From: Aleksey Gurtovoy (agurtovoy_at_[hidden])
Date: 2002-07-25 05:42:07
Paul A. Bristow wrote:
> Minor observations:
> 1 I find iter_fold a most unhelpful and obscure name.
I think it's not much worse than 'iter_swap', which somehow got into the
> Would iter_operate be better?
I would prefer to keep 'fold' as a part of the name.
> 2 Table of Content should surely be plural Contents? (Occurs twice?)
Right, thanks. I always forget which one is the right spelling.
> 3 I find the _c suffix less than obvious. (The paper uses it
> without prior definition which left me fast forwarding to find what it
Will try to make a more gentle introduction of it :)
> It would be more in keeping with current trends towards full words to
> make the suffix _constant.
It would ruin the essential shortness of many names such as 'int_c'. I am
afraid one or another suffix notation is unavoidable here.
> hat does _impl mean?
"implementation". Remember Sutter's "Pimpl"? :)
> Why is forward abbreviated to fwd but backward spelt out?
> 4 All the documentation would benefit greatly from MANY more
> These could be provided in-line, or as hyperlinks.
Just to clarify, do you mean the examples like at the end of the reference
page of every component, or something else?
> I find the examples much more easily understood than the
> rather standardese descriptions
> (though these are essential).
> 5 The structure of the documentation left me confused. It
> doesn't seem to start with a top-level index.html and link to
> everything else?
It does, actually, only 'index.html' has the contents on its own :). I'll
look into reorganizing it.
> 6 Comments do not follow the style guide of starting with
> capital letter and ending with a period.
I do not subscribe to this one :).
> Overall I support adoption of MPL. It appears to have great potential.
> Being based on C++ and STL are great strengths, but MPL will
> need a lot of selling through more and better documentation and
> examples to really catch on.
Thank you for your comments,