Schema languages are among the simplest formal languages we use. Yet even with their simplicity, they offer valuable lessons for programming language design. This presentation explores various schema languages, focusing on two main aspects. First, it examines the cognitive impact of using a unified syntax versus separate syntaxes for defining values and types, analyzing how these different approaches influence developers' cognitive load and ease of understanding. Second, the talk delves into the language-theoretical differences in the semantics of these schema languages, specifically comparing deterministic parsing schemas and nondeterministic validating ones. This involves analyzing how variations in their formal structures affect computational properties such as expressiveness and decidability. By investigating both the cognitive and theoretical dimensions, this talk aims to provide a comprehensive understanding of the trade-offs involved in the design and use of different schema languages.

Gary has a PhD from UTS, researching the design of programming languages from the perspective of innate human cognitive abilities. He is a full-time senior software engineer and chief scientist at The Helix Collective, a venture studio which specialises in getting start-ups off the ground.