Re: [HACKERS] A small problem with the new inet and cidr types

> Well, maybe not, but don't beat yourselves up about it.  An awful lot 
> of small but good things have happened in the last month in the way of 
> bug fixes and configure improvements.  Not to mention docs.  If you'd
> released 6.4 a month ago, it would've been of substantially lower
> quality than what will go out the door this week.  I don't think the
> delay was such a bad thing.

That's been pointed out :)

But every release we try to learn something new (or are forced to
relearn something old). In this case, imho we delayed focusing the
developer's group on release-specific issues while waiting for these
additions, and could have/should have covered the same territory
starting Sept 1 rather than Oct 15. otoh, you did a lot of these fixes
and improvements over the last month, and only you can guess if you
would have been prepared/willing/in the right frame of mind to do the
same in September. Also, I got some great docs contributions since Sept
15 (about when I had planned on freezing the docs), and those things
would have had to wait for v6.5, at least for hardcopy versions.

So, the only issue really is trying to be a bit more decisive about
which features will go into a release, feature-freezing when we had
planned to (more or less), and then doing the right thing for testing
and packaging once we have frozen.

And then move on to the next release cycle with clear heads and great
plans ;)
                    - Tom