It's a pretty unusual situation - at least, I can't think of another one where someone who has made such a huge impact in terms of code that probably the majority of us use daily (oh hey, mocha, express..) has decided it's time to move on to focussing on another language. Express is one of the libraries that almost anyone who starts playing with node touches at first, no? Has something like this happened in other parts of the coding world?
He made some really great points about the ecosystem - it can definitely be hard to assess the relative merits of various libraries, without really digging into the code, or playing with something for a while and seeing how it does. is that laziness? maybe. looking at the npm downloads graph can be useful, but those numbers can be a bit artificial (Forrest, if you're here, mentioned on twitter that they include numbers for people replicating to their own npms, for example). I can imagine npm have a ton of ideas about how to improve that experience of discovering what modules are available ;)
It was interesting that he made the comparison between the standard library in node and go too - maybe with native promises and generators more or less landing in core (please correct if wrong) then things like the profusion of async libraries might start to die back a bit, and a canonical approach emerge. Not sure. It could be great if you knew that when you were dealing with an async operation, then the return was always a promise, for example, or you had that option. I know you can wrap almost anything in a promise and go crazy, but its not quite the same thing.
I wonder if the NPM people could produce some stats about relative popularity of repos that he's had a hand in? Maybe try and figure out if there is some scope for a GH organisation to collect this stuff, and then the volunteers for maintaining them could work together under that umbrella for things like code reviews, merging PRs etc.
If he's reading this - really huge thank you for all the work, and best for the future!