Are they kidding? IDC is predicting that Microsoft is giving up the home user and gamer market? Do they own stock in Apple? Certainly, businesses will be examining the feasibility of thin clients for desktops -- it just makes business sense to not provide every executive, clerk and janitor with a standardized, whiz-bang, costly, state-of the-art PC. Laptop users, not so much! They'll need the full-blown fat capabilities so they can work in a vacuum (or cheap hotel room).
My opinion is that some shops will move to thin clients as soon as they think they can. Those same shops now lock down every workstation so the user can't do anything at all beyond work on authorized files stored on the corporate network, and they already store all their corporate data on servers they own and fully control. Without their networks, they are already shut down, so thin clients make sense. Handing all their data to Google does not.
If Windows 7 hangs on as long as XP (about six years), that takes us to about 2015. I would be surprised if by then, the rest of us are so confident in Google and its competitors that we are willing to trust every important piece of data we own to them. Networks are vulnerable to minor disasters (a carelessly operated backhoe, for example), major disasters and human malice (anything from terrorist acts to abrupt changes in terms of service; check the mail you've gotten from your bank in the last couple of months for examples). I am not convinced that individual users and smaller companies will be confident enough in the cloud to put that much trust in it by 2015. I may be wrong, of course.
While I don't see Windows 7 as the end of the line for Microsoft's Windows OS, I do think that the times, they are a-changin'. The one flaw with always-connected computing is that it's not always on. With the problems that people have getting good cell phone service, the notion that you can have all your data and apps dispersed in places other than your local PC looks foolish.
What we need to see is a synchronized local/remote hybrid. When connectivity is present and bandwidth permits, then the data can be synced between local and remote storage. When connectivity is absent or of sub-standard quality, you process locally. This notion isn't really rocket science, you know. Until the data networks that permeate our nation get considerably more robust and reliable, and bandwidth gets immensely more abundant and affordable, the notion of a totally cloud-based computing paradigm is so much pie in the sky (to mangle a metaphor). And we haven't even talked about the security aspects of this yet.
The end of fat clients and the move to the cloud all depend on an infrastructure that can support what these technologies represent. And what's a better example of that infrastructure challenge today than the ongoing "There's a map for that" battle between Verizon and AT&T? Quite frankly, from what I read, even Verizon would have a difficult time attaining the level of service that AT&T has achieved, even though the level of service at AT&T falls somewhat short. Make everything essentially a thin client and put it on the 'Net or in the cloud, and just who would be the infrastructure provider that could stand up to that challenge? I don't think the two of them combined could muster up the dollars to deal with this one in the seven years that IDC seems to think this will happen.
And what about the rest of us? I'm referring to the 30 percent (maybe more, maybe less) of the population that don't live in major cities and face virtually no investment on the part of providers because there isn't enough profit where we live. A lot of us in the trenches live in localities with a single ISP that provides DSL at "blazing speeds" of 384KB to 1.5MB on an intermittent and geographically spotty basis (actually, they call this "high speed"). I can't even imagine trying to convince one of my clients at this point to consider any kind of thin client cloud computing. Can IDC and all the other cloud pundits help me sell this technological nightmare to more of my customers? I think not. Fat clients will live for many more years than the folks at IDC think. And if Microsoft doesn't provide that client, someone else will.
I think you're right about the premature obituaries on fat clients, but I think that your reasoning only touches on part of the issue. I feel that there are at least two other related issues that need to be addressed. One, not everyone has unlimited Internet access, not by a long shot. Even though AT&T would like the FCC to believe otherwise, there are large numbers of people who do not have unlimited broadband access because the phone and cable companies are not willing to run the wires out to the rural areas of the USA. For some of us, the only games in town are dial-up and cellular modems with their 5GB/month caps before you are charged overage fees.
And two, security. There are some things that people just do not want on someone else's computers. There are things that are just too personal to have floating around in the cloud. There are liability issues with things like medical records or student grades. There are privacy issues.
You are right, and all the clowns saying we will all have dumb devices and store and access everything we need in the cloud have their heads in a controlled-substance cloud. I have worked in the security industry and with the NSA and I will NEVER trust the cloud with any of my data or for any of my apps.
IDC is dead wrong. The proponents of the thin client model have been predicting the demise of local processing almost as long as it has been around. Thin clients have come a long way but there are still things that you cannot do with them. Today's thin clients face pricing pressure from netbooks and from entry-level laptops that make them a poor choice for most users and in most settings. Lots of baseline services can be delivered to a thin client but these are largely in settings with a few dedicated application needs. Basic personal productivity? Browsing? Sure! But what about gaming? Real-time audio and video streaming?
Demand for bandwidth continues to grow exponentially. If for no other reason, this means that today's thin client will become crippled and obsolete long before today's entry-level PC -- which, feature-for-feature, costs perhaps $100 to $200 more to purchase than its entry-level PC brethren and can do a great deal more than that thin client. Can a netbook be a great thin client? You bet! But it can also be a really good full-function computer when you need one and insufficient bandwidth is not available.
I think you're right -- with the lapses in security causing data to be hacked or stolen, users are more likely to hold their data close to their chests.
This may well be the time for the OS-on-a-stick to come into its own. After all, the desktop, laptop, tablet and mobile devices are just interfaces that we use depending on our requirements at a particular time. Let's start pushing for the separation of the OS, applications and our personal data files from the hardware. From a practical perspective, perhaps the mobile device should be the core, since this is the interface most people carry everywhere, with seamless links to other hardware as necessary.