OK, understood. I think it's a psychological important step to release the code, even when its half-baked and early. If necessary, give it an alpha label. This is enough warning for most devs.
I don't understand too much on the current codebase, so I can't judge what is the best first step. Separating between Commons/Client sounds horrible awful task. Can you give an overview of what would be the actual steps here? I imagine it's critical, but is it possible to have multiple hands working on it? Thanks! On Tue, Mar 31, 2015, at 11:55, Yuri Z wrote: > Build system - Actually things like GWT compilation and protobuff > generation make it harder to move to another build system. Also, > restructuring wave project into more standard form that maven likes to > work > with is imho hard. I think first we should just split Wiab into separated > projects - Client/Common/Server or Client/ Common + Server and preferably > with more elastic build system than maven -than can be more easily made > to > work with current folder structure. > Release - imho what stops us is just bad feeling about releasing half > baked > project, technically we can create new RC candidate today and vote. From > what I know all legal issues are already solved. > > On Tue, Mar 31, 2015 at 12:43 PM Christian Grobmeier > <grobme...@apache.org> > wrote: > > > Hi, > > > > from the huge thread "Roadmap" I took a way the following potential > > action items: > > > > - Client/Server split, maybe common > > - Setup testing > > - Setup CI > > - Replace GXP templating system > > - Improve buildsystem with using Maven, Gradle or SBT > > - Replace custom configuration framework with something more conveniant > > - Rewrite of the concurrency handling. > > - Make Wave work with modern browsers > > - Shortcuts are broken > > - Inbox search is needs improvments > > - GWT superdev mode is broken > > - Upgrade to GWT > > - GWT-tests broken > > - Gadgets broken > > - Bonus: understand the changes from wiab.pro and discuss if they > > should be upstreamed > > > > It seems we have some infra, some architectural and some frontend > > issues. > > From the Frontend it seems we have no conclusion if GWT is the right > > track or not. > > > > What I understood is that most people agree that we need to have some > > kind of Client/Server/Common split. > > > > We have some issues in the backlog also, which makes refactoring hard > > like build system, broken tests or CI not being setup. > > > > This is quite a bunch of work. > > > > I have two questions: > > > > 1. why did we went stuck with the first release? Sorry Ali to ask > > again, I forgot and couldn't find it in my mails quickly. Could we > > solve this right now and get it done? > > > > 2. What would be the next, small step? In my opinion it actually should > > be the build system. Once this is done, it can be explained how "easy" > > it is to setup Wave, which might help to attract approaching devs. > > > > Both items are small enough to be resolved with in April. Volunteers > > should learn how to make releases at Apache, also learn and improve the > > build process. I think this is a great step for new volunteers. > > > > Thoughts? > > > > Christian > > > > > > > > ----- Original message ----- > > From: Christian Grobmeier <grobme...@apache.org> > > To: "wave-dev" <wave-dev@incubator.apache.org> > > Subject: Roadmap > > Date: Tue, 24 Mar 2015 12:12:32 +0100 > > > > we have volunteers for the next months. Why not discussing what we > > should do? > > > > My first preference would be: craft a release. > > I forgot what was missing back then, but it would be great to find out > > from the mail archives and create jira issues for the open things. Maybe > > Ali could help here, as he was the RM for the last try. > > > > The next thing would maybe be more technical. Can you throw in some > > concrete ideas what could be achieved in small steps? > > I guess "refactoring everything" is not a good start :) > > > > Regards, > > > > Christian > >
