Saturday, July 19, 2014

Skateboards, rockets and math

This slide from Spotify has been popular for a while:
It explains how a product can be built iteratively, satisfying first the need for transport with lesser means and then evolving to a more powerful platform. In this model feedback such as business model validation and satisfaction from the project sponsors can arrive early, even when they're negative (especially so).

From what I read about Spotify, they're also well-aware that incremental development can only take you so far: you don't get a car by making a better bicycle. Sometimes you have to take a leap to a new platform; or if it's clear that simpler technology won't support your vision, start from an higher level of essential complexity.

Here's someone that didn't start from a skateboard:
Imagine telling Spotify to install WebSphere (or some other technological terror) as the first step when starting a brand new project; or telling SpaceX teams "Come on, Elon, just give us a bicycle and we'll get some first sales!"

Or telling Larry Page that programming isn't math:

Keeping in mind this strong dependency on context, where do the competitive advantages of your product lie?
In finding a better fit with the needs of users, maybe a lower time to market? In solving technology problems to carry humanity into space at an acceptable cost? In algorithms that can find high quality information in the web ocean? In fooling VCs in giving you free money?

From your vision, your choices of education, process, and technology.