With this lang team, I think Rust is going to die quickly. It doesn't worth investment to learn it.
With this lang team, I think Rust is going to die quickly. It doesn't worth investment to learn it.
There was some debate, but it's been under pretty intense debate since at least November of last year. This has been the most discussed aspect of Rust ever. The tracking issue has 296 comments. There have been numerous threads on the internals forum, and recently, multiple threads on reddit per day. The team (which, I should note, I am not on) repeatedly asked for feedback, and read these threads, as well as participated in them.
It's been going in circles for ages. There are a few options with the syntax, but there's no clear standout winner; it really depends on how you weigh each pro and each con. At some point, though, it's the team's responsibility to decide. And so, on the sixth, they finally reached internal consensus. That's actually enough to just make the decision, but instead of just making the decision, they decided to go the extra mile, and say "here is our decision, and here are our reasons for reaching that decision. If you have something new to add to this incredibly long debate, please let us know, we'll be making the decision on the 23rd." : https://boats.gitlab.io/blog/post/await-decision/ The internals thread on this post alone is 420 posts long: https://internals.rust-lang.org/t/a-final-proposal-for-await...
So yes, there are some vocal people who preferred a different syntax than the one that was chosen. Heck, I actually mostly prefer a different syntax. But at some point, a decision has to be made. Some people are not going to be happy with any decision. That's just how it is. It's not about "not listening to the concerns"; the concerns were listened to, but in the end, a different decision than (apparently) your preferred one was made. That's just how it goes sometimes. There are people who are equally passionate about this that would have been just as mad as you are now had the decision gone your way.
(I personally still wouldn't have made the call they did, but the call is reasonable. I've written code using it, and it's totally fine. I understand why the decision was made. We'll just have to see!)
I still remember the lifetime syntax (&'a T) discussion. await syntax discussion felt really similar.
For example, I don't know if this was a new idea or an old one, but standardizing the general concept of dot-keyword before following through with dot-await, even if it ultimately doesn't change the decision, would make the decision easier to swallow (for me).
Yes, that has been discussed. It hasn’t been committed to in any way.
This is the most discussed feature of Rust ever. The team spent a lot of time and energy on it, way way way more than any other feature. Several team members flip flopped positions several times.
(For disclosure, I preferred postfix punctuation similar to `?` that most people hated - so I'm not gettng my way either.)
Those polls were incredibly non-scientific, and don’t really mean anything.
Yes, they were bad, self-selecting polls. You have nothing better, and I hear far more complains than people being happy with the decision. The polls both tended to agree though so that would lend some credibility.
Maybe you should have just explained "We aren't really asking the users since we have way to show that any poll would be valid, but we will take some input and make up our mind from that."
I think that is a very precise (and charitable) description.
The teams always make decisions. That’s why they exist.
In that case, you as a team basically saying that you were afraid of punctuation was intellectually shallow, lazy, and a terrible decision. It wasn't like a tailing @ or # would have turned rust into APL. You should go back and revision the `?` decision then under the same reasoning.
The reality was that _every_ time some discussion about the async syntax came up _every_ proposal and concern was uttered again. It's not that any direction was particularly more popular in the community than any other one, so it's not that the lang team decided against what the community wanted. The only thing that seemed certain in the end that all options have some downsides.
I'm very happy with the Rust lang team. They brought us this far, and I'm very happy with the vast majority of their decision. And apart from the async drama, they are delivering a lot of other nice goodies.
If you're looking for a language without warts, good luck.