The progress of D since 2013

[Maxim Fomin via Digitalmars-d]

Hi!

Good to see D is progressing! I was active forum and bugzilla 
participant in 2011-2013. Since then I have not touched D.

What is the progress of D (2014-2017) in following dimensions:
1) Support of linking in win64? AFAIK Walter introduced win64 
support in circa 2012 which was the big progress. However, 
support for win64 linking was limited because dmd defaulted on 
old dmc linker, and Walter didn't plan to do anything with this.
2) What is the support of other platforms? AFAIK there was 
progress on Android. From my memory recollections, the full 
support of Android was expected at that time.
3) What is the state of GC? AFAIK there were some improvements 
for GC sent as GSOC projects but they were not added in 2013. I 
see in the changelog that there are some improvements in speed 
and configuration support was added.
4) What is the state of GDC/LDC? GDC team was actively working on 
including gdc in gcc project. Do gdc and ldc still pull D 
frontend, so there is essentially 1 frontend (where gdc and ldc 
frontends lag several versions behind) + 3 backends? I see in the 
changelog some dmd backend improvements. How the dmd backend is 
compared with C++/GDC/LDC? AFAIK in 2013 there was a tradeoff: 
either you use dmd with brand-new frontend or gdc/ldc where 
performance is comparable to gcc, but frontend lags behind. Is it 
still true?
5) What is the progress with CTFE? I see a lot of discussions in 
forum archive devoted to the development of CTFE. What is the 
summary of CTFE development in recent years?
6) I don't see any significant changes in D core from dlang 
documentation (except those mentioned in changelog for 
2014-2017). Is is true or is the official spec as usual delayed 
:)? Is dlang spec fully and frequently updated or is it sparse as 
in the past? Is TDPL book still relevant?
7) Is UDA still compile-time? Are there plans to make it also 
runtime?
8) What is the progress with shared and immutable? AFAIK the 
compiler support for shared was not complete and Phobos library 
itself was not 'immutable-' and 'shared-correct'.
9) Does D gains popularity?
10) Anything else 2013 D user must know? :) I don't ask about 
Phobos because according to the changelog the progress is 
enormous, incremential and targets several directions - I doubt 
it can be easily summarised...

Thanks!