This year, the humble television set is going to transform more radically than it has in the previous 60-odd years of its existence. More than it did with the coming of colour, or HD, or even 3D. Those changes merely upgraded TV's basic function: improving the image and sound. This year, TV's basic function itself will change.

For the first time in the history of television, it's not going to be about who has the best picture, or the biggest screen, or the thinnest bezel. With the advent of "smart TV," market superiority will no longer depend on hardware, but on software. And that could be a problem.

A few years ago, I was speaking with Bryan McLeod, creator of the indispensable Harmony Remote, and now CEO of Storage Appliance, makers of easy-to-use Clickfree backup drives. He pointed out to me a fact that I'd been dimly aware of, but had never really isolated in my mind. To wit: hardware manufacturers almost never do good software.

It seems kind of obvious. Nobody can be good at everything. And software, after all, is fiendishly difficult to do well. Projects tend to spin out of control. Bugs become impossible to eradicate from millions of lines of unreadable code. And then there are the visual interfaces, which need a whole other mixture of talents and disciplines.

When hardware manufacturers dabble in software, mediocrity is typically the best one can hope for. As McLeod pointed out, the backup software included with disk drives is usually both limited and awkward. Or, if it's a licensed version of a product that's managed to establish itself on its own merits, integration will be lacking.

Hardware manufacturers don't tend to have the expertise to do great software. They can hire it, of course. But they can't simply hire a deep-down understanding of software design. They don't feel in their bones the virtue - the necessity - of lean, compact code. Of elegant architecture. Of clever integration. These concepts aren't embedded in their corporate culture.

So now we come to the age of smart TV. Not just "Internet TV," but TV with actual smarts. TV that absorbs many functions that were formerly far outside its domain: Skype video calling; search, especially of program listings; Web surfing; playing games... You name it.

Each of these things is tough to do well. Building a platform that will support them all (as well as any other conceivable application that may emerge) is comparable to putting a man on the moon. It needs a tightly focused organization like NASA.

Software isn't just one more feature you add to a spec sheet. It's fundamentally different from hardware, in ways that are hard to even define. Creating great software is an art, and building the teams to do it is both difficult and slow. Creating a great software platform is the pinnacle of the art, a feat that's been pulled off successfully by a very limited number of companies.

Sony is one, and should have a significant advantage with the PlayStation (1, 2 and 3) under its belt. But a lot of the paradigms that work in the narrow world of console gaming won't translate to the wider realm of TV. Even if individual companies like Sony do a great technical job on their new TV ecosystems, they'll encounter other issues.

A big one is standardization. Even if every manufacturer does a fabulous job on its own software platform, history shows that consumers are likely to arbitrarily reject all but a couple of them.
In game consoles, whenever three operating systems were competing, you knew at least one of them was heading for a cliff. (It's happening again right now.) Sony's timing was lucky enough, and its up-front investment huge enough, to make it one of the two dominant players in that world. But that same success won't automatically follow in the new arena of smart TV.

On tablets, we're seeing two dominant platforms (iOS and Android), with other players being jettisoned left and right (HP's elegant WebOS is gone, and RIM's equally clever PlayBook OS is hanging by a thread.)

In the desktop world, two major operating systems proved to be about all the market would tolerate. In smartphones, we saw a huge proliferation of software platforms and user interfaces... then an overnight implosion that again left two dominant platforms: iOS and Android.

The living-room TV is about to acquire all the complexity that consumers thought they'd left behind in the den, on their desktop computer. Only now it won't just be a question of "Mac or Windows?" It will be a choice between a half-dozen OSes and UIs, each with its own unique combination of hardware controllers, on-screen apps, tablet-integration apps and online services.

And appealing to consumers is still only half the battle. Reaching out to app developers, acting as custodians to a global software ecosystem, requires yet another esoteric skill set. Many people find Microsoft's huge success to be either magical or diabolical, simply because they have no idea what a superb job the company has done over the years in making developers happy.

Faced with a half-dozen TV platforms, developers will be forced to choose allegiances carefully. No matter how fundamentally similar all the platforms are under the skin, there's going to be a significant expense in accommodating their differences. The inevitable consequence is fewer apps for everybody.
There are traditionally two alternatives to this sort of quagmire. Neither of them has so far been adopted for smart TV.

One, manufacturers can establish an industry-standard platform. This approach tends to be slow and cumbersome, but it has worked often enough in the past (especially when given a governmental assist, as with NTSC and HD specs).

Two, they can farm the whole mess out to a relatively neutral corporate outsider. In the computer world, that company has been Microsoft. In the tablet and smartphone world, Google has taken the lead (with somewhat mixed success, technologically speaking).

Unfortunately, it looks like Google TV is being adopted rather slowly and half-heartedly, and mingled with each manufacturer's own proprietary ecosystem. And it's not clear that Google TV would be a total solution in any case. (We'll know more by mid-year, hopefully.)

Meanwhile, we keep hearing that Apple is getting ready to launch its own "Apple TV." Not a new generation of its existing media box, but an actual Apple-branded TV.

Does this not sound familiar to anyone? Multiple idiosyncratic software platforms, a reluctance to adopt standards, user confusion about user interfaces? And Apple riding in on its shining white horse, offering consumers a really nice user interface, where everything is really polished and integrated, by a company that's made this its one and only core competency since the early 1980s?

Of course, as of early February 2012, this possibility remains safely in the realm of pure speculation. Apple is tight-lipped as ever, and probably has many other things on its mind. But there are signs that Apple (including discussions with big Canadian communications companies) that Apple is moving in this direction. The fact that it's even possible to contemplate this kind of development should be some sort of warning signal.

When it becomes plausible that a total outsider could waltz in and snatch a big chunk of a crowded, highly competitive market, isn't it time for a bit of a re-think? History tells us a lot about software platforms, but is anyone listening?

Article Tags:  smart TV, IPTV, Internet, Web, Apple TV, iOS, Android, Google, Microsoft, videogame, console, Sony, PlayStation, Xbox, Nintendo, developer, app, HP, webOS, RIM, BlackBerry, PlayBook, television, HDTV, MacLeod, Harmony, Clickfree