Blockchain is still in its infancies, but people are already predicting a new paradigm shift coming with this new technology. I have read lots of articles these past months where they explain how the Internet used to be centralized with the first main frames, shifting to a distributed model with commercial PCs, to return again to a centralised paradigm with the appearance of cloud computing. What people seem to be predicting is that with IoT, and the blockchain as an enabler, this will change again and we will return to a distributed Internet. So, what is my view in this matter?
All the Internet products and services we use right now are connected to a data center or a cloud computing provider, from AWS to Google Cloud Computing or Dropbox. This means that all of our data is confined in a few computing infrastructures from a small number of providers. These providers have complete control over all of our data. Even if they have good and honest intentions bad guys could hack them leaving our data in hands of people with not such an honest intentions (take as an example the recent Equifax hack). If this is worrying now that IoT haven’t really exploded imagine how worrying it could become when even our toaster and our home is constantly uploading data to the Internet in order to be stored in these cloud providers. You see? The current Internet paradigm is completely data centralised, even with smart phones making distributed computing in our pockets.
The Internet’s current model is completely data centric
However, this may completely change with the blockchain (and it should be this way if we want to avoid the single point of failure that cloud providers involve for our data). Let’s forget for a minute all the current technical limitations of blockchain technology (as they will hopefully be fixed in the next couple of years) and let’s think of the blockchain as a distributed database which is self-regulated and persistent and where non-trusted parties can store and share data without worrying about hacks or any other malicious behaviour (we’ll see how blockchain evolves but this would be a good outcome).
Now let’s replace current centralized cloud providers by a centrally distributed “cloud storage provider” enabled by a blockchain technology which has nodes all over the world and is not regulated by a central authority as a company or a government, but it self-regulates through all the nodes and participants of the system. Furthermore, this system can not be hacked in the same way traditional database are, and we will be able to have our data available anywhere (just like with our traditional cloud that we learnt to love). You would feel a bit safer dumping your data in this new system compared to the current scenario, right?
And these “nodes” and “participants” that enable the operation of the system? Who are they? Easy peasy, all of our devices. Imagine all of our awesomely powerful smartphones, tablets and laptops dedicating part of their unused resources to make the blockchain work. Even more, imagine that the blockchain is able not only of storing data but also of computing operations and running code through smart contracts. Then we would have a global distributed cloud system as we know it today but enabled by every single device in the world instead of by centralized data centers regulated by central authorities and prone to attacks.
But let’s scale a bit more this utopian system. IoT has been a promising techology for quite a few years now, and it shouldn’t take long until it completely explodes. IoT means having sensors and little computers everywhere. These little computers could be storing data and making computations for the blockchain. Not at the same scale as our laptops but in a way that the storing capacity and computing power of the Internet is huge and completely decentralised.
A huge, secure and decentralised computing power for the Internet, please.
This may finally not outcome exactly like this but what I am trying to show with this post is how the trend is for systems to become distributed more and more for scalability and in pursuit of many other advantages (and if you don’t believe me, do you remember how Hadoop’s map-reduce algorithm revolutionised the way of Big Data computing?).