In the Electronics industry, there has traditionally been a strong preference for choosing components which have a second source, that is an provider who can independently deliver functionally identical and interchangeable parts. Businesses prefer this as it helps to protect against supply side disruption, and the need to change designs because a part can no longer be purchased.
A second source for Twitter, for example... would be API equivalent, and would hopefully be able to perform the same functions should Twitter close up shop. However, this is not an attractive option, because Twitter is not a component, it is a complete service and infrastructure.
A better way to provide second sourcing for Twitter would be to provide the tools to build a Twitter equivalent ourselves, so that we maintain the infrastructure (or pay a hosting provider to do so). If any part of the system goes down, we could route around the nodes that have failed.
I'd like to call this idea offsourcing, for lack of a better Google tag.
We should start figuring out ways to offsource the main things we all depend upon daily. To figure out components and services that could be used as replacements. Then we need to start offsourcing to these new systems and networks.
Merely replicating an existing system is not enough, however. There needs to be an through analysis of what the benefits are of a system, and how it really interacts with things. For example, twitter has evolved into a notification platform. When you post a blog entry, you can tweet it. But it's more than that... it's also a voting platform, because you can retweet someone's notification about a blog post, in effect giving it your endorsement, and a vote.
The downside of twitter is that it's a big stream, and people tend to piss in it. As pointed out by Merlin Mann in his post, Better, we need to be more mindful of what we put out there, and we need to add value. It's impossible to edit your posts once they are out there, so they are biased towards low value, high flow sewage.
An ideal Twitter replacement would allow notification, but would be more nuanced. It would allow better curation, and support editors. it wouldn't be Twitter anymore, but would keep the best features.
I believe it will take a while to find the necessary components to build a useful, decentralized slower but more useful Twitter replacement. I don't have a quick fix to offer here. I just want to get the idea of doing it into play.
Maybe I'm getting too tied to creating a new term... but I think this type of analysis needs to be done for a lot of things, and I'd like to be able to find those ideas with google, so it needs a new term to help.