I'm a bit confused about the premise of the article. Does anyone assert that typing speed is a bottleneck at all? I've been in the industry for years, and have never heard that claim.
I do agree about the whole "less code is not always more", but I get confused when the author keeps bringing it back to typing speed.
I'm confused, what exactly is going downhill? Hacktoberfest, or open source in general?
Ah, you're right! Somehow I missed that, good catch.
Most sites load no content at all if JS is disabled.
I don't think that necessarily holds true for OSS. The average user with no development experience wanting to use an open source project doesn't mean it will always develop faster.
For #2, there isn't anything stopping a separate auth system not through GitHub. Really just needs someone to own the implementation. See https://github.com/rust-lang/crates.io/issues/326#issuecomment-216662599 for past discussion surrounding this.
Same quote of the week as last week?
What the heck is this paywalled article doing here? That's some reddit-level shit.
I was originally introduced to the idea by this RustConf 2018 keynote: https://kyren.github.io/2018/09/14/rustconf-talk.html. It's rather dense though.
I did find this random article that outlines just the concept of generational indices pretty concisely: https://lucassardois.medium.com/generational-indices-guide-8e3c5f7fd594
If you want to go one step further, a lot of game development uses a generational index, where the index is both a value and a generation, allowing you to know whether the index you currently have stored references an object that has already been destroyed and replaced by another object. Basically every ECS framework I've ever seen uses this pattern.
Hard to say without being able to see the comments. I suspect that if that were the case, the entire post would have been removed.