The Democratization of Storage

How do you store your data? How do you protect the electronic essence of who you are? The way things are these days you have two choices. First, you can horde data onto your own computer. If you are diligent you can encrypt and backup and all will be well. Not quite. Alternatively, you can pay someone to let you put your data on their equipment. You won’t have to do the backup thing, and you’re clever enough to encrypt before you send them your stuff, so it’s protected, right? Not so.

The first approach plays into the myth of physical security. If it’s on your own machine, in your own home and you own a big lock you may think it’s safe. The trouble is that your computer is connected by a big fat pipe to an enormous underground industry of organized crime who are using (someone else’s) clever attacks disguised as email attachments, free games and useful utilities. There is enough profit motive in the theft of your identity that they have amassed an army of thousands of computers to make attacks – many of which belong to innocent owners. Google ‘botnet’ if you want the scoop on that. The long and short is that your data is not all that safe unless you disconnect the Internet.

The second approach seems safe, assuming you trust the people on the other end to maintain all of their hardware and diligently test for virus attacks. The trouble is that you aren’t likely to visit their facility and meet their staff, so you have to take their promises on faith – the premise being that they wouldn’t let you down because their reputation would suffer. But you just don’t feel 100% safe with your personal stuff because there is always the nagging worry that you’ll get stuck with the one disgruntled employee that releases your data in an act of spite. Worse yet, new technologies on the horizon such as Quantum Computing are threatening to make a joke of conventional content encryption.

There are larger philosophical issues at play here. You pay for computer hardware, Internet connectivity and some of you pay for on-line storage but you don’t get what you want. You don’t get privacy. On the flip side, when you do want your stuff to be in the public domain, you have to give it to the big content holders such as FaceBook, My Space, Flickr, etc., and they decide how it is distributed. You have to play by their rules, and once it’s out there you have lost editorial control. Why is that? Why is data storage in such a mess?

I believe this is the result of an eccentricity of evolution. Digital data was born of colossal mainframes, owned by big corporations and centralized for maximum efficiency. That is where we got the concept of the ‘backup.’ Every machine was so expensive and the manpower to maintain it so large that there had to be a quick way to get back to a working state in the event of a breakdown. Fifty years later we are still thinking in terms of complete backup and restore operations even though the computers are dirt cheap but the labour to perform these complicated processes is through the roof. This is a ‘Machine-Centric’ view in which the data *belongs* to the machine. You have to figure out which of several machines you might operate contains the latest version of something you are working on.

Welcome to the 21st century and a new approach to data: All of your data, broken into tiny anonymous grains of sand where they reside alongside billions of others in the big public beach known as the ‘Net. Why is this safe? It’s all about the tyranny of big numbers. Sure botnets or Quantum Computing might be able to decrypt one of these ‘grains’, but how would an attacker know how to find the others needed to make up the rest of the file? Even if the decryption were instantaneous, the network delays of retrieving billions of billions of guesses would add up to forever. Only the author of the file has the ‘key’ to determine where all of the grains exist to fetch them, reorder them and recreate the original file, and the container of that ‘key’ is small enough for the author to carry it exclusively on their person.

It gets better. This is a democratic architecture. Lots of people contribute to maintain thousands of simple servers on the ‘Net. Each of these servers is eligible to store any ‘grain’ of data, no one server is more important than any other. These servers are analogous to the routing equipment of the Internet. No one owns the ‘highway’ but everyone maintains a little piece of it and permits everyone else’s traffic to flow. Like Internet connectivity, where the emergent behaviour produces self-correcting, robust traffic for all users, this solution will provide total data reliability for all users without anyone having to perform tedious 20th century backup and restore operations.

More importantly, your data is now stored in the ‘Net cloud but is independent of any of the servers that hold its bits and is also independent of the computer you might be using to interact with that data. That is the promise of ‘User-Centric’ storage: Your data, anytime, anywhere and with complete privacy. It is trivial to manufacture the keys that decide how your data is distributed across the ‘Net. That means you can create and distribute as many as you want. You can make them read-only so that your audience can view but not modify their contents, or you can decide to let them change their contents. You can control who gets those keys or you can put them in a public place for all. You’re in control.

If you would like to learn more about our implementation of this new architecture, Google ‘InfiniDrive’ by ‘dataSentinel’ or visit our public site at www.datasentinel.com. We are looking for lots of feedback on how we can make this even better for you. Check us out.