joy and touting the virtues of competition, as well they should be. But there should be no naiveté about this launch: the blood bath will have impact on the general market value (value destruction) on investment capacity and on employment. There's nothing inherently wrong there, these are all standard free market effects. The issue will arise from the brutal change resulting from a combination of a cartel exploding in the face of a disagreeing competitor and said competitor betting on volume to make up for low margins. The impact will occur in a very short period of time and the industry and some of its employees will inevitably suffer. There was even an ironic (considering their free-market advocacy) short note in today's Challenge (Free va coûter 4 milliards à l'état) arguing that the state would likely loose €4bn in tax revenues as a consequence of the introduction of real competition in the market.

There is however one way in which Iliad's offer could be considered not revolutionary per se but driving a revolution, and that is the perception of hidden costs. Iliad has done its utmost to avoid all the hidden costs that customers tend to hate but could not avoid until now. Consumer advocacy group UFC Que Choisir published a list of all the noteworthy specifics in the terms and conditions, and while there are still a few aspects that might be considered dodgy (especially around roaming in certain countries), by and large the contract is as short, clean and understandable as they come and certainly much cleaner than anything the industry has seen so far.

That's exactly the point that Rudolf van der Berg argues in this article in Les Echos entitled Grâce à la clarté de ses forfaits, Free ouvre en grand les robinets du mobile. Basically by making their offers transparent, they are going to force transparency onto the market and forever change the perception and expectations of customers.

I'll buy that. That is a true revolution, and will likely happen no matter how well Iliad fares...

Email and beyond

2012-01-04T19:02:27+01:00

I haven't posted about much unrelated directly to the coverage of this blog in the last few months. Although you might see that as a good thing, I feel like musing on a topic that is in no way directly related to NGAs: email.

I have long thought that email was badly implemented in businesses, mostly because users were not guided into how it should be used optimally. The consequence is email glut, which leads to information noise and, ultimately, miscommunication. There are other email related issues in organisations, but that one is most likely the biggest.

I was reminded of this today by a short post on Gareth Spence's Online Scribbles and - just a few minutes later - by a retweet of an article written in August by Techcrunch's MC Siegler entitled Inbox 10000: Some thoughts after a month away from email. Over lunch my mind started churning again.

The backdrop to Gareth's point of view is the recent announcement by ATOS' CEO Thierry Breton of a zero email policy for the company within the year. It's not the first such announcement, but it maybe is the first for such a large organisation (80 000 employees worldwide). As Gareth points out, email is already partly substituted by other tools (he mentions Salesforce Chatter and Rypple, I've used or seriously investigated neither), and Gareth bets that these tools will continue to grow.

I have no reason to doubt that, but I have reservations as well. The results of Siegler's experiment, when he stopped using email for a month are interesting. Although Siegler makes light of the fact in his article, he couldn't actually ban email completely. He didn't use it much, but he did use it. He estimates though that he received over 10k emails in that month, and just not reading those and not responding to them was hugely liberating. I can totally see that.

Still, I'm not sure banning the tool is the right way to go. Here are a few reasons why:

We're clearly not using the tool properly, and have little to know guidance on how to use it well, especially in a work environment. Very few companies actually tell you when you should or shouldn't email, when you should CC or not, etc. All of that contributes hugely to the noise.

As sophisticated as our inbox tools are, they are still unbelievably klunky. Gmail is probably the best tool around, but if you want to be a gmail power user and actually set it up to optimise your emailing flow, you have to learn the intricacies of a complex tool and spend a heck of a lot of time setting it up. Something that regular users will never do. One of the comments to Siegler's story had a link to http://zeromail.com/ and that looked interesting and promising. Considering I know nothing about them, I'm reluctant to try it out but intrigued at the same time.

Finally, and most importantly, the email and the internet share a very important feature: they are universal. Emails can be written and read no matter which client(s) you use, who your ISP(s) is/are, which device(s) you're writing from or reading on. How do you replace a universal communication system without the replacement being just as open or universal? To take Gareth's examples, chatter and rypple are dependent on a contract with a software company. Social networking relies on accounts being open with facebook, twitter or others. IM also requires specific (ie non-interoprable) clients and accounts. As a consequence none of them will replace email. None of them can. They will divert some of its traffic for sure, they already are.

So email is here to stay. I'm curious to see what Breton comes up with to replace email. That's the part ATOS has been stangely silent about. My suspiscion is that it won't eliminate email completely (if only for external communications) and will replicate a lot of its failings as well. But maybe I'm too pessimistic.

My own view is this: if email has to stay, even if it's only for 10s of messages a month, that's one more tool. Why not rethink email instead of trying to add more tools and complexities to the communication processes ?

If that's what www.zeromail.com does and they turn out to be reputable, I'd like to try that first...

The #1 Connectivity Topic for 2012 is NOT digital rights!

2012-01-02T15:05:34+01:00

It's the new year, usually a good time to reflect on what's been happening in the past year and what we hope to see in the coming one. These last few weeks I have been reflecting, but the fruits of these reflection is not, I'm afraid, very positive.

I have seen with a increasing worry the crucially important topic of IT and internet development get political and mainstream attention these last few years solely through one prism: digital content rights. Whether HADOPI in France (which although voted in prior to 2011 kept making the headlines last year), SOPA in the US and numerous variants of either in other countries, that is the only IT topic that has been visible politically in 2011.

Not only is it depressing, it's non-sensical. Don't get me wrong, dealing with issues related to digital rights is important. It's a lot more important in industrial IP than it is in content (in my opinion) as the current patent wars in mobile technologies fully illustrate. The framework for intellectual rights hasn't really been altered in over 100 years so clearly it is in need to rethinking. That's exactly what politicians accross the world are not doing with laws like HADOPI and SOPA.

More importantly though, digital content rights, which are at the heart of all these political discussions (and as a consequence are often seen as the only IT related issue by both politicial decision makers and the public) are a really really tiny problem.

The content industry, if you aggregate music and TV/Cinema (other content industries like books or photography are not sexy enough to be considered as part of these discussions... or maybe they're not lobbying hard enough!) is a really tiny industry in terms of jobs and wealth creation. If you don't believe me, read the timeless Content is not King by Andrew Odlyzko. The numbers may be outdated, but they're actually evolving in a direction that reinforces the trend (ie. content is getting smaller relative to networks).

Meanwhile, the internet as we know it (and more generally ubiquitous connectivity) is an economic powerhouse. It creates jobs and wealth, it has led to the emergence of a new generation of entrepreneurs all over the world, it has driven huge efficiency gains in businesses, with about as much or more to be gained still. It has opened new markets (and yes, disrupted a few).

As far as politicians are concerned though, all of that doesn't count! The internet is evil, because it's not centrally controlled. And that allows artists' works to be pirated. Incidentally, it also allows for politicians' snafus to be broadcast far and wide, and I suspect that's in part why they like the idea of a centralized control so much.

So the trend is there: kill the internet to protect the artists. Politicians would never agree publicly that this is what they are doing, but it is. In order to protect a few (generally already wealthy) artists and their labels, they are willing to sacrifice the golden egg goose. That's bad enough, but it's not the heart of the issue as far as I'm concerned.

Politicians and governments should be focusing on transforming society with connectivity in mind instead of viewing it as the serial killer. How come 20 years after the emergence of the internet administrations and governments all over the world are still being run like they were in the 80s? We're neck deep in debt, isn't it time for the efficiency gains that we've seen in private businesses to make it into public institutions?

We're tackling huge issues collectively of resource scarcity, pollution, unbalanced welfare systems, looming healthcare crisis, excessive urban development, etc. and our politicians miss the forest for the tree. Instead of looking at networking as a means to address some of these issues, streamline the antiquated processes that are at the very least a hindrance to finding solutions - if not part of the problem themselves - they choose instead to demonize it to protect a tiny corner of the economy.

The irony is even starker here in Europe where most of the efforts to bridle the internet are deployed to protect Hollywood and large US record companies. At the same time our politicians complain that most of the internet economy is happening in the US and hamper European companies in order to protect an essentially US-centric industry.

And the worse is that we - the public - fall for the trap. We get distracted from the real opportunities as well, by all this pandering and posturing. There are major elections in many countries this year. Presidential elections in France and in the US. Let's make them matter. Let's challenge candidates not on how they intend to deal with digital rights but on how they intend to make IT and the internet count. How they expect it to shake tomorrow's world.

I don't expect them to have convincing answers. I do expect it to shape how I will vote. This is not a single-issue platform, this is a societal issue. It's about time we collectively embraced it. It may be a crucial key to extricating ourselves from the economic quagmire we have let our politicians facilitate.

Xavier Niel on French TV

2011-12-22T14:28:27+01:00

There was a feature on Xavier Niel on a French TV News show this week. Reproduced here, but it's (obviously) all in French...

The other side of the FTTH Ecosystem

2011-12-19T15:10:09+01:00

Last week in NYC I met with Hunter Newby, the founder of Allied Fiber and of Telx before that. Not only is hunter a super-sharp and very nice guy, he's also a forward thinker when it comes to telecom infrastructure.

Allied Fiber, for those who don't know, is a relatively recent business that will be offering long-distance fiber in the US and colo- exchanges along the routes. This will for the first time introduce genuine competition in a market that was until now largely held by a small number of players, with the inevitable consequences on prices.

As you probably know by now, we recently released a report on what actually happens in wireline broadband networks when consumers all try to access resources at the same time (see Do data caps punish the wrong users?) One of the interesting aspects of that report was focusing on where the variable costs are in operating a broadband network. Our key assumption was that in competitive markets, only peak load demand has an impact on your costs. But not everywhere is a competitive market, even in America.

When we initiated the "bandwidth hog" debate back in late 2009, one of the things that came again and again in the comments from Wireless ISPs in particular was that the cost of long-distance fiber was such that even relatively low levels of bandwidth usage were prohibitively expensive. That's part of the market opportunity that Allied Fiber sees and Hunter is keen to bring competitive rates to those markets (an another aspect of the market opportunity is that fiber that was deployed in the 90s isn't necessarily optimal for today's communications. Even though was excess capacity was deployed back then, it doesn't translate into affordable connectivity today...)

So interestingly, I found that Hunter and I were looking in a very similar way at the two sides of a same ecosystem. Many municipalities are looking into ways to get high-quality broadband into their territory, and sometimes are willing to go so far as to invest in the infrastructure themselves. But what's the cost of aggregating all that traffic back to an internet hub if there is no competition in the market and/or if the infrastructure in place is itself strained because it's too old?

It's not every day in America that I meet people who understand the dynamics of open access dark fiber. Hunter is one of them. I had a great breakfast conversation with him and have no doubt our paths will cross again.

Sonic.net looking to expand FTTH to San Francisco

2011-12-14T10:56:41+01:00

Californian ISP Sonic.net has filed an application to deploy street cabinets in San Francisco in order to deploy FTTH in the City. This could be an important development in a geography where few options are available to get good quality broadband. However, according to this article in the Chronicle, sonic.net could face challenges as AT&T has for environmental reasons. AT&T has been constrained in developing its U-Verse FTTC service in San Francisco due to these same issues with their cabinets. Note that Fiber cabinets are normally substantially smaller than copper cabinets and I haven't heard of any FTTH developments hindered by these issues, whereas other FTTC operations (like BT's) have faced such challenges.

In any case, this is an interesting development that suggests that sonic.net's Fusion product is successful and that their FTTH trials in Sebastopol are promising...

(Note: the Chronicle article is paywalled until later today, I'm hoping the url will stay the same.)/

Sonic.net looking to expand FTTH to San Francisco

2011-12-14T10:56:39+01:00