I think of, for example, "promise cancellation", which was a long long long running saga of topics discussed in the old es-discuss mailing list. Today there is AbortSignal and AbortController, but these are actually specs from the DOM, even though they don't really depend on any other DOM technology & even though they are widely used as "the" way to do promise cancellation. Trying to evolve the language was just too hard. Also a lot of those ideas from back then were kind of not as good- we learned a lot & AbortController kind of works decently, with what we've got (that said, AbortSignal is core proof that promises being async only was kind of bone headed- we immediately had to go invent an async+sync alternative, someplace a couple miles away, to make promise cancellation something doable in a reasonable fashion! (yet that "Signal" sync+async tech remains a single-use one-off!! eek!!)).
So we have a wider distributed set of folks doing "standards" in different places. Almost nothing gets folded back in to the langauge. Instead of having processes & eventual growth of the language, taking the things that have in fact won & bringing them in, there's a very rag-tag acceptance. Governance is split up across a couple major entities. Changes percolate at varying rates to downstream specifications. WinterCG is interesting as like, the first people to get together to say, here's what things from the sea-of-standards we should agree on. It's the first union. Which has power to embrace, but still, has little power right now to shape & influence the upstream standards, from a wide body of places, that it is consuming/blessing. Wild stuff.
