A sad day for Rust: actix-web is dead (2020) | Hacker News Reader

A sad day for Rust: actix-web is dead (2020)

steveklabnik.com

21 pointspabs310 comments

For those who aren't aware, this wasn't actually the end of the story; the original author ended up deciding to let someone else take over after all, and actix is still actively developed and used: https://github.com/actix/actix-web

It is mentioned in the epilogue at the bottom, but probably still good to highlight

... I was about to say! I use Actix to great effect in production. Don't do me in like that.

I saw "(2020)" and then realized something was up, because Actix has been incredibly active.

You gave me a real Halloween scare.

I don't really understand what the purpose of sharing these types of old articles is. I've been using Actix for the better part of 2 years now in production at my place of employment. I've regularly discussed with the maintainers about my questions and issues, and they have been extremely helpful.

Maybe there's better Rust web frameworks out there, but the Actix web project seems alive and well to me.

I’m confused too.. when I imagine the purposes behind these type of shares, it always brings out the cynic in me.

It's history, and it's an interesting example of community dynamics and of Rust's journey. Just because the conclusion turned out to be wrong, I don't think that makes it worthless

I mean, it wasn’t so much a “conclusion” as it was “here’s what’s happening right now.” At the time I wrote it, it was correct.

There are lots of cases where projects are so successful and get so complex that the best step for the maintainer who created it is to step down from making the final technical decisions. I can imagine that it’s super hard to build a super fast web server that is also memory safe.

It seems like the good thing was to focus on speed first, but at some point the server became somewhat of a standard, and at that point memory unsafety can create huge damage and has to become the priority.

In this specific case I would have added a benchmark to the CI and not accept pull requests with regressions of speed.

That would have put the pressure on the contributors to find solutions that improve safety without compromising for speed.

Not one comment complaining about C and memory safety? C'mon guys, step it up.

It’s not dead it’s alive and well.

Reply on news.ycombinator.com