Google's stab at Chrome operating system has had many a pundit thinking more and more about the direction of the OS market. Microsoft's Windows has been suffering heavily since Windows XP's inception. It's widely reported that it took 10,000 people and over five years to produce Vista. yet it was so badly received that Windows 7 was released just two years later.
Such are Microsoft's struggles, it's behemoth has hardly experienced change between the two versions - despite the hype, Windows 7 is Vista with a new task bar and a few bug patches.
Legacy driver and software support, a creaking registry, a near-infinite number of hardware combinations, and the necessity to provide multiple flavours of the same underlying OS to computers ranging from a netbook to a hardcore server, has meant Microsoft's releases are inevitably slow to arrive, buggy, and perform poorly next to other highly-tuned host operating systems.
Linux has seen usability increases since the inception of projects such as Ubuntu and further driver support for legacy and new devices. However, it's still struggling to make much impact on the consumer market, failing to overcome the novice user's fear of using a command line to drive their software and hardware.
This doesn't mean it doesn't have a future - Linux is highly adaptable and can be slip-streamed to provide the bare basics needed by the end-hardware used to inter-operate with the cloud. It'll also provide the underlying sever operating systems needed on the Cloud itself.
This is why Google is using an implementation of Linux under-the-hood of Chrome OS. From a Microsoft perspective this could be where we see a new use of MinWin, Microsoft's cut down kernel and OS - which allegedly already provides the underlying core components of its current operating systems.
It's also easy to see that a highly constrained hardware base provided by the manufacturer of the underlying OS will serve to mitigate the huge problems of driver compatibility, legacy device support, and stability issues.
This why Google is aiming to control the hardware specifications used as the underlying technology for it's Chrome OS. There are additional advantages to having your OS in the cloud on a constrained hardware implementation and minimal on-silicon operating system. Your desktop and data can be literally taken anywhere, usable on any device available - even more important in the real world of hot-desking, working from home, and international travel. You programs will be the same - not a plethora of competing programs or radically different versions and implementations.
Your data will ultimately live in a highly redundant network supported by high-end technicians and data gurus (if you ignore Microsoft's recent Danger debacle). You will ultimately have enormous peace of mind that you'll be able to work without the hassle of software and data management.
From the Cloud's side, the companies providing the network applications are selling a service to their tools and remote software which has many benefits over the usual install and use software.
These include maximising the removal of the retail channel middleman, cutting down on possible software piracy, producing greater end-user analytics of software and services use, allowing effective targeting of online advertising, locking-in end-users to their tool-sets and data storage, increased collaboration support, and supporting software on only a minimal set of hardware implementations. By providing all available software, the Cloud can also protect users against viruses - the software will be much more constrained than is allowed today, and any tumours will be much more rapidly discovered. Their are obvious problems with an operation system in the cloud.
Internet connections aren't always available, but with our increasingly on-line world, and the increasing availability of web-applications that operate in an off-line capacity, the weaknesses of the Cloud OS are limited further. As the Cloud OS matures, so will the availability of our Internet access - to the point of constant connectivity in the developed world.
Cloud OSes are going to boot fast, provide the basic needs of the majority of users and will be incredibly stable due to a limited pre-specified constrained hardware base - and cheap hardware at that. It's early days, and Google may seem to have jumped the gun - but by releasing now and delivering it as open-source project, it allows Chrome OS to mature from its infancy and gain a major foothold in what will become a highly competitive market.
It's not hard to see the future of operating systems is in the Cloud, and it seems Sun was right all along - the network will be the computer.