Article Preview
TopIntroduction
As a new generation of data exchange language, JSON (JavaScript Object Notation) has the advantages of being lightweight, having fast parsing speed, and high transmission efficiency compared with XML (eXtensible Markup Language) (Lin et al., 2012). Nowadays, JSON has been widely used in various application fields. A distributed evolutionary computing system, for example, is designed and implemented in (Merelo-Guervós et al., 2008), which is based on the usability and efficiency of asynchronous JavaScript and JSON to improve the utilization of computing resources in distributed systems. In addition, the JSON network interface is based on its concise grammar format and efficient transmission, which can simplify message communication between heterogeneous systems (e.g., ACT-R and EPIC1). With Big Data's development, NoSQL (not only SQL) databases have been widely studied and applied. The document databases in NoSQL databases (e.g., MongoDB and CouchDB), for example, are designed and implemented based on JSON (Boicea, Radulescu, & Agapin, 2012). Moreover, JSON adopts a language-independent text format, which makes it easier for ones to read and write and at the same time retains the grammatical habits similar to the C programming language. These characteristics have made JSON an ideal data exchange language.
Time widely exists in various practical applications. Typically, data generated by applications are closely associated with time: data may be available or correct at a certain time point or a given time interval (Dyreson et al., 1994). In addition, some historical data need to be recorded. From a philosophical point of view, time is ubiquitous. For example, for Bunge, a state of anything is a list of its properties at a particular moment in time. Hence, every application would require the consideration of time, explicitly or implicitly. Therefore, it is essential to manage temporal data (Nascimento, Sellis, & Cheng, 2015; Claramunt, Schneider, & Wong, 2017; Zheng, Li, & Shim, 2018). A core issue in temporal data management is establishing a temporal data model (Nascimento, Sellis, & Cheng, 2015). Some temporal data models have been proposed for representing the temporal semantics of data. Traditionally, temporal data management is mainly based on temporal relational databases (Tansel and Tin, 1998; Tansel et al., 1993; Chomicki, 1994; Clifford et al., 1995; Clifford, Croker, & Tuzhilin, 1996). With the emergence of new application paradigms and data models, several temporal data models are proposed beyond the temporal relational databases. In the context of the Web, for example, temporal XML models are developed (Wang, Zhou, & Zaniolo, 2004; Rizzolo and Vaisman, 2008; Snodgrass et al., 2008), and temporal RDF (Resource Description Framework) models are proposed (Gutiérrez, Hurtado, & Vaisman, 2007; Wang and Tansel, 2019). Furthermore, to deal with Big Data with time (Cuzzocrea, 2015), temporal NoSQL database models are devised (Hu & Dessloch, 2015).