Predicting OSS trustworthiness on the basis of elementary code assessment

Background. Open Source Software (OSS) provides increasingly serious and viable alternatives to traditional closed source software. The number of OSS users is continuously growing, as is the number of potential users that are interested in evaluating the quality of OSS. The latter would greatly benefit from simple methods for evaluating the trustworthiness of OSS.

Objective. This paper aims at finding a quantitative relationship between the perceived quality of OSS and a few simple objective measures.

Method. the users' and developers' evaluations of trustworthiness and reliability of OSS products were collected and correlated to static code measures, called "Elementary Code Assessment" rules, which check very simple rules that well-written code should satisfy.

Results. The result of the analysis is a set of quantitative models that link static measures of the source code to perceivable qualities of OSS. These models can be used by: 1) end-users and developers that would like to reuse existing OSS products and components, to evaluate the level of trustworthiness and reliability that can be expected based on the characteristics of code; 2) developers of OSS products, who can set code quality targets based on the level of trustworthiness and reliability they want to achieve.

Conclusions. The perceivable quality of OSS seems to be predictable on the basis of simple static code measures. However, only a part of the many measures produced by tools appears actually correlated to the quality of software that are perceivable by users.