Presentation: Tweet"I Throw Itching Powder at Tulips"

I don't write software like any of you. Software development methodologies are nonsense. Literally. Here is my manifesto:

* Nature over Individuals and Interactions over Processes and Tools
* Insights over Working Software over Comprehensive Documentation
* Problem Engagement over Customer Collaboration over Contract Negotiation
* Grappling with Mystery over Responding to Change over Following a Plan

My guiding principles:

* Create opportunities for change
* Continuous engagement with software
* Code and scientists must work together
* Build projects around mysteries
* Stare at the problem staring at you
* Puzzling software is the primary measure of progress
* Surprising development
* Contentious attention to insight and diversion
* Simplicity is beside the point
* Self-organizing scientific method
* Continuous revolution

I write software to make machines do what they cannot do. I work from no requirements nor specifications. My only collaborators are human knowledge as it exists today (at a distance) and the code I am working with (which constantly stares blankly back at me). For standard programmers, design and coding are, at most, puzzles; for me design and coding are mysteries, at least.

In this talk I'll tell you about the software I am working on and why no software engineering process has anything to do with it.