[Moved to the <bug-hurd@gnu.org> mailing list.]
Hello! Matthew Ayres <[EMAIL PROTECTED]>: > This might seem odd, but my curiosity was aroused during a conversation > yesterday. I was wondering if someone could tell me what development > lifecycle is used in the Hurd project. I thought perhaps the Waterfall > model might be a good choice for a microkernel system, but it doesn't look > like waterfall. I would say that most small- to medium-sized Free Software/Open Source projects (and even a bunch of the real big ones) don't have a real development model. Everybody simply does what he feels like working on. Locate a thing YOU consider broken, or a functionality YOU consider missing, then fix/implement it. And a set of maintainers tries to coordinate that ``process'' a bit and tries to combine the individual works into, for example, a new release. Whether this is an effective process surely is a discussion on its own. But with only volunteer workers you don't have much other possibilities, at least not until the workers demand more steering/leadership from the maintainers. I'm quite sure that enough PhD students (or other people) have written nice articles about that. My memory on those tries of development model categorizations is a bit rusty, but instead of the Waterfall model, I think we're rather using somthing incremental or iterative. Of course, feel free to discuss this topic w.r.t. the Hurd! Regards, Thomas
signature.asc
Description: Digital signature
