Good names act as an introduction to a domain's core concepts, making code more approachable for newcomers.
We process code using the same mental shortcuts as natural language, creating meaningful chunks and making assumptions based on names.
A function's name must not misrepresent its behavior, such as hiding side effects or returning unexpected types.
Overly general names like 'process' or 'handle' obscure a function's true purpose and should be replaced with more descriptive alternatives.
Maintain clarity by assigning exactly one name to each concept, avoiding both synonyms and using one name for multiple different ideas.
The Knight Capital case illustrates how a seemingly small naming ambiguity, when combined with other system faults, can lead to massive financial loss.
APIs that are consistent with domain nomenclature or established frameworks are easier for developers to learn and use correctly.
Names must be re-evaluated and updated when underlying technology or business domain concepts change to prevent them from becoming misleading.
The core principles of good naming are to avoid being misleading, consider the reader's perspective, respect context, and maintain names over time.
Bitpanda
Vienna, Austria
39:49Mustafa Toroman
24:52Hendrik Lösch
21:39Johannes Hansen
27:27Christian Seifert
43:13Carola Lilienthal
25:03David Whitney
28:17Hendrik Lösch
26:59Benjamin Bischoff
Jobs that call for the skills explored in this talk.
Wolters Kluwer
Alphen aan den Rijn, Netherlands
