What are the worst parts of D?

[Chris via Digitalmars-d]

On Wednesday, 24 September 2014 at 09:57:06 UTC, ketmar via 
Digitalmars-d wrote:
> On Wed, 24 Sep 2014 09:15:27 +0000
> Paulo  Pinto via Digitalmars-d <digitalmars-d at puremagic.com> 
> wrote:
>
>> Most of us cannot afford to be a "Technology X" developer.
>> Every project, every client is a complete new world.
> yeah. and so there is *no* *reason* to stress c++ interop, heh. 
> 'cause
> "client dictates language" anyway. i like it.

I do understand your concerns and I once mentioned in a thread 
that companies who use D should not dictate the way D evolves, 
instead it should remain community driven. Say for example D were 
used on web servers and a huge amount of effort was directed 
towards turning D into a web server language, while other 
important features/improvements were neglected, that'd be bad.

However, in the case of C++ I must say that it is important. One 
of the reasons I opted for D was (and still is) its seamless 
C-integration. It allowed me to use so much existing code in C, 
libraries I would and could never have rewritten myself in D. 
There are loads of C(++) libraries out there you might want to or 
have to use for a particular project. When I started using D the 
Unicode module lacked certain features I needed. I just used an 
existing C library and could go on with the real program in D. 
Hadn't this been possible, the project in D would have died right 
then and there. Now D itself has the features I needed back then, 
but the C library was essential to bridge the gap. I think a lot 
of C++ programmers would do the same. Start a new project in D 
resting assured they can still use their carefully built code 
bases in C++. So I think interop with C++ is important. And don't 
forget that the reality is that most people interested in (yet 
reluctant about) D are C++ programmers, at least that's the 
impression I get here.