On Wed, Aug 12, 2020 at 12:07 PM Gregory Nutt <spudan...@gmail.com> wrote:
> > How/who would prioritize and > schedule implementation of roadmap items? How could accomplish any of > this with only a volunteer organization and no project management with > any authority? Open source tends to be driven by people scratching their own itches, so I think we should try to flow with that rather than against it. The roadmap will likely begin as a jumble of everyone's wishlist items, and things would gradually get prioritized based on how popular and urgent the items are to those who do the work. But who will do the work? In the spirit of open source, whoever needs/wants it to get done, whether they're already committers or people new to the project, whether volunteer or paid. Importantly, if something gets implemented ahead of schedule (because it's urgent to someone), that's okay. Things shouldn't be delayed because the roadmap says so. The roadmap should be a guideline, not a hard rule. Unless there are technical prerequisites where one feature must be implemented before another becomes possible. Regarding "no project management with any authority," although the PMC doesn't have authority to order people to work on certain things at a certain time, the way a company's management does, it does have authority, through discussions and voting, to decide what kinds of changes fit the project's needs. So if, for example, someone wants to merge a change that cranks up our memory usage to the stratosphere, we can say sorry, it's not a good fit. And the road map, in conjunction with the INVIOLABLES, can be a good way to articulate what is likely a good fit and what isn't. Just my 2¢ Nathan
