Chris Crawford considers Zimmerman's definitions.
Zimmerman does justice to the task. Eschewing the conceit of formal definition, he concentrates on utility rather than form. The sole test of his success then lies in the answer to the question: how useful are Zimmerman's definitions? To what extent do they bring us closer to understanding the concoction of game and narrative? Unfortunately, the concluding suggestions he offers don't seem to get us very far; no grand answers leap from the page. Perhaps this is too harsh a standard by which to judge his contribution. Perhaps we should settle for a more lenient standard of judgment, to wit: had these ideas been widely accepted ten years ago, would we have been spared some of the many disastrous marriages of narrative and interactivity we have seen?
Consider branching stories on the computer. After many years and hundreds of attempts, most with dismal results, many old pros have abandoned this design concept (although it retains a hard core of followers). If we apply these definitions to branching stories, will we unearth a fatal flaw? I think not. Branching stories don't violate any of the terms of these definitions, nor do they run against the grain of the further elucidations Zimmerman offers.