On Sun, May 3, 2009 at 2:25 AM, Richard Hirsch <[email protected]>wrote:
> I agree with Vassil in that the main issue deals with the creation of > exceptional open-source code and the different ways to build the community > to support such code. Much of the debate revolving around the ASF concerns > unspoken expectations - what should / can ASF provide emerging projects. As > Gianugo puts it > > > The actual community building is however a task for the project itself: > the > > ASF isn't Midas and won't be able to > > turn an unattractive project into sexy stuff that gathers time and > > enthusiasm from volunteers. > > > IMHO, the ASF provides the structure - based on years of experience - and > infrastructure to support such communities. I > think all open-source projects want to succeed. There is always some hidden > hope that the ASF's Midas touch will lead to a stream of new developers > contributing to this success. I think in the ASF the focus is on doing > things the "Apache" way as a means of creating this community. Although > ASF > can provide guidance based upon what has been successful in other Apache > projects, it can't be expected to do the grunt work for all its projects. > We could expect more "lessons learned" from other ASF projects coming from > the mentors but the actual application of these ideas has to come from us. > > Speaking of grunt work, we should probably be considering what to do about > the necessity of rewriting the ESME codebase as David and Erik describe. > We > can have the best wiki in the ASF but ESME is a software project and > without > a solid code base we aren't going to get very far. Dick, I'm a fan of burning down code. Code has little value compared to institutional knowledge. The problem as I see it with ESME is that the institutional knowledge is slipping away because the momentum is slipping away. The code is a lot less material to the process. The project momentum is a lot more valuable. Put another way, I think about 50% of the Lift code needs to be burned down and over the next year, I expect that will happen. I also expect that once that's happened... in a year, another 50% of the Lift code will need to be burned down. There is no code base that I've ever seen that doesn't need some substantial re-writing. Doing this while preserving backward compatibility is always a fun challenge. As opposed to you, I expect that part of the structural value of the ASF is the structures that help build community. Please see my prior post. Thanks, David > > D. > > On Sun, May 3, 2009 at 12:22 AM, Vassil Dichev <[email protected]> wrote: > > > Without trying to get into David's mind, I'd like to point out that > > David's blog post was more of a reaction to defend the Rails > > community. I must say it's possible to get the point across even > > without the unfortunate comparison with the ASF. The point is this: > > it's hard writing exceptional software. I think you both agree on one > > count: even guidance and support don't guarantee a groundbreaking > > software project. If success was easy to reproduce, someone would have > > discovered a way of generating groundbreaking software projects on a > > mass scale. > > > > Now I don't think that a software project has to be groundbreaking to > > be useful. I have no illusions that ESME is destined to be as > > groundbreaking as e.g. Rails. I still hope it has the chance to be > > useful. > > > > With that said, I hope that any heated arguments originating from the > > Rails scandal are over soon, because there are probably no two people > > who agree on which software is useful or groundbreaking. And the time > > and effort spent in a discussion like this could be spent creating > > software. > > > > Vassil > > > -- Lift, the simply functional web framework http://liftweb.net Beginning Scala http://www.apress.com/book/view/1430219890 Follow me: http://twitter.com/dpp Git some: http://github.com/dpp
