On Friday, 8 December 2023 at 09:06:34 UTC, Hors wrote:

I couldn't agree more. D needs to pick its poison :)

I don't think there is enough people interested in a high level language in the D community. If anything, I see D becoming more and more of a system language over time.
Therefore, I am thiiiiiiiiiiis close to just forking it and tweaking it to be more high-level language. I'm mostly stopped by needing to write code completion though

On Thursday, 7 December 2023 at 19:21:10 UTC, Paulo Pinto wrote:

Computer games and GC is really a good match if you have room for certain performance hit (regardless of type of GC). Many computer games have a lot of objects, these objects also often have some kind of relationship to each other. The lifetime and relationships are usually totally unpredictable. If the developers can do away with lifetime tracking and all associated bugs that's a big win, especially when investors today seem to have unrealistic deadlines. There are usually enough bugs as it is anyway.

GC in computer games today is probably standard I would presume. GC in the D language is a great thing and I don't want it to go away, just make it better.

On Friday, 8 December 2023 at 11:42:22 UTC, ryuukk_ wrote:

10000% Agree! Dlang has potential in the gaming industry.
Still dominated by C++ but it will be interesting to see if Odin or Jai rise in popularity in the coming years.

Is there any lessons they are doing that could elevate D? Odin, for example, seems to have a lot of tools "out of the box" to get you start easily.

Back to the point - D has soooo muh potential in this field!

On Friday, 8 December 2023 at 13:27:46 UTC, GrimMaple wrote:

I thought we're way past that decision after 10yrs. I doubt D is changing to anything other than what it is now. The language is mature and set. It's already being used put there.

On Friday, 8 December 2023 at 09:06:34 UTC, Hors wrote:

FWIW, IMO, apart from the development and maintenance effort, there is nothing wrong in 'trying to be everything'. Indeed it's a laudable effort.

There (was) a specific problem with the GC debate, in that there appeared to be a real effort to 'force' developers into the GC camp, rather than simply presenting it as an option (and I personally think that for business application programming it is a very good idea).
Trying to force people to do anything, even if you sincerely believe in what you are advocating, is simply a bad idea.

On Sunday, 10 December 2023 at 17:47:52 UTC, DLearner wrote:

Everything here means being a c/zig replacement, c++ replacement, python replacement and for the hell of it rust replacement

and no, there is just no way to do that; the rust replacement plan of making a """"borrow checker"""" went nowhere, importC isn't going to implement a 1 to 1 preprecosser so it won't ever be useful and zig must be an ok c compiler given what people say is done.

D should focus on being a c++ replacement; we have better templates they just aren't well used by the std and it's not being communicated.

On Sunday, 10 December 2023 at 18:07:35 UTC, monkyyy wrote:

I agree 100% and that is why it was named D to begin. The world has since then moved on and C++ as well (but you can argue it became Frankenstein monster). D should remain a competitor to C++ but it needs to move on as well in order to remain competitive.

On Friday, 8 December 2023 at 13:27:46 UTC, GrimMaple wrote:

That D treats those two domains equally is supposed to be its unique selling point that no mainstream language does. I still fail to see why so many people pick D and then advocate for killing that fundamental feature. Aren't you excited about the thought that you can use one language to excel in both domains?

Consider the two probably biggest companies using D. Symmetry investments used D for high level purposes. Weka.io on the other hand vendors a file system, which is a systems langauge job. If we were to declare either domain a second class citizen, we have to anger one of those companies.

On Monday, 4 December 2023 at 09:31:16 UTC, GrimMaple wrote:

D isn't tripping you here, rather the other languages sweep the dust under the rug. For instance, if you implement a library in Java, you're automatically failing to support GC-less programming. You don't have to support @nogc/-betterC scenarios in D either (and probably are best off not supporting them in most cases, at least for initial versions), it's just that the language gives you the choice.

On Monday, 11 December 2023 at 07:18:44 UTC, Dukc wrote:

Clarification: this is a typo. I meant "uses", not that Symmetry would have stopped using D.