Data consistency is very desirable because strong semantic properties make it easier to write correct programs that perform as users expect. However, there are good reasons why consistency may have to be weakened to achieve other business goals. In this CIDR 2009 Perspectives paper, we present real-world reasons inconsistency may be necessary, offer principles for managing inconsistency coherently, and describe implementation approaches we are investigating for sustainably scalable systems that offer comprehensible user experiences despite inconsistency.