Just in case people are browsing the comments trying to decide whether this is worth spending reading time on: it is. The headline is rather off-putting, but the article is a somewhat qualitative and nuanced discussion of a number of failed and successful software projects, and contributing factors.
An overview: any project can succeed or fail regardless of the design/development methodology used; more successes using lighter, iterative techniques. Individuals have a huge impact on project outcomes. Furthermore, individuals are inconsistent and your rockstar/ninja developer may turn into a plodding average Joe and visa versa depending on management style, methodology and other circumstances beyond your control.
Very good review and still highly applicable (published 1999)
Cockburn places his finger on the fact that the enormous variability of human beings feeds into wide variability in project outcomes.
I sometimes the analogy of cooking (because that's not played out, amirite?).
Sure, you have all the cook books, you have the sharpest knives, you have perfect mise en place and so on. But ask a master chef what the most important factor is in fine food. He or she will say "quality, fresh ingredients".
And so it is with software. Without quality engineers you cannot make the soufflé rise, no matter what recipe you try to use.
And to torture the (apt) metaphor further - put a master chef in a large-chain fast food kitchen, force him to use the cheap frozen ingredients and the company's standard recipes, and you'll still end up with the same low grade hamburger you've been churning out all along.
is alistair c still active? he wrote my favourite process book (cooperative game) and i was looking at his wiki just today (to be sure i had the difference between user stories and use cases right), but i don't seem to hear his name mentioned much these days.
he's a smart guy, who seemed to be right more often than most. i hope he's still around.
An overview: any project can succeed or fail regardless of the design/development methodology used; more successes using lighter, iterative techniques. Individuals have a huge impact on project outcomes. Furthermore, individuals are inconsistent and your rockstar/ninja developer may turn into a plodding average Joe and visa versa depending on management style, methodology and other circumstances beyond your control.
Very good review and still highly applicable (published 1999)