On 2016/06/08 22:41, Daniel Jakots wrote: > On Wed, 8 Jun 2016 22:21:20 +0200, frantisek holop <[email protected]> > wrote: > > > Daniel Jakots, 08 Jun 2016 21:01: > > > Previously, last stable flask release was published on 14 Jun 2013 > > > and then 10 days ago they finally released a new one. Yesterday they > > > released a bug fix release, here's a diff to update to this one. > > > > it is nice to see a port update so fast, but flask > > being "stable" for so long, i think we could wait a bit > > until the "dust settles a bit". a wholy new test > > framework and other questionable changes (like click, > > the stupid theme error), it will be interesting to see > > how the flask community will respond on the longer > > run... i know we won't be upgrading in production > > for a while. > > Well, I think the best for the ports tree is to have up-to-date > software, not to produce artificial latency between upstream and us > whether someone (btw, who decides?) like or not what they did.
For a port with a maintainer, this sort of thing is usually up to them, somebody familiar with the particular software is likely to have a better idea of whether it's a good idea to update quickly, or hold off for a while. For example, for many ports it's often a good idea to wait a release or two after a release that has major changes, to wait for upstream to unbreak things ;) > One solution would be to have one newer release (0.11.x) and an older > release, as they planned to release a bugfix release (0.10.2), like we > do for django but I don't know if we want to do that. This sometimes makes sense where it's a single port that is involved, but where you have a chain of depencies involved it can get rather complicated..
