
Bill Stone is CEO and president of Handango has a reality check article on RCR . He has an article that in the main I totally agree with, as far as OS’s go on the handset. But that is as far as I agree with him. –
It feels like every day we are seeing articles about the “operating system wars” and who the winners and losers will be. The questions and analysis seem endless. What about Nokia’s purchase of Symbian? Will Google’s Android gain traction? What is the impact of Apple’s proprietary OS on the marketplace? Is Palm going to survive? Is Windows Mobile 7 going to allow Microsoft to finally break out in mobile? What about LiMo? Will Adobe be able to do for mobile what Flash did for the Web? RIM: mainstream or destined to be only 10% of the market? Then there are the “Mobile 2.0” widget/browser technologies. How many widget containers (AXcess, Plusmo, Widsets, Opera, Plaza, etc.) will make it? There’s lots of confusion out there when it comes to finding answers to these questions.
And this of a 5 point observation –
1) Fragmentation is here and we better get use to it.
The “winners” will have to support multiple OSs. We all know applications run best in a native environment. Developers are going to have to build their applications across multiple environments in the future (vs. primarily Java or BREW today for feature phones). Prioritization of which environments to support and near ubiquitous coverage of the OS will be key to success.
I agree with the what of his observations. We diverge however when it comes to why which is the crux of the issue. The failing in my view is in his first point. Fragmentation maybe here but the carriers aren’t going to like that for the simple reason of — portals. For three iterations now Apple has convenietly (iTune, iPhone, iPhone2.0) utilized a portal with considerable success. The Verizon’s, AT&T’s and other carriers will do the same. But wait! Can one really expect the carriers to write portal code for a dozen OS systems? Perish the thought.
The second one comes right out of the Desktop environment — thin clients. The desktop world has realized for at least a decade that one of the largest cost centers is the ongoing support required to keep a discrete compute system running. The usual approach to deflect this support has been thin client developments — Citrix, RDP, etc. Those same costs are evident in the handsets to a greater or lesser degree. It would be beyond comprehension for a Verizon who has visions of portals in their heads to face the daunting task of managing this for four handset suppliers.
The final realization is that if one has a reliable high speed connection why not use it? Not only that but a great deal of the carriers game plan is to assure lock in. Hence the reason the ETF has long outlived its financial need. So how would this play out? –
-
A very small layer to provide boot code and lock on to the network.
-
A minimalist IO layer is downloaded to handle the various ports.
-
Finally a X-like layer is downloaded followed by the desktop presentation. But the desktop is not downloaded it is handled as session layer traffic and the actual compute is done on back end servers.
The benefits? You code to a single IO/UDP layer on the handset. You don’t care about the OS! A whole cost layer is removed so the handset could be considerably cheaper. The support goes to the datacenter where it is the cheapest it can be. Bottom line, the handset is nothing but a fancy terminal.
Dog what about my data when not on the network? I still want my data! That’s actually already resolved as well. The carriers use a AJAX based, ala Google Gears layer to provide persistence at the handset. When you get back to the network a sync occurs. The IO sublayer handles making sure that the data in RAM stays that way even when the handset is turned off.
Benefits to the customer? Well there is a big one. You can sit at your desktop or laptop thru a remote access connection work on that data that is on the server just like it is on the desktop. Syncing becomes somewhat easier as well. If done right one could have this as a data store like gSpace.
The bottom line is, done correctly the OS is toast. The irony is Google with or without Android is better positioned to pull this off than anyone. Though to be honest the technologies to be used are company agnostic.
Bill, the fact is in a high density wireless world the need for a permanent discrete OS on the handset is superfluous. Its better for the carriers. The carriers can deliver a cheaper handset. It has features the users would like. Its where we are headed.
Linky