That said, I would like to see us keep the door open for Android dev too. It keeps coming up and people keep trying to use jclouds on Android so I think it’s worthwhile to try it, if someone is willing to run with it. But if nothing significant happens with it in 6 months or so, I think we would want to reconsider that and just forge ahead with a newer version of Java. I can’t argue with that. If we keep this opportunity open and encourage the exploration, and still nobody steps up for it then I think it is fair to reevaluate this again.
Chris -- Chris Custine On October 16, 2014 at 4:13:21 PM, Everett Toews (everett.to...@rackspace.com) wrote: On Oct 12, 2014, at 6:30 PM, Chris Custine <chris.cust...@gmail.com> wrote: > * On promoting labs providers in minor releases: One of my main concerns > about labs providers is that they frequently get ignored when doing > refactoring in the upstream jclouds repo and are left broken for periods of > time until someone has time to bring them in sync with upstream jclouds. The > chances of this happening are probably much higher while working on a major > release like 2.0, so I feel that if the provider is ready to promote, why > delay until a major release when you can bring it in sooner and give it more > visibility for the inevitable refactorings of a major release? +1 to promoting providers whenever they’re ready regardless of major/minor release to avoid refactoring pain. > * Java 7 vs Java 6 API level: I had not considered the implications moving to > Java 7 API level would have on any potential Android development. I don’t do > any Android development but I understand the try-with-resources issue and I > think it is fair to reconsider and revise some of the earlier decisions in > order to keep this door open. I don’t think this is a major issue, we can > still use most Java 7 language features with specific exceptions such as > try-with-resources, and these issues are fairly clearly documented and well > known from the brief googling that I did. I hadn’t considered the Android aspect either. I dipped my toe into Android development a while back (I even attended AnDevCon) and learned that serious Android developers are very cognizant of the performance and memory footprint of the dependencies they bring into their apps. Both of those things would need to be addressed with jclouds on Android. That said, I would like to see us keep the door open for Android dev too. It keeps coming up and people keep trying to use jclouds on Android so I think it’s worthwhile to try it, if someone is willing to run with it. But if nothing significant happens with it in 6 months or so, I think we would want to reconsider that and just forge ahead with a newer version of Java. Everett
