> > "We who are not Eclipse fanatics should work more together and not dilute > our > > resources." > +1, beside that, most people know each other so we are already working > alongside each other, regardless of the community we are committing > most to.
Exactly :-) > > My main concern revolves around the community model, which is dictated by > the > > Apache Software Foundation to be a meritocracy for Felix, and a "No > Barrier" > > approach here. Now, I think it was Richard who suggested that perhaps Pax > > should be the sandbox of Felix, for quick and open collaboration outside > the > > current committership. In principle I think this is Ok, although this adds > > some paperwork overhead (called Software Grant) when importing the > codebase > > to Felix. My interpretation of this would then mean that codebases not > > directly relevant to the specification suites, present or future > Mmh if we add the paperwork to be "upstream compliant" with Apache, > would that even add some at least theoretical benefit when it comes to > legal help etc, strengthening the position of the copyright holders > even before the code enters the ASF, if it ever does? The most important thing is that a clear IP trail is kept i.e., it must be clear who contributed to the codebase and what. I think that is already happening no? > > The other suggestion (which I think was from Karl), was that Pax would > > re-brand selected Felix stuff, which is confirmed inter-operable on other > > platforms. > Would be great since there is a number of stuff that could benefit > from greater even occasional care :) Thats for sure. I think it would be great for both communities if that would lead to less duplication of effort. > > Apache Felix is a community strongly committed to the OSGi specification > > suites and intend to build fully compliant implementations of these > > specifications, current and future. > +1 > > > > > > The Pax community is strongly committed to OSGi framework independency, > > interoperability and open participation. > +1 > > > Now, to get more concrete, I would like to propose the following action > plan; > > > > 1) Pax Logging and Pax Web codebases are moved to Apache Felix and > becomes the > > Felix implementation of these Compendium Specs. The primary developers > of > > these will become (if not already is) Felix committers. > > > > 2) Pax Logging and Pax Web remains "Pax" branded, and will continue to be > > released out of the Pax project, possibly not in sync with the releases > > from the Felix project, as ASF release rules are more rigid and hence > > slow. > > > > 3) Pax Web Extender & Co stays in the Pax project, at least for now. If > there > > is specs heading in that direction, we can bring this up again. > > > > 4) An open invitation to all Felix and other OSGi developers to join the > Pax > > project at OPS4J. OPS4J is a "No Barrier" community, what we call "Wiki > > brought to Coding". OSGi stuff that are not related to either the > current > > specification or ambitions to become specifications are probably better > > served at Pax. > > > > 5) Pax will continue to encourage experimenation, and people interested in > > OSGi will do themselves a favour of doing the experiments at Pax, as > > Felix PMC will consider Pax community members for committer status at > > Felix. > > > > 6) Felix "configadmin" and "fileinstall" are "imported" into the Pax > ConfMan > > suite. Work will start to ensure the interop and full spec compliance. > > Others may follow as people have itches. > > > > 7) Felix and Pax will cross-reference each other on their websites. > Sounds like a good distribution of work and responsibility, adding > more traction to both communities without hindering the participation > from other sources and thus optimizing the overall outcome for OSGi > and the developers involved. +1 > > Well, this is a proposal, mostly to the Pax community (Felix community is > > CC'ed) and not written in stone. What do you all think? > > IMHO, some projects at Felix should probably be moved over to Pax, to > > encourage more participation from others. I leave it as an encouragement > to > > those who work on such candidates to bring this up themselves. > > I think that is a great starter from both sides. If this gets > successful, there might even be pattern there in aligning early > ideas, wild experiments etc with OPS as a form to foster them and then > take "serious" development into a bit more rigid model like the ASF. I > am not sure this is an accurate description but let's see what effects > we can see of trying closer collaboration between Pax and Felix! Sounds great! regards, Karl > /peter > -- > GTalk: neubauer.peter > Skype peter.neubauer > ICQ 18762544 > GTalk neubauer.peter > Phone +46704 106975 > LinkedIn http://www.linkedin.com/in/neubauer > > http://www.neo4j.org - New Energy for Data - the Netbase. > http://www.ops4j.org - New Energy for OSS Communities - Open > Participation Software. > http://www.qi4j.org - New Energy for Java - Domain Driven Development. > -- Karl Pauls [EMAIL PROTECTED]
