I'm curious (and please answer with as much nuance as you can muster): How would you feel if decorators in JS could never be standardized because classes with decorators would have an unavoidable performance penalty relative to the same code written by hand?
-
-
To me, the other ordering fits the spirit of EWM better. Let the ecosystem experiment with decorators (with no performance loss compared to status quo). Then we can spec and optimize common decorators that profiling in the wild tells us would benefit from optimization.
- 3 more replies
New conversation -
-
-
I don't think that's necessarily agreement. Decorators falling into a few patterns doesn't mean that they're all exactly the same. My point is that common patterns aren't free-for-all and I'd like to hear if they specifically are optimization-killers vs anything decorators can do
-
It works for @bind. There are probably some users who think that's the 80/20 case and who don't care about the other use-cases. They'd probably be happy

- 1 more reply
New conversation -
-
-
I don't think this perspective is very close to any of the folks who are strongly interested in this feature, but the details matter. I'd encourage us to avoid coalescing on a bumper-sticker slogan that makes us feel better about progress. We need to roll up sleeves together.
-
Agree that details matter. I think Mathias's idea might be slightly too narrow, but that there may be some point in the middle that is just as static without requiring changing the way people use decorators.
- 3 more replies
New conversation -
-
-
It's important to be able to compose and partially apply built-in decorators, and export the result, to match the ergonomics of existing use cases. I think we can do this in a way which remains statically analyzable by tools and JS engines.
-
Programming language researchers have long thought about where composing functions can take us. See John Backus's Turing award lecture https://www.thocp.net/biographies/papers/backus_turingaward_lecture.pdf … . This paper spawned both
#concatenative languages like Factor and point-free style in languages like Haskell. - 10 more replies
New conversation -
-
-
Co-sign this
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
That doesn't sound like how new features have come about in JavaScript in the past. Hasn't it always been like engines trying to find a way to optimize something that theoretically can't be optimized? And new features come from the users of the language? Like .map() and .bind()?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
JavaScript, HTML, CSS, HTTP, performance, security, Bash, Unicode, i18n, macOS.