The pendulum may be swinging, but I'm not sure I see it. Mobile computing isn't really different from the PC/web of the 90s. The shift is in the access pattern, not the access paradigm. Napster/Kazaa/BitTorrent is an example of the pendulum swinging. The Web was originally envisioned as an open access community, but we have yet to move it beyond the client-server (centralized) paradigm. Home servers preserve this paradigm. Many of us here run home servers because we're geeks, but that doesn't apply to most people on Facebook. Look at Diaspora: great idea (wrest control of our data from Facebook), but still the centralized (seed/server) paradigm.
Distributed computing at the level we want is still WAY too hard for the end user. The goal should be to make it as simple as using Facebook today, but retaining control in a much better way than Facebook or Google let us do now.
I agree with you, and my great hope is that there is (or will be) enough of a market opportunity for someone to exploit. The market would be for a productized distributed cloud solution that's simple to set up and maintain.
The target demographic would be people who are concerned enough about privacy, ownership of online identity, and government access to corporate cloud data to pay for a self-operated solution.
The prerequisites for this to be viable as I see it are: cheap enough hardware; cheap enough full-access pipes; enough demand; good enough UX. The first two seem fairly obviously on the right path. The third I'm not sure, and the fourth is under the control of the vendor.
From my point of view, full-access pipes are the problem. Asymmetric connections, bandwidth caps, incoming port filtering, etc are all de rigueur for large fixed line ISPs these days. This is especially true in North America. The mobile connectivity picture here is even worse.
Demand can be generated, if you do it correctly. The hardware is less of a challenge if you do the distributed computing on several layers at once. Apple has shown us the way on UI/UX, as has science fiction literature for decades now.
We'll never get completely away from client-server (especially with mobile computing), and that needs to be part of the design. But we have seen the power of peer to peer networks, and even some kind of proxy network between the p2p base and mobile clients could be done in a non-evil way.
Cheap hardware and cheap bandwidth are easy to get... in datacenters. And if you have reasonable scale you can amortize away the cost of high availability, too. The cloud isn't the problem here, it's SaaS. If you have a personal cloud server, you can not upgrade all you want.
That's an interesting angle that I don't know enough about. For example, do the laws governing SaaS data provision to law enforcement also apply to data centers? If they do, then you're giving up some privacy from the gov't for the convenience of outsourcing server management.
I guess what's interesting about your idea is that it presents control as a continuum as opposed to a binary. There are varying degrees of control/convenience possible, but most of the current options are heavily weighted away from user control.
Also, I don't think this thread is really about the cheese-moving/silent-upgrade phenomenon so much as data and identity ownership.
We're not getting away from SaaS without an underlying infrastructure for something along the lines of Software as a Peer to Peer Network. PaaS and IaaS have the potential to exhibit some of the same problems as SaaS. Look at the AWS/Wikileaks thing. There may not be privacy issues in the same way, but Daddy can still take the car keys away.
Also, bandwidth is only as good as the size of the total pipe between communicating entities. The cheap bandwidth has to go all the way to the end-user or it doesn't help too much. Think about backing up personal data. The initial upload takes forever in cloud backup systems.
I was excited about Opera Unite because it did make setting up a small webserver really easy for end users. Unfortunately slow residential upload speeds, lack of a must-have app to run on it, and Opera's small user-base mean that it didn't go anywhere.
The responses to Opera's marketing also indicated that a lot of people didn't understand what a server was. It launched about the same time as Google Wave, and both products had marketing campaigns which emphasized how easy each product made communicating with friends. As a result, many of the comments I saw on forums were people asking if the products functioned identically.
I feel that if it had launched with a peer to peer social networking application that synced messages and photos between friends, it might have been significantly more popular.
Peer to Peer social networking suffers from a massive chicken and egg problem. Something like this isn't going to be built on the backs of Facebook users seeking privacy or freedom from UI updates.
The key to solving this will be to find the killer app for always-on peer to peer home systems that won't get you in legal troubles.
Distributed computing at the level we want is still WAY too hard for the end user. The goal should be to make it as simple as using Facebook today, but retaining control in a much better way than Facebook or Google let us do now.