On Thu, Nov 6, 2008 at 2:14 PM, Simon Laws <[EMAIL PROTECTED]>wrote: > I want to start a vote(s) to help clarify how we move forward. I want to > separate out some of the issues that have been raised into sparate votes. > > 1) Create two development streams* > One (in a branch with a 1.x-SNAPSHOT label) to support our OSOA code > base** > One (in trunk with a 2.x-SNAPSHOT label) to build support for the OASIS > versions of the SCA specs*** > > 2) If we agree with 1 then decide how we go about developing the OASIS code > base? **** > > > > * We should take this opportunity to take a step back and discuss what we > would like Tuscany to be going forward. Let's do this on a separate thread. > I recognize that there is an element of cart before the horse here but I'm > taking the liberty of proposing that future support for the OASIS level of > the SCA specifications (which we hope will become standards) is such an > obvious watershed that we should head in the OASIS direction and use it as > the catalyst that allows us to address any wider Tuscany strategy issues > (TBD). > > ** Our 1.x OSOA version to be maintained to support our 1.x users and be > extended where the need arises (hopefully by merging compatible changes from > the OASIS code base if possible). There is a danger here that we create > spearate 1.x and 2.x groups of developers which is not the intention. > > *** Developing OASIS function does NOT necessarily imply that OASIS > compatibility will be developed at the expense of OSOA compatibility. The > two may be able to coexist. > > **** A number of ideas have been raised, for example, > Start with what we have in trunk > Start from an empty trunk and build up from modules in trunk and from > the equinox branch. > I don't feel confident enough to formulate the vote for this part yet, > particularly w.r.t timing. Let's discuss this on a separate thread. > > > > Comments are welcome. If no one suggests this is completely mad I'll start > a vote thread for point 1 tomorrow. I'd like us to agree to the way forward > in consumable chunks. I fear will will go round in circles otherwise. If > someone wants to start talking strategy then please go ahead otherwise I'll > start a thread shortly. > > Simon >
Good, sounds like a fine approach to me. ...ant
