There are some functions that only governments perform, such as law enforcement, but even here the fact that the worker is a government employee doesn’t really affect the drivers and success factors that much. Mobility and immediacy (a time sensitive information need) are still the hallmarks of a good mobile application. It doesn’t matter if the user is a policemen checking fingerprints by the side of the road or a salesman trying to convince me to buy a hundred new photocopiers. The principles are the same, you identify a need, choose devices and technologies, deliver an application and probably re-engineer the business process to exploit the new opportunities that mobility brings. It’s true that government mobility sometimes involves unique details – a client once asked for a mobile device for prison use that was rugged, but not so heavy that it could be used as a dangerous weapon if it was forcibly taken away from the prison officer. But the private sector can have equally bizarre requirements – once when I was young I worked on a financial system for a very traditional insurance company which was like stepping back in time, imagine “the office” but set in 1940. One design constraint was that certain admin staff would never enter data for certain brokers because they’d had a row years before and hadn’t spoken since.

From my undoubtedly cynical perspective there seem to be a fair number of government applications delivered for no reason at all, other than the fact that money was available or a politician thought it was a good idea. When I look at some of the EU funded / sponsored mobile applications I feel that if there was a genuine need for them many would happen even without government sponsorship. It’s hardly necessary for the EU to fund location based applications showing points of interest for tourists when Lonely Planet and Nokia have already got together to deliver the same thing. I’m not dogmatic that everything should be done by the private sector, very far from it. But I am very aware that these applications are being funded by my taxes, and I’d rather they were spent on fundamental research that the private sector might never do.

Once we whittle away all the applications which are just normal m-business, and those mobile applications that should never have been developed in the first place, we get left with only a few cases where governments have a unique advantage or are delivering a special type of service that no-one else would provide. For example m-voting or public safety alerts for floods, tsunamis and so on. There are also some applications where the private sector would never co-operate without being leant on by government, such as accessing the traffic logs of mobile operators to find which handsets were in the vicinity of a crime and sending messages to them to ask for potential witnesses.

So I’m coming to the conclusion that 90% of the applications labelled “m-government”, “G2C” or “G2E” are actually just normal mobile applications subject to the same design and delivery rules as the private sector. The only difference is that in some countries they’re developed by the public sector. But that alone doesn’t change the principles which determine the technology that’s selected and whether the app is cost justified, sensible etc. Maybe 10% of mobile apps developed by the public sector are genuinely unique to government in some way. So perhaps we should stop talking about m-government as if it’s something different and special; most of the time it isn’t.

Mobile software innovation is so far ahead of mobile hardware innovation it’s painful. We’re generating tens of thousands of new mobile applications a year and finding all sorts of cool things to do with smartphones. But hardware design is stuck in the stone age. When did you last see a genuinely innovative new form factor? Surely there must be something waiting to be discovered beyond those boring candy bars, tablets, PDAs, flips and sliders. Here in the UK it’s London fashion week and the newspapers are bombarding us with pictures of amazingly creative clothes design; I wish the handset industry would harness some of that creativity.

As an industry we waste massive amounts of time and money agonising over questions such as “what do people want from netbooks”? We build useless devices such as MIDs that no-one buys. Why? It’s because we don’t have a discovery machine for mobile hardware that operates as efficiently as the discovery machine we have for mobile applications. So the solution is simple, create one.

I’m not underestimating the challenges, it’s a lot harder to design a mobile phone than a t-shirt, But on the other hand crowdsourcing is being effectively used for amazingly complex tasks like finding organic synthesis paths (Innocentive) so why not mobile hardware design? Maybe not all the designs would be practical, I don’t for a moment believe the phone design I sketched out in a blog a few months back is totally feasible, but the goal is to generate as many ideas as possible because somewhere along the way some of them will be both feasible and amazing.

This would be a great opportunity for one of the struggling tier 2 handset manufacturers like Sony Ericsson. They’ve got nothing to lose, their attempts to use traditional tactics to compete against the giants like Nokia and Samsung haven’t worked. It’s time to try something really radical, and crowdsourcing might just be it.

First the cost of e-books. A quick calculation based on bookshelf capacity suggests my wife and I have over 3,500 books scattered around the house, I’d guess at least two thirds are recent enough that they’re still in copyright. Probably half the books are ones we both read, so if we both had an e-book reader we’d need two licenses for current volumes. So if we assume $10 per e-book that works out at about $34,500 to replace p-books with e-books assuming equivalent volumes were available. And that also assumes the out-of-copyright books are free, which is probably over-optimistic.

E-book readers are going to get a lot of use, so they’ll probably only last a couple of years, therefore we’d have to pay a few hundred dollars each every two years to get a new reader. (In reality we probably need two readers each, a big A4 one at home and a small one for planes and travelling, but the DRM models haven’t caught up with that need yet). And when e-book file formats evolve or we switch reader brand to something with different DRM then in the worst case we’d have to spend tens of thousands again to replace the e-books with new more modern ones. No way.

Buying the e-books isn’t the end of the story. Even if we make the unrealistic assumption that we could obtain e-versions of all the books that we currently own that’s a lot more books than will fit on a reader like a Kindle. So we’re going to have to maintain offline copies somewhere. I could (maybe) trust a cloud service provider with this job, but I don’t like that option for reasons I discussed a while back. If I own e-books on digital media I’m going to have to maintain backups, and keep refreshing them. My colleague Valdis Fikis pointed out that so-called “persistent” media isn’t really that persistent compared to the life of a book. He estimates the following media lifetimes:

Hard disk: 3 to 5 years

DVD or CD: Up to 7 years if you buy a good quality product, less if you’re a cheapskate

Tape: (yes, it still exists), 7 to 10 years if you treat it kindly

So every few years I’ll need to remember to copy my e-book library. iPod has shown us just how painful it can be to back up and restore media libraries or transfer DRM protected media from one device to another. And I just don’t want that hassle. Books don’t deteriorate, just blow off the dust and they can be as good as they were a hundred years before.

In Gartner we talk a lot about total cost of ownership (TCO), but I think in the consumer product space we ought also to talk about total pain of ownership (TPO), And for me both the TCO and TPO of e-books are way too high. Old-fashioned p-books are more robust, lower cost and a lot easier to own. I’m much more likely to buy an e-book reader as an alternative to all the magazines and newspapers which I read once and throw away. That would not only be more convenient but also more environmentally responsible.

The E series enterprise devices are doing very well, informally Nokia is claiming much higher growth rates than RIM for QWERTY device shipments into corporates in Europe, so if the numbers are to be believed we’ll see RIM facing serious competition here. The Microsoft deal will also give E series a lot of extra corporate credibility in both Europe and the US.  The new E72 looks great, it’s a significant upgrade of an already very nice handset. If you want an official Gartner perspective on the prospects of E series vs RIM talk to my colleague Carolina Milanesi who is the expert on all things involving handset shipment predictions.

Less exciting events included a presentation from Deutsche Telekom’s Rene Obermann who delivered a somewhat lacklustre keynote which confirmed my suspicion that most operators are struggling to visualise their role in the future communications ecosystem. One of my Nokia acquaintances pointed out an interesting semantic clue; in his talk Rene referred to “users” and not “subscribers” which maybe hints at the future business relationships between operators and ourselves, although personally I’d have been happier if he’d used the term “customers” which to my mind has more connotations of service.

However, Rene Obermann made one excellent point – he stressed the need for simplicity because many people are confused by the complexity of mobile devices. He quoted a survey which claimed 59% of users think devices have too many functions. I know all about this because I’m married to one of these people.

I’ve finally seen and touched the 3G Nokia netbook. It’s a very elegant design and as a fashion statement is a product you’d be happy to be seen using. But despite the high spec it’s expensive – EUR 575 is a lot for a netbook. Nokia can probably have a modest success even at that price because operators will subsidise it as part of a contract. I wouldn’t be surprised if Nokia tried some creative sales strategies such as “comes with music” versions or maybe a bundle including a handset and a netbook. However, I am now  convinced that my previous sceptical reaction was right. At the end of the day, however elegant, this is a “me too” product; just one more netbook in a very crowded market.

If we invoke quantum uncertainty and infinite numbers of parallel universes there must be one where this vision materialises. However for those of us stuck in this one specific universe I just don’t think it’s going to happen. The three screen story is full of holes, it’s a bit like saying that a consumer will use the same maps in a car, on a mountain bike and on a skateboard.

Consumers don’t watch video on mobiles. Nielsen’s “three screen” report shows that in Q1 2009 around 285 million Americans watched TV, but only 13 million watched video on a mobile. And those 13 million only averaged a bit over 3 hours a month, compared to the 153 hours of couch time put in by a TV viewer. Mobile video viewer hours are insignificant.

Mobile phones aren’t good video devices. Mobile phones aren’t well designed for video. Unless you have a physically large handset the screen is too small and many batteries can’t last the duration of a Tarantino epic. You can walk down the street making a phone call, but when did you last see someone walking down the street watching a mobile video?

There are so many things more important than video. There’s a much more important lesson here which the app stores are teaching us. Mobile phones are application platforms, they are tools, they are toys, they are micro blogs, they are our gateway to social networks and the internet, they are the way we get dates (well, not me, I’m married), they are e-book readers, they are a million things which are more important to us than paying to watch video on an inadequate screen. Even when we do use mobiles for video we tend to snack on sources like Youtube, we don’t pay to watch Hollywood blockbusters. There are several million application developers working in the mobile area world-wide, and every week they discover new things to do with a mobile that are more compelling than watching video.

There are lots more problems facing video, such as the difficulty of running a profitable business when your primary demographic hasn’t even grasped the concept of paying for content yet. The three screen vision is missing the point. The future of mobility isn’t recycled media business models forced onto unwilling mobile users. The future of mobility will be a lot weirder than that, and a lot more fun than that. And we’re lucky to be part of the generation that’s going to discover it.

One of the reasons for this future battle is that mobile business is going to become complicated. Or, to be more precise, what’s simple for the user could get very messy behind the scenes. It’s easy to construct interesting m-business scenarios, for example: I could be walking down the street, my phone beeps and Julia Robert’s voice says “look at that movie poster across the road, it’s me, you know you like me” and when I stop to look, the phone realises I’m interested gets movie reviews, details of where it’s showing, and which cinema has free seats. I might then choose a venue and performance, book seats & a parking space nearby, pay, get directions loaded onto my handset GPS and when I arrive use NFC on the handset as a ticket to enter the cinema. And maybe I’d also get a reminder inserted  into my handset calendar so that when it’s time to leave home for the cinema Julia’s voice tells me “start out now or you’ll miss me”.

Let’s ignore my personal Julia Roberts fantasies for a moment and analyse that scenario. It requires someone whom I trust to track my location in real time and act as what we call a Context Service Provider (CSP) to trigger event based applications. It brings together a mobile payment service, movie reviews which might come from a fan magazine, a cinema, a parking garage, maybe a ticket booking agency, a geo services provider and probably an advertising agency. It also requires someone to develop and deploy widely accepted standards for NFC ticketing. This is all getting very complicated. If we wait for market forces to deliver the necessary standards in an open and non-proprietary manner Julia will likely have retired from the move business, or home entertainment systems will have advanced so far that no-one goes to the cinema any more.

So we have a complex multipartner business model, with a need for new de-facto standards to be established. This sounds to me like an opportunity for a few big vendors to take a lead. It’s easy to imagine a scenario where someone like Google, Nokia, maybe Yahoo or perhaps a mobile outsider like eBay creates some de-facto standards and provides some sort of mobile backbone architecture to enable such scenarios. The backbone vendor would have to become a CSP (for more details read our context research) and would have to be trusted by consumers like you and I with personal information such as location and preferences; but that’s achievable. At least, it’s achievable for some vendors, although I suspect this issue of trust will be a key challenge. I’m personally amazed that people trust Google as much as they do given that Google’s business model is based on snooping on your behaviour and email. However, I digress. The backbone vendor will also likely need some enabling functions like mobile payment and advertising as well as the commercial power to establish de-facto standards, so this isn’t going to be a job for lightweights. So what I expect is that we’ll see a number of the heavyweight web vendors, plus maybe some outsiders starting to set up new types of mobile business ecosystem.

Such ecosystems will need lots of willing victims (sorry, I meant consumers) like you and I who are prepared to forgo a little privacy in return for some interesting applications and services. Some of the biggest corporations in the world will soon want to be your mobile friends. How nice.

So what’s going to differentiate the ecosystems of the future? Well, one factor will be that developers and users will likely congregate around those which have the largest customer base because they’ll be more profitable and have more content. And that’s a challenge for someone like LG because if you sell a Symbian or Android handset the customers will probably go to Ovi or the Android marketplace rather than your store. There are other benefits of scale, for example the big ecosystem app stores will be able to do things such as personalisation more effectively, this benefits both developers and users and is one reason I still go to Amazon rather than other web stores.

But there’s another big piece in this ecosystem puzzle which will make it particularly hard for companies like LG to compete – cloud services. The winning ecosystems of the future will host services which will empower their developers to create innovative apps that they couldn’t build on their own. Embryonic services include notification engines like Apple’s MobileMe, some of the emerging range of Ovi services like IM and the map player, Google maps, Latitude and so on. These are mashable foundations for developers to create the next generation of mobile apps using technologies such as location and context. And what we see today is just the start, in a year there will be some even more amazing cloud services out there for developers to build on. I believe that in the future users will gravitate to the ecosystem owner’s store, because they’ll have to sign up to that ecosystem to run apps that access the services. And that puts the ecosystem stores like Ovi, Android and maybe Microsoft in a stronger position than independents.

So a small scale app store without an ecosystem and cloud services looks pretty weak. Which explains my scepticism about LG, and the title of this blog.

Which wireless? Smart meters will need wireless both to transmit readings to the supplier and to connect to in-house energy monitoring systems. But which wireless? Meters are installed for decades, but wireless standards are seldom stable for more than a few years. Home wireless technologies such as domestic meshes are a still a battleground between standards such as Zigbee and proprietary systems such as ZWave. Some of the wireless technologies which will likely be important in the home such as Bluetooth LE aren’t even deployed yet. But will a government understand this and mandate the necessary flexibility?

Unknown future use cases. No-one – least of all the government or the energy companies – can predict how people will use the information collected by smart meters. Today most smart metering involves wall-mounted displays that show energy consumption. And I don’t want those. At the very least I want real-time monitoring on my mobile so I can see if I left something switched on by mistake. I’ve already seen researchers connect their house to twitter so it can tweet about energy consumption and other events such as phone messages. I’ve also seen cool display devices like the ambient orb used to monitor energy consumption. In a decade my energy consumption data could feed into dozens of social networking systems and new services. And some of those services like real-time supplier selection might not be very attractive to energy companies. Technical and commercial openness would be good. 

Data privacy and data ownership.  My government has a sloppy record on data protection, they managed to lose the personal details of about 25 million people in one single incident a while back. Energy consumption data must by rigorously protected, a hacker with access could easily work out when I was on vacation, or what times I was regularly out of the house. Burglars would love this. And what else might a “smart” meter do? Will the energy companies be able to cut off people over the air when they don’t pay? Sounds like a good idea until you have a pricing dispute with your energy supplier.

Unintended consequences.Complex initiatives like this always have unexpected consequences and it will be interesting to see how these emerge For example regular visits from a meter reader might be an important source of human contact for some seniors. Real time metering will allow me to pay for exactly what I use, when I use it. This could reduce the level of pre-paid balances and negatively impact the energy company cash flow. Or as I mentioned above, if I could switch energy suppliers at 5 minutes notice that would have a huge impact on how energy is priced and marketed.

Smart metering could be anything from a disaster to an infrastructure that will enable a range of innovative future services. The devil is going to be in the detail, so I’m going to watch this carefully.

(note to self after reading that sentence, stop sounding like ‘sex and the city’)

The overall stats aren’t impressive. Take Sourceforge for example. Out of 336,000 projects only 731 are tagged as “mobile”, around 0.2%. Black Duck logged around 17,000 new open source projects in 2008 of which they reckon 120 were mobile (0.7%), and a third of those were for iPhone.

There are a few high profile successes – but more for tools than applications. Eclipse is very popular, lots of mobile browsers use open source components, platforms including Linux and (soon) Open Symbian are gaining ground. But there doesn’t seem to be much here if you’re an application user.

Mobile applications are popular, but mobile open source applications aren’t. Apple recently claimed 500 million downloads from their store which they claim contains over 15,000 applications. An outstanding showing in just 8 months for a platform that’s well under 2% of total handset shipments. So both developers and users adore Apple applications and the app store. Mobile developers usually want to get rich or famous, ideally both. Users may be motivated by sex, fear and greed; but not by open source. So from one perspective you could say the question doesn’t matter. Mobile application developers don’t need open source to make money, mobile users don’t care about it.

But that still leaves me with a nagging feeling that there’s a lost opportunity. And I think the answer is to change the question. The powerhouse of mobile application, content and service development is going to be ecosystems; communities of users and developers under the umbrella of an ecosystem owner. Some will be associated with platforms, some with networks, some with services. The challenge is to make the ecosystem a success and to use open source to accelerate that success. I think ecosystem owners could do a lot more to leverage open source; perhaps as components more than end user applications.

So I think that while open source in mobile isn’t yet a success it’s too early to say it’s a failure. There are unexplored opportunities, especially if you’re an ecosystem owner. As usual, if you want to discuss this add a comment, or if you’re a client, call me.

My pessimism is heightened because after a decade of e-commerce we still can’t make the basics work. A few days ago I tried to renew the subscription to my Kaspersky anti-virus system. Kaspersky is a nice AV tool, but they have outsourced the e-commerce to a third party who handles payment. As I decided to pay by Visa, the bank forces itself into the loop as well with a nasty little “verified by Visa” popup which requires me to enter an additional password. So we now have three actors making my life complicated when all I want to do is pay a small sum to Kaspersky.

Naturally something went wrong, probably a glitch from my somewhat unreliable broadband provider. I lost contact after the “verified by Visa” step and couldn’t get back to the e-store to complete my purchase. So my first problem was that I didn’t know whether I’d made a payment or succeeded in purchasing anything. After a day I got a response to my email to the e-commerce provider saying it was OK, the purchase hadn’t gone through. So fired with naive enthusiasm I decided to try again. And now Kaspersky’s web site refuses to let me renew, saying the license number in question has already been used for a renewal.

I’m now faced with emails, calls to premium rate support numbers, and wasted time just because some designer can’t correctly implement a distributed transaction involving three parties. It’s not even obvious that all of these parties are necessary, as The Register has pointed out there are many people who doubt the effectiveness of “verified by Visa”. But let’s leave that one for another day.

If we still can’t implement simple distributed networked transactions when the participants are well defined, what hope do we have with complex SOA and clouds?