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

[Dave]

Bruno Medeiros wrote:
> Walter Bright wrote:
>> Kyle Furlong wrote:
>>> *Standing Ovation*
>>
>> Yeah, that's concerned me as well. But it isn't just me trying to make 
>> it perfect, everyone's got their favorite bug/feature that must get in 
>> before 1.0.
>>
>> So what do you say we just call D right now *1.0* and move on? It's 
>> not like D will stop undergoing improvements.
>>
> 
> Here comes the contrarian view:
> I think the people that want 1.0 ASAP (I'm not one of them btw) want 
> more than just the branding "1.0", they want some guarantees that the 
> language is good enough to be usable as is, and they likely also want 
> "1.0" to mean that 2.0 won't be radically different from 1.0 . For 
> example, let's consider this:
> 
> clayasaurus wrote:
>  >
>  > This will make two groups of people happy,
>  >
>  > #1) People who are waiting for D 1.0 for very large / commercial
>  > products, as well as perhaps a D 1.0 book to start the publicity tour
>  >
> 
> In this case of wanting to write a 1.0 book or doing very large 
> commercial products, then "1.0" actually should indicate that the 
> language is good and polished enough as is. They want a finished 
> product, and likely also want that 2.0 won't be radically different from 
> 1.0, so that the book won't quickly become mostly obsolete, or that the 
> large-scale product will need a lot of work to be updated to 2.0 .
> 
> Taking too long to reach a true 1.0 is a slightly bad in my opinion, but 
> I think it is *much* worse to shove a "1.0" product that is flawed, 
> unpolished, inconsistent, incomplete, etc.. And as is clear to all here, 
> D still has many design issues that need to be worked out (not to 
> mention Phobos).
> 
> 

Good points, but as with everything there is a tradeoff between polish 
and release and IMHO there's been just about enough polish. I think the 
general consensus is that D is currently good enough *if*: The import 
usability issues and the broken/inconsistent protection attribute issues 
are taken care of.

There are also some who'd like some form of 'const' but the trick is or 
will be to make that better than what C++ offers (and readily 
implementable). Heck, I'd be happier than a pig in mud if we could also 
get some form of 'const' reference-type function arguments, but that can 
wait for v1.1 if need be <g>

- Dave