Hi guys,

Just wanted to remind you that, D maybe isn't that bad.

We're very good at bashing our own language, but we should also remember sometimes what it has given us.

I have spent the last months going through other languages, and I can say, the grass is always not so much greener on the other side.

Yes, there are more mature languages.
Yes, there are languages with better ecosystems.

But, just as an example - Zig - which is getting attention, is according to the community itself (including its creator) not in 1.0 until about 2025.

And still people use it, and might even think it's better than D.

Some information from their community (not my words)
It does not have a standardized package manager and build system.
It does not have an official registry of packages.
It is unstable.
It should not be used in production (actively advised against).
It changes so often that you can not rely on code to work even in 1 month from now.
etc

And still, people still think Zig is better for some reason.

Yes, D has it's flaws, true. But it's far from unfixable? Or is that what people believe?

Forget about Jai, Odin, Beef and all those languages.

Go - Welcome rheumatism 👴
Rust - Welcome brain tumor from not even being able to prototype something in less than 2 years 😩
C++ - Welcome to hell 🔥
...

The only real language out there that is close to what D is/could be is Nim and I respect it.
But, its syntax is not that kind to those who loved the curly braces.

All I'm saying is - maybe it's best if we just fix D?

There is some valid criticism, like the risk for attribute soup etc. But maybe it's fixable?

Remember what D gives you in terms of UFCS, CTFE, metaprogramming, performance, package manager, prototyping, inline assembly, 3 compilers for different use cases etc.

Is D really that bad?

Hi guys,
Is D really that bad?

On Friday, 28 October 2022 at 09:51:04 UTC, Imperatorn wrote:

I don't think it makes much sense to talk about better or worse without a use case. Zig seems to aim for embedded like settings, but I don't like the language design regardless, as of today. But that could change.

You cannot predict the future, but if you go 10 years back and identify things that ought to be fixed (in the sense that it would make the language more broadly appealing) and find that those issues are still there, then there must be something making it difficult to address. Could be the current compiler code base, could be other things.

Keeping the same process makes it improbable that there will be a significant change, for good or bad.

Why forget about Jai? If it gains traction in a specific domain like games, why not use it?

(There are also plenty of others in the works: Carbon, Circle, C++2.0, V etc)

On Friday, 28 October 2022 at 09:51:04 UTC, Imperatorn wrote:

The language itself isn't bad, it actually quite alright, when I bough Andrei's book, I thought to have found a modern version of Modula-3 and Delphi.

However in all these years, the direction was never clear, and its use at Facebook and Remedy didn't do much to help it grow adoption.

Nowadays although D the language is quite nice, for my line of work, the ongoing improvements in Java and C# languages for low level coding + AOT + ecosystem, mean that in no way I would be able to convince my peers to use D.

On top of that, for better or worse, Go and Rust are also creeping in into my line of work, as we are adopting frameworks written in those languages, making it even harder to try to advocate for D.

So for me, D remains one of the languages that I have fun doing hobby coding.

Hi guys,

Just wanted to remind you that, D maybe isn't that bad.

[...]

On Friday, 28 October 2022 at 09:51:04 UTC, Imperatorn wrote:

There is something to be said for adding static break and static continue to the language, and we could do away with classes as a reference type, but D is the only language that provides the metaprogramming facilities that I need.

On Friday, 28 October 2022 at 16:58:20 UTC, RubyTheRoobster wrote:

Write a DIP 😍