New Cultures of ProgrammingRewriting the software development rulebookA Workshop held at OT2000, Oxford, March 29 2000Peter Marks, David Harvey |
 |
Contributors
Contributors to the workshop included| Sarah-Jo Baldwin, Andy Carmichael, Bob Davison, Nestor Gonzalez, Ken Harper, Tuomas Kassila, Richard Lawlor, Hamish Lawson, David Leibs, Keith Lewis, Andy Longshaw, Ewan Milne, Gwyn Roberts, Josh Sacks, Colston Sanger, Mark Skipper, Nicholas Smith, Yvonne Wilson |
Many thanks from Peter and I to these and the others involved.
Download
Download the presentation (zipped PowerPoint).Premise
Development is often understood as an exercise in command and control - from the structures imposed in any programming language, up to methodology, process and management. At all levels above the machine (for now :-)), it's clear to us that what matters in software development is more richly described (and certainly more readily achieved) by taking a systems view, in which creativity, invention, collaboration and feedback are more important then proscribed process, and architecture and quality are emergent and not imposed. A brief presentation at the mid-point of the workshop considered four approaches (there are more!) as evidence of this new thinking: the Programmer's Stone, Dynamics of Software Development, Extreme Programming, and Adaptive Software Development.Process
The process of the workshop led through examination of the rules we program by, to consideration of when and how we successfully break these rules. Groups then worked to uncover the truths or values behind each rule and its breaking.Each truth was written on a separate (large) card, then a 'value space' was created by arranging these on the floor, the whole group refactoring the map several times to cluster related values. (The map was photographed, we hope to reproduce either the photographs or a schematic at some stage).
Smaller groups were formed around these clusters, each of which took a localised set of observations to produce a poster with a single insight, with justifications and implications for the way we build software.
Insights
Here are the five posters produced (as closely as my HTML/PSP skills can match them...) with some commentary.- Keep going
- Software development is a social activity
- Code talks
- Software development is the debugging of half-baked ideas and half-baked code
- Simplicity enables change
Some reflection...
(1 April 2000) Both Peter and I felt this was the most rewarding session either of us had run at any OT conference. I'm still blown away by the focus and commitment everyone brought to the workshop, and enormously impressed by the output. Even within the constraints of a 3-hour workshop, what we've ended up with is direct, expressive and highly relevant.It's important to note that most of the contributors to the session were much less familiar with XP and similar approaches (if at all) than with more traditional processes and environments. It's not surprising that the insights expressed converge on XP values - two important goals of the session were (1) to let people discover these values themselves, and (2) to demonstrate that XP does not have a monopoly in values-based development.
Please feel free to mail me if you'd like to add
to these observations
Contact
Created 1 April 2000
Last modified 1 April 2000