Dr. Jones and I are probably more in agreement than his letter suggests. However, he appears to have conflated his deep distrust of guideline developers with the act of developing them and with the guidelines themselves. Here, we clearly disagree. Osler aptly defined medicine as "the art of probability and the science of uncertainty." Few medical questions are settled forever or, for that matter, for very long. Although the assertion that medical knowledge has "a half-life of five years" [1] may be extreme, the erosion of today's certainties makes practicing medicine difficult and explains its wide variability. Thus, I favor the development of guidelines for managing important clinical problems, regardless of the existence of controversy.

Dr. Jones prefers the term guideline "replacement" to the term "works in progress." This distinction is mainly semantic. Nonetheless, I view the process of guideline development as analogous to our education—a "work in progress" despite such milestones as diplomas or certificates of completion. Done correctly, guidelines can aid clinical decision making. Of possibly greater importance, to paraphrase Marshall McLuhan's famous dictum about the medium being the message, guideline development may be an instance in which the process may be the outcome. Developing guidelines forces physicians to come to grips with how little is truly certain and how much practice variability is reasonable (rather than inefficient, ineffective, or even harmful). It is this process that serves as a "beacon into the vast scientific unknown"—not the guidelines per se.