Scientific study on code readability – Looking for participants

5 points by cessor 11y ago ↗ HN

The Department of Psychology of Heidelberg University, Germany, is conducting an experiment on the psychology of programming.

We are researching how programmers read source code. At the moment we're running a pretest for the main study for which we need the support of programmers who are fluent in C#.

If you're a C# programmer and would like to help science, please follow the link below. Your task will be to find a bug in a piece of C# source code. The pretest will take approximately 20 Minutes.

We are very interested in your comments, so that we can further improve the design and the software used to collect data. Your input would be much appreciated, so that the final experiment can be further improved.

It is the first step of a larger project on the psychology of programming that will include experimental studies, eye-tracking, and electro-encephalography (EEG) studies. If you are interested in finding out more about this project, please contact me. You'll find my contact details on the provided website.

To participate, please visit the following url:

http://code.psychologie.uni-heidelberg.de

Regards,

Johannes Hofmeister

3 comments

[ 3.5 ms ] story [ 21.9 ms ] thread

andypalmer 11y ago ↗

Well, that was fun. It took a lot less than 20 minutes though

horstschneider 11y ago ↗

not sure I got that second thing right :D

dalke 11y ago ↗

I found two errors, so I did the one which was a more obvious flub, rather than the first one I found. But it made me worry about the rest of the code.

The spec is also underspecified for what to do on duplicates. The behavior of last-wins is quite common, but I would prefer a comment that highlights the implementation-defined resolution of that ambiguity.

With experience, I've learned to be very distrustful about what goes in query strings. Users put in unexpected values, and there can be security holes when, say, the input verification code resolves ambiguities one way and the actual code resolves it another way.