This is a stupid idea - looks like Google have lost the plot completely now!

If the iPhone shows us one thing, it is that future applications on Mobile phones will be like future applications on Desktop machines: they will be delivered through a browser, with the underlying OS being immaterial.

"If the iPhone shows us one thing, it is that future applications on Mobile phones will be like future applications on Desktop machines: they will be delivered through a browser, with the underlying OS being immaterial."

This may be the future that Apple wants for the iPhone, but it certainly isn't the case at present - as evidenced by the hoards of people who want to be able to develop for the iPhone without the "limitations" imposed by only being able to develop within Safari. This may be where a certain niche of applications are headed (and also currently are) in general across all platforms but the majority of development will still be on the OS level for the foreseeable future. The reason Apple wants developers doing their stuff inside Safari only is because they still want overall control of the hardware and software development on the device because it's proprietary. What Google is proposing is completely open, and so the forced "protection" layer is unnecessary and contrary to the philosophy they're aiming for. If the OS itself is designed specifically for the platform and for providing developers with the tools to develop, why insert another layer?

I'm not backing the Google horse quite yet, but I do believe that it will wake up the industry on a software level (much as the iPhone has done on a hardware level) to the point of pushing everybody foward.

Future apps on desktop machines "Delivered through a browser"

Bollocks it will. You have lost the plot in your web 2.0 fugue.

If the iPhone shows us one thing it's that no matter how crap something is, Apple can market it well.

Regardless of that however, there's still the more blatant issue of the problem with proprietary browsers and platforms that actually prevent widespread adoption of web applications. We still have no agreed standard on what access to hardware a web application should have, if it's limited some applications wont work, if it's open you have a security risk. Web applications developers also already face the issue of non-standards compliant browsers and everyone's to blame there including Microsoft with IE, Mozilla with Firefox and Apple with Safari - none of them correctly implement the various web standards and a proprietary platform and browser only increases that problem. So even if we are going towards the whole web applications route with mobile devices an open platform and open standards shared by all device manufacturers is still the way to go - if only for the sanity of web applications developers such that they don't have to write!

The point is, the more we move to web applications the more the browser and OS are tied together or at least forced to communicate so that the browser can do the things it needs to do to be able to support as wide a range of web applications as we have standard applications.

Finally, It's also worth noting that some applications certainly wont even work over the web, specifically applications that have to shift large amounts of rapidly changing data such that a networking connection can't handle the amount of resources shifting back and forth - many modern games for example. All these are reasons why an open OS is still important - to ensure standardisation across the industry so you can be sure your applications be they web or normal are accessible on whichever mobile device you use.

And there was me thinking the iPhone taught us that OS design and user interface was of central importance.

I'm very impressed with Nokia's Maemo project and the results of the N800 & N810. They've put in a LOT of time, effort, and money, whereas all I see from Google is huff and puff.

Until Google has a piece of hardware and a dev kit sitting on my desk, it's vaporware and FUD.