On Fri, Jan 8, 2016 at 3:31 AM, <ulrik.peder...@diamond.ac.uk> wrote:
> As a user (and small time contributor once) of log4cxx, I would vote for a > move to a central hosting on github. I don't mind what happens to the > project in terms of the apache organization as I use log4cxx as a > stand-alone library - and I guess many others use log4cxx in the same > fashion. However, putting it in a read-only repo is effectively declaring > it dead and forcing everyone who uses it to either create individual forks > or rewrite code using alternative libraries. That will be a shame for a > library that probably for most users (me certainly) Just Works :-) > In the ASF, "creating open source software for the public good" means releasing it. I am happy to sign up to mentor the project out of the incubator, but that should be a 12 week project, tops. We would need to see commits from at least three and preferably five committers. Who are these people? > The trouble on github at the moment is that there are up to 40 potential > forks on there with no central one - as the github.com/apache/log4cxx is > very out-of-date... > Forty forks means 40 prospective committers. Nothing is solved by "moving" the project to github if their changes are never moved back to the ASF. It's not enough to simply apply all of the merge requests; the ASF strongly holds that we don't do 'one man shows'. We work to attract many committers and prevent the benevolent-dictator-for-life and hit-by-a-bus conundrums. So if the project demonstrates that they can 1. recruit at least a handful of the github fork maintainers to participate at log4cxx and sign on as committer/ppmc members, and 2. assemble a 0.11 release candidate, review and vote to release it, I'll jump on to ease the transition to a full fledged PMC. If that PMC wants to move from svn to git as their primary development history, work is in progress at the ASF to facilitate that. There is a functional place to further develop the software here. But if these two things can't happen, this is past time to land in the attic for others to run with elsewhere. Either it is effectively in the attic with no code changes anyways, or these code changes aren't being released to the public, but users are being directed to use the bleed dev repo, which is against the spirit of the ASF's release early and release often policy.
