Designing API for Using Publicly Accessible Data Sets
The paper explores two aspects of using data which is shared in clouds. First of all, authors present steps of data preprocessing which allow the usage of nearly each data source available in public data sets. Authors give hints in which situations it is recommended using replication and synchronization mechanisms. Furthermore, authors discuss possible ways of publishing own API from middleware tier for client application, mainly REST API, GraphQL and OData. Static REST API gives more control of the way the data is being queried. On the other hand, if you have many different client applications making use of data the GraphQL gives much more flexibility for API usage. OData gives common way query, filter and structure response data but adheres to server domain model. Our findings can be useful by organizations while exposing new data sets for public usage. Moreover, the authors summarize the paper and outline directions for further work.
KeywordsData sets Data sharing Application programming interface (API) REST API GraphQL OData
- 3.Byars, B.: Enterprise integration using REST. https://martinfowler.com/articles/enterpriseREST.html
- 8.Górski, T.: UML profiles for architecture description of an integration platform. Bull. Mil.Y Univ. Technol., LXI I(2), 43–56 (2013)Google Scholar
- 14.Postel J.: Transmission control protocol. IETF, RFC 761 (1980)Google Scholar
- 16.Requirements for Internet Hosts – Communication Layers. https://tools.ietf.org/html/rfc1122 RFC 1122 (1989)