Breaking D2 language/spec changes with D1 being discontinued in a month

[bearophile]

SiegeLord:

> With that kind of attitude D2 will never stabilize.

Currently in D there are unfinished parts like vector operations 
that are not finished, shared that maybe needs to be redesigned, 
foreach loops and switch statements that need some design fixes, 
but also its standard library has rebindable, emplace, Typedef, 
Nullable and Algebraic that are not good enough and need 
improvements or rethinking. Plus some missing collections and 
memory allocators. Plus there are several corner cases that must 
be addressed, I am following Bugzilla closely and if you want I 
can list you one hundred of significant bugs. Fixing some of 
those "bugs" will break user D2 code. Not fixing those bugs will 
be a worse fate.

D is a quite complex language, composed of many parts, and Walter 
is not just designing a language and developing a front-end, but 
he's also keeping its back-end, there is also the standard 
library, and the garbage collector. And the amount of work force 
is limited. So D develops and stabilize rather slowly. Developing 
D takes many years. I think there is no escape from this.

On the other hand I used to find a significant new bug every 
20-30 lines of my D2 code, this today happens much more rarely.

Bye,
bearophile