Measuring the understandability of WSDL specifications, web service understanding degree approach and system

Abstract

Web Services (WS) are fundamental software artifacts for building service oriented applications and they are usually reused by others. Therefore they must be analyzed and comprehended for maintenance tasks: identification of critical parts, bug fixing, adaptation and improvement. In this article, WSDLUD a method aimed at measuring a priori the understanding degree (UD) of WSDL (Web Service Description Language) descriptions is presented. In order to compute UD several criteria useful to measure the understanding’s complexity of WSDL descriptions must be defined. These criteria are used by LSP (Logic Scoring of Preference), a multicriteria evaluation method, for producing a Global Preference value that indicates the satisfaction level of the WSDL description regarding the evaluation focus, in this case, the understanding degree. All the criteria information required by LSP is extracted from WSDL descriptions by using static analysis techniques and processed by specific algorithms which allow gathering semantic information. This process allows to obtain a priori information about the comprehension difficulty which proves our research hypotheses that states that it is possible to compute the understanding degree of a WSDL description.