Naming Things

There are 2 hard problems in computer science: cache invalidation, naming things, and off-by-1 errors. This session is about naming things. In software, we need to name classes, methods, properties, libraries, and solutions. We also have to give names to variables, parameters, namespaces, interfaces, services, resources, etc. And most of the time, they can't conflict. Sometimes it's easy but sometimes it can be hard. We'll do a quick review of some of the easy guidelines that avoid some (somewhat easy) decisions then look into the things that require names that involve a deeper understanding of context. What the contexts are, their granularity, and how that applies to alleviating naming issues will be covered. How the types of contexts and their motivation and good compartmentalization makes for easier naming will be detailed. We'll dig into some details of English grammar that should relegate naming is hard as a code smell—something that hints at deeper problems in the structure of code. And finally, time permitting, some techniques and ideas to relegate *naming is hard* to problems of the past will be outlined.