I dont think that such an aproach will succeed.
Defining a programming language and writing a lexer/parser is a
compromise between conventions used by humans and technics to
analyze a program automatically. Analyzing expressions with infix
operators with priority and associativity is an easy task. Humans
and computers can do this without problems.
OTOH: When something is complicated to parse by a program it is
probably also hard to understand by a human. Such things must be
investigated in detail before they are included in a language.
Especially technics which claim that they read the mind of the
programmer are suspicious. There are programs that pretend to be
smarter than the user and do some corrections (e.g. spelling)
without being asked to do it.
Most heuristics are not obvious and fail to some degree. In programs
this could introduce bugs.
Complicated algorithms like type inference can involve big parts of
a program and a human must also use the same algorighm and the same
big parts of the program just to determine a type.
Greetings Thomas Mertes
Seed7 Homepage: http://www.yqcomputer.com/
Seed7 - The extensible programming language: User defined statements
and operators, abstract data types, templates without special
syntax, OO with interfaces and multiple dispatch, statically typed,
interpreted or compiled, portable, runs under linux/unix/windows.