Future of D

[Guillaume Piolat]

On Monday, 30 October 2023 at 14:30:25 UTC, Imperatorn wrote:
>
> Do you have an explaination to why the number of contributions 
> to phobos has decreased so much since about 2018-2019?

A combination of:
- std.experimental.allocator and std.experimental.logger faced 
increasing scrutiny, it took a long-time and ended up underused 
anyway.
- Phobos being well used and mature, with very-high quality stuff 
that is hard to produce.
- disinterest for Phobos additions (driven by "no way to remove 
stuff") versus the reactivity in DUB, where API can evolve more 
easily
- also being somewhat not possible to use in all cases (-betterC, 
alt. runtimes)

Personally I think the STL Stepanov-style is a dead end and we 
were oversold composable algorithms, that fail to create 
understandability.