Wow, that is about the worst example you could come up with. You know as well as I do that rpteilla couldn't explain that feature good or bad if his life depended on it.

We have added patches from BBR when it has made sense to do so. On multiple occasions, and will do so in the future. Some user-visible features from BBR will likely be adopted in the future. Though probably not that one because we don't buy into the tiny constant factor benefit as important enough to give up verifabilty, and also, as i've explained before, we know how to do much more effective pruning (for when it matters, if that ever happens).


Wrong again, and again, if you want to discuss something you are certainly ignorant about, you should ask questions instead of assuming you know something you are actually entirely ignorant about.

What actually happened is that I suggested we consider adopting that feature. Tacotime was strongly opposed to it because of the change to the trust model. We hashed things over quite a bit, I considered his points, as well as the extremely limited benefit (as you now acknowledge), and decided that he was right and I was wrong. No one was unwilling to admit that BBR did something innovative. We considered it, evaluated the technical merits (and also the effort to actually merge the code and adequately test, which is somewhat non-trivial), and decided against it.


You obviously have no real experience in this area. For one thing, they don't say that because they don't understand or pay attention to design issues closely enough to do so, if at all, and for another, if they did, we would still dismiss such comment from "investors" as not interesting. By contrast, we do pay attention when you or Zoidberg or other knowledgable people offer technical feedback. See how this works?