Using Readability Metrics in Estimating the Readability of REpresentational State Transfer State Transfer Uniform Resource Identifiers Schema

Abstrak

Uniform Resource Identifiers (URIs) may have a direct impact on the understanding of REpresentational State Transfer State Transfer (RESTful) functionality, and thus, on the discovery of final RESTful product. RESTful Web Services (WS)/Application Programming Interfaces (APIs) are designed to expose data and functionality through resources accessed by dedicated URIs over HyperText Transfer Protocol (HTTP), which recently represents the direct descriptions schema of what functions does the concerned RESTful WS/API present. Furthermore, the discovery of suitable RESTful is heavily rely on the simplicity of understanding their URI schemas, which recently suffer from critical issues in how to measure their readability. For that, WS/APIs developers aspire to measure the readability of RESTful URI schemas before exposing them over the Internet to estimate their usability. Consequently, this research proposes four readability metrics for the stated purpose namely: Flesch-Kincaid (F-K), Flesch Reading Ease (FRES), Simple Measure of Gobbledygook (SMOG), and Coleman Liau Index (CLI). The research identifies the variables required to calculate the readability metric and formulate the equations for them. Four experts in linguistics were asked to validate the proposed metrics and their identified variables. The research successfully conducted empirical research on 8 well-known RESTful WSs/APIs of the dataset, and the proposed metrics were implemented on 6952 URIs schemas. The average values for the aforementioned metrics were 7.41%, 59.63%, 6.73%, and 17.55% respectively, where in certain metrics, a low average value signifies easy readability, but in others, it signifies hard readability, and vice versa.