Analysis If necessity is the mother of invention, maybe it helps to be really, really needy when you have to come up with something great. That would seem to apply to Palm, which was considered down and out when it was developing its new mobile operating system, webOS. But as further details of the system and development process emerged this week, it looks like a very smart design choice indeed.
Dave Pike, well known in the Palm developer community, makes some intelligent speculation on his blog based on leaked screenshots of the development environment. Dave spikes Linux running a web server, but also suggests that Palm has used the Open Services Gateway Initiative running on a Java VM.
"OSGi makes Android's IPC-based component model look slow, weak and restrictive. OSGi communicates with its bundles by ordinary method calls, a much better way to enable the kind of mash-ups that have made Android attractive to developers," he suggests. Well, there is an OSGi framework for Android - we covered it here - but they come from third parties and it's not Google's preferred approach. So a really ancient specification designed for embedded devices finally gets its moment in the sun.
We have an early glimpse of webOS development from internet radio service Pandora, over at PalmInfoCenter here. Pandora's Tom Conrad points out:
So what can we conclude?
Overnight, large industrial-strength operating systems, the kind Microsoft and Symbian want to license you, now look like overkill for simple local applications and applications that access a web service. This leaves a small but important class of applications for which there is no alternative but to use the native frameworks.
Apple's iPhone OS is also a complex industrial-strength OS - Unix, obviously - but thanks to its very well thought out APIs, developers don't find development as challanging as Symbian, even though Objective C may be new to them. Android seems to suffer all the drawbacks that writing for a non-standard Java would suggest, but it does have the might of Google behind it.
Palm's careful hybrid approach makes a lot of sense. Make local services really rich and easy to use - and make accessing web and "cloud" services trivial. This is really the territory Adobe should own. Adobe's Flash for mobile once looked to solve a lot of these headaches, but the company could never capitalise on it. It's not uncommon to see Flash Lite today, but rarely as anything other than a presentation run time.
Of course Palm remains the neediest, and least capitalised of all the players: Apple, RIM, Nokia, Google and Microsoft have far deeper pockets. But after all the mis-steps over the years - anyone else remember Palm CTO Bill "Mad" Maggs explaining that mobile devices didn't need a memory management unit? - Palm has stumbled on what may be a real sweet spot. ®