Phobos addition formal review: std.experimental.allocator

[jmh530 via Digitalmars-d]

On Friday, 26 June 2015 at 13:35:34 UTC, Dicebot wrote:
It is ok to use that time for any
> additional comments :P

I would second the suggestion for improvement in the 
documentation. The overview linked to in the original post is 
written assuming advanced programming knowledge. My understanding 
of allocation is pretty much limited to something could be on 
stack or heap (I had to Google what an allocator was). So I would 
say it could be improved by adding some explanation of the basics 
and why to use allocators instead of new or other techniques.

I had seen this when it was originally posted two weeks ago, but 
I came back around to read it again today. What brought me to 
look it over again was that I had come across the 
std.container.array package and I didn't understand how it was 
different from the built-in array (I still don't have a very good 
understanding there, but a little better). Anyway, as I was 
searching through the Learn forum, I came across a post from 2011 
saying that std.container might change in the future based on 
Andrei's work on allocators. I take it this is the work it was 
referring to (especially after looking at the wikipedia page for 
allocators).

With that I think I only have two questions. Am I correct that 
the makeArray function is for allocating a dynamic array and not 
an std.container Array? Are there plans to refactor std.container 
with std.allocator in mind?