Re: Time to branch 1.9

Hi Greg,

I'm really happy to see you joined to the log-addressing discussion.

> That is no "decision" because it didn't occur here on the list.
Personally, I think that most of the new features should be developed
incrementally in trunk.

But the log addressing feature was already developed in a branch (by a single
developer) and I treat some kind of voting as the only way to get the
peer review for such big code.

I think we should make things clear with that policy and apply it in a
consistent way (or do not apply it at all). Note that there was an
*request* to apply this policy for a relatively small fix in the
FSFS code [1].

> As Branko notes, "voting" is not typically done in this community. There
> have been only a few in its near-15 year history. I see no reason to change
> that now

That was not mine suggestion to make a vote [2] and there was lazy consesus
on this matter. But anyway I think it was a good idea because it actually
catalyzed the through review of the code bomb committed in the r1547045.

[1] http://svn.haxx.se/dev/archive-2014-08/0198.shtml
[2] http://svn.haxx.se/dev/archive-2014-09/0081.shtml

-- 
Ivan Zhakov