Lambda the Ultimate

The Programming Language Wars: Questions and Responsibilities for the Programming Language Community by Andreas Stefik & Stefan Hanenberg

video of presentation @ CodeMesh 2014
slides in PDF format
the paper in pdf

A presentation of the results of a number of randomized controlled trial studies of the affects programming language designs on users.

He spends a great deal of time on the need for the programming language creation community to do more of these kinds of trials and why he thinks they are so necessary.
He also spends some time defending the expected attacks that either their goal is either one language for all or a different language for everyone, both of which seemed like odd accusations to me.

Among the conclusions are

• On average, static typing improves developer productivity under a wide variety of conditions
• Threads, compared to software transactional memory, leads to approximately 8-fold more bugs among those in the sample
• Some languages (e.g., Perl, Java) have symbols so poorly chosen that a randomly generated language is just as easy for a novice to initially use
• Overall, measurements of the impact of Aspect Oriented Programming appeared to be limited and most useful in simple situations like logging
• C style syntax hurts comprehension compared to alternatives

He says that languages shouldn't be created or changed without this kind of research to back up the ideas.

Only half the interesting detail is in the slides.

questions start @ 34:00

the Perl vs java research wave previously discussed as Perl vs. Random Syntax here on LTU