628	 On a Floating-Point Number Representation For Use with Algorithmic Languages
