You might say "Let's just give it a try anyway, it's ready to go, just drop all objection to it." I'd like to point out that once a feature is in, it's harder to remove.

and

The goal is rational: give more incentive potential to comments to increase user participation, engagement, and retention.

I'm not sure if separating out a comment pool will indeed support intended goal. However, my opinion is, we shall test and see the results.

I understand the fact when some feature is implemented in the code, it may not be that easy to get it disabled again. But I still think we should give it a try and give x weeks, or y months to determine the effectes and results. Would it not be possible to agree with Steemit Inc a clear "what to do" plan with clear criteria? When the real purpose of this feature is not resulting in the wanted effects, Steemit Inc commits to remove the separation of the comment pool from the reward pool again in z days/weeks/months?

I would also add the ability for a configurable split, ie the 38/63 split can be adjusted by just a configurable parameter to eg 20/80, 30/70. With such ability, a series of test can be executed in sequence trying different settings/percentages of split. This may give all of us a much better view how to progress, then just based on theories define our way forward.

In the end, we need to get the basics right before Steemit is ready to grow, and with that Steem. It will benefit all of us. To be able to create a great solid foundation to Steemit and Steem, we really have to test a lot of different ideas. IMHO, the split of the reward pool is one of them.