For those interested, James Renegar also had some great stuff on Newton's method ("A polynomial-time algorithm, based on Newton's method, for linear programming

For those interested, James Renegar also had some great stuff on Newton's method ("A polynomial-time algorithm, based on Newton's method, for linear programming" 1988). It seems there are a lot of great algorithmic opportunities in specializing Newton's method to a given problem domain.