If D becomes a failure, what's the key reason, do you think?

[Sean Kelly]

Don Clugston wrote:
> Walter Bright wrote:
>>
>> So the question I have is, does D 1.0 have to be bug-free? I don't 
>> think  so.
> 
> No, it doesn't. But for developing libraries, when there are bugs like 
> that with serious implications for library design, you have to make a 
> decision. If it's a bug that won't ever be fixed in 1.0, it's worth 
> spending a lot of time developing a workaround. But, if it's a 
> limitation that's likely to disappear in the next few compiler releases, 
> you're wasting your time. Major bugs actually shape the language; in the 
> C++ world, Visual C++ 6.0 had so many template bugs that it was a 
> language all of its own. So anything you can tell us about what your 
> current vision of 1.0 is, will be helpful.

Exactly.  For ifti I've been operating under the assumption that it was 
intended to be like C++ and have simply been putting off any serious 
metaprogramming in D until the compiler matures a bit.  The 
clarification regarding aliases (that they aren't valid for referencing 
the result of an expression) was also quite valuable information.  About 
the only remaining issue for me is whether any other template features 
might be added, specifically:

* whether the template specialization mechanism will be extended to 
support type lists or 'is' expression logic

* whether D will gain a structured way to identify or specialize for 
static arrays (currently, specializing for T[] seems potentially not useful)

Hrm... that's all I can think of at the moment.


Sean