What Nokia (and everyone else) needs to learn about smartphones
Lessons from Star Trek
When Gene Roddenberry was devising the original Star Trek in 1965, he predicted humans would make common use of teleportation and faster-than-lightspeed travel. But one 23rd century innovation was either missing, or absent because nobody wanted it: converged electronic devices.
Cosmonauts who needed a data device got a tricorder, which was basically a PDA. Those who need voice calls got a communicator, which was a mobile phone. Sometimes you wonder if Roddenberry wasn’t on to something, and that these huge industries have been chasing a phantom, something that nobody really needs, for many years.
I don't think that's quite the case - but things have been more problematic than anyone envisaged. And the answer, perhaps, has been staring us in the face.
It's Convergence, Jim - but not as we know it
Yesterday I wrote  how Nokia executives sound peevish and a bit fed up on their big day on Tuesday. Perhaps you would too, if you’d invented the smartphone 15 years ago, and you still sold the most smartphones - then saw somebody else come along and pocket all the profits. But there’s a factor here that almost everyone forgets, which has nothing to do with technology, design or logistics. It becomes obvious when you look at the history, because getting everything else “right” still doesn’t help. Only by taking a long view does it become quite striking.
Let there be data
Let’s go back to the early 1990s, with huge investments going into the first 2G digital cellular networks, which opened for business in 1992. It was clear to a lot of people that data would become part of the mobile deal. This informed thinking in Europe, Japan and the USA, and the development of CDMA and the first UMTS meetings in 1992.
Consumer deployments would be years of wrangling away, but once GSM took off in the mid-1990s it was evident that mobile data would most probably be brought to us by the same people who brought us those mobile networks. They would have such a strong economic and organisational advantages (they made money, had access to capital, had the best sites, a fast growing user base, and knew how to co-operate) that new market entrants touting some whizzy new new technology (for example an Iridium, or a Wi-Fi network) would almost certainly fail.
This tended to focus minds wonderfully. The PDAs of the early 1990s were also sprouting comms ports, even though the destinations were a PC (via a serial link) or a CompuServe or AOL. The Psions and Newtons were already “communicators”, they just weren’t integrated communicators. So the leading phone companies began to think about how to put data on a phone.
A smartphone from late 1997: Sharp's G1
At CeBIT in the mid-1990s, Nokia and Ericsson would happily show you could see “concept designs” of integrated touch screen communicators. When it began writing a new OS in 1994, Psion had the foresight to anticipate what phone manufacturers might one day need, and the resulting Epoc had an advantage in resource consumption and performance, as well as a rich and flexible framework for applications. The three largest phone companies all agreed to pool development on this joint venture, Microsoft swore vengeance, and a long, phoney war followed for several years.
But this happy consensus tended to obscure some issues which were acknowledged but not really addressed by anyone in the industry. The Roddenberry conundrum was still alive and well.
The data people wanted to make the experience as rich as possible, thinking of sophisticated devices that were really mobile computers, that also did voice. The other school of thought envisaged data as being a service, a bolt-on to a device which was still a voice phone first and foremost.
Naturally operators took the second view – and not always for Machiavellian reasons. They thought the untamed World Wide Web looked a little too much like the Wild Wild West, and not enough like a service. Current mobile data usage patterns show that they were as right as they were wrong. So they steered everyone towards a portal – their own portal, of course.
But while the operators thought in terms of services, they were uniquely incapable of delivering interesting or innovating services, lurched from WAP to walled gardens. Hopping out of the walled garden wasn't so difficult, but it was a minority pursuit.
Nokia, being Nokia, tried to please everybody. And why not, it had the scale to do so.
The mid-Noughties crisis: nobody wants a smartphone
It took an almighty crash for the leading phone manufacturers to readdress the problem, and when they realigned, it was in the interests of their biggest customers, the operators. What happened was that the intense focus on smartphones between 2000 and 2003 hadn’t paid off.
The rollout of GPRS and 3G networks had been behind schedule, which negated the advantage of a phone being “smart”. And while it poured billions into multimedia and enterprise device development on Symbian, Nokia neglected its more humble feature phones. The result was a dramatic collapse in market share  , ten percentage points in some markets. Nokia steadied the boat, focussed on making these midrange handsets (like the 6230) desirable, and made sure it never made the same mistake again – the feature phones are its bedrock today.
Nokia's first mass market smartphone, the 7650 from 2002
The networks weren't ready, so there wasn't much you could do with it. 
But the resulting realignment saw the original idea of a sophisticated converged data device take a back seat. Sony Ericsson and Nokia abandoned the rich data-centric UIs they had developed and concentrated on smartphones optimised for one-handed use.
In a bonfire of the UIs, Nokia ditched two platforms within a year: it would no longer maintain the S80 Communicator, nor would it develop the touch screen capable S90, that was its designated successor. Microsoft and a host of others filled the gap with Windows Mobile devices aimed at vertical markets and tech enthusiasts, that in truth didn’t do either very well.
There was a huge hole in the portfolios of the leading companies when Apple announced the iPhone. And Apple, along with RIM, today scoop up almost all of the profits made by anyone making high-end devices – and more than half of the profits the mobile handset industry makes.
The reasons for this are numerous and varied - but the overriding one I think is that both RIM and Apple made the sticker shock go away. Both bundled a data plan with the device. And this was the biggest breakthrough in the history of the smartphone.
This looks so obvious now to an outsider that a Martian would wonder why we obsess about “platforms” and chip speeds and megapixels so much. Whether somebody is using a phone primarily as a phone, or mostly as a data device with a bit of phone thrown in, the bundling remains key factor to its success. It doesn't guarantee success, but it provides the vendor with a huge advantage.
You knew that once you’d got either device, you wouldn’t get a nasty surprise for actually using it. RIM went a step further, and with capped roaming charges overseas. Lo and behold, people actually starting using iPhones and BlackBerrys for something other than phone calls. People bought them. Amazing, innit?
To be surprised that people don't flock to mobile data without this kind of assurance is like imagining the teabag wouldn't be invented. But all things look obvious in hindsight.
Old-fashioned sales just aren't enough
RIM and Apple got away with it at a time when the operators were desperate that somebody might come along and use their networks. They’d got into debt equipping their networks with data capacity and buying licenses, but nobody was using them for data.
I’m reliably informed that Apple even extracted a revenue share from the operators when the iPhone first launched – the networks were paying somebody else to come along and use their infrastructure. But the punter who got either device got something new: an insurance policy against penalties for using the device as the maker intended it to be used.
Now the problem for Nokia – and everyone else – is whether they can come to some sort of similar arrangement. Given Nokia’s inclination to partner, and not to upset the networks, it could go either way. The history of the past five years has rightly seen the UX put in its proper place. But creating data bundles for its users – whether they’re in France or India – seems to be far more important to me than just flinging more phones or services at the market.
This isn't going to be easy for Nokia, or anybody else; the same dilemma faces Samsung, Motorola and HTC, all handset manufacturers which basically sell their proposition on the basis of megapixels and megahertz. You may argue it will be harder for these manufacturers to come up with bundles, since they don't have Nokia's reach or influence.
But it augers a fundamental realignment of the industry, traditionally with operators over here, buying phones from device manufacturers over there. It's one that neither side will be comfortable with. But unless they want to be stranded, they'd better come up with something. ®
Not being a Trekkie (honest) I have no idea whether Gene Roddenberry envisaged whether 23rd century humans used teabags or loose tea. Perhaps it was tele-transported from the plantation instantly, and brewed nicely along the way. No doubt you will tell me.