Future of D

[IGotD-]

On Monday, 14 December 2020 at 13:12:55 UTC, Dukc wrote:
>
> I think `@live` without that differentation does the job 
> better. It is a relatively cheap solution that does not even 
> try to be perfect. The managed-unmanaged pointers are better in 
> theory, as they can be more foolproof, but they have many times 
> as much complexity. Since most of the code can avoid memory 
> corruption simply by using the GC, I doubt whether closing that 
> gap in `@live` is worth that much.

I would really like to know more about how @live together with 
the supposedly reference counting without fat pointers is 
supposed to work. I haven't seen any comprehensive document that 
explains it. Reference counting without fat pointers is an 
oxymoron as far as I understand. Maybe someone can explain.

I think that is what's missing, what the plan is for this.