On Sunday, 30 October 2022 at 02:15:06 UTC, Walter Bright wrote:
> Besides, if it was so easy to do, why has nobody produced a C compiler that uses a GC instead of malloc/free?

Well, it has been done with Bohem.

But anyway I wrote up my sketch design in my blog:
http://dpldocs.info/this-week-in-d/Blog.Posted_2022_10_31.html


It is more like bounds checking than a new type. Sure, I do propose a new type, but just to locally bypass the thing, not as something mandatory. With a compiler switch, you can turn it on or off to experiment with different strategies.

On 10/29/22 04:02, Walter Bright wrote:
> On 10/28/2022 3:20 AM, IGotD- wrote:
>> D is a good language but it has some obvious flaws that are easily recognizable and these can be fixed. The problem is the management and the total tone deafness/insight who refuse to recognize these often because of self manufactured dogmas. This is really frustrating as if the odds and ends could be fixed in the language it could bring it to a very competitive state. It's like falling over right before the finishing line. The only way out of this is if D is forked.
> 
> You can fork it if you like. Feel free!
> ...

It is pretty standard procedure for contributing to D to fork the official repositories.

DMD alone has been forked over 600 times on github. I guess coordinating DMD development is just not so easy and we don't really have people with the requisite skills and motivation.

> Meanwhile, last week I spent fixing some gaps in the XMM semantics (relational operators are now supported).
> 
> Before that, I went through every DIP1000 bugzilla issue, and submitted PRs for the problems.
> 
> Before that, I went through (again) the list of outstanding issues with ImportC and fixed the top problems.

Thank you! I wish I had more spare time. (And push access to DMD master ;] )

On 10/31/2022 5:47 AM, zjh wrote:
> `D` needs a reasonable `todo` list with `priority` !.

The thing is, we don't assign people tasks from a list. People work on what they want to work on. Sponsors can fund specific tasks they need done.

On 10/31/22 20:52, Walter Bright wrote:
> On 10/31/2022 5:47 AM, zjh wrote:
>> `D` needs a reasonable `todo` list with `priority` !.
> 
> The thing is, we don't assign people tasks from a list. People work on what they want to work on. Sponsors can fund specific tasks they need done.

The value of such a list is that people who do want to work on one of the items on the list anyway receive some assurance that what they want to work on aligns with the goals of the project and has a reasonable chance of being upstreamed. This is important for people who have limited resources to spend on open source projects not immediately related to their day job. For example, your and Andrei's very clear, concise and consistent communication making clear that that `static foreach` is desirable but difficult to implement is exactly what prompted me to start working on the implementation.

I think the foundation has generally been moving into this direction? (E.g., vision statements.) Of course, keeping a detailed list up to date also causes some overhead, so it is a tradeoff.