Good article, and I agree with the author's enthusiasm. However his points that "cheap parts are a great enabler" and "shipping costs are high to Trinidad and

Good article, and I agree with the author's enthusiasm. However his points that "cheap parts are a great enabler" and "shipping costs are high to Trinidad and Tobago", while true, don't actually counter the OP's points that "feature determination needs to be universal" and "the instruction set is poorly designed". Others have said application authors will code to the specific platform, but that's not true of library authors -- they need portability. Why can the RISC-V team not get basic stuff like feature determination right? No matter how stupid the rest of the instruction set is, I'm with OP that the responsible individuals should retire from the committee.