People are alive -- they behave and respond. Creations within the computer can also live, behave, and respond... if they are allowed to. The message of this talk is that computer-based art tools should embrace both forms of life -- artists behaving through real-time performance, and art behaving through real-time simulation. Everything we draw should be alive by default.
Part 1 talks about the potential of the computer as a new visual art medium. I show a collaboration between art and artist, with the art behaving through simulation, the artist behaving through performance, and the two of them working together, responding to each other.
Part 2 demonstrates a tool for "programming" how art should behave and respond. The artist directly manipulates art objects on the canvas, the way that visual art has always been created since the time of cave paintings. The tool is based around direct, geometric construction rather than indirect, algebraic "code".
Part 3 is a short performance.
This talk was originally presented to SF SIGGRAPH on May 16 2012, and was recorded at the Exploratorium on November 27 2012.
One of Go's key design goals is code adaptability; that it should be easy to take a simple design and build upon it in a clean and natural way. In this talk I describe a simple "chat roulette" server that matches pairs of incoming TCP connections, and then use Go's concurrency mechanisms, interfaces, and standard library to extend it with a web interface and other features. While the function of the program changes dramatically, Go's flexibility preserves the original design as it grows.
We all love our tests to speak in the language of business and not in the language of implementation details. In this short presentation I will show how we can get closer to this goal by writing custom assertions with AssertJ. Prepare for discussion about pros and cons, and live-coding session.
What do deployment pipelines look like when your system consists of 10s of different types of services? How do you know what to test before deployment? Should you release a service at a time, or bunch them up? This talk goes into the nitty gritty of managing build,test and release of micro services and also covers the often ignored tradeoff between testing before deployment, and testing afterwards.