I don’t understand why you’re getting downvoted. While I don’t share your conviction, I do admit it’s certainly a possibility.

The advantage of doing things that way is that code becomes much more portable. We may finally reach the goal of “write once, run anywhere”, because the AI may write all the platform specific code.

It does make a big assumption that the AI output is reliable enough though. At times people will want to tweak the output, so how are they gonna go about that? Maybe if the language is based on Markdown, you can inject snippets of code where necessary. But if you have to do that too often, such a language will lose its appeal.

There’s a lot of unknowns, but I see why it’s a tempting idea.

tsc is (very) slow and there are also no convenient ways to interact with it from Rust.

So it saves a lot development and CI time to roll our own. The downside is that our inference still isn’t as good as tsc of course, but we’re hopeful the community can help us get very close at least.

I see this argument repeated here every now and then, but it’s not true: https://aoav.org.uk/2020/the-effects-of-strategic-bombing-in-wwii-on-german-morale/

To summarise: Bombing had a profound effect on depressing morale through inducing defeatism, fear and apathy. Bombing did not harden popular resolve against the US. The hate and anger it aroused was directed against the Nazi regime, not the Allies.

It’s that the compiler doesn’t help you with preventing race conditions. This makes some problems so hard to solve in C that C programmers simply stay away from attempting it, because they fear the complexity involved.

It’s a variation of the same theme: Maybe a C programmer could do it too, given infinite time and skill. But in practice it’s often not feasible.

Which one should I pick then, that is both as fast as the std solutions in the other languages and as reusable for arbitrary use cases?

Because it sounds like your initial pick made you loose the machine efficiency argument and you can’t have it both ways.

I’m not saying you can’t, but it’s a lot more work to use such solutions, to say nothing about their quality compared to std solutions in other languages.

And it’s also just one example. If we bring multi-threading into it, we’re opening another can of worms where C doesn’t particularly shine.

Well, let’s be real: many C programs don’t want to rely on Glib, and licensing (as the other reply mentioned) is only one reason. Glib is not exactly known for high performance, and is significantly slower than the alternatives supported by the other languages I mentioned.

I would argue that because C is so hard to program in, even the claim to machine efficiency is arguable. Yes, if you have infinite time for implementation, then C is among the most efficient, but then the same applies to C++, Rust and Zig too, because with infinite time any artificial hurdle can be cleared by the programmer.

In practice however, programmers have limited time. That means they need to use the tools of the language to save themselves time. Languages with higher levels of abstraction make it easier, not harder, to reach high performance, assuming the abstractions don’t provide too much overhead. C++, Rust and Zig all apply in this domain.

An example is the situation where you need a hash map or B-Tree map to implement efficient lookups. The languages with higher abstraction give you reusable, high performance options. The C programmer will need to either roll his own, which may not be an option if time Is limited, or choose a lower-performance alternative.

Of course, but it needn’t be black and white. You can also diversify, make yourself less reliant on a single platform. And by doing so, enable your audience to follow you elsewhere. Or diversify into different activities altogether. And when it’s no longer half your income on the line, then switch.

But doing nothing and saying, “but half my income!”? That’s not only a choice, but also complacency.

Great points, except:

People can’t leave for anything smaller.

They can and some do. It’s still a choice.

I’m not arguing against that. Merely providing some counterweight to the idea that the author was “flinging shit in the trenches” 😅

I found the title of that section slightly triggering too, but the argument they lay down actually makes sense. Consistency helps you to achieve correctness in large codebases, because it means you don’t have to reinvent what is correct over and over in separate pockets of the codebase. Such pockets also make incremental improvements to the codebase harder and harder, so they do come back to bite you.

Your example of vendors doesn’t relate to that, because you don’t control your vendor’s code. But you do control your organisation’s.

There is a serious attempt for that actually: https://www.assemblyscript.org/

It doesn’t offer full compatibility with the regular TypeScript though, despite being very similar.

Feel free to just use React on the frontend if you’re more familiar with it, but make sure you couple it with Redux. Then when the time comes you want to bring some Rust into the frontend, you can do so by writing your Redux reducers in Rust.

PS.: The blog post mentions using fp-bindgen for WASM bindings, but nowadays you’re probably better off using wasm-bindgen.

Yes, it has a few APIs of its own. I merely think they are negligible in this discussion because they only provide a minimal superset over Node.js’s own APIs and are also very minimal compared to what Deno provides.

I’ve updated my post to mention “noteworthy” APIs.

You’re ignoring the fact that for many projects it does work.

It only needs to be perfect if you want to run 100% Node.js software unaltered. While that may be a lofty goal, it’s also an infeasible one.

That doesn’t mean imperfect support is futile though. By your logic, Bun has no right to exist because it only supports Node.js APIs and doesn’t have noteworthy APIs of its own, and they’re not perfect either. Yet they seem to be at least as successful as Deno is.

Or for an example in a different domain: Your argument would state that a project like WINE shouldn’t exist because it doesn’t have perfect compatibility with Windows, and it disincentivizes development of Linux games. Yet it is largely thanks to WINE that Valve has been able to make the Steam Deck and that Linux gaming is finally taking off.

I think what your argument fails to take into account is that you need a significant amount of users to make any impact on the market. And many users have legacy requirements that they can’t throw out overnight, so you have to support those legacy environments. And even with imperfect legacy support you can support your users, especially if the users are willing to make a few changes here or there. But if you have no legacy support, you also get no users except those that have niche greenfield requirements.

So instead of trying to replace NodeJS or offering an upgrade path for existing Node projects, incentivize formation of ecosystem around Deno

They are incentivizing their own ecosystem. That’s what Jsr.io is all about. But the world isn’t black and white. They can do more than one thing.

I dunno, I still see a blog post. Which is hosted in their own issue tracker, which is of course odd, but also the point.

Maybe it went down for a bit?

I would love to see Go, Rust, Swift and Kotlin added to this. Anyone willing to take a shot?

Using smart pointers doesn’t eliminate the memory safety issue, it merely addresses one aspect of it. Even with smart pointers, nothing is preventing you from passing references and using them after they’re freed.

I think that’s very team/project dependent. I’ve seen it done before indeed, but I’ve never been on a team where it was considered idiomatic.