Walter Bright Wrote:
> Derek Parnell wrote:
> > The current development model is no longer the best one for a better D.
> 
> If you put 100 people working on the D compiler, it probably wouldn't move forward any faster. Take a look at the D compiler progress vs compilers for other languages with teams working on them.

I agree the D compiler is already pretty good. But just 10-20 people working on it would definitely polish D to a new level. Please Walter, I urge you to some how allow more developers on the D compiler. If you can't manage people, you have to elect someone you really trust to take care of it and discuss the key things weekly.

Homer wrote:
> Walter Bright Wrote:
>> Derek Parnell wrote:
>>> The current development model is no longer the best one for a better D.
>> If you put 100 people working on the D compiler, it probably wouldn't move forward any faster. Take a look at the D compiler progress vs compilers for other languages with teams working on them.
> 
> I agree the D compiler is already pretty good. But just 10-20 people working on it would definitely polish D to a new level. Please Walter, I urge you to some how allow more developers on the D compiler. If you can't manage people, you have to elect someone you really trust to take care of it and discuss the key things weekly.

IMHO, I don't think the compiler is holding D back at all. Languages have had great success with compilers that aren't nearly as good.
Sure, there's many bugs to be fixed, but people can already submit patches. That's not a bottleneck.

Bigger limitations are in the rest of the tool chain, documentation, and
successful major projects. D needs polishing, more than anything else.

I'd even suggest that already, too much of the D community spends time worrying about language issues (such as const) when there's far more useful things that could be done.

On Thu, 2008-04-17 at 09:53 +0200, Don wrote:
> Bigger limitations are in the rest of the tool chain, documentation, and successful major projects. D needs polishing, more than anything else.
> 
> I'd even suggest that already, too much of the D community spends time worrying about language issues (such as const) when there's far more useful things that could be done.

I think this is evidence that a lot of issues still exist.  The language needs some polishing.

== Quote from Don (nospam@nospam.com)'s article
> I'd even suggest that already, too much of the D community spends time worrying about language issues (such as const) when there's far more useful things that could be done.

My worry with issues like const is that if the community is simply focused
on development then new features that are potentially not appealing will
be added as we'll be stuck with them once D 2.0 stabilizes.  Walter and
crew are very talented people, but they aren't infallible and neither do
they represent an accurate cross-section of the D community in terms
of desires or experience.  Were I in their place, I'd want as much feedback
as I could get.  Heck, I do for Tango :-)


Sean

Homer wrote:
> Walter Bright Wrote:
>> Derek Parnell wrote:
>>> The current development model is no longer the best one for a better D.
>> If you put 100 people working on the D compiler, it probably wouldn't move forward any faster. Take a look at the D compiler progress vs compilers for other languages with teams working on them.
> 
> I agree the D compiler is already pretty good. But just 10-20 people working on it would definitely polish D to a new level. Please Walter, I urge you to some how allow more developers on the D compiler. If you can't manage people, you have to elect someone you really trust to take care of it and discuss the key things weekly.

How hard would it be to make something that just translates D 1.0 source into intermediate C/C++ code before feeding it through the users compiler of choice? It would fix object-format incompatibility issues and we'd have a compiler with a real optimizer.

Sure its just papering over the issue, but its probably unrealistic to expect that theres going to be several vendors producing D compilers any time soon.

Neal Alexander Wrote:

> 
> How hard would it be to make something that just translates D 1.0 source into intermediate C/C++ code before feeding it through the users compiler of choice? It would fix object-format incompatibility issues and we'd have a compiler with a real optimizer.
> 

Far better to generate LLVM code