I hope WG1 will reconsider lifting the restriction that "any internal syntax definitions in a body must come before any internal definitions."
Sometimes I write macros that expand into both definitions and syntax definitions. If this restriction stands I won't be able to reliably use those macros in <bodies> (and nor would anyone else).
I suggested that this rule only applied to overt definitions, not ones that were the product of macro expansion. Andre van Tonder points out that without macroexpanding, you cannot tell which forms are syntax definitions, which are definitions, and which are neither.