GregHowley.com

Cloud Decentralization

August 27, 2013 -

I believe that the future of the internet lies in decentralization.

Google's products have done a marvelous job of showing us the value of the cloud. In many ways, the usage of cloud services is entirely invisible - their autocomplete, voice-to-text, and many other common Google services all use cloud processing. But the centralization required by the cloud carries disadvantages. That which has been made most clear by recent events is data's vulnerability to mass theft when the data of many users is stored in a centralized location. Although Google has been both vigilant and successful in preventing theft via technological infiltration, or "hacking", they've failed to stave off theft of their users' data when that theft was perpetrated through surreptitious legal channels.

I've been considering for a while now which of Google's service I could transition myself away from with a minimum of inconvenience, at first because I don't like having all my eggs in one Google basket, and more recently for the above-stated reasons. I've begun using DuckDuckGo rather than Google search, and have been very happy with the service. I'm now considering alternatives to gmail, which won't be so difficult a switch for me as it may seem given that I mostly use @greghowley.com email addresses rather than the @gmail.com address. A good alternative is hard to come by though - gmail is an excellent product. Being able to quickly search through every email that I've sent or received over the past few years from my phone is a convenience the magnitude of which I cannot overstate.

What would be ideal is a suite of cloud-like services that lives on my own machines rather than on a Google server. I imagine a distributed storage solution similar to BitTorrent's Dropbox alternative that stores its data on and syncs its data between my home PC, my phone, my home media server, and perhaps a NAS box. If this theoretical software suite could handle my email history and search history, providing full-text email search and search engine term autocomplete, that would be huge.

So who's gonna code this thing?