I Just Don’t Understand Google: Comp Sci REthink

Yesterday I had the pleasure of see Peter Norvig, Director of Research at Google, speak about his industry work and 2009 Journal “The Unreasonable Effectiveness of Data.”

With his multi-coloured polka-dot button-ups, Peter is one of the most colourful men in Computer Science, but more importantly he is the Father of modern AI. Today his algorithms write themselves. The breakthroughs have not come from an arduous process of programming line by line, when observing the code it is relatively simple when compared to the accuracy of their results.

So how did Peter reinvent the proverbially wheel?

Through a REthink of the computer science design model

OLD

The classical view of computer science design was and in some areas still is program-centric. Meaning, the world of a developer was to spend an arduous amount of time on a top-down process focusing on every little thing until there was a level of perfection that was acceptable. Data was used as a means to an end to test the program. This process was rigid and very susceptible to obsolescence.

GAME CHANGER: We now live in a world where data is ubiquitous and we can process it all with a click of a button

NEW

The amount of data available to us has opened doors for all of us. It has created a world of real time insight. Peter has taken this game changer and  applied a REthink to the programming-centric world of Comp Sci. It turns out the simplest programs can do just as well as the finest programs. In the past the great divide was data. Here lies the REthink. Peter now sees the Comp Sci design model as data-focused. The program is now secondary, with enough data the errors of the program are negligible. Utilizing the World’s dynamic data Peter has given birth to a dynamic entity. His lines of code are no longer exposed to the risk of obselcence. They are perfectly adapted to their environment, because the environment drives the process.