Am 16.07.2014 17:39, schrieb Vic:
On Monday, 14 July 2014 at 10:13:43 UTC, Chris wrote:
On Saturday, 12 July 2014 at 10:27:12 UTC, Russel Winder via
<snip>
I think we need to address these issues, because they are of a
psychological nature and not really language issues. I'm sure that if
we fixed GC and had the best implementation ever, people would find
something else to complain about "D doesn't have blah, I don't like it!"
<snip>
I'm sort of getting the idea that D goal would to be a better Java.
I'm running away from Java (after 10 years). I hope that someone at D
has power and can say NO to a feature the way Linus does as opposed to
adding more 'JCP' features, pushing such stuff downstream. Adding more
features to be good at everything, aka a submarine that is a law mover.
It's all done w/ best intentions. But forcing GC into base library of a
system programing language? Maybe D is not a system programing language,
but a enterprise app productivity lang. At least give us a choice, to
use D why do I have to re-write the base lib.
Cheers, Vic
Yes, it has been done many times before.
Starting at Xerox PARC, those beautiful systems were many ideas of the
modern web were pioneered.
Interlisp-D, Smalltalk and Mesa/Cedar, all had a mix of RC/GC.
Olivetti was playing around with Modula-3 with the SPIN OS, before
Digital closed their R&D unit.
Niklaus Wirth and his colleagues created Oberon at the Swiss Federal
Institute of Technology, which was an workable desktop OS used by a few
at the informatics department and OS research topics, specially version
System 3 with its gadgest toolkit. This spunned quite a few derivatives
namely EthOS and AOS. Active Oberon on AOS already offered a concurrent
compiler before that was a theme.
Microsoft created Singularity with Sing#. Although the project was
cancelled, many of its outcomes live on WP8 native compiler and on the
upcoming .NET Native.
Apple is stating that Swift is a C replacement ("Swift is a successor to
the C and Objective-C languages." - https://developer.apple.com/swift/).
We just need a successful mainstream OS vendor to push a RC/GC enabled
systems language to anyone targeting their OS, to finally break the
stigma that GC enabled systems programming languages don't leave the
research lab.
--
Paulo