Decimal floating-point: algorism for computers | IEEE Conference Publication | IEEE Xplore