Our interaction with computers has been largely unchanged since the late 1960s. It’s easy to argue that progress has been made with multitouch displays and voice-based user interfaces, but the underlying interaction paradigm remains the same. Computers exist to serve us and execute our directions… but little else. It is frustrating to have the world’s information in my pocket, teraflops of compute, and a dumb interface that can barely determine when I’m looking at the screen.

Under the umbrella of Cajal, I am building a set of technologies that change how computers understand and interact with their users. There are fundamental weaknesses to overcome in the Engelbart approach, but I believe it is possible today to build systems that can read for you, tell you what you know, show you what you don’t know, and help you learn new things rapidly.

The working title on a manuscript that I hope to finish by 2020, this starts with a simple question - how did we arrive at the current Scientific Method, and better yet, why haven't we found something better? The first insight is that the current method we use today is one of many different methods presented through the course of history. Our current method was first described by Francis Bacon in the 1600s. On a deeper read of his works, you find that not only was he a fountain of brilliance, he also foresaw many of the necessary complications of having humans in the loop. The method we use today was explicitly engineered to manage the biases introduced by humans.

Where We Are Going presents this historical view then asks the question: “What prevents us from realizing Bacon’s vision for a knowledge generating machine?”. The second section presents an NLP-based approach that aims to codify a researcher's intuition and make it computable. The third and final section integrates the computable substrate with modern software development techniques like version control, data federation, and continuous integration to outline a scalable knowledge system with a proposed meaningful incentive alignment across all types of scientist.