In SQL Server 2008, you can have what is called a "Filtered Index".  What does this mean? 

An index may be created for a subset of rows in your table.  For instance you could create a filtered index for a "state" column if the value is "CA" for California; the rest of the states would not have a value.  When you create your index, simply add a WHERE clause to activate this feature. 

In the state example, it probably does not make sense to have an index just for California, but there are cases where this could save a lot of space.  For example, if a table had a column that for most rows had no data, a filtered index could be built on it where the column is not null. 

I used FoxPro in a previous life.  This was a feature available to us way back in the 90s.  We used FoxPro and this feature in DOS, WfW3.11, and SCO Unix.  It has been a while, but there was a time when we created an index just for this one particular customer because of the proceessing time it took.  Disk space was more scarce back then and filtered indexes helped us squeeze the most out of it.  I am happy to see this feature in SQL Server 2008. I have only started to play with this and expect that some limitations will pop-up.