…4732)

A new note explains that `emplace-deref` does not require materialization
of the result of `*i` before the initialization. The normatively expressed
requirement in terms of an invented initialization of a variable was prone
to confusion, since such a declaration would indeed require `*i` to be
copyable/movable, whereas the actual use of `emplace-deref` does not
cause any intermediate materializations of prvalues.

…his object"

Avoid saying "this object" when not talking about `*this`.

…re accurately

Class templates are not types. Function templates are not function signatures.

The definition of 'common initial sequence' in [class.mem] only applies
to standard-layout struct types, which excludes unions. There is no
reason to define is_corresponding_member in terms of standard-layout
types (which includes unions and scalars and arrays of such types) when
the common initial sequence is only meaningful for standard-layout
structs.

The former includes names introduced by alias-declarations,
the latter (arguably) does not.