This mirrors my feelings about Ruby. Especially the lack of type hints. It’s a huge problem when trying to work on large Ruby codebases, e.g. Gitlab or Asciidoctor. Easily doubles the time it takes to get anything done. Sometimes I’ve tried to make a change to Gitlab but had to give up entirely simply because it’s impossible to follow the control flow.

That’s very rarely a problem with statically typed languages. (It can happen with excessive use of interfaces that are resolved at runtime but it’s much less common.)

So aside from Rails I can’t really see any reason to use it over even Python, let alone actually good languages like Rust, Go, Typescript, etc.

Not buggier than Python is a really low bar.

You forgot about the lack of static type hints. That’s a serious flaw.

If you ignore the performance and lack of static types, then I don’t think there’s too much wrong with Ruby but apart from Rails there isn’t really a compelling reason to use it over Python either. And that’s not saying much!

None of the points in the article about the flaws of Ruby are because of Rails. In fact the article says the exact opposite - the only reason Ruby is still relevant is because of Rails!

I would say it’s the other way around. The only reason monorepos aren’t used more is because the tooling for monorepos is inadequate, or because people don’t know how to use them.

I can’t deny that the tooling for multi-repos is crap too though.

They are simpler, but they do not scale.

I don’t think many of us need to worry about scaling beyond the size of Google or Microsoft.

a push to a repo doesn’t mean “deploy changes to everything” or “build everything” any more

What do you mean? I have yet to work for a company that’s organised and sophisticated enough to actually use a monorepo but my understanding is you’d set up something like Bazel so it only builds & tests (and I guess deploys) things that depend on your change.

in the react-server package used by React Server Components (RSC).

Is Deno not convenient and fast? I am also interested in knowing why I would want to use Bun over Deno.

I’m glad this is still alive. It’s a good idea. I keep thinking about doing the same for TCL.

I do wonder if Bash is perhaps just a bit too insane to make this really feasible though.

Reversed range support

This is a misfeature.

I didn’t say they are equivalent. I said I dislike them both.

I wouldn’t say those are “abominable” either. AI is fine. Microsoft actually cut off Israel’s access to Azure…

They’re operating infrastructure. They aren’t going to morally vet every single customer. Imagine how many dubious things run on AWS that we never hear about!

I must have missed those abominable things…? You don’t mean the ICE stuff?

It’s “relationship with ICE” is that they haven’t banned an official US government agency from buying their software. I might not agree with what ICE is doing but I also don’t agree with every corporation in the world having to morally police all of their customers for fear of being pilloried by cancel culturists.

The “created by monkeys” seems to be a minor bug. What system doesn’t have those? I’ve certainly had plenty worse bugs with Gitlab CI.

“Completely neglected” is complaining about the lack of FreeBSD support!

I do think GitHub is relatively neglected. There are quite a few big issues they could fix with relatively little effort but they seem to go years with no comment.

It’s not really much better with Gitlab though; the only difference is you see more “a large premium customer is requesting this” comments!

I doubt Forgejo really have more resources to fix bugs than GitHub or Gitlab.

Why? Microsoft gives them a ton of money for CI and infrastructure. Unless they are having serious technical issues I don’t see why they would move to a more expensive and probably less well integrated CI provider.

The audacity to be nostalgic for Eclipse!

I don’t think you can blame anyone except him for not understanding that AI sometimes hallucinates. Hell basically every AI tool makes you read that before you start using it. I’m normally very reticent to blame users for using things incorrectly, but if it took him 3 hours to realise it was wrong here then I have to say that’s on him.

Come on has anyone else ever persevered with a hallucinated answer for 3 hours before realising it was a hallucination? Longest it’s taken me is like 5 minutes, and that’s only for things that aren’t easily googleable.

I suspect in the real world it’s frustrating enough for restaurants that it wouldn’t have worked out.

You’re pretty much tricking restaurant workers into one of those awful voice-based phone trees.

Plus there are so many things that can actually happen when you try to book a table on the phone - they don’t have exactly what you want but can offer you this time instead… they only have outside seating available… etc. etc.

Plus, just having a proper online booking form is clearly a better option and not totally uncommon these days.

This is silly, and not in an interesting way.

Actually chars() is pretty simple - it’s just UTF-8 decoding which is elegant and simple.

The complexity is all around unicode, not UTF-8.