Factmint have worked with Telegraph Media Group (TMG) to create an umbrella API for all content access. This enables detailed subject-specific pages to be generated on the fly and published immediately.
The project uses a Factmint-powered ontology and a self-describing API provided as linked data (serialised as JSON-LD) to drive the creation of subject pages. Articles are tagged with data from the ontology, informing the system which entities they relate to; the ontology can then drive the creation of pages about an entity and other entities that are directly related. As an example, if a page about the actor Johnny Depp is generated, it will draw in data relating specifically to him, but also other entities that are related to him: these might include films he has appeared in, co-stars, and directors he has worked with. Related entities can be listed on a page and formatted as links permitting a reader to traverse the generated subject pages like nodes on a graph.
The ontology that provides information about the class of entities is also annotated with publication information, to inform the system as to which pieces of information should be used for page generation. The authoring process for these pages shifts from an editorial role to a data-modelling role, drawing in facts from data APIs and feeds.
The API is agnostic as to the subject of a page: the data that will be drawn in and the template used to format the layout of the page are dependent on the class of the entity that represents the subject in the ontology.
This project demonstrates the successful coupling of editorial, enterprise and CRM data.
To bring this about, a RedHat Fuse Enterprise Service Bus (ESB) has been used, with Apache Camel components integrating Factmint software. There is then an Apache CXF service layer with Swagger-generated documentation.