On Thu, Aug 13, 2009 at 11:36 AM, Aaron Winborn<winborn@advomatic.com> wrote:
great ideas. however, i suspect that no matter how much we try to plan a road map, real life will take over and development will continue haphazardly in relatively uncontrollable spurts, at least as long as we keep pushing the envelope (which i'm all in favor of).
Which is why I suggested the second alternative, of intentionally longer and shorter release cycles. Dries generally sets the date for code freeze, and so he could have some control over the cycle length. Influencing developers to focus on API vs. UX is indeed a lot more difficult, and road maps are similarly problematic as Aaron points out. But I, at least, find it conceivable that the development community could be slightly influenced enough to eventually reach a point (in a year or 2, perhaps) where they would actually be thinking "API release" and "UX release" while engaged in those haphazard spurts (alternative 1). And I only mean on the average across the entire development community. The standard deviation is expected to be quite large, and that's ok. :-) I don't intend to tell any developer what to work on.