Archive | Technology RSS feed for this section

Boosting copper: necessary and dangerous

22 Sep

BBWF Europe next week will clearly be about copper. Virtually every vendor or PR agency that contacted me wanted to brief me on VDSL, vectoring, bonding and all that jazz. Just this morning, Alcatel-Lucent announced their big push in this direction (see Alcatel-Lucent accelerates the availability of superfast broadband.) My coverage of these technologies on fiberevolution in the past might have seemed somewhat simplistic, and it may have come accross as me disagreeing with them without nuance. That's actually not where I think their role is. 

Broad FTTH deployment is a long-term endeavour. It's unlikely that any country other than really small ones (like Singapour) or really authoritarian ones (like China) will achieve any broad scale deployment in less than ten years. The question of the quality of services delivered to the people who will not be in the early waves of fiber to the home deployment will therefore be crucial. And FTTC, enhanced by Vectoring and Bonding if necessary can be a stop gap to ensure that while these people don't get as good service as the ones connected with fiber, they will at least get services that allow them to continue experiencing internet services in tolerable comfort (let's not forget that as average capacity to the home increases accross the world, the average load of websites and online services will also increase…)

That's why I don't oppose VDSL2, bonding, vectoring and other phantom modes on principle. I do however believe that there are a couple of real dangers associated with these technologies. The first is clearly that they give the opportunity to incumbents who hate the idea of deploying FTTH to simply substitute VDSL2 for FTTH and not undergo the necessary network investments. That's a really short-term view that can only have dire consequences in the long-term. Instead of being a stop gap on the way to FTTH, VDSL2 becomes an end in and of itself. That's clearly the path that AT&T, BT and Belgacom (to name a few) have chosen, and while it might look like it's solved the "problem", wait a couple of years and we'll have to go through the whole rigmarole of insufficient service levels and slipping OECD broadband rankings again.

Furthermore, I have some real questions about how these technologies can be implemented in countries with progressive access regulation without jeopardizing the models that have proven their worth. In the UK, we have already seen quite clearly that the implementation of FTTC pushes the market from a proven unbundling access regime to an unproven bitstream access regime. It's quite clear why incumbents would want that, it's a lot less clear why competitive operators should. And it's very unclear how regulators are dealing with these issues. A few years ago, a number of European NRAs  (ARCEP and OPTA spring to mind) VDSL was considered not to be a solution that would sustain a competitive market  for just these reasons, so I'm a little puzzled as to what has changed now…

How do these technologies cohabit with unbundled copper regimes, I wonder. That's really what I'll be investigating at BBWF: is there a regulatory risk associated with VDSL and how can it be dealt with?

BBC's click on IPTV (and Free)

4 Aug

A fairly low-level short documentary on IPTV in France. I can't seem to embed the video, so I'll just hotlink it.

A few quick comments:

Analogies are evil: are we driving on a dedicated fast lane or reserving a seat on a train when we watch IPTV? Neither.

Micro-targeted ads don't make economic sense: this has been a huge promise of IPTV, but it's not going to happen because niche markets will never be big enough to sustain the ad production.

Or maybe we'll go back to ads like this (in French and NSFW:



Getting to know Huawei

4 May

Last week I was in Shenzhen and Shanghai at the invitation of Huawei. Huawei, like most other large telecom vendors, organises annual meetings for analysts: the aim is to present their strategy to us and get some feedback from the analyst community.

Huawei's success is undeniable. The numbers speak for themselves: they are the #1 in volume in many markets (including FTTH), and their annual profit increased by 30% last year at a time when most vendors are struggling. Their growth is not just in China, of course, but China is an important market for them and so it made sense to go and meet them on their turf to try and get to know them better. 

The first thing that struck me is the sheer scale of their organisation. Of course, again, the numbers are striking (130 000 employees worldwide), but I never really got a sense of that scale until I visited their fixed line HQ in Shenzhen. The word "offices" or even "campus" doesn't describe it. It's a city within the city…

The analyst conference itself had its share of big announcements, but two of them stand out as trhe kind of bold moves that most established players in the market no longer dare make: 

  • Huawei is going to start marketing its own brand devices. They are currently the 8th mobile device manufacturer worldwide (if I remember correctly), but until now they only sold white labelled models to carriers. From now on this will no longer be the case. I don't know the mobile device space well enough to assess their chances of success, but it seems to me at least that the timing, as the Nokia giant is teetering on its pedestal, is appropriate.
  • Huawei is also this year launching an official enterprise business with the clear aim of eating up a chunk of Cisco's market. Just to give you a sense of the muscle Huawei can bring to bear to such an endeavour, they migrated 10 000 people (8% of the overall workforce) overnight to this new entity. Not exactly something that many players in the market can do or even would do if they could.

Will boldness pay ? I don't know. What I do know is that boldness is clearly part of their corporate culture. I was surprised, when I attended the first of their European analyst conferences three years ago at the tone of the delivery. They don't sugar coat their messages and they boast unabashedly at their achievements. The kind of message we would hear three years ago was "in this particular market, we trounced ALU and Ericsson". It still is.

While this may seem a little crass to our Western ears used to polite corporate messages and considering corporate humbleness as a virtue, it's also somewhat refreshing to hear a vendor being fairly transparent as to what their goals and achievements are. While their messaging has become more sophisticated in the last couple of years, the brashness has remained. 

Which doesn't mean Huawei isn't facing a number of issues in its expansion, issues that might not be overcome by just being bold.  

One of these issues was that of a corporate culture too centered on China. That's changing, with an open admission at the European analyst conference last yeat that the globalization of the employee mindset was necessary and being addressed. That's not to say that there aren't hurdles remaining in that space, but their management is clearly becoming more global and increasingly I find myself having real conversations with their executives as opposed to the one-way broadcast of their messages which used to be the case a few years ago.

Another issue, particularly in North America is the perceived affiliation with the Chinese state. This is in part rumours that through the Huawei founder, an an ex-Chinese PLA man, the Chinese army really controls all these routers being deployed all around the world. That smacks to me as guttter level conspirationism, and now seems to be actually believed only in the US (and India).

An accusation that it more difficult for Huawei to brush off is their access to effectively "free" loans from the Chinese development banks and sovereign funds. I have no real way to assess whether that's true or not (and if true it would (or at least should) probably constitute an issue with WTO rules). It's both an asset and a problem though, as vendor financing by Huawei has allowed them to win many a deal these last few years.

Huawei themselves pointed to other macro-economic issues, particularly the gradual appreciation of the Chinese Yuan which means that their cost of labour advantage from building in China is diminishing. While true, it's still far from really hampering their stride at this stage.

I don't know if Huawei will succeed at these new endeavours or even continue to be as successful on their established markets as they are today, but what I do know is that they are certainly moving with great momentum towards these stated goals. They are largely unencumbered by some of the western constraints (especially around employee management) and they are acquiring a global culture, faster than many would have anticipated.

It was worth meeting them on their turf, if only to learn that. And since China will likely be the first FTTH market worldwide within a very short period of time, knowing Huawei better is worth its weight in glass…

In which I rant against the all-mobile utopians

13 Dec

Bell Tells it Like it Is (in Chinese)

26 Jul

HKBN’s ad creativity is a big part of the success. This new one ads a new twist, but it’s just as funny and in your face as their previous endeavours:

Google holds a trenching race!

15 Jul

When I was in US last, I heard that Google had organised a fiber trenching contest in a parking lot in Mountain View, and I just found out today thanks to my friends at that there was a video released of the 'event' if you can call it that:

It's an interesting (and fun) way to select a provider for trenching work, butit doesn't alleviate a particular concern of mine which is that in the context of their FTTH trial, Google focuses too much on the technical side of things and too little on getting the business model right and the ecosystem aligned.

Then again, it's harder to do a video of a business model contest…

Reflections on Cable, FTTx and future-proofing…

17 May

Last week, I had the opportunity to speak with the CTO of a European Cable Operator who shall remain nameless, at least for now. You see, I've been trying to ascertain the upgrade path of cable for the last few months, cutting through the vendor bull and the boastful (and often silly) declarations of cable associations here and there. I also had the opportunity for an interesting talk with the technology strategy people at an incumbent operator who shall also remain nameless.

These were enlightening discussions. They made me realise a number of things:

  • not all cable operators are made equal: now this is an evidence, the same way that not all telecom operators are made equal. Clearly though, technology choices made 15 years ago are either coming back to haunt some cable operators or (as seems to be the case with the company I'm thinking of) opening new opportunities in the internet age. These guys have a huge proportion of their footprint very close to the end user (less than 1/2 a km on average) which means that they can deliver high speeds today without too much effort and have multiple upgrade paths to very high speeds without excessive reinvestment.
  • a sound long-term technology strategy is the underpinning of a sound commercial strategy: interestingly, talking to this CTO, I realised that his company (and in a large part I think it was down to him) has had a lot of foresight. This tied in interestingly with the discussion I had with the incumbent's strategy people. They seemed gung-ho on FTTC and argued that FTTH was not commercially viable. That's a sensible position, but I asked them what they were intending to do in 5-7 years when they will have squeezed the last drop of bandwidth out of their FTTC. They didn't have an answer.

There's an overall issue with the industry today which I have already touched on (and it's not just the telecom sector, it's accross the board) and that is that businesses no longer know how to think and make decisions for the long term. That doesn't mean that you have to walk into dead ends knowingly though. When a CTO presents his technology strategy to the board, it should look good in the short term and keep options open for the long-term. Ultimately, it's not even about your technology choice (in this instance
cable vs. fttx), it's about how intelligently you implement them to
avoid these dead ends…

You'd be surprised how few cable and telecom companies I have seen think along those lines…

Definitive Views on GPON vs EP2P…

5 Feb

If you were one of the early followers of Fiberevolution, you know that one of the first "visible" posts explored the issue of incumbent GPON deployment.I did not know it back then (although I found out real fast!), but I had stumbled upon one of the great religious wars of the NGA world (the other being cable vs. fiber.) I revisited the issue several times on this blog, and over the course of 18 months, I talked on this issue to numerous service providers and vendors.

Last November, I started to write a report that would summarize my views. This report, entitled GPON or EP2P: Choosing an FTTH Architecture is the result of these discussions and a string of specific interviews. Vendors interviewed include Cisco, Alcatel-Lucent, Hitachi, LG-Nortel, Ericsson, Huawei, PacketFront and ADVA Optical. Service providers interviewed for this report include Reggefiber, Telecom New Zealand, Amsterdam Citynet, BT, PBC, Free/Iliad, i3, Swisscom and Fastweb, Lafayette Utilities System, Verizon and more.

For obvious reasons, I can't share this report or even its conclusions here. It's available to Yankee Group Link Research customers and can be custom purchased on the Yankee Group online store as well. What I can do here is tell you what I tried to achieve, and hopefully convince you that the report is worth looking into.

The starting point of my thought process was that the debate as it is often framed by vendors is sterile. Both sides more or less claim that theirs is the superior technology at lower costs. That's not helpful to service providers (and, I suspect, not to vendor credibility either…) A technology, after all, is only as good as what you do with it. What I endeavoured to do therefore was to frame the technology choices in context by answering these two questions: which technology choice is best aligned with which strategic objectives, and which deployment model should be chosen (for GPON in particular) to maintain that alignment.

I hope you find it interesting and challenging and would welcome comments!

Are Street Cabinets Such a Good Idea ?

21 Jan

Numericable Cabinets As you will know, I have been skeptical about the potential of Fiber to the Cabinet, mostly due to what I perceive as a lack of future-proofness (yeah, I know such a word doesn't exist…) of the deployment. One aspect that I honestly haven't investigated much is the impact of having cabinets in the streets. After all, I thought, cable operators have had them forever…

I guess I should have looked more closely at some of the posts James put up on eurotelcoblog. The photo on the right is a photo of one of Numéricable's cabinets in Marseille. Last year, a group of people went through several districts of Marseille and tracked down cabinets to photograph them. The results are here for the 14th district and here for the 3rd district.

This is truly frightening stuff. And the worse is I see no reason why this would happen to Numéricable and Virgin and not to others, including FTTC deployers. If you see similar sights, please send me photos!

A third P2P incumbent in Europe

12 Oct

Ch_dp Last week I mentioned Telia Sonera's announced P2P deployment saying that it was the second incumbent in Europe after KPN to choose Ethernet Point to Point for their national roll-out. Turns out I was wrong. For some reason I was under the missaprehension that Swisscom was deploying GPON, but I was wrong on that one. I spent some time with people from the Swiss incumbent at ITU in Geneva, and they fixed that error for me.

It's not an insignificant roll-out either since the plan is for 100.000 homes to be connected at the end of this year and 1 million by 2015. Interestingly, last week when I posted the news about Telia Sonera I again received a number of emails and direct tweets saying that incumbents would never accept EP2P if only because it makes unbundling too easy. It seems that argument alone may not be sufficient for those incumbents who have done the math and realise that open access makes economic sense