On Friday, 29 June 2018 at 07:03:52 UTC, Dmitry Olshansky wrote:
I never ever (I think) did something provocative, something to
finally see:
- who in the community WANTS D language to succeed?
- who are just these funny “people” let’s call th this, that
are I don’t know “just hang around”
Because shame is a weapon much like fear (of death esp), pride
can be used as weapon but ehm better shame the bastard...
And so on.
So - until we all understand that these donations are not
because we are begging fir money.
I will send ~ 10$ each day _specifically_ to see who WANTS D TO
SUCCED and WILL NOT BE SHAMED LIKE THAT FOR ONCE!
It is because it’s (soon) your last chance to invest into the
Future.
P.S. I mean what you think the future of native code is???
Rust? Crystal?? Nim???
I know most people here don't agree with me, but I think you're
fighting an already lost battle ;)
As you know, I'm convinced that D could be marketed as the
perfect language to develop native web servers and mobile
applications, and have its core libraries somewhat extended in
thqg direction, like Go and Crystal which allow "plug'n'play" web
server development for instance, but obviously the D "leadership"
remains convinced that D must be sold as the best alternative to
C++.
Personally I'm a complete D fan because it is SOOO MUCH better
than JavaScript/Python/Perl/etc for file processing...
For engine and game development I'm still using C++, despite I
prefer D, and believe me this won't change for a while.
Game development is a very special use case, but personally I
don't think that many of those who use C++ for close-to-the-metal
development should be that much interested in switching to D,
because most of its standard libraries depend on the presence of
a GC...
And to answer your question, IMHO the future of native code
probably remains C++ (not Rust) for system programming, and
(unfortunately) Go for web development (great ecosystem, db
drivers, often faster than Java, C#, Dart, etc) despite it lacks
several core feature many developers need (generics, etc).
Once Crystal integrates parallelism (at 1.0), it should become de
facto one of the best alternative to Go, Java, C#, etc, because
it's actually "Go-made-right". For instance it's genericity
system works well, and its type inference system natively support
union types.
Nim disqualifies itself because contrarily to D and C# for
instance, it doesn't manage mutual dependencies automatically for
you, which is a pity.
And Rust, despite it has perfect C/C++-like performance and
doens't need a GC, its borrow checker made it a hell to use at
first, as unfortunately Rust hasn't integrated strong/weak
references as a core feature of the language (Rc/Weak are
templates, RefCell is needed for mutability, etc), despite that's
actually what many C++ developers use today for resource
management, and would be more than enough for them to get their
job done once they switch to Rust...
