On 5/14/07, Jorge G. Mare (a.k.a. Koki) <koki@xxxxxxxxxxxxxx> wrote:
But it seems that others have also tried and failed in the past, so it may well be fair to say that something in Haiku makes it difficult for non-devs to make it to the higher ranks and be productive, and that unless something is done, that is most likely not going to change.
I think you make a reasonable point above, but I don't know if real effort to address this will be done until the work is done on Haiku R1. Right now the developers (including myself) have sort of tunnel-vision in replicating BeOS R5 in the form of Haiku R1. This tunnel-vision has proved quite effective since Haiku R1 is close to completion, something many people predicted was impossible. But I think once Haiku R1 is released, the project will face a make or break point. The future direction of Haiku will need to be decided, and it won't be as simple as saying "we will replicated BeOS RX." GlassElevator generated lots of ideas, but the momentum on that died a long time ago, and I don't think all the ideas present a nice cohesive whole. So the project will need to become more organized in the form of better project management. Decisions made by the admin group or Haiku, Inc BOD will need to be more visible. Definite milestones and goals will need to be articulated. Simple "to do lists" should be provided to allow new developers the ability to jump in as quickly as possible. I also think marketing and financial management will become crucial at this point. Whether they like it or not, I think Haiku, Inc will be the best party to produce the official Haiku distribution. This can become a great form of financial support, but will require proper marketing and money management. Now the question we need to ask is, are we going to wait until Haiku R1 is about to be released to sort out all of the above? Or maybe we should start now, which I think is more or less what Koki is arguing (though I must admit your arguments can become rather long-winded at times.) So I suggest some action items: 1) The members of the CURRENT admin list and their general roles there should be published. i.e "Michael Phipps, Project Administrator. Axel Dörfler, Lead Developer." 2) On the same page as the above the current general discussions and/or decisions in the making should be shown, along with estimated decision dates. i.e. "We are currently in discussion about Haiku distributions and the use of the name Haiku in such. We hope to have a final decision by May 21st, 2007." 3) The general guidelines for how to get into the admin team should be published as well. This will vary based on the roles, for example "Developers will need to have been committing code for X months before being considered for the admin team. After applying to the admin team a vote by the current admin team will be used to decide whether the new developer will be added. Marketers will need some portfolio work to demonstrate their experience in this area. A vote will be used to decide if they should be added to the admin team." Etc. 4) The Haiku, Inc board of directors should also be published. Like the admin team, their should be some guidelines as to how someone could join the BOD. 5) Someone should be chosen to help Michael with the Haiku, Inc financial management. Simon Porter volunteered, but maybe other "applications" should be taken before deciding. 6) Some consideration should be made NOW as to the direction of the project post R1. If this work has already been done by the admin team (in private), some of it should be published. Hope this helps, Ryan