On 11/5/14, 1:32 PM, "Justin Mclean" <jus...@classsoftware.com> wrote:
>Hi, > >I'll also point out that what is currently being looked at it identical >to RC0 that I called a vote on a week ago, so there would be no need to >cancel that vote and call another one. I think the MD5s probably changed. SWFs built at different times from the same source have different MD5s, unfortunately. But yes, this one took a while because we’re working out the kinks in the new process. Some release was going to be subjected to that. Plus, we now have nightly builds of TDF thanks to Erik. > >There's only been one change made to the release branch which was an >addition of a MD5 check by Alex and IMO probably should be been checked >into develop not the release branch but not really an issue either way. I wanted to get the MD5’s published so the approval scripts can work and more people might test. I reverted from develop and re-checked into the release branch. I’m getting set up to run the approval script now. > >I do see this as a bigger issue on more complex releases if we go down >the "noRC" path, releases only require 3 +1 and more +1 than -1, in a >discussion it can be hard to know if someone considers something a >minor/major or blocker issues unless they state what their voting >intention would be. Without RCs/vote threads the discussion could >conceivably go on for a long time without any clear consensus if a vote >should even be called or not, prolonging the voting process and costing >more time for everyone involved. Folks who find blockers should declare them as such in the DISCUSS thread. -Alex
