The unigrids project is planning to roll the Usage Record into its monitoring system, and it is largely suited to that 
task. However, there are a few minor things that it would be nice to have changed.

1) Following the practices of GFD.58 is strongly recommended
2) Adding a generalized extensibility capability to the UR definition would be very nice, since it would permit non-
resource information to be described much more easily.  It is recommended that this be done through updates to the 
definition of the UsageRecord and JobUsageRecord elements such that each allows an arbitrary number of extra elements 
not in the UR namespace after the main content of the record. It is not recommended that this be done through updating 
the UsageRecordType type because that leads to problems with some XML deserializers (apparently; I'm no expert on this 
aspect). Thus, for example, the JobUsageRecord schema definition would become:
	<xsd:element name="JobUsageRecord" substitutionGroup="urwg:Usage">
		<xsd:complexType>
			<xsd:complexContent>
				<xsd:extension base="urwg:UsageRecordType">
                                     <xsd:sequence>
                                         <xsd:any namespace="##other" minOccurs="0" maxOccurs="unbounded" 
processContents="lax"/>
                                     </xsd:sequence>
                                     <xsd:anyAttribute namespace="##other" processContents="lax"/>
                                </xsd:extension>
			</xsd:complexContent>
		</xsd:complexType>
	</xsd:element>