Language designer's notebook, Quantitative language design

Using real-world data to drive language evolution decisions

From the developerWorks archives

Brian Goetz

Date archived: December 19, 2016 | First published: April 12, 2010

For any given programming language, there is no shortage of new feature ideas. Language designers must not only perform the difficult task of deciding which of many possible (and often incompatible) language features should receive priority, but they also must consider that new language features can interact with existing ones in surprising, and sometimes incompatible, ways. Language evolution often requires making a trade-off between the benefits of enabling desirable new patterns of coding and the costs of potentially breaking some existing "weird" code. In this situation, being able to quantify — using real-world data — just how unusual that "weird" code is can provide valuable clues to which way a decision should go.

This content is no longer being updated or maintained. The full article is provided "as is" in a PDF file. Given the rapid evolution of technology, some steps and illustrations may have changed.

Zone=Java development
ArticleTitle=Language designer's notebook: Quantitative language design