Scope
The essential features that characterize an Ed-Fi REST API implementation are the data model that serves as its basis and the REST architectural style.
Data Model
The Ed-Fi Unifying Data Model (UDM)[1] provides the basis for the data transferred and manipulated by an Ed-Fi REST API implementation. The Ed-Fi UDM is a structured, conceptual model of common K–12 education data. The model includes entities that are easily recognized by educators and administrators: schools, students, teachers, attendance, grades, assessment results, and many others. These entities contain attributes (i.e., properties) that are also easily recognized. For example, assessment results contain data, such as a score and the date the assessment was administered. The UDM also includes associations (i.e., relationships) between entities, such as the association between students and schools.
REST interfaces are built around Resources that define nouns. In the education domain, these nouns include such things as schools, students, and teachers. In the Ed-Fi UDM these nouns have been rigorously defined as “entities,” with specific attributes and associations. Compositions of entities, with their attributes and associations, are called “domain aggregates.” These are identified from the Ed-Fi UDM according to the principles of Domain-Driven Design (DDD).[2] Domain aggregates are the Resources for an Ed-Fi REST API. These concepts are discussed in more detail later in this document.
An Ed-Fi REST API may cover a subset of the full Ed-Fi UDM that is exposed and exchanged in a particular system or implementation. The API need not be implemented for the entire scope of the Ed-Fi UDM in order to be aligned.
Architectural Style
The REST architectural style[3] is a convention-based approach to defining APIs. HTTPS (Hypertext Transfer Protocol Secure), using the HTTP operations (GET, PUT, POST, DELETE, etc.), is used as the application protocol.
REST-style architectures consist of clients and servers. Clients initiate requests to servers; servers process requests and return appropriate responses. Requests and responses are built around the transfer of representations of Resources. As depicted below, a data store or server-based application implements, exposes, or is wrapped with an Ed-Fi REST API to allow client applications to exchange and manipulate education data.
Figure 1. Interaction between REST API and client application
APIs can be thought of as a “contract” between data sources and client applications. The underlying platform and application choices are unimportant in terms of this contract. An Ed-Fi REST API follows this pattern. An Ed-Fi REST API levies no technical requirements on how data is internally stored or how it is used by client applications. The API must only provide the technical contract between a provider of data and its consumer applications, externally representing the exchanged data Resources in a way that is aligned with the Ed-Fi UDM.
The same resource may be represented to different clients using different representations. For example, an API may represent a resource as JSON for an application that is performing transactions, but may use XML to represent the same resource to another application for bulk data export. The representation is a way to externally represent the resource, but is not the resource itself.
There may be circumstances where an Ed-Fi REST API would diverge from a pure REST approach to support specific use cases, for example, to support application-specific operations.
While the Ed-Fi XML Data Exchange Framework allows for file-based exchange of education data between systems, an Ed-Fi REST API extends this capability to include real-time and transactional exchanges of information.
1 For more information about the Ed-Fi Unifying Data Model, see here.
2 See, e.g., Evans, Eric, et al. (2006), Domain-Driven Design Quickly, C4Media Inc., for a brief outline of Domain-Driven Design principles.
3 The key principles of REST are outlined in Fielding, Roy Thomas (2000), Architectural Styles and the Design of Network-Based Software Architectures, Doctoral dissertation, University of California, Irvine.