iPhone available with a year's prepaid connectivity on O2

I saw an advert in one of the Sunday newspapers today from O2.

It is selling the 3G iPhone with a prepaid SIM for £350 (about US$540) for the 8GB version, with the price including a year's data access via 3G and WiFi (excluding roaming or voice calls). After 12 months, an extension is possible at an extra £10 per month.

Interestingly, this needs a dedicated SIM card (presumably to stop people swapping the flatrate SIM into other devices), and the service excludes the availability of visual voicemail. Roaming data in the EU is £3 per MB, and elsewhere at £6.

Full details are here .

Although very conspicuously, there's no link from the line "Unlimited text and minutes are subject to an excessive usage policy. Terms apply" to a web page detailing exactly what the policy and terms actually are. Which is slightly ominous, given all the other aspects have a "View More details" link.

Mobile broadband PCs with embedded 3G - pricing comparisons

Interesting to see that Dell in the US has announced the availability of Mini 9 netbooks with optional embedded 3G.

The optional extra cost is an interesting one though - $125. Of which $120 is refundable through AT&T if you take out a data plan starting at $60/month for 24 months. (The base price of the PC is $349 plus other options and taxes).


Makes the UK plan from Vodafone look pretty good value in comparison, at £25 / £30 (for 1GB or 3GB per month respectively) for 24 months - plus you get the embedded-3G PC for free.


Although it's still better value just to buy the PC and add a separate 3G dongle....

As a side note, the UK Dell site has an advertising banner from Vodafone which offers a free 30-day trial and half-priced optional internal modem. It says "Select the option to include the half price mobile broadband module with your notebook". Initially, all the laptop configuration pages I checked out on the site, didn't give a mobile broadband option, full-price or otherwise. Then I found one which suggested the option would cost £50, but it was unclear whether this was the half-price or the full-price amount. And then when I selected it, it told me "Errors: The selection - Wireless Label (Dell Wireless Cards)- Core 2 Duo is incompatible with Wireless Networking ". Yes, this embedded module approach is so much easier than using a dongle.....

Will "iPhone" be used as general slang for "smartphone"?

Patrick Smith (whose new blog SMS is the New Black has got a plethora of thought-provoking posts) has written about an interesting phenomenon: the use of the term iPhone as a generic word instead of "smartphone".

It makes a lot of sense to me - I don't really hear "normal" people ever talking about smartphones. They certainly don't care about technical distinction like open operating systems. Frankly, the term "smartphone" is pretty geeky, and almost guaranteed to turn off a sizeable part of the potential market.

No doubt that Apple's lawyers and branding mavens will hate this as much as their peers at Nokia, RIM and others. But I can certainly imagine it happening, especially in Europe where technical jargonese often tends to be viewed as deeply uncool.

Nokia gets HSUPA.... blurring the featurephone/smartphone divide

Thanks to the anonymous commenter on this earlier post of mine, who draws attention to the new 6260 Slide, which is due for release in early 2009 at around €299 unlocked.

The interesting thing is that this is technically a featurephone, as it's based on Nokia's proprietary S40 platform rather than its open smartphone S60/Symbian OS. Although with a 5MP camera, 320x480 screen, GPS and maps, WiFi and presumably a full-spec browser, it's really a smartphone in all but name and "native" openness.

It's devices like this that really blur the line between smart and non-smart. The iPhone has already shown that full multi-tasking and background applications aren't really necessary if the device is fast enough. And this phone should be capable of running some pretty decent apps in the browser, Flash Lite or Java.

Sure, something like the similar (but more expensive) N85 has Symbian if you really want to download complex native software, but in the absence of a Nokia equivalent of the AppStore, I'd guess that most customers will go for the better radio and screen of the 6260.

The new phone's full spec is here.

When is UMA not UMA? Part 2

I mentioned last week that I'd been confused by T-Mobile US, which was using "UMA" as its network name for inbound roamers to the US, irrespective of whether they had WiFi in their phones.

Since then, I've also met a couple of (generally knowledgeable) industry folk who are starting to use "UMA" as a generic term for any dual-mode WiFi/cellular solution. I even heard it used in the context of enterprise-centric solutions.

I suspect that this is purely a US thing, as T-Mobile's Hotspot@Home is probably the highest-profile dual-mode service in the market at present, and is often compared against the Sprint Airave femtocell service.

Other non-carrier dual-mode solution providers (Cisco, Divitas, fring, Truphone et al), should perhaps start deploying some marketing firepower to correct this misapprehension.

Assorted interesting tidbits & observations

I liked the Nokia presentation at the Open Mobile Summit, which talked about using aggregated data from embedded sensors in phones. They're doing one experiment that uses GPS from 10,000 handsets to show traffic delays in realtime, and they also talked about possibilities of using motion sensors as seismometers to track earthquakes, or temperature sensors to watch the pattern of influenza epidemics. Cool stuff - and which could yield new Telco 2.0-style revenue streams from Government rather than end-users.


I went to the enormous new Westfield shopping centre in West London yesterday. I was surprised at how poor indoor wireless coverage way, given that it's a state-of-the-art facility that I would have assumed had a state-of-the-art distributed antennas or other coverage solutions. Both O2 and T-Mobile were very spotty, especially for 3G.

There's a fascinating comment at MobileBurn about the possibility of Nokia starting its own MVNO in Japan, as a way of driving uptake of its handsets without relying on the existing incumbent operators. If this is true, it also makes me wonder whether it would ever try something similar in its other under-strength major market, the US.

There's a good post on the uneasy mix of VoIP and LTE at 3G4G Wireless . This issue is something I identified a year ago with my VoIPo3G report - what happens to voice telephony when we move to an all-IP network? The easy option is to just use 2G/3G circuit connections in parallel - although that means having two radios working simultaneously, which could impact battery life. The other options include

• using a UMA-like approach to tunnelling "circuit voice over IP"
• using IMS as a VoIP platform - albeit without having a standardised IMS VoIP application defined at the moment, so probably having to use a "standard" NGN VoIP SIP server instead
• using NGN VoIP without IMS, which would need a lot of work on both handset and core network, possibly proprietary
• using proprietary VoIP solutions like Skype or fring, with lots of work needed in interoperability, numbering etc
• assume everyone with an LTE device will also have a normal cellphone, and forget out the problem entirely.

BT's recently-acquired web telephone company Ribbit has just announced its "Bring Your Own Network" strategy, along with its developer platform. Along with some femtocells and the Vodafone Facebook SMS app from a couple of months ago, this represents the first signs of incumbent operators selling their own-brand services over other operators' access. I'm waiting to see BT or Vodafone or H3G or Verizon applications running on iPhones and T-Mobile G1's....

My local branch of Carphone Warehouse was today advertising a promotion along the lines of "Buy a Nokia XXXX handset and get a free Asus eeePC netbook". (Sorry, can't remember which Nokia it was, or the contract details).

Thoughts on white space spectrum

One of the panels I moderated yesterday at the Open Mobile Summit was focused on US regulatory trends around open access, focusing on this year's 700MHz auctions, and also recent moves to free up spare spectrum in what are known as "white spaces". One of the speakers was from the FCC, so I got a pretty definitive view on what's going on and why.


For those people not familiar with the US White Space phenomenon, I thought I'd give a quick round-up of what it's all about:


- In the US, there are many areas that have spare (unused) channels among the TV broadcast spectrum. Because of the regional nature of the US TV industry, there's a geographic patchwork of "white space" frequencies that could be used .
- Some of these channels are already used by wireless microphones, eg at conferences or sporting events
- There is a strong push by some organisations to exploit the spare spectrum, especially by Internet players like Google and Microsoft, for purposes like mobile broadband, or perhaps a wide-area equivalent to unlicenced WiFi
- There has been push-back against the concept by incumbent network operators (particularly those like T-Mobile that worry about interference), and special interest groups like microphone users (a constituency notably championed by Dolly Parton!)
- The FCC has been broadly supportive of the concept, on the basis that "wasted" spectrum is undesirable, although so is interference. It has tested a variety of prototype devices from assorted manufacturers, which have demonstrated that avoidance of conflict with other users of those bands is possible, albeit with mixed initial results
- After lots of testing and lobbying, the FCC has essentially now given a green light to the general principle of white space utilisation, with a lot of caveats about using a combination of geolocation and databases of used channels, plus device-based monitoring, to avoid interference with TV broadcasts or wireless microphones
- there are various proposed rules that will limit how (eg fixed vs mobile, different power levels) and where (eg not around microphone-laden sports venues or theatre districts like Broadway), the spectrum might be used.

Various companies have already professed huge enthusiasm for the move, hailing it as a new era in WiFi-type business models.

My take?

Well, yes, sort of. At the risk of sounding like a broken record, my initial opinion is that utilising white spaces in a meaningful way is "a great idea in principle, but not as easy as it looks in practice".

There is clearly going to be a lot of necessary work first - not the least of which is going from several proprietary radio types to something which is at least roughly standardised. In particular, the location look-up / sensing approach will need a lot of work, especially for fully-mobile devices. (It should be easier for fixed devices which can be registered / activated centrally). The location accuracy won't need to be as good as GPS for emergency E-911, though.

At the moment, it's not even clear what the radio candidates are, of who is promoting them. It would be incredibly unusual to see just a single united approach, so we'll probably be in something of a battle to get to market. It will presumably also be necessary to ensure that white space devices don't interfere with each other, as well as with TV and microphone signals - which could be a challenge if there are two or more radio technologies involved.

I'd expect the main initial beneficiaries of white space to be people living in rural parts of the US - there are large areas without 3G or WiMAX coverage, or decent fixed broadband. Next may be inclusion of white space capability in large devices like laptops (Dell has already made positive noises). But given all the variables involved, I wouldn't expect to see massmarket urban users touting white-space handsets any time soon.

On a side note, I reckon this trend towards sharing little-used spectrum will become increasingly important on a more general basis in the next few years. It sort of fits with the moves by Ofcom in the UK to get access to swathes of frequencies used by the Ministry of Defence, Civil Aviation Authority and the like. To use a specific example - some bands are mostly used for maritime radar, such as the coastguards, pointing out to sea. In theory, that spectrum should be re-usable for commercial broadband for users who are inland.

A lot of this also fits in with general moves towards software-defined or "agile" radio technology, which is still at very early stages of maturity. But given the likely congestion of established mobile bands in coming years, it's good to see moves like White Spaces as smart ways to squeeze more from a limited resource.

Every time I see the Android pitch, I like it less and less

I've seen various pitches from Google about mobile. In particular, I've seen several Android presentations over the last year.

Each one looks less convincing than the last. I've just been watching Rich Miner speaking at the Open Mobile Summit, and although the developer story looks fairly well-articulated, I just think that some of the wider back-story about mobile looks tired, and the assumptions and world-view seem too simplisitic.

One slide seemed to suggest that hardware cost as a % of handset bill-of-materials is falling, making the software cost more important as a way of driving down overall costs. Really? That was the prevailing thinking four years ago, but since then we've entered a world of QWERTY keypads, 480x800 touchscreens, 16GB of memory, 8MP cameras and multiple radios.

Another slide trotted out the usual tired array of irrelevant statistics about 3bn mobile users vs 1bn PCs, and the stereotypical myth about people in developing markets only accessing the Internet on a phone.

It's possible that Google really does know how the industry works (Google Maps on mobile is fantastic, after all), but for some reason has not updated its messaging, and tries to paint too primitive a picture for reasons best known to its communications department.

Having played with the T-Mobile G1 a bit, and canvassed opinion from a few knowledgeable users, it seems to be what I'd class as a 7-out-10 device. Not bad, but nothing like the leap forward of the iPhone. A bit clunky, and although the network speed was impressive on HSUPA, I felt the user experience was quite frustrating and counter-intuitive. And certainly not what you'd call a sexy device - sitting in a bar playing with it a few days after launch, there certainly wasn't anyone coming over and going "Oh wow, is that the new Google phone? Can I see it".

It will also be interesting to see how it fares for T-Mo and other operators in terms of support costs, because of the branding and lack of direct consumer support from Google. One theme coming out of this conference has been the worry among US market players that they will end up wearing the costs of supporting users, when they download random applications to their phones and they don't work. (Or even if the built-in functions are poorly performing).

Yesterday, an AT&T representative said that they'd not experienced people calling their customer support hotlines about iPhone software questions - because it was clear to users that the software (and AppStore) was Apple's. But Rich Miner essentially took the opposite stance, and said that any problems with Android phones or user experience fall to the carrier or OEM to fix.

It will be interesting to see the relative costs - AT&T subsidy on iPhone, vs T-Mobile support costs on the G1.

GSMA trying to force adoption of NFC in phones

The other day, I wrote a post about the need for some form of forum representing end-users of mobile phones, which could define true requirements for handset manufacturers . I contrasted this with the current, highly-biased requirements-setting bodies, whose demands are derived from the desires of the mobile operators, rather than the actual preferences and best interests of real customers.

The GSMA has kindly illustrated my point for me, pronouncing its views on the "need" for handset manufacturers to hurry up with putting NFC into phones - based on a model which would see the SIM card playing a pivotal role.

As far as I can see, the handset-buying public neither needs nor wants NFC. Ignore the inevitable sponsored survey results in the press releases, which are clearly designed to prove a point rather than be dispassionate analyses of real requirements. NFC is a mildly interesting but fairly unimportant innovation, beloved only by large system integrators and the type of blinkered "mobile convergence-ist" who likes to spout nonsense about the phone absorbing the function of your wallet. Most of the NFC applications are "solutions looking for problems", especially the ones around mobile payments.

If the GSMA really wanted to help the manufacturers by "potentially boosting sales at a time when forecasts of device sales are looking significantly down" it would have told them to stop wasting time and money on things like NFC, and instead make sure they prioritise put decent web browsers into handsets, or GPS, or memory card slots, or xenon flashes for the cameras.

Carnival of the Mobilists at Mippin

Some good stuff this week on the Carnival, plus a big thank-you to the author for the kind words about this blog. I really liked Andrew Grill's post on mobile advertising.

Actually, I have to confess I didn't realise what Mippin actually was - basically an easy way to reformat sites like blogs to work better on handsets. Now I've checked it out, I may use it myself.

Question: When is UMA not UMA?

Answer: when T-Mobile uses the word "UMA" as its network ID for all roaming customers, instead of "T-Mobile US".

No idea why this happens - initially on an unlocked Nokia E71 with a T-Mobile SIM (I thought it maybe had a "hidden" UMA stack and auto-logged on to the SFO Airport network, until I switched off the WiFi). Then I got the same tag on my non-WiFi SonyEricsson, and realised the answer must be simpler.

I'm pretty sure it's always just said "T-Mobile US" when I've been here in the past.

Next week in SF - Open Mobile Summit & Telecom Council

I'm going to be in the Bay Area next week for a couple of events and various client meetings.

I'd exhort anyone who's interested in the whole "openness" scene to consider attending the Open Mobile Summit   on Weds/Thurs 19th and 20th.

I'm chairing the track on the first day that's examining the implications of "open" for existing mobile operators, with speakers & panellists from some notable organisations like the FCC, AT&T, BT, TeliaSonera, Orange and NextWave, as well as well-known individuals like David Isen and Chris Sacca. A parallel track looks at the implications for upstarts & challengers like Skype and Truphone & the Kindle guys at Amazon. 

The event's also got other luminaries from Google Android, Nokia, Intel, Motorola, Qualcomm, Accel, RIM and various others. 

Should be awesome, and I'll try and blog during it if I get a chance.

If you can't make that, I'm also one of four speakers at the Telecom Council lunch on the Tuesday 18th. Again, I'll be pontificating on openness in mobile.

Lastly, I've got a few hours spare on Tuesday 18th before and after that event. If you're in the Bay Area and want to set up a last-minute briefing let me know via [FIRSTNAME.LASTNAME   AT disruptive-analysis.com ]  I'm particularly interested in companies involved in mobile broadband (devices & services), femtocells and enterprise UC. And if you've got any urgent  requirements for analyst research or consultancy on these areas, so much the better.  

The need for end-user forums to start specifiying handsets

There are already a number of organisations that capture operators' requirements for handsets and feed them back to the OEMs. The OMTP is probably the most obvious, along with the OMA and some of the activities of the GSMA and CDG and WiMAX Forum. Standards groups like 3GPP also are clearly important in defining the standards for "UE" (user equipment) behaviour.

These groups do sterling work in a lot of areas - from making sensible decisions about standard screen sizes & power connectors, through to defining APIs for browsers, and certification mechanisms for radios. They put out reams of documentation stating that manufacturers MUST do this, or SHALL do that, or OPTIONALLY MAY do the other.

But there's something missing.

As far as I know, there are no consumer advocacy groups defining what the end user wants to get from their phones, and how they expect them to behave. There's no group setting desirable specifications that "average" people want, such as "It SHALL be possible to delete useless apps like videotelephony from the homescreen", or "There MUST be a highly-visible indicator of monthly data usage, configurable in X, Y & Z fashion"

Particularly as handset subsidies fall - or handsets become cheap enough for normal people to buy "vanilla" - it shouldn't be just the mobile operators that get to define features & behaviour. Yes, they might be the handset producers' main direct customers, but it's about time that the people paying at the end of the value chain started to throw their weight around.

The only reason that people have to "put up with" hobbled Bluetooth, annoying operator-branded UIs, locked-down operating systems and so on is because they aren't organised enough to try to change it. Complaining on web forums doesn't cut it. But if a delegation from a "Phone Users' Forum" representing 100 million people turned up in Helsinki or Seoul with a list of demands, I suspect they'd get a good hearing. (Although probably not in Cupertino, California....).

There is already precedent for this in the enterprise world. Organisations like the EVUA represent powerful groups of telecom and IT-using multinationals, that collectively exert leverage on their technology suppliers.

Perhaps the starting point for mobile handsets is also in the enterprise space. Although the volumes are lower, there is a greater bias towards smartphones and "vanilla" devices. There are various organisations like systems integrators and unified communications vendors that also have a lot of pull here. Collectively, the business community needs to make sure that the phone manufacturers continue to enable "over the top" applications in the best way possible. An influential group articulating & defining that requirement in concrete terms would be hugely beneficial.

A particular point of future contention might be around the IMS RCS client, which if pre-installed on handsets could become the "default" IM and presence application, hooking them into operator IMS cores. This clearly needs to be delete-able from devices used as part of enterprise-controlled unified communications solutions. There need to be variants of handsets available that are optimised for server/PBX-anchored corporate alternatives.

Over time, there may be parallel consumer advocacy initiatives, but I suspect they be more difficult to organise, especially internationally. But I can see great benefit in mandatory requirements for easy sideloading / backup of content and contacts, for example, and better information about costs and balances. The best way for end users to avoid future "lock in" by their operators is to specify workarounds to the handset manufacturers upfront. Make them remember who the real customers are.

The bottom line is that the billion or so handset purchasers need to play the system better. Otherwise, the system will continue to play them.

Mobile Broadband - the cracks are starting to show

Over the last 2 weeks, I've encountered a steady stream of indicators to suggest that the mobile broadband market, and especially 3G USB dongles, might be starting to overheat.

The problem is that operator and retailer marketing departments have suddenly found a new source of revenue, and are engaging in a massive land-grab to sign up new customers, roll out new propositions, and offset declining voice revenues as we go into recession.

But I'm starting to hear some creaks. I initially missed this article, about an increase in the return rate for 3G dongles which had been sold "over-enthusiastically". I've noticed a distinct uptick in marketing hype, especially around headline speeds for HSDPA. I've also noticed a distinct lack of warnings about the fact that coverage is patchy, especially indoors.

I suspect what's happening is that the market has shifted from people like me, who already have ADSL and use their mobile service as an adjunct when they're out (and typically out in areas with good coverage like airports or cafes with big windows). The customers coming onboard now are those who want to use the products at home - perhaps instead of ADSL/cable, and perhaps in buildings with a couple of walls between their room and the outside world. Unsurprisingly this makes for unhappy results with 3G at 2.1GHz.

I've also noticed slowing speeds on 3's network in central London - although whether that's because of capacity limitations or throttling of certain apps (I really notice it on streamed audio) is hard to tell.

But above all, the alarm bells have started to ring with the rate at which network capacity is being apparently used up. It was only about 6 months ago that I heard a presentation refer to an operator that had "fired up a second carrier", ie had filled up the initial 5MHz chunk of their 3G spectrum, and had started using another. Then I spoke to a large vendor a few weeks back, who said they knew of a few places where people were on their 3rd carrier.  And then another last week who mentioned somewhere they had heard about a 4th carrier - which is apparently outside the original UMTS specifications.

So given that most operators only have 10MHz or 15MHz paired allocations for 3G, it's no surprise to see the panicked interest in femtocells, 900MHz refarming, 2.6GHz auctions and various approaches to adding or splitting cells. LTE offers the chance of some more headroom too - but only if you've got convenient 10MHz or 20MHz chunks of decent spectrum spare. As if. The problem is that none of these is going to be ready for prime-time in most markets in 2009. And anyway, the backhual is still another bottleneck for many operators.

I'm predicting that next year is going to see some fairly ugly examples of mobile broadband "capacity crunch". And given that capex budgets are going to be a bit thin, I reckon we'll see quite a few more dissatisfied customers. The problem is that the €10 flatrate genie is out of the bottle, and it's going to be very hard to step up the prices now.

Another open question is how this will start to impact the useability of all those nice new smartphones & broswer/widget frameworks that are starting to get traction as well.....

Last point to raise, that I'll tackle in another post soon..... what does all this mean for upcoming spectrum auctions in 2.6GHz next year? Lots of capacity, but auctions at exactly the wrong time (hmm, shades of 2001 again?). Will the auction-tuned game theorists controlling the bidding remember that 2.6GHz and bricks/concrete don't always make a happy combination, I wonder?

Mobile data revenues more than messaging?

There's plenty of commentary today about Vodafone's half-year results out today. However, given the currency fluctuations that have been going on, you need to be a professor of accountancy to back out some of the more granular underlying trends. Nevertheless, there's plenty of evidence of both the economy and tightening competition and regulation squeezing some of the historic cash cows like roaming.

On the upside, there's plenty of healthy growth in European mobile broadband, which is taking the edge off the declines in voice revenues somewhat.

One little data point I do find a bit strange is tucked away on page 14 of the results release. Apparently, the EMAPA region (Eastern Europe, Africa, Middle East & Pacific)  has seen particularly spectacular increases in mobile data revenues, more than doubling compared to the same period in 2007. But what's really odd is that for Africa & Middle East, data revenues now exceed messaging , by £127m to £107m. That compares with £48m vs £88m last year. Given that number is mostly made up of Egypt, South Africa [Vodacom] and India, I guess that indicates a huge ramp-up of 3G modems at Vodacoms, possibly aided by mobile payments.

The release says "Strong growth in data revenue was driven by the increased penetration of mobile PC connectivity devices, as the absence of fixed line alternatives makes mobile data a more attractive offering. Data revenue also benefited from the launch of South Africa’s first high speed uplink packet access (‘HSUPA’) and the launch of Vodafone M-Pesa/Vodafone Money Transfer service in Tanzania"

I don't think I've seen any other operators reporting 3G dongle revenues > SMS, so that's definitely a first if true.  Vodacom reports its detailed group financial report next Monday, so maybe we'll get some deeper insight.

Carnival of the Mobilists 149

Thanks to Mark Hooft at Ubiquitous Thoughts for hosting this week's Carnival, and highlighting a couple of my posts. Some other really good reference to others' blog posts as well, like Mjelly's post on the impact of the recession on mobile. Check it out here.

PC vs mobile Internet access - where's the asymptote?

Clearly, access to the Internet from mobile devices is increasingly rapidly, from smartphones, lower-end featurephones and (depending on your definition) notebooks/netbooks.

As I've written about many times before, there's a huge debate about whether the future of the web lies in the mobile domain, or whether it will remain PC-centric.

Usually, the example that gets held up by the mobile fraternity is Japan, which clearly has been in the vanguard of wireless Internet adoption since the early days of i-Mode. The interesting thing to see is that in fact, most people in Japan use BOTH mobile and PC access. About 4% even use TV or game consoles as well.

The chart on Page 3 of this document from the Japanese Ministry of Communications is probably the most clear representation I've seen. I'll be interested to see what the end-2008 numbers look like, as there will undoubtedly be more mobile phone users, but also more people with netbooks and other mobile-broadband enabled notebooks. That said, the 3G dongle phenomenon doesn't appear to have been as broadly popular in Japan as elsewhere.

Of course, Japan could arguably be treated as an exception rather than an example -it is *different* to most markets in many ways. The operators control handset architecture & capabilities much more tightly, it was a very early adopter of 3G, and everyone is on poast-paid data plans with easy access to data. On the other hand, PC Internet is much more attractive as well, with ultra-fast and cheap DSL, and rapidly rising penetration of fibre to the home.

Looking at the stats, I wonder if the asymptote, for a country with affluent population and fantastic mobile and fixed networks, looks something like 15/70/15 . That is, 15% are mobile-only Internet users, 15% are PC-only, and 70% use both.

As I said, I'm hesitant to hold up Japan as an inevitable end-point for other country, because much of the technology market is pre-defined by "historical accidents" like government policy, specific operator strategies, or early PC/mobile uptake shaping users' behaviour.

ARPU is a lousy metric for tracking mobile broadband

The limitations of ARPU as a metric for mobile phones are fairly well-understood - people have multiple accounts, it's skewed by certain things like BlackBerries and business users, and has a hard time dealing with pre-bundled services. Yet observers and investors continue to use it, despite these caveats, because they now have a "feel" for how to interpret it against this complex background - and also because there isn't really something else convenient to use. Particularly when you separate out between prepaid and postpaid, ARPU can still convey something of use, if you're sufficiently cynical about it.

Unfortunately, I think this complexity is going to get worse with mobile broadband, rather than better. At least with phones, people tend to use them regularly, whether with prepay or postpay subscriptions. And phones tend to be either actively used for making calls/SMS, or not used at all. Nobody uses a mobile handset exclusively "offline".

In the early days of mobile broadband, it was much easier - there was usually a fixed monthly subscription, plus any extra megabytes and roaming charges. No really onerous accounting judgements to make.

Now, things are changing - various models of prepaid mobile broadband are popping up, either with traditional "top-ups", or pre-packaged. I saw an offer yesterday for a HSDPA dongle with 12GB of data pre-bundled, usable at any point up to 12 months from purchase. Also, there are various "free notebook" offers that massively hike the apparent ARPU by bundling in the implicit consumer finance of the computer with the data access charge.

That's a bit like buying a car with a built-in cellular tracking system, and charging the whole monthly finance charge against ARPU.

Going forward, the picture is going to get even more complex. There will be more "connectivity built-in" devices like the Amazon Kindle. If I (say) pay $300 for a mobile-enabled games console, with zero monthly fees, how does that count in ARPU calculations? Something based around the wholesale fee charged by the MNO to the device manufacturer? But what happens if it's offset against (say) using the device for the operator's TV and advertising channel?

What happens with instances of free or sponsored use? Let's say the Tunisian Tourist authority offers to sponsor 10MB of roaming data as long as a visitor checks out their restaurant guide & completes a survey form? What about free broadband given to delegates at a conference? How does that fit into either the "revenue" or "number of users" calculation?

What about bundling? Fixed+mobile broadband, with a DSL modem plus integral 3G dongle so I can take my broadband with me when I go out? That should keep the accountants busy.

If I get an embedded-3G notebook with a 2-month "free trial", am I counted as a user? And what happens if I sign up to iPass, which has various MVNO deals in a number of countries - am I one mobile broadband user, or six? And what happens where I offload traffic to a femtocell, or WiFi - am I double-counted, and how is the revenue split recognised?

And all this is before we get to any of the really clever Telco 2.0 style business models. If an MNO sees clusters of active 3G notebook users around certain locations, and sells the data to an advertising agency planning a campaign for Dell or Toshiba, does that count as mobile broadband "revenue"?

None of this lends itself to a nice, convenient headline metric like ARPU unfortunately. But that's the nature of convergence, so we have to ditch it before it obscures the reality of the market.

I'd suggest that a more useful metric might be the number of active individual users, or active devices, coupled with some quantifier of service revenue split between end-user and wholesale/3rd-party. Hardware sale/subsidy/financing should be stripped out and reported separately.

And there needs to be absolute clarity on definitions and assumptions - I see too many charts which give bland numbers of "mobile broadband subscribers" without defining what a "subscription" actually is.

And if you're an investor - now is the time to press CFOs on this. There isn't much legacy data on this market, so now is the ideal time to pick a useful metric to start with, rather than a superficial but increasingly meaningless one.

Work and personal profiles on a single mobile phone

Various companies in the mobile business have pitched me recently with the idea of separate personal and work "profiles" or "personalities" on mobile phones. The Nokia E71 has a way to toggle back and forth between these personae, and various enterprise unified communications (UC) suites are also talking up the idea.

In theory, there are some benefits:

- ability for user to personalise the phone with ringtones, wallpaper etc, but also have it "locked down" for business use
- ability for enterprise to distinguish between work- and personal-related calling, perhaps for internal accounting or benefit-taxation reasons
- supposedly helping the employee to manage their "work-life balance"
- helping the enterprise "own" their worker's business mobile number, so that if they leave employment it can be rescinded.

All these are worth goals, but largely fail to take into account typical user behaviour:

- The phone that the employee wants is probably different to the phone the employer wants to supply (5MP camera, looks cool, available in pink or green etc)
- Phones (and services) are cheap. There is an increasing likelihood that employees with company phones or email device will also have a personal mobile
- Possible legal liabilities of holding personal data on a business mobile device, especially if backed up in a corporate UC server. In the UK, for example, this is definitel a grey area when it comes to the Data Protection Act.
- Increasingly, the end user will want to do more on their mobile than just phone calls, and separating work & personal profiles makes it especially difficult to manage things like web browsing, personal email, data plan costs etc
- Privacy concerns of the individual, especially regarding deeply-personal call, SMS etc made from a "work" device. What is the legal/contractual status of using the device to call a recruitment consultant, for instance? Is it grounds for dismissal?

Overall, my view is that the dual-personality concept is vastly overstated in importance for this use case. However, the concept can be modified to help blend "work mobile" and "work fixed" numbers on the same device, which is much more useful.

Do you want HSPA modules in device, if you're aggressive about LTE?

I just had a call with the always-interesting Actix, which provides planning and optimisation tools for 3G networks. They're enthusiastic about LTE, commenting on its ability to lower the costs of mobile broadband provision compared to alternatives like HSPA+.

Certainly, some operators appear to agree - T-Mobile is cited as being an advocate of leapfrogging HSPA+ and going for early LTE deployment. Others like AT&T and Telstra are more enthusiastic about HSPA+ .

(Yes, officially it's called HSPA Evolved, but everyone I speak to outside Ericsson thinks that HSPA+ is a snappier term).

For the LTE enthusiasts, one of the prime attractions is lower cost-per-MB for data compared to HSPA, especially important given the price and use curves of 3G in notebooks and smartphones. There's even a suggestion that they might switch off UMTS 3G networks before turning off voice-optimised 2G GSM.

If that's really the case, I would have thought that having a long-lasting legacy of devices only using the older, more expensive network technology would be undesirable - you wouldn't be able to transition their traffic to the more efficient LTE until much later. Now phones have a natural lifecycle of a year or two, especially at the high end. But notebooks endure much longer.

If you're rolling out LTE as early as you can, do you really want a load of HSPA-embedded notebooks and other devices lingering around on your older network for 4-5 years?

[I suppose you could just give these people an extra LTE dongle nearer the time, although that may mean having 2 SIMs in the notebook operational simultaneously if they don't remove the embedded one].

Partner conference - Femtocells USA

A lot of the recent buzz around femtos has headed West across the Atlantic in recent months.

Sprint is continuining to roll out Airave, Verizon has been talking about launching its own 2G CDMA femto, and even AT&T has come out and said it's quite keen on the possibilities. And T-Mobile, while better-known for its dual-mode UMA service based on WiFi, has in fact been one of the pioneers in deploying femtos' bigger brothers, picocells.

Against that backdrop, the upcoming Avren conference in Dallas next month should be pretty lively. Details are here - tell 'em I sent ya.

Redefining "Telco Grade" for the Mobile Internet era

I've been at a couple of conferences this week - covering Telco 2.0 business models and femtocells. At each one, an "old school" telecom speaker has made reference to "Telco Grade" infrastructure and capabilities, either in terms of security models or reliable voice services.


This isn't new - I always hear representatives of incumbent operators or traditional infrastructure vendors talking about "Five 9's" and QoS. Voice quality scores (MOS) get quite a few mentions too.


It made me start wondering whether these concepts, which are heavily rooted in a world where telecoms was all about voice sessions and dialtone, are in need of a serious update.


In particular, any future metrics for reliability and quality need to cast their net much wider. Marek Pawlowski over at the Mobile User Experience has a coruscating post about his nightmare experiences in buying a T-Mobile Android G1 phone, for example. The fact that ordinary circuit calls might work quite well is clearly almost irrelevant to his view of either the "quality" of the operator or the phone itself.

Similarly, if we're talking about IP services like VoIP, IM or even full IMS over a mobile connection, the fact that one link of a long chain has managed QoS (and even differentiated billing) is totally irrelevant if 30% of the time I'm outside decent 3G coverage. That's the gating factor on quality, not what's going on in the middle of the NGN - you might as well just use the Internet.

It's like posting a letter from London to New York, and getting a QoS guarantee that the central London to Heathrow Airport leg of its journey is 99.999% certain to take 47.2 minutes. But there's a 30% chance the flight might get delayed until tomorrow, and another 17% chance the letter will get sent to Boston by mistake. And a 75% chance the guy at the Post Office sold you the wrong stamp in the first place.

It's also critical to think about metrics other than basic coverage and availability when it comes to "mobile Internet grade". Latency, jitter, packet loss and speed of connection setup can be critical for many applications. And it's amazing that many of the so-called "telco grade" mobile broadband networks have distinctly non-Internet grade DNS lookup capabilities. Loading a MySpace page with all its plug-ins might mean resolving 70+ IP addresses, and again, if you do it over mobile the critical factor isn't the operator's core network.

The bottom line is that vendors and operators continuing to use the "carrier grade" and QoS arguments are often missing the point, or are being disingenuous. Why over-invest in "quality" in one part of the network, when basic radio coverage, IP network internals, or customer service are the "weakest links" on your customers' overall perception of service reliability and performance?

Swisscom - the worst offender in public WiFi?

I'm still staggered that the paid-WiFi hotspot industry has managed to squander a solid four year lead, to become virtually a running joke in provision of mobility access. Awful user experience, shocking pricing and limited roaming are still the norm.

Yes, there are some good experiences to be had with premium-grade providers like iPass or Boingo. But for the one-off user, it's only a matter of time before cellular data roaming kills the business dead.

One of those hammering nails into its coffin is undoubtedly Swisscom. I always wince when I find out that they're the provider of what's euphemistically called a "service" at a conference venue.

Today I was at the Femtocell Deployment conference at the Hilton Amsterdam. The irony of a 27 Euro daily fee for WiFi access at an event talking about cheap & effective 3G indoor coverage was palpable. Of course, I could have opted for the "basic" access of 256kbit/s and no VoIP, for a bargain 22 Euros. In the end, I used neither, and picked up email headers via cellular roaming at a (comparably) sane 2 Euros per MB - itself a travesty, but that's another story.

But the real kicker was this - during a discussion last night, a certain mobile/WiFi operator was named as the villain by a separate conference organiser, this time wanting to charge $54000 (yes, you read that right) for providing delegate access at a proposed future event.

For that money, he could probably persuade one of the carriers to install a full HSPA macrocell in the building, with full-spec backhaul, and give everyone a free dongle & SIM card.

Don't get me wrong, I really like using WiFi, and will generally prefer it to mobile broadband if it's available on a free/cheap basis. The latency is better, the speed is better and I generally get better battery life than with a 3G dongle. But it just amazes me that the industry has such a suicidal attitude to pricing.

Conference organisers: choose your venues carefully. Or just put a few 3G routers or dongle-docks around the room, with some cheap prepaid 3G data plans, and bypass the "house" WiFi altogether. Frankly, I'm surprised the people who rent out lighting & conference equipment aren't supplying these already.

Mobile broadband - O2 highlights dissatisfaction and returns management

Last week, O2 announced a refresh of its mobile broadband offerings - an area in which it has been pretty weak in the UK so far. As part of its press release, it also yielded some findings from a consumer survey it had commissioned.


In particular, it highlighted the concept of "mis-selling" of mobile broadband - for example, customers' irritation at extra heavy costs for roaming that they hadn't been made aware of at the time of purchase, or what happens if promises of 3G coverage don't match reality.


Disabling "default" roaming privileges, forcing consumers to call customer service and get a lecture on the pricing realities obviously isn't as good as just lowering the cost to sensible levels, but should at least mitigate the risk of serious bill-shock.


More interesting to me is the idea of a 50-day money-back "happiness guarantee", which will apparently "allow customers, who purchased directly from O2, to return the device within 50 days of purchase with no termination fees being charged and any costs for purchasing the device being refunded".

It will be interesting to see whether this forces the competition into reciprocating. And even more interesting to see how this plays into the USB dongle vs. Embedded 3G notebook debate. It's one thing returning a dongle after a month - it won't need much "refurbishment" before it can be re-used. It's another thing entirely dealing with a notebook with an internal 3G module - returns management could be a nightmare for both customer and operators.

Just think about the practicalities for a minute - you buy an embedded-3G notebook. You configure it, download various new applications, start using it, set the passwords, register your copy of MS Office, start using the mobile broadband. Then maybe it doesn't work properly immediately. Or maybe it works OK in the study where you unboxed it, but not in the bedroom when you move it the next day. Or a week later, your network operator tweaks their local cell tower's set-up, or you move house, or you suddenly get congestion in your cell as 10 neighbours sign up.

So you go back to the shop. You complain about poor coverage. Maybe call customer services. You have a dilemma - you need to find another notebook, another operator, or both. If it's an "unlocked" 3G notebook, bought from an independent retailer, you can probably just try another network's SIM in the same device, without too much extra hassle besides sorting the cancellation and new sign-up. But if you've got a subsidised or locked notebook - perhaps bought from the MNO's own retail outlets - you have a problem. You would have to cancel the contract, download your data temporaily onto a USB stick or old PC (if you have one), delete your data & apps from the hard drive, and rebox it before returning it. Then the store would have to reformat the hard drive, check inside the PC to make sure you haven't added/changed anything (memory etc), reinstall all the bundled applications, test it..... and then find someone who will buy a "shop-soiled" laptop. And deal with all the back-office things like cancelling the MS Office licence.

Bottom line - I'd be very surprised to see anyone - including O2 for that matter - offering an equivalent return guarantee for embedded-3G notebooks. (If I'm wrong on this and someone's already doing such a policy, please let me know).

No WiFi in the Vodafone BlackBerry Storm - a serious mistake?

In the past, I've not been convinced by some of the conspiracy theories around some operators' alleged rabid diapproval of WiFi. Yes, Verizon has been a bit of an outlier on this, and some parts of Asia (China, Japan, Korea) have not been especially enthusiastic dual-mode WiFi/cellular.

My general impression is that when it comes to smartphones, most of the more progressive operators are now relatively open-minded about WiFi. It works OK, battery performance has improved, it's not generally used for canibalising applications - most VoWLAN is usually incremental not substitutive - and it's seen as an important utility by a fair proportion of users and developers. Not only that, but WiFi is starting to be valuable in offloading data traffic from the macrocellular network.

The situation with featurephones has been a bit different - WiFi has been seen as less valuable, software complexities reduce its utility, and the customer base is perhaps more likely to make expensive technical support calls about relatively trivial problems that smartphone-using peers could fix themselves.

The iPhone has demonstrated how appreciated it is. And almost all other top-end smartphones now have WiFi - Nokias, various Windows devices, the Android G1, quite a few of the recent Samsung Symbian phones and so on. So does the BlackBerry Bold.

Which makes its exclusion from the Verizon / Vodafone BlackBerry Storm all the more mystifying - and, to be honest, it seems rather cynical. Yes, I know it's got EVDO and HSPA in it, so it's a fairly complex RF platform, but that's not a sufficient excuse to hobble what could otherwise be legitimately seen as a proper iPhone peer.

Given that Vodafone has quite a few other WiFi-enabled devices in its portfolio, I guess the finger of blame must point at Verizon on this, which has a lot of "prior" when it comes to hobbling handset features. A real shame, in my view.

Mobile search doesn't exist - a good proof point

For the last two years, I've been saying that "there's no such thing as mobile search". I've completely disagreed with the rhetoric that "people don't want to search on mobile, they want to find things".

Instead, I believe that broadly, people just want to use Google on their phones. If I want to find an address, or a restaurant, or cheat in a pub quiz, there's a perfectly easy way to do it, without some newfangled search tool that guesses my context wrong.

So Symbian CTO David Wood's write-up of a Google presentation last week made me feel validated:

"The surprising thing is that the spider graph for mobile-originated search enquiries had a very similar general shape to that for search enquiries from fixed devices. In other words, people seem to want to search for the same sorts of things - in the same proportion of times - regardless of whether they are using fixed devices or mobile ones."

Unlike David, I'm not surprised.