Skip to content
rlaemmel edited this page Aug 20, 2012 · 1 revision

Software as a complex system

Proposal by Tom Mens

Can software-intensive systems be regarded as complex systems (as known in physics, biology, chemistry, sociology): a system composed of interconnected parts that as a whole exhibit one or more properties (behavior among the possible properties) not obvious from the properties of the individual parts? If no, why not? If yes, can results and ideas from complex systems science and network science be exploited to better understand, maintain and control software systems?

My personal response to the above questions would be "yes", since several researchers have already shown that many aspects of software systems witness power laws, scale-free network behavior, and the small-world phenomenon. It would therefore be interesting to discuss about these ideas as part of a working group.

Clone this wiki locally