I think this decade will push further towards a partial decentralization of software.

Some fields that will bring impactful changes:

  • Remote computing:
    • Local input —> remote computation/execution —> local output. Apply this to any computationally heavy task and your personal computer will merely serve as an interface to your real computer in the cloud.
  • Remote work tools:
    • Remote work is redefining almost every aspect of how we get things done. I think there’s still a lot of work to do when it comes to abstracting other aspects of the office to software. There are some concepts we were physically limited to that we could simply omit or drastically change.
  • NFTs and applied digital scarcity:
    • Applications can be anything from provable digital identity to non-reversible file exchange. The Nakamoto consensus algorithm is about publicly proving ownership of something and NFTs apply the same idea to the ownership of a unique digital item.
  • Cryptocurrency:
    • In an era where information moves across the globe in a matter of milliseconds, moving value as data packets is inevitable. The replacement of an antiquated and over-centralized financial system is a natural consequence of the internet.

Interestingly, Moore’s law should play at the advantage of decentralization. Today, you need to explicitly spin up an EC2/GCP instance to train a deep learning model, but tomorrow you’ll most likely be able to do it locally –and cheaply–, or meta-locally using your local computer as an interface to your cloud computer, which will perhaps not be hosted by one of the four horsemen of the Internet (GCP, AWS, Azure, Alibaba).

Decentralization here doesn’t necessarily mean a more evenly distributed computation network, it just means that more than one company/entity has control over something.