Tuesday, September 30, 2008

Singapore and Malaysia: worlds apart

Singapore and Malaysia are walking two very well distinguished roads toward NGA (next-generation access) - and Singapore proves vastly superior, we believe. Not only do they aim for nationwide FTTH (OK, that would be a little easier than in Malaysia), the network will also be structurally separated and hence providing full open access. Malaysia on the other hand opts for an extended regulatory holiday for Telekom Malaysia: until 2015! In the process, it will create a huge digital divide.

To freshen up your mind, here are the specs (comments welcome to add more).

  • Name: Next Generation National Broadband Network.
  • FTTP, but apparently no decision on active (P2P) or passive (PON) yet.
  • Targets: 1 Gb/s, reaching 60% by 2010 and 95% by 2012; universal service obligation from 2013.
  • Passive layer (NetCo): to be built by OpenNet, a consortium of SingTel (30%), Axia (30%), Singapore Press (25%) and Singapore Power Telemedia (15%); maximum state subsidy SGD 750m. Wholesale prices: 15 SGD/mo for residentials, 50 SGD/mo for businesses, no connection fee. Furthermore, SingTel will be allowed/forced to spin-off ducts and other infrastructure into an AssetCo. OpenNet will lease those assets from AssetCo.
  • Active layer (OpCo): winner to be announced 09Q1 (bidders: BT, DT, SingTel, Axia, StarHub, MobileOne, City Telecom); maximum state subsidy SGD 250m.
  • Service layer (RSPs): to be announced.
  • Name: High Speed Broadband (HSBB) Project.
  • For high-density areas only. Target: 1.3m homes passed by 2012.
  • State to subsidise around 20% of the total cost ($3.3bn), Telekom Malaysia to be the rest.
  • No open access (regulatory holiday) until September 15, 2015.
  • More details 09Q1.

Consumer 3.D

Here's another development from video land, underpinning the need for more bandwidth: a 3-D camera from Fujifilm. For now a prototype, but it will be aimed at consumers, and will generate both stills and clips.

Video is not 'just' YouTube or IPTV. There is lots more to it, such as 3-D. Hence, what we are starting to see now is 1 Gb/s networks, whereas formerly 100 Mb/s used to be the benchmark. And that, dear reader, can only be done using FTTH.

Wednesday, September 24, 2008

New Poll: Structural separation?

My third poll (How to solve the broadband incentive problem) has ended, and here are the results:
  1. Usage-based metering: 9%
  2. Service-based tiers: 18%
  3. More bandwidth (FTTH): 31%
  4. Allow throttling: 15%
  5. Accept that access is a dumb pipe business: 53%
  6. Bundle access with other services: 25%
I suppose answers 1-4 are remedies of sorts, with FTTH fortunately by far the most popular. Answer 6 has surpisingly few proponents, or put differently: surprisingly many people still feel that access is good business (answer 5), despite the fact that it is quite 'dumb' (in other words: a utility service).

Of course, one could easily vote for all 6 answers since none of them excludes another, as a valued reader remarked.

Now, do take the new poll (see right).

Thursday, September 18, 2008

Changing the web

Two intersting links on the nature of the web and what we do with it:
  • How the Large Hadron Collider Might Change the Web (Scientific American). The LHC produces such a vast amount of data, that it requires a new kind of grid (the LHC Computing Grid) and middleware, called Globus, to analyse them. "Bader imagines future middleware allowing home computers to provide instant weather forecasts by accessing information from nearby environmental sensors. Or it might help sift through a life's accumulation of personal medical records or years of home video footage looking for dimly remembered events."
  • Where Are They Now? A Decade Of Google Products And Deals You've Forgotten About (Silicon Alley Insider). A concise overview of the betas that Google produced. There are 11 categories (search, maps, mobile & telecom, advertising, consumer products, social, for the public good, for developers, research, enterprise, statistics), and each contains both products and acquisitions. Let's hope they keep this portal up to date.

Wednesday, September 17, 2008

Land of war and mystery

Being part of the Fiber Ring sometimes brings along a Lord of the Rings feel. Fiberland too is full of war and mystery.

Take Italy. The government is preparing to pump EUR 1bn into a nationwide broadband effort at a total cost of EUR 10bn (at this price level one can only assume we are talking NGN or FTTC/VDSL, not NGA or FTTH - unless of course the government is thinking of creating a nice and big digital divide by only hooking up the north). Yesterday AGCOM added that Telecom Italia indeed needs this kind of government and PPP support, because it cannot do it on its own.
Interestingly, the whole effort is supposed to add 1.5-2.0 pp to GDP.

Can anyone explain to me how any such investmmnet could not be justified, in light of this GDP enhancement? (Doing a little Wikipedia research, it looks like the economic benefit could be something like EUR 28bn a year.)

Or take the Netherlands. UPC is usually fighting the threat of superior fiber bandwidths by launching extremely aggressive pricing. This time around however, things are turned upside down.

UPC (part of Liberty Global) is rolling out DOCSIS 3.0 (advertised as 120 Mb/s, but we all know this is shared and not symmetrical), and at the same time the Reggefiber/XMS combo is waiving installation costs (normally EUR 150) and maintains a very competitive promo (20 EUR/mo for the first 6 months).

Monday, September 15, 2008

Telefonica: no takeovers?

Telefonica, for the umptieth time, denied any interest in E-Plus (owned by KPN), or any other European telco for that matter.

I feel however that a 'minor' transaction in Germany (not E-Plus) cannot be ruled out. The market is fragmented and any next-generation investment requires a lot more scale than operators currently have. Still, China and Brazil seem more likely areas of interest.

Here are my comments:
  • They have been denying explicitly since the summer of 2007. Telefonica has a lot on its mind (China, Italy). Also, you obviously don't want to make the share price of any potential target skyrocket. Anyway, in the long run anything may happen (see this article from last year).
  • Europe isn't exactly appealing right now for any large telco to go shopping. Broadband and mobile markets are much more saturated than in most other countries, and regulation is currently under review. Telcos prefer high growth emerging markets, such as those held by TeliaSonera (itself sniffing at Poland's Play) or Telenor (itself looking to Africa). Telefonica is much more likely to focus on Brazil, through Telecom Italia (thanks Stefano).
  • Looking at Germany and including Telecom Italia assets, Telefonica already has a significant presence: O2 Germany, Telefonica Deutschland, and (via Telecom Italia) Hansenet. Still, Germany sooner or later will consolidate around DT, Vodafone, Telefonica and the current coalition of freenet/United/Drillisch/Versatel. It doesn't take a genius to predict that the latter grouping could be a target for Telefonica.

IBC2008: 3-D HD television is the future

No surprise for our readers, and no surprise coming from Jeffrey Katzenberg (in a 3-D HD address by satellite, quoted by B&C and TV Predictions): 3-D HDTV is "the most exciting thing to happen to the visual experience in 70 years".

I'm not sure how much bandwidth is needed for such a stream, but 100 Mb/s all of sudden isn't looking like a hell of a lot anymore.

Friday, September 12, 2008

Terahertz waves - now aimed at IEDs

Gaiacomm hasn't been able to deliver on its 4G promises, but now the technology is under scrutiny from the Office of Naval Research and the U.S. Naval Research Laboratory to help fight improvised explosive devices (IEDs).
"A practical system is years, maybe decades away", writes IEEE. Well, maybe not quite.

Thursday, September 11, 2008

Vodafone NL: real flat fees, but limited speeds

Vodafone NL is launching new flat fee internet access. Downloads speeds are lowered (from the current 3.6 Mb/s maximum), and above a certain cap there are no overage fees, but the speed drops even further (to 128 kb/s for all three tiers):
  • Standard: 20 EUR/mo, 0.768 Mb/s
  • Plus: 35 EUR/mo, 1.5 Mb/s
  • Super: 50 EUR/mo, 2 Mb/s
Vodafone is saying people prefer bill certainty over speed. I'm sure they're right.
At the same time, KPN is doubling its download speed maximum to 7.2 Mb/s (with Dongle). They say people want to save time. I'm sure they're right too.

Let's see who is more right.


Yesterday was D-Day on multiple fronts. Parties to celebrate and also some predictions made over here coming true.

First of all, my daughter turned 5, which is a major achievement in a child's life. It kept me very busy, and I'm sure it was a day to remember for her and everybody else present.

Over in Geneva people had their own party. CERN's Large Hadron Collider went live and anybody with an interest in physics, such as myself, is anxious to see results coming out of those 30 years of preparation and 6 billion euros. Check out this newspaper article (in Dutch), with a film clip (in English).

Talking about speed. The Amsterdam FTTH network, now connecting 40k homes, yesterday demo'ed a 1 Gb/s connection (imagine using that full bandwidth - it will blow you through Comcasts's monthly allowance of 250 GB in half an hour). A trial will go live shortly, but a commercial launch isn't planned yet. I somehow felt this was coming ...

On the side: this is a typical OA PPP project: open access public/private partnership. The passive layer is run by GNA, in which the city, housing corporations, banks and Reggefiber participate. The active layer is run by BBned (part of Telecom Italia) and InterNLnet (same Italian parent) is the service provider for this trial (there are other SPs active on the network).

On the side: I mentioned a study on the benefits of broadband and the impact of availability on usage in my previous post (the Eindhoven University report). Ventura Team of course come to a similar conclusion in March. Now the Milken Institute published a report, showing that Provo, with its troubled FTTH network, is the 'best performing city' (in terms of job creation). Even Lafayette (whose munifiber isn't live yet) is among the top 25 towns. Finally, check out this article on the benefit of broadband for Africa.

Something else I kind of predicted was DTT coming to the Daily Media box (see this post). By the way: I also talked about adding a DVR with the United Content Distributors guys, but at the time they didn't think that was necessary. We'll see.
Let me first bring the box back to your memory: it helps place-shift internet video streams to the TV (as other boxes from Apple, Sony and Sezmi do), but there is much more: VoD, interactivity, ease of use, targeted ads, several trials and a live deployment, and last but not least: a unique business model.
Guess what happened yesterday? They demo'ed a new version of the box, that includes a DTT (DVB-T) tuner. After I first wrote about the box, the people behind it acknowledged that adding such a tuner could be a smart strategy for making a deal with the local incumbent telco over here (KPN), which also happens to own the country's single DTT license and operations (extended with DVB-H technology). Or with any other DTT operator worldwide.
Apparently, there is even more. They also added DVB-C (cable) and DVB-S (satellite) capabilities. O, and FTTH is no problemo either.
In short: the box is a great tool for adding video or enhancing current digital telco TV offerings. We'll keep tracking the box, because yesterday a lot of cable execs were present at the demo.

Tuesday, September 09, 2008

FTTH in the UK: small leap of faith needed

Yesterday Analysys Mason published its fiber-in-the-UK report for the Broadband Stakeholders Group. Recently, there has been a wide range of FTTH related developments. Let's first make a little list of them:
  • Incufiber: KPN is steadily, if not stealthily, rolling out through its Glashart ('heart of glass') joint venture with privately owned Reggefiber. New towns are coming on board on a near daily basis. Telefonica is launching its network October 1. Swisscom is getting serious too. Makedonski Telekom (owned by Deutsche Telekom) is rolling out in Skopje. SureWest (USA) is progressing too.
  • Munifiber, utility fiber, etc.: Greece and Mauritius launched a big plan, whereas Australia (primarily targeting FTTC) and New Zealand are still stuck in the debating phase. Saudi Arabia is building a new city for 2m people, with FTTH from Ericsson. In the US, progress was made in several towns. Localised initiatives are found in Ireland and Australia too.
  • Altnetfiber: Smart Comp is building in Brno. Over in Korea, Hanaro Telecom is getting its act together.
  • Cable. There are several initiatives worldwide, most recently in Hawaii (Time Warner Cable), Japan (Suo Cable) and the US (Corn Belt Communications).
  • Open access. Europe is fighting for the extension of open access obligations from copper to fiber. Companies as diverse as KPN and Telstra are saying they are in the OA mood. But not Telefonica. Interestingly, in Utah the iProvo network was acquired by Broadweave, which subsequently tried to end competition by buying up two independent service providers. But those deals fell through.
  • Upgrades. 100 Mb/s isn't the end of it, 1 Gb/s is now in sight. FastWeb (controlled by Swisscom) still has to upgrade to 100 Mb/s first. Somehow, I have a feeling we will see more of that tomorrow ...
Here are my very easy comments:
  • Of course the Brits need to do FTTH. Is anybody listening? Access networks are bottlenecks - FTTH is the end game - it takes 20 years to build - video is coming - and there are indeed socio-economic benefits, as this very convenient study from Eindhoven University shows (in relation to the well-known networks of Nuenen and Eindhoven in the Netherlands).
  • Check out the new (second) Akamai report on the State of the Internet, as observed through their network. There is a lot about security, but Akamai also ranks countries by the percentage of connections above 5 Mb/s. Comparing the Q2 report with the Q1 report, some minor things catch the eye. The top 10 is pretty much the same, with South Korea #1 with an unchanged 64% of connections faster than 5 Mb/s. Belgium and the US make a big leap forward, both to 26% (from 21 and 20% resp.).
  • Things are complicated. Analysys Mason produced an impressive report, but it's just a cost model, in other words: one half of the equation. There are so many variables, a decisive report, including a revenue model, is totally unrealistic. Why not then make a little leap of faith and play the end-game?

Friday, September 05, 2008

Video: place your bets

It's Polling Season. Please do take my Poll in the right hand column (multiple answers, you may want to read this post first), and while you're at it anyway: please do fill out the new survey on Internet Video Distribution from my friends over at STL Partners. For that, you may want to read this post first.
On the side: I am quite sure we haven't seen the end of video coming to the web and it poses real threats to ISP business models. But let's not forget that primo it is a wonderful development.
Which brings me to United Content Distributors, my love-baby and maker of the Daily Media box. I know, there are lots of boxes like the Apple TV around. Some, like Sezmi, even come close to Daily Media. But the latter still has the advantage of its unique business model. And: it's not just a press release or a prototype. The Daily Media box is up and running in several (trial) locations in the Netherlands, one of which is a full-blown deployment at a hotel (fiber, IP - the works). Read about it here or here (in Dutch, but translated here and here).
On the side: you may counter that an application like Daily Media initially will not add to the pressures on ISPs and last miles. However, one thing it is supposed to do is eat the video store's lunch by offering VoD.
Which brings me to the new application I found on UCD's web site: Daily Care. Very much in line with KPN Health's strategy, and yesterday even Telstra (!) stepped on the bandwagon.
I look at the KPN strategy as very much of a blueprint for the entire telco business:
  1. Traditional ICT services.
  2. Connecting professionals and adding (third-party) VAS.
  3. Consumer services.

Thursday, September 04, 2008

... fit to print?

In my days of working for a Rabobank affiliate, a subscription to local business newspaper Het Financieele Dagblad was thrown into the secondary payment package. Frankly, I cancelled my subscription many years ago. The paper recently adopted the salmony color that so easily distinguishes the Financial Times, but let me reassure you: you don't want to waste your time and money on this pale imitation.

A valued reader in the heart of Amsterdam's Fiber Heaven pointed me to another inferior story (in Dutch, translated here) that was printed today. The writer manages to take some of the Telecoms Reform package currently in the European Parliament and Council of Telecoms Ministers, adds some broadband penetration numbers, throws in Bret Swanson's exaflood debate and rounds off with a touch of net neutrality and FTTH. The thing is titled 'Internet Under Pressure' and suggests it may come to a screaching hault because of traffic overflow.

And what do you get? Exactly, a lot of BS. Sorry Mr. Livestro, things can get quite complicated.

It's a shame, because just one day earlier some interesting statistics were released by Telegeography: international internet traffic grew 53% (a year ago: +61%), but capacity grew even faster. The result is that the utilisation rate averaged just 29% (from 31% a year ago). We're talking backbones here, dear Mr. Livestro, not last miles. And for some extra info, do read Rudolf's piece from a day before over here.

To round off, my reader adds (in my translation):
The real bottleneck is the backhaul between Central Office and the Internet Exchange. It uses expensive ATM technology and is practically a monopoly. Hence, it is not in the interest of the operator to switch to cheaper ethernet technology, since the regulator has set tariffs based on that relatively expensive ATM technology.

Wednesday, September 03, 2008

I Will Bring You Down, Baby

Everybody seems Himmelhoch jauchzend over Ben Verwaayen as Alcatel-Lucent's new CEO. Following a very sound tradition, here are some second thoughts. After all, he's Dutch and so am I; he's getting his hands on millions and I'm not. So, I will bring you down, baby.
  • Didn't BT's success largely hinge on the exploding broadband market? And can Ben take the credit for that?
  • Did it not also hinge on BT's functional separation? And wasn't that only accepted when the regulator (erstwhile Oftel) threatened to refer competion issues to the Competition Commission?
  • Didn't Ben come with some pretty outrageous targets in April 2002 (2 months after he started work at BT), such as a revenue CAGR of 6-8% for the years through 2004/2005 and mobile revenues of GBP 500m after 5 years (it was less than GBP 300m in the year 2006/2007)? It wasn't long before Ben became a little more cauteous and said he merely wanted to raise revenues, EBITDA, EPS and dividends on an annual basis.
  • A new MVNO deal with Vodafone (May 2004) was supposed to bring in GBP 1bn in revenues after 5 years. Well, it wasn't Ben's fault that BT sold off mmO2 (back in 2001) - and with it all sense of the mobile market.
  • Isn't Britain stuck in broadband middle ages called ADSL2+, with FTTH not really in sight?
Of course, there was lots of good stuff too, such as broadband and EBITDA margins.
So congrats, Ben!

Peek is to T-Mobile what Kindle is to Sprint

Peek is a New York company launching a specialized email device (no other functions) under the same name, including service. Read this article for all the details.
The Peek device will cost $100 and service will be 20 $/mo (flat fee, no contract). T-Mobile USA is the network partner and Target is Peek's distribution partner.

Some quick comments:
  • The target demographic may be well-defined (middle-aged women with kids), but competition is severe (BlackBerry, smartphones, not to mention the iPhone etc.), but it could be a very well targeted strategy. However, the barrier-to-entry seems low for operators or device manufacturers (Samsung, Nokia, Sprint) to copy this product/service. Interesting so see if Peek will fly.
  • Amazon Kindle comes to mind as a look-alike, but for books. Very focused and no dealings with the network partner (in that case Sprint Nextel).
  • From T-Mobile's perspective, this is another smart way to improve utilisation rates and create a new high-margin revenue stream. This type of partners (like Peek and Kindle) is a whole new breed (a downstream partner for the network as a platform - in STL speak).
  • Email messages are generally small (in terms of bits) and not as time-sensitive as voice service (hence latency is no problem) and will not easily clog up the network.

Tuesday, September 02, 2008

FTTH: will they ever learn?

This is a remarkable statement coming from an alternative operator, TelstraClear of New Zealand. The telco says the main result of faster broadband links to the home may be more downloads of pornography and movies rather than improvements to productivity, quotes the New Zealand Herald. "At the moment we don't believe that putting fibre into every home is economic or necessary."

My comments:
  • TelstraClear is owned by Telstra, which explains a lot of the above. Telstra too thinks FTTN (instead of FTTH) is sufficient for the Australian consumer market.
  • TelstraClear proposes FTTB (business market) instead of FTTH. I'm not sure about traffic congestion in New Zealand, but there is a lot to say for teleworking. That too is a driver on the demand side.
  • I've heard the argument before: why build FTTH if all they do with it is illegal file-sharing? Who are you to say what people may or may not do with their internet connection. Let the people rule!
  • There is some reference to wireless as an alternative. I just don't believe it.
Here is a reminder to Mr. Freeth of TelstraClear that video is not just about illegal or otherwise questionable material:
  • Growth rates are still high. Think YouTube and other user-generated stuff.
  • Most markets need some serious TV competition, and not just for live broadcast TV, but for VoD and catch-up TV as well. Telcos are all upgrading to offer IPTV. For that, even VDSL2 isn't enough.
  • Telepresence, videoconferencing, monitoring, telehealth, teleworking, cloud computing, video calling, etc.: they all require huge bandwidths.
  • Screens are getting bigger; movies and games are going HD, 3-D and holographic.
  • Place-shifting (e.g. Slingbox).
And to be sure, there is more:
  • Fuel and carbon savings.
  • If you don't, somebody alse will build and grab some extra GDP growth (like Mauritius).
  • Cablecos are upgrading to DOCSIS 3.0. If you want to keep up, you might as well leap ahead of them and acknowledge that FTTH is the end-game.
  • The build-out of a nationwide FTTH network takes at least 10 years to reach a good portion of the population. So, you better start today.
  • Many applications require symmetric connections. Only FTTH will be able to offer that.

Structural separation: spread the incumbent benefits

Very disruptive new regulation are coming to the telecom sector. Incumbents will suffer, both former PTTs and mobile operators.

Dominant operators are finally attacked. The synergistic benefits of running an integrated operator will be spread among all service providers.
The stranglehold that mobile operators hold on the market may end too.

Here are two of the most significant regulatory developments today.

1. Structural separation
In Australia and New Zealand several people call for structural separation of Telstra and Telecom NZ respectively.
Australia is preparing an NBN at the cost of AUD 4.7bn. At that price, it will probabaly be a FTTC network, but FTTH is still a possibility. It prompted opposition spokesman Bruce Billson to propose structural separation: "the natural monopoly that will be produced requires that kind of clarity".
Telstra retorted that structural separation "increases costs, reduces efficiencies, limits future innovation, and most importantly, kills off investment". Telecom NZ also resists.

I am very much in favor of structural separation on these grounds:
  • It will be good for competition. One-time costs are something we'll have to live with.
  • You bet it will reduce efficiencies that are linked to operating a vertically integrated monopoly that controls all three network layers (passive infrastructure, active infrastructure, services). Reducing efficiencies is not the purpose of introducing competition, but it is inevitable. There needs to be symmetry between all service providers on the telco network. Only then will there be true equivalence. Also: why should the incumbent be the only operator reaping all the synergetic benefits of integration? And don't forget: one of the main tasks of any NRA is to make the telco incumbent less dominant and hence: smaller! After many years of competition, these incumbents still dominate the market. Put differently: if the incumbent says it's bad for them, it must be good for the market!
  • Innovation and investment theoretically suffer because the operator of the passive layer most likely would be a monopolist. However, I don't buy this argument, in our brave new co-opetitive world. Still, I suppose value-based management, regulation, incentive schemes and ownership structure of the passive layer could help solve the problem.

2. Bill and keep
British NRA Ofcom launched a consultation on the future of mobile communication (until November 6). Among the new regulations could be a move from termation charges (currently around 15% of mobile revenues) to a bill and keep regime (by 2011).
At about the same time, Vodafone released a report claiming that 40m Europeans would cancel their subscription if the sector moved to US style interconnection (i.e. bill & keep).

It has been stated before: lowering interconnection rates, or indeed moving to bill & keep, is meant to increase usage and lower prices. So how does Vodafone arrive at their claim of 40m people (10%) pushed out of the market? Could it be the way they structured their inquiry? "Suppose we would be forced to double our rates, because we would have to bill you for both making and receiving calls, would you still subscribe to our service?"
Or does Vodafone fundamentally disagree with the expected price elasticity? That seems odd, since mobile substitution still has a long way to go.
It remains to be seen how much pricing power mobile operators really have in the retail market. But it sure looks like they have to do another round of slimming down.
A different way to look at it is the fixed line alternative operator perspective. These operators look upon mobile termination as a subsidy for mobile operators building out their networks. Again, after so many years it is time to do away with this subsidy.