The following paper
http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2009/n3010.pdf
proposes to separate expressions into 3 kinds: rref lvalue, non-rref
lvalue, and rvalue.

I can imagine 4 kinds of an expression:
1) ref lvalue (which is non-rref lvalue in N3010)
2) non-ref lvalue (new kind)
3) ref rvalue (which is rref lvalue in N3010)
4) non-ref rvalue (which is rvalue in N3010).

According to this concept, every expression is either an lvalue or an
rvalue, and every expression is either a ref expression or a non-ref
expression.

-- An rvalue expression designates an entity that shall be treated as
temporary (for example: an enumerator, an object returned by value or
by rvalue reference). Any other expression is an lvalue (for example:
a function, a local variable, an object returned by lvalue reference,
a bit-field).

-- A ref expression is a reference or an addressable expression that
can be converted to a reference type using reinterpret_cast (for
example: a function, a local variable, an object returned by lvalue
reference). Any other expression is a non-ref expression (for example:
enumerator, bit-field, an object returned by value).

Like some rvalues, bit-fields are not addressable. However, bit-fields
can be treated as temporary and as non-temporary:

     struct X
     {
         X() : m(1) {}
         int m : 10;
     };

     template <class T>
         void f(T &);
     template <class T>
         void f(T const &) {}

     int main()
     {
         X x;
         f(X().m); // OK: X().m is an rvalue
         f(x.m);   // ill-formed: X().m is an lvalue
     }

This example demonstrates collision between two concepts of lvalue:
designation of an addressable entity and designation of a possibly non-
temporary entity. It seems, the model with 4 kinds of expressions
looks more consistent. For the purposes of the overload resolution non-
ref lvalues could be treated like rvalues.

--
[ comp.std.c++ is moderated.  To submit articles, try just posting with ]
[ your news-reader.  If that fails, use mailto:std-c++@netlab.cs.rpi.edu]
[              --- Please see the FAQ before posting. ---               ]
[ FAQ: http://www.comeaucomputing.com/csc/faq.html                      ]