I like his idea, but it sounds a little hypocritical coming from someone who advocated a change that broke many distros builds of the Python cryptography packag

I like his idea, but it sounds a little hypocritical coming from someone who advocated a change that broke many distros builds of the Python cryptography package. Personally, I think he spent his "backwards-incompatibility budget" for about 5 years with that change because that's probably how long it will take for the new version of cryptography to be widely adopted and supported long-term.