A Mathematician looks at D

[so]

On Tuesday, 19 February 2013 at 00:57:31 UTC, Joshua Niehus wrote:
> On Monday, 18 February 2013 at 23:55:46 UTC, bearophile wrote:
>> Most exploratory mathematics systems have a REPL, because for 
>> some people and for some kinds of problems, it's much better 
>> to have it. It's not for everyone nor for every kind problem.
>
> I'll take your word for it, in my narrow experience, I've found 
> REPLs slow me down.
> And as for graphing math problems, I found MatLab, Mathematica 
> or even OSX's "Grapher" to be sufficient for my usages (when I 
> was a student).
>
> In any event, I wouldn't consider a language "rubbish" because 
> it doesn't have a REPL.

Assuming we are talking about same thing (as all language trying 
to replicate the lisp based REPL, and failing, it is possible you 
haven't seen the real thing), it can't possibly slow you down, 
quite quite (could add more) the opposite. It is not forced, it 
is a gift. You hit the same compile key but you don't have to 
compile everything, you can compile a single function or a file 
and you don't have to restart anything, you can do all this in a 
running application and work as you see the changes take effect.