I liken it more to Evan investing "hammock time" (https://www.youtube.com/watch?v=f84n5oFoZBc https://www.youtube.com/watch?v=f84n5oFoZBc) into Elm because he h

I liken it more to Evan investing "hammock time" (https://www.youtube.com/watch?v=f84n5oFoZBc https://www.youtube.com/watch?v=f84n5oFoZBc) into Elm because he has loftier goals than just another SPA abstraction. Which is a bit different than perfection. There are just too many cats that can be released from the bag that would be too hard to reverse once out in the ecosystem. Classic example would be allowing native modules into http://package.elm-lang.org/ http://package.elm-lang.org/. This puts you in a position of treading cautiously with the long-term in mind while your users may clamor for quick fixes that jeopardize the overall strategy. So I agree with you. This results in the "something good vs nothing at all" release cycle that you point out, but I think its a more accurate explanation of why that is. The rest of the issues seem to be related to his limited bandwidth. And I suspect Elm is tied so much to his bandwidth because such fundamental design is still being decided.