Closed kevin940726 closed 4 years ago
Oops... sorry... accidentally created.
Can you (or someone else) edit the title of this and #2861 to something shorter so that it doesn’t blow up the size of the GitHub issue number completion window?
(That, or delete it outright.)
@estebank I would say discoverability, guessability, and -- maybe the biggest issue to me -- every one-more-thing you have to explain risks being the one thing that blows a newcomer's cognitive budget. Don't get me wrong, overall I've been pleasantly surprised at how quickly my colleagues have gone from zero to productive users of Rust. But there are still so many little surprises in the onboarding experience, that it's easy for a learner to just blow their stack and give up. (I think I'm just basically giving an exegesis for why I'm so excited about all the effort Rust is putting into ergonomics, at least the learning side of it.)
@SimonSapin My only point about JavaScript was that it shows us that it does in fact work in practice, at scale, to make de facto compatible changes. JavaScript of course does not tell us what we should do. From my perspective, doing it with language editions means we wait several years longer to solve a pain point simply because we have a mechanism that lets us do so, but that's obviously biased by my experiences I mentioned. Can you help me understand the rationale for why you wouldn't want to make the change now? Are you concerned about de facto risks that we haven't actually uncovered in our empirical investigations? Or is it an aesthetic about clarity of policy?
Originally posted by @dherman in https://github.com/rust-lang/rfcs/pull/2544#issuecomment-422193876