On Thu, Aug 9, 2012 at 4:21 PM, Ewan Mellor <ewan.mel...@eu.citrix.com> wrote: >> -----Original Message----- >> From: David Nalley [mailto:da...@gnsa.us] >> Sent: Wednesday, August 08, 2012 6:17 PM >> To: cloudstack-dev@incubator.apache.org >> Subject: Re: [DISCUSS] Binaries (jars) in our source tree/source >> releases. >> >> On Wed, Aug 8, 2012 at 9:07 PM, Brett Porter <br...@apache.org> wrote: >> > >> > On 09/08/2012, at 3:01 AM, David Nalley <da...@gnsa.us> wrote: >> > >> >> On Tue, Aug 7, 2012 at 11:32 PM, Brett Porter <br...@apache.org> >> wrote: >> >>> If this is a direction CloudStack decides it'd like to go, I'm more >> than happy to help :) >> >>> >> >>> - Brett >> >> >> >> How much help are you willing to provide? :) >> >> I am not sure we have a lot of maven expertise around, but maybe >> this >> >> provides an even cleaner way to get rid of waf. >> > >> > >> > As you'd note from my being behind responding to several threads, I >> don't have a lot of bandwidth :) I can help get it started and provide >> assistance, but someone else would need to be driving it - and as >> others have noted, someone actively developing the project should >> understand how it works going forward anyway. >> > >> > I wasn't thinking this was something you planned to do before the 4.0 >> release. Is the thought that this might save some of the time on >> release-related activities, or is it an incremental improvement to >> consider beyond that? >> > >> >> I think that folks are thinking after the initial release at this >> point. At least I am. > > I was thinking the opposite. Didn't this thread start because the current > system isn't fit for purpose? I don't want 4.0 going out with a whole bunch > of new complex dependencies, and have users unable to resolve them. That > would be a backward step from what they have today. > > Ewan. >
The thread started because the current system isn't really a system, it's just a folder full of binary jars which we were advised against (and saw another incubator project taken to task by the IPMC). As I said elsewhere, I am not opposed to do something else (Maven, Ivy, Gradle, $somethingelse), but who is going to pick it up, get it running, and educate the rest of us between now and a 4.0 release? The proposed intermediate solution is admittedly inelegant, ugly and little better than a shell script, but I doubt there is anyone here who doesn't understand it, and it keeps the build system that does at least build. I personally wouldn't mind picking this up after 4.0 (and learning something new in the process), but just have no cycles at the moment to do so. I am happy for someone else to do it, we just haven't had anyone jumping to own the task. --David
