J3/13-311r1 To: J3 From: Van Snyder & Malcolm Cohen Subject: Editorial wording improvement Date: 2013 October 15 Reference: 13-211r1 1. Introduction --------------- The wording in 5.3.4p4 and 5.3.19p2 permitting an object to have the ASYNCHRONOUS or VOLATILE attribute within a BLOCK construct scoping unit might be confusing if there is some other requirement that prohibits the object from having the attribute. 2. Edits to 10-007r1 -------------------- {Avoid even the appearance of giving permission.} [90:21-23] Replace sentence "Within ... construct." with "Specifying the ASYNCHRONOUS attribute within a BLOCK construct for an object whose scope is wider than that of the BLOCK construct does not confer the attribute on the object whilst outside the BLOCK construct.". [102:14-16] Replace sentence "Within ... construct." with "Specifying the VOLATILE attribute within a BLOCK construct for an object whose scope is wider than that of the BLOCK construct does not confer the attribute on the object whilst outside the BLOCK construct.". Note: An alternative formulation using more words would be "If an object that is not a local variable of a BLOCK construct is specified to have the XYZ attribute in the of the construct, the object has the attribute within the construct even if it does not have the attribute outside the construct.". 3. No wart removal ------------------ The claim in 13-311 that the VOLATILE VALUE proposal was not added to 13-244r1 is mistaken; it was US proposal 11 in that paper. The further claim that it was "therefore not taken up at Delft" is also mistaken, as the entire WG5 committee explicitly voted to defer consideration of US-11 along with many other proposals: that vote passed without any dissent. ===END===