Garfinkel. FEBS Letters, 2:S9-S13, 1969.

"Attempts are being made to establish a standard form of what a model should be like, how it should be described and documented, as well as to find out how efficient different computers are in computing its behavior. this is intended in part to facilitate communication of information about models, which may be difficult owing to their complexity. In building a model one has to keep track of why one made the decisions one did, what mistakes were made, what was learned along the way, etc. It may be desirable in describing a model to also include its unsuccessful predecessors, and the reasons for their failure to fit the data or other inadequacies."