D parsing

[Andrei Alexandrescu]

On 11/12/13 12:02 AM, Jacob Carlborg wrote:
> On 2013-11-12 08:52, Andrei Alexandrescu wrote:
>
>> Fine, although a sense of futility is hard to shake seeing as we won't
>> replace those existing features. I think a much stronger point would be
>> made if the power of the feature were demonstrated on problems not
>> accessible with the existing ones.
>
> You just said we shouldn't replace existing features.
>
> "The point here is that there is
> significant difficulty to remove features that already exist"
>
> http://forum.dlang.org/thread/bwsofbnigfbrxwouiobj@forum.dlang.org?page=9#post-l5s44b:242c36:241:40digitalmars.com

Yes. So I said. I don't get why you'd provide a link - it's in my text 
that you quote. Indeed, we shouldn't replace existing features.

>> About DIP 50: I will say "no" but please do not take it personally. It's
>> great there is discussion about this, and I encourage it, but at this
>> time I think we should make no plans to add macros to D.
>
> I don't think we should add macros now either. This proposal is far from
> ready. If Martin hadn't suggested I should create a DIP, I wouldn't
> have, at least now at this time.

Fine.

> BTW, just saying "no" doesn't help a bit. You could just have said
> "foo". That's extremely annoying. You're shooting down very many
> suggestions/proposal/ideas with just a "no", or the more elaborated
> answer "no, never going to happen".
>
> On the other hand when you have a proposal it should be consider
> pre-approved and is a more of a FYI.

So how could we express a "no" that doesn't annoy you in the extreme? In 
case the answer would be "you haven't explained why", allow me to retort.

I've mentioned the argument before: at this point we should focus on 
quality of implementation and making good use of the features we have. 
In fact I am repeating myself: http://goo.gl/1thq1j. As has been 
publicly known for a while now, our strategy has been to improve quality 
and to double down on the assets we have. People ask for a roadmap, and 
what's missing from a roadmap is as important as what's there.

This is a strategy that Walter and I agree with, have been transparent 
about, and that may work or not, with various degrees of success. 
Reasonable people may disagree what the best step moving forward should 
be, but at some point some step must be made and we can't adopt your 
strategy, with which we disagree, as our strategy, just to be nice and 
not offend your sensibility. (I'm using "we" here because Walter and I 
discussed this at large.) There must be a way to say "no" that doesn't 
offend you. Please advise what that is.


Andrei