I was kind of wondering the same thing about parsing an MS-Word document into HTML on a linux box...Any ideas?
I imagine that if you can parse Excel, Word would be a similar function in the same package.
I can assume, however that, since MS-Office file formats are propietary, the chances of finding a solution are slim to none...
However, you might want to try asking your users to store the data as a CSV (Comma-Separated-Value) file. I do know this option is available in Excel and might do the trick...
Chris