- From: Henry Zongaro <zongaro@ca.ibm.com>
- Date: Thu, 1 Apr 2004 16:21:36 -0500
- To: public-qt-comments@w3.org
Hello, The following is a comment from IBM on the Last Call working draft of Formal Semantics. Section 4.4 The normalization rule for the unary arithmetic operators is problematic for values of type xs:float and xs:double. If the operand of unary minus is positive zero, the result produced using the usual IEEE.738 rules for arithmetic ought to be negative zero, but the result of the normalized operation 0.0-(+0.0) will be positive zero. Similarly, if the operand of unary plus is negative zero, the result ought to be negative zero, but the result of the normalized operation 0.0+(-0.0) will be positive zero. Thanks, Henry (on behalf of IBM) ------------------------------------------------------------------ Henry Zongaro Xalan development IBM SWS Toronto Lab T/L 969-6044; Phone +1 905 413-6044 mailto:zongaro@ca.ibm.com
Received on Thursday, 1 April 2004 16:22:10 UTC