std.getopt suggestion

[Jonathan M Davis]

On Sunday, October 02, 2011 01:35:09 David Nadlinger wrote:
> On 10/2/11 1:01 AM, Andrei Alexandrescu wrote:
> > On 10/1/11 4:17 PM, Nick Sabalausky wrote:
> > [snip]
> > 
> > Not buying it. Sorry.
> > 
> > Andrei
> 
> Sorry Andrei, not buying your opinion either.
> 
> David
> 
> 
> 
> And how exactly did this get us any further? :P

I think that it's quite clear at this point that no one is going to convince 
Andrei and that Andrei is not going to convince anyone else. I think that it's 
horrible that we have mutable global variables in std.getopt, but the actual 
impact in this particular case is fairly minor, and I don't think that it's 
worth continuing this argument that is clearly going nowhere. If it were a new 
module, we could vote against its inclusion in Phobos, but it's already in 
there and has been for some time. And since Andrei is the original author, and 
I'd prefer to not just change things like that when he's clearly in strong 
opposition to it (particularly in a module that _he_ wrote), we appear to be 
pretty much stuck with how it is. If anything, this is an example of why 
having a formal review process can be valuable.

- Jonathan M Davis