Comment 4 for bug 1413592

I think the idea and implementation here look fine. In regards to the Cond. Negative Balances branch, the original report described this as "a stop-gap toward the same end if that should happen to slip past 2.8." They will definitely overlap in function, and I understand hope might be fading (or gone), but I am posting to ask that any would-be reviewers not push this in for a few more weeks. Feel free to test and sign-off, of course, but this branch is simple and targeted enough that I don't think an extended review period will be necessary, so I hope others don't see any harm in waiting a bit.

On the flip side, I see definite merit in an approach (like this) which emphasizes avoiding negative balances rather than correcting them. It would be painful given the time already invested in the other branch, but is this a path which can eventually meet all the various needs? Has anyone thought it through that far?