Just a few hours after I posted the original version of this article, HP showed how much they agreed with my analysis by cancelling the TouchPad and putting WebOS on ice. Will they now follow the rest of my advice and open the platform up? Read on...
At OSCON, I got the chance to speak with Phil Robb and Joshua Marinacci of Hewlett-Packard about WebOS. It was a long and entertaining discussion with two smart and interesting people. I was impressed both by WebOS and by the people who introduced it to me. You'll recall that WebOS was developed by Palm as their next-generation device platform, and the company was then acquired by HP.
Phil explained to me that HP bought Palm in the light of "the consumerisation of IT", presumably with the goal of providing a vehicle to connect HP's enterprise services business (they own EDS) and a consumer brand that associates HP with printers and laptops. Whatever the intent of the company when Palm was purchased, the strategy was short-lived, blown away yesterday by a change of CEO and his moves to build a competitor to IBM. But does WebOS still have a future?
WebOS itself is fascinating, albeit proprietary. The architecture is pretty much what a Moore's Law believer with an eye on open standards would devise. Simplifying hugely, the user interface relies on HTML 5, so any apps designed to be platform agnostic (like Amazon's new cloud Kindle, or the FT's app) would work there fine if their owners so desired. The HTML 5 angle means WebOS stands to benefit from Apple's control instincts, which are already driving app designers to the standard. There is also access to the underlying Linux kernel and to the graphics hardware, both provided via C/C++ toolkits, so games can also be made to shine here, but I doubt that will be a feature of the platform until adoption grows for other reasons.
HP didn't send me a trial device to evaluate, so I've no personal experience of what the actual user experience of HP's touch devices is like, but there's no doubt this architecture is well considered for an edge-of-cloud device. HTML 5 for the user interface and node.js for the system already seems to be the destination most cloud geeks aspire to reach. So why didn't it sweep the world? Why were there mountains of the things stacked up in US retailers, ignored despite a round of price cuts and publicity, even before HP discontinued the TouchPad?
Technology Is Not Enough
To win in this space takes more than just technology; it take critical mass in multiple dimensions. The naked device is rarely compelling. There need to be apps to run on it, networked experiences to have with it, data ready to sync with it. While HP was able to attract around 8000 apps in the last 18 months, the money to be made on other platforms - notably from Apple iOS and Google Android - meant WebOS support was likely an add-on rather than a target for most developers.
The HTML 5 aspect means the technical barrier to delivering content on WebOS is low, but it will be tough attracting the business attention at a small app development shop. HP did itself no favours in attracting that development community either. In creating WebOS, they used a great deal of open source software and implemented open standards. They thus stood to benefit from the freedoms these both create - great software at low investment, and knowledge of the patent landscape around it. But they've so far failed to make WebOS itself open source in an open community. So the software freedom stopped at HP. After the project cancellation, all that goodness remains trapped.
The result was they gained none of the spontaneous adoption from which open source software benefits, derived in part from the confidence OEMs gain from the openness. At it's end, WebOS was only available on HP devices, and that was the death of the platform. HP may have huge consumer clout with devices we have to buy, but it's been a long time since there has been an HP device we wanted to buy. Phil told me HP was relying on the quality of its execution to win in the market - "software quality and in-store experience".
Clearly that wasn't enough. It takes organic growth, rippling through a huge constituency that has been won over by a multi-dimensional compulsion. HP squandered all the opportunities, keeping the platform closed and controlled, keeping it HP-only and using only its traditional channels to sell. I've seen this story before (as has Josh, actually). I'm in no doubt that it's a technically interesting platform, but without community participation through genuine open source communities and availability on non-HP devices - it was dead end even before they cancelled it.
So what now? What will happen in the (not unlikely) event that Google is forced into a settlement with Oracle that demands per-device royalties, preventing it from using a "less-than-free" model in the market? What will happen if Google's purchase of Motorola makes the generic device manufacturers - who have been flocking to Android - seek an alternative that does not develop the market for a competing device manufacturer?
Released as open source, with a focus on bringing software and standards freedom to the market and not just to HP, WebOS could become the darling of both the mobile and open source worlds. It could be the stalking horse waiting to benefit from Android's misfortune. Clearly someone at HP has explained all this to their CEO, as he indicated in the post-announcement press conference that he's open to ideas on what to do with WebOS. I'd encourage HP to set it free and disrupt the mobile market even more than Android. The question is whether HP has the courage to live the koan, trading control for influence, rather than keeping this potentially world-changing platform under tight control. Time will tell.
[Updated to reflect the news HP discontinued the touchpad, which broke hours after original publication]