Can We Crowdsource Language Design?

Preston Tunnell Wilson, Justin Pombrio, Shriram Krishnamurthi

SPLASH Onward!, 2017

Abstract

Most programming languages have been designed by committees or individuals. What happens if, instead, we throw open the design process and let lots of programmers weigh in on semantic choices? Will they avoid well-known mistakes like dynamic scope? What do they expect of aliasing? What kind of overloading behavior will they choose?

We investigate this issue by posing questions to programmers on Amazon Mechanical Turk. We examine several language features, in each case using multiple-choice questions to explore programmer preferences. We check the responses for consensus (agreement between people) and consistency (agreement across responses from one person). In general we find low consistency and consensus, potential confusion over mainstream features, and arguably poor design choices. In short, this preliminary evidence does not argue in favor of designing languages based on programmer preference.

Paper

PDF


These papers may differ in formatting from the versions that appear in print. They are made available only to support the rapid dissemination of results; the printed versions, not these, should be considered definitive. The copyrights belong to their respective owners.