Determine this repo's direction and policy for 2019

[cgruber]

There are a few teams which use MOE, and which need fixes, outside of Google. Some teams inside google use MOE, others have migrated to Copybara (github.com/google/copybara). Officially, Copybara is the successor to MOE, but has a different set of use-cases and assumptions.

We should decide whether this open-source repo is going to be updated, and accept patches, or whether it is officially sun-setted, as its earlier maintainer (@cgruber) is willing to continue to patch and update it, but is not at Google, and needs either to spin up a fork or to be re-empowered to make merges on this repository.

[cgruber]

cf. @ngzhian @ronshapiro

[ronshapiro]

I think we can continue to keep this in maintenance mode and accept patches as necessary. I don't think we necessarily need to sun-set as there are still users of MOE that aren't ready to use Copybara as you say. This has been relatively stable and when patches come in (which is infrequent) we can figure out how to address them. As for making official releases, I'm happy to do that as well as it should be relatively easy.

So in sum: keep alive, keep changes minimal where possible but still fix bugs and add important features. Encourage Copybara or other tools unless there's a good reason to desire otherwise.

[cgruber]

Not sure how I missed this, but it makes sense. I may evolve my fork as a personal hobby, but I can see maintenance mode as being reasonable here.