The Programming Language Wars

  • Andreas Stefik | University of Nevada, Las Vegas

Modern society is built on software platforms that encompass a great deal of our lives. While this is well known, software is invented by people and this comes at considerable cost. Notably, approximately $331.7 billion are paid, in the U.S. alone, in wages every year for this purpose. Generally, developers in industry use programming languages to create their software, but there exists significant dispersion in the designs of competing language products. In some cases, this dispersion leads to trivial design inconsistencies (e.g., the meaning of the symbol +), while in other cases the approaches are radically different. Studies in the literature show that some of the broader debates, like the classic ones on static vs. dynamic typing or competing syntactic designs, provide consistent and replicable results in regard to their human factors impacts. For example, programmers can generally write correct programs more quickly using static typing than dynamic for reasons that are now known. In this talk, we will discuss three facets of language design dispersion, sometimes colloquially referred to as the “programming language wars.” First, we will flesh out the broader impacts inventing software has on society, including its cost to industry, education, and government. Second, recent evidence has shown that even research scholars are not gathering replicable and reliable data on the problem. Finally, we will give an overview of the facts now known about competing alternatives (e.g., types, syntax, compiler error design, lambdas).

Series: Microsoft Research Talks