I already establised a working solution for better dependency management based on ant+ivy. Resulting in a reduction of zip size to 1/5 of the checkout at that time (35 MBs). And it seems with less effort/less complexity than is now is being shown in the OFBIZ-6172 branch...
I suggested a dev branch back then ( https://issues.apache.org/jira/browse/OFBIZ-5464) so that others could evaluate. Unfortunately it didn't gather momentum at the time. Does that mean that it is a worse fit? I dare say: not! Pierre Smits *ORRTIZ.COM <http://www.orrtiz.com>* Services & Solutions for Cloud- Based Manufacturing, Professional Services and Retail & Trade http://www.orrtiz.com On Wed, Apr 22, 2015 at 10:32 AM, Ron Wheeler < [email protected]> wrote: > Perhaps it would be a good idea for some of the key people to take a close > look at what has been done. > > This is potentially a big step forward in modernizing the product. > > Having a working solution takes a lot of the FUD out of the discussion and > allows the approach to be tested by the people who are building OFBiz every > day. > > Even if it actually does everything that Adam claims and the consensus of > the committers is to move to Maven, it will still be a good idea to support > the 2 build methods until everyone important is ready to commit to Maven. > It may take a while to get the Maven approach sold to everyone even if they > know that at some point they will be forced to move. Some will be early > adopters and some will be late but if you don't have to force everyone to > move at once, it does make the transition easier. > > If it is the consensus that the Ant build is still better, the Maven stuff > is easy to remove without damaging the Ant build. > > I suggest leaving it in until everyone who needs to test it before the > decision is made, has a chance to test it. > It is unreasonable to expect each of the committers to make their own > Maven build to test the idea. > > Adam has saved us a lot of speculation about what it means to move to > Maven. > > Give the supporters and skeptics some time to test before removing it. > > Ron > > > On 22/04/2015 2:52 AM, Jacopo Cappellato wrote: > >> On Apr 21, 2015, at 10:37 PM, Adam Heath <[email protected]> wrote: >> >> My commit is not breaking anything. Why remove something that is >>> harmless? >>> >> Hi Adam, >> >> The fact that a commit is harmless is not enough for its approval. >> I know that your commit doesn't cause any side effects and I appreciate >> that you are now doing your work in a feature branch. >> I am asking you to revert that commit to trunk not because its quality is >> bad or I see potential issues but only because the decision about the >> official build tool for the project must be taken by the community and we >> are not planning to maintain more than one alternative options in the >> official repository. >> Just to make it super clear, I restate my request: please revert 1674216 >> (it is the only commit to trunk) then let's continue the work about Maven >> in the release branch you have created. >> In the meantime the discussion about "ant vs ant+ivy vs maven vs gradle >> vs ..." will go on and its outcome will determine the final decision; since >> there are clearly different points of view for the different tools we all >> have to be open to consider other's opinions: crystallized positions will >> not help much in this context. >> The branch you have created is valuable because it provides a reference >> implementation for the discussion, but it is important that you appreciate >> that it may not be merged into the project (based on the outcome of the >> ongoing discussion). >> >> Regards, >> >> Jacopo >> > > > -- > Ron Wheeler > President > Artifact Software Inc > email: [email protected] > skype: ronaldmwheeler > phone: 866-970-2435, ext 102 > >
