Geoff Cohen, Independent Consultant
Track: Rich Internet Applications
Date: Friday, April 25
Time: 3:15pm - 4:00pm
Concepts programmers hold dear modular design, encapsulation, reliability, versioning, performance, human readability, certainty, and predictability. These are not only becoming impossible to achieve, they are dangerous and ultimately damaging ideas.
We've gotten away with it for decades because when we write software for a single box, those ideas make for decent software. But the network isn't just a giant box. A thousand programmers on a project isn't just more than a hundred, and a hundred million (or a billion!) users aren’t just more than a million.
We need a new set of ideas, a different set of tools, and a different way of speaking of and thinking about software if we're to rise to the challenge of building (or growing) software relevant to modern needs.
But all is not without hope. This talk details a number of intriguing and promising approaches (aspect-oriented programming, intentional software, biological models of computing, protocol-oriented programming, etc.) and places them in context.