Prepare for 0.9.0 release

[travisbrown]

It's been a little over two months since 0.8.0, and while there haven't been many big changes since then, I think it's probably time for a new release, and at least three other people agree. 😄

There's nothing more in particular that I personally want to see in 0.9 besides #1506 (flatMap syntax for Either on 2.10 / 2.11 that's consistent with Either#flatMap in 2.12), but if other people have other things they'd like to wait for, we can expand this list:

• Finish up Partially revert 4530d3aa93131ec28096968a3c903ed1016dbf1b. Add back v… #1506 (flatMap for Either, etc.) and get it reviewed and merged.
• Merge Reduce stack depth in StateT #1466 (miscellaneous StateT stuff).
• Merge Add Reducible for Eval and Id #1475 (Reducible instances).
• Merge Update to Scala.js 0.6.14 #1509 (Scala.js update).
• Merge Add biased operators to Ior #1500 (new operators for Ior).
• Merge Add FunctionK and #1480 (FunctionK#and).
• Merge Add NonEmptyList.reverse #1455 (NonEmptyList#reverse).
• Review Make catchNonFatal on Future async #1510 (async catchNonFatal for Future).
• One more review for Add Foldable.foldMapM and Reducible.reduceMapM #1452 (Foldable#foldMapM).
• More eyes on Diverging Inject instances #1505.
• Review release notes and update before release (Draft of 0.9.0 release notes #1511).

I can write up the release notes on Monday (the 2nd), but if someone beats me to it I'd be happy to review instead.

[johnynek]

I want #1466 and #1475

[travisbrown]

@johnynek Added!

[notxcain]

I don't know if #1505 is fixable at all?

[johnynek]

I'd like to add #1500 too. If someone can give it one more +1.

[johnynek]

would like #1510 in too.

[johnynek]

I think #1480 is ready too.

By the way, I would mention we don't seem to be very proactive on reviewing and merging code. :/

[johnynek]

#1455 looks good too.

[johnynek]

#1452 also

[travisbrown]

@johnynek Updated again.

Maybe we should make it a matter of policy that if there's no obvious controversy in the PR the second maintainer to 👍 should be responsible for merging immediately? I know that personally if I review something with multiple 👍s that's been open for a while, I tend to assume that there's a reason it's not merged, and I don't merge it (which is of course self-perpetuating 😦).

[johnynek]

yeah, I'm +1 on that. revert is there if we want.

The bigger issues are the ones where there is 1 +1, but no other comments. Like no one else has looked and it should be simple in many cases.

[travisbrown]

The flip-side obligation for PR openers will need to be that if there's been related conversation on Gitter it must be linked in the PR. Again, I haven't personally been doing a good job of keeping up with Gitter recently, so if I see something with no comments I tend to assume I missed something there.

[travisbrown]

@notxcain I just took a look at #1505 and don't really have any idea how to help. I think we should spend some more time trying to fix it, but not before 0.9.0. Does that sound reasonable?

[philwills]

Is this still likely to be released soon? I've been planning to do a release of a downstream project, but have held off on the assumption that this was imminent.

[travisbrown]

@philwills Same here. I think we just need to get someone to merge #1511 and someone else with publishing rights to pull the trigger.

[adelbertc]

We should also be sure to update the README and website with the new version after.