On Tue, Feb 12, 2013 at 10:54:42AM -0600, Joe Brockmeier wrote: > On Tue, Feb 12, 2013 at 05:25:11PM +0100, Sebastien Goasguen wrote: > > After 2/28 we will just fix doc bugs, not add new docs, even if a feature > > is included in 4.1 if its docs are not in by 2/28 then too bad. > > -1 to this. > > The reason for stopping feature development with a hard freeze is that > new features have the potential to disrupt the codebase badly enough > that it's a major issue. > > Adding new docs makes it a problem for translators, but it doesn't carry > the same level of disruption as new features. For example, think about > the impact if Javelin landed a week later than it did into master vs. > the impact of adding another section to the install or admin guide. It's > not really comparable. > > Also - leaving a feature out of a release means it can be picked up in a > later release. Failing to document features is, basically, a bug that > should be fixed (if possible) before a release. >
Unless that feature's lack of documentation effectively just means that it's not used (but available). > > That's a bit of tough love, but that's what will happen if we want to > > stick to the schedule. > > I'd like to stick to the schedule, but we should have more flexibility > here than with features. It's also worth noting this is the first > time through with a hard feature release and we're still finding our way > - and what we seem to be finding is that ~1 month for documentation is a > bit tricky when a lot of features land (or don't) right on the feature > deadline. I actually agree with Joe for 4.1.0. Right up until 2013-03-22, but only as a way to give our docs folks a bit more time to catch up. The more that's done earlier the better. The implication of this is that our translations will suffer (i.e., not be done). We could have a post-release translation update planned, since the published docs site doesn't have to be part of a release officially. BTW, do we have anything beyond en-US published to the website anyway? If not, we should! All that being said, I think that we need to work out a better schedule / plan for the next feature release. I think back to the discussions about the schedule, and a general desire to consider "done" for a feature as meaning "coded, tested, documented". Perhaps that has to become the criteria for next time? > > Best, > > jzb > -- > Joe Brockmeier > http://dissociatedpress.net/ > Twitter: @jzb >
