On Mon, Nov 4, 2024 at 3:11 PM Klemens Morgenstern via Boost < boost@lists.boost.org> wrote:
How would you evaluate the technical merits of such a review?
I'd ask myself if the person doing the review had joined the club of Asio implementers like Niall and I. Interestingly, Niall and I had a wonderful discussion about "reimplementing Asio" (should be the name of the thread) but kind of surprisingly to me, none of the resident Asio fans actually chimed in. But back to it, the thing is, other people on the list had actually stepped in and said that, "Hey, maybe a protocol lib is a good idea and worth talking about". Thomas Fowlery above noted:
Meanwhile, the two rejections (and some non-review comments) brought up major design questions which haven't been addressed by either the review manager or the author(s). These facts make the outcome way less clear, in my opinion.
This was a bad Boost review because the author of the proposed library chose to ignore interacting with any and all "negative" discussion and the review manager did the exact same. There was no fostered discussion by the people who should've been fostering it. It mostly focused on me for dismissing a library for not hitting the level of "interesting" that I use when evaluating a library for inclusion into Boost. I think what makes Boost so great is that we all have our own ideas of what makes a library a Boost library. In this case, I think just implementing a protocol in Asio isn't interesting enough to warrant being a Boost library. Instead, I think a pure protocol library for MQTT is quite interesting and would be worthy of the Boost moniker. But what I think as an individual doesn't really matter. The point of Boost and review is to foster discussion and toss ideas around about how feasible it is to implement things. In this case, my impression is that because the library was an Asio library and because we like Asio, we'll just accept it. And that's basically exactly what happened. - Christian