A D vs. Rust example

[ryuukk_]

On Sunday, 23 October 2022 at 14:55:56 UTC, Imperatorn wrote:
> Yeah, that's why I don't like "hyped" languages.

> We use C# at work too, mainly because it's a well designed 
> language (quite easy to learn), has good support (ecosystem) 
> and IDEs. Sure it's a bit verbose, but the upsides make it 
> worth it.

> D has so much potential to be a replacement for many other 
> languages.

> It could replace Python, C# in some cases, C++ for sure, C with 
> betterC, maybe even R or Julia.

> What is needs though is to be stable and have an LTS branch.

> Companies don't really care at all what language you use, they 
> care about economics and risk mitigation.

> If I say to my team "let's write the next thing in D", the 
> first question will be about ecosystem, IDEs and other things 
> surrounding the language, not the language itself, no matter 
> how good it is, unfortunately.

> I really hope that one day we will have a bigger following 
> (which will enable us to make better tools), because there are 
> so many things D is better at than many other languages.


That's not all true, TLS version is not needed

Go became the cloud native language without that BS

Java was held back because of it, wich helped C# penetrate the 
bloat driven corporates

As they figured it out, Oracle started to speed up development of 
the language:

- multiple GCs
- pattern matching
- var
- data class
- green threads
- graalvm

Now C# try to fight back

- record
- pattern matching
- green thread
- nativeaot

Go now is being held back by corporates 
(amazon/google/microsoft), just like rust

D should be free to experiment, not every language needs to be 
taken hostage by corporates

As you said, corporates don't care about what language they use, 
they'll dump it if they find something that suits their cost 
requirements better


- simplifying the language

- finishing the existing features

- drastically improving ergonomics in some areas

- invest massively in tooling

- keep experiment with new ideas and take a look at the backlog 
of DIPs

That's what i'd do if i knew how to make and had to work on a 
language

Mistakes of the past need to go, otherwise people will constantly 
look for alternative, that alternative needs to be D+1 that fixed 
past mistakes, and not double down on them, because demography 
mean the people who rely the mistake will disappear, and the new 
generation of devs will choose something else