Current relational database management systems (RDMSs), such Microsoft (MS) SQL Server, have the ability to sort data for major languages, but they do not have the ability to sort properly for many of the lesser-known languages around the world. Moreover, MS SQL Server sorting uses MS Windows collations, and collations cannot be added to Windows. A software developer is limited by the operating system.

We have chosen to use IBM's International Components (ICU) for Unicode to address these shortcomings. More specifically, we are using the Collation Services of ICU. The services can be "tailored" for different languages and locales; theoretically any language of the world can be sorted. We are accessing the Collation Services by extending the capabilities of the database. In the case of MS SQL Server, we do this by means of an "extended stored procedure". This extended stored procedure is a C++ project compiled as a dll.

Although MS SQL Server is used in this example, the approach can be used with other RDMSs. It has the potential to be a powerful factor in bridging the "digital divide" encountered by many computer users in developing nations.