I spent many years working for NASA on Oracle databases. At that time, in the late 1980's and early 1990's, spatial indexing was a novel and difficult idea. We worked heavily on the notion and, with significant help from Oracle, enabled our databases to store specially designed datatypes for spatial data.
Microsoft is researching similar or even greater capabilities: http://research.microsoft.com/research/pubs/view.aspx?msr_tr_id=MSR-TR-2005-122
At first, it wasn't immediately obvious to me who might want to store spatial data except perhaps an organization like NASA. But over the years, I grew to understand that many organizations the work with important data elements that are fixed in time and space (e.f. oil exploration, mining operations, real estate applications) can benefit.