On Sat, 14 Jul 2012 14:39:04 -0400, John Gilmore wrote: > >Some years ago this situation changed dramatically. Mike >Cowlishaw---he who designed REXX---devised what is now ANSI decimal >floating point (DFP). DFP behaves consistently in ways that do not >surprise accountants. (All three floating-point formats are supported >by zArchitecture hardware.) > I understand (I have very little firm knowledge of these things) that DFP supports setting an absolute rather than relative rounding point, and that is required by GAAP. I don't know that Rexx has such support (but I haven't sought it -- I'm a numeric rather than accounting kinda guy. And I notice you merely cited a common authorship without asserting that Rexx has the facility).
Much of Rexx's numeric architecture was driven by Cowlishaw's objective of supporting no data type other than string in Rexx. Surely there must be GAAP conformance testing suites. Have DFP, Rexx, Cobol, PL/I, etc. been vetted against such? The multiplicity of choices for rounding and truncation provides a strong incentive for OO design. I could imagine different classes with methods for the various rounding conventions. -- gil ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
