---

• Replace a general function by a simpler function, and perform the computation exactly for the simpler function.
• Given an expression in terms of something simple plus a remainder, generate a numerical approximation by dropping the remainder term. (and adjusting the notation, perhaps)
  • This is where we introduce Truncation Error
• Use computable estimates of quantities that cannot easily be computed.

These thoughts taken from An Introduction to Numerical Methods and Analysis, by James F. Epperson, John-Wiley and Sons, 2002.

---

When attaining bounds for error analysis:

• The bound itself is usually the least important part of an error analysis. The main object of such an analysis is to expose the potential instabilities, if any, of an algorithm so that, from the insight thus obtained, one may be led to improved algorithms.

This thought taken from Numerical Linear Algebra and Optimization, by P. Gill, W. Murray and M. Wright, Addison Wesley, 1991. The authors credit this remark to J.H. Wilkinson -famous (in certain circles any way) Numerical Analyst.

---

• Finality is not attainable in science.

This thought taken from Relaxation Methods in Theoretical Physics, by R.V. Southwell, Oxford at the Clarendon Press, 1956.

---

• Mathematics is the Queen of Science, and Arithmetic the Queen of Mathematics - Carl Friedrich Gauss (famous mathematical BIG DOGG)

This thought given to me by Dr. Jennie Luebeck right here in our own department.

---

• Nobody likes a math geek, Scully.

This thought taken from Agent Fox Mulder , X-Files Episode, circa the Y2K episode.

---

• Just because you CAN fix something doesn't mean that you HAVE to.

This thought taken from Christine Latulippe , over tea and a cookie at the Leaf and Bean.

---