On 3/17/17 2:34 AM, Andrzej Krzemienski via Boost wrote:
Hi Everyone, In the context of the discussion around review process improvements, I wanted to report another problem that I do not know how to solve. But maybe you do.
Extending the review period by allowing the review on the BLIncubator looks like a good idea, but it also comes with a problem. The review applies to a particular commit (if we use GItHub terms), that is, to the state of the library on a particular branch at a particular point in time. If I see a review on BLIncubator (not that I see many of them), how do I know what commit it applies to and if it is even irrelevant.
It might be partially relevant, but it puts me in an uncomfortable situation, that I will be investing my time in something that is likely to turn out to be useless.
Also, I might consider submitting a review into BLIncubator, but what chances I have that it will be considered in the formal review, given that the formal review may work on a different commit: possibly on a quite different library by now?
This is really a criticism of the blincubator and it's the way that it handles reviews. It's a valid issue that did not occur to me when I made it. Since the blincubator hasn't gathered many reviews as I had hoped it would, it hasn't been a big issue. The original idea/hope that people would review submissions independently of the boost formal review process/timeframe hasn't really taken off. Until that starts to happen it's kind of a moot point. Robert Ramey