Tell Me a Story Data?: The Linked Data Platform Use Cases and Requirements
The W3C (World Wide Web Consortium) Linked Data Platform Working Group is working on “a specification [that] describes the use of HTTP for accessing, updating, creating and deleting resources from servers that expose their resources as Linked Data. It provides clarifications and extensions of the rules of Linked Data.” Last week they released Linked Data Platform Use Cases and Requirements. This new Working Group Note builds on the working draft they released in October 2013 and aims to “motivate a simple read-write Linked Data architecture” through a collection of stories and use cases.
We all love stories right? Unfortunately these are not the kinds of stories you are likely to read before you go to bed. They have provided 15 short user stories that touch on a variety of topics including maintaining personal and business relationships, enriching metadata, sharing binary resources, project membership information and managing cloud infrastructure. I am drawn, of course, to the fourth story in their list: Library Linked Data.
They begin by refering to the work of the Library Linked Data Incubator Group, a report that I reviewed here. They’ve organized their comments here under two broad headings: “digital objects cluster” and “collections.” Elements from these groups are then linked to the 9 Use Cases provided in their report.
In particular they relate the Library Linked Data “story” to these three use cases:
For each of these use cases they provide a “primary scenario” and an “alternative scenario.”
The ‘aggregate resources’ use case (UC6) talks about the need to add a new concept to an existing skos:ConceptScheme (primary) and also looks at a resource that is a member of more than one collection (alternative). Each use case provides an example of how the RDF might look like using Turtle notation.
‘Filter resource description’ (UC7) is also related to collections of resources. In this case providing the ability to retrieve parts of a collection without having to retrieve the full list of the collection. The alternative scenario here explores getting a list of all the items in a collection but retreiving a limited set of data elements, e.g. title, creator and links for each resource. This uses the FRBR ontology to allow navigation to the individual members of the collection. Very similar to a typical catalogue search where a results list provides a brief record that leads to the full record with links to the resources.
It’s a useful document providing “realistic examples describing how people may use read-write Linked Data” and gives a feel for the potential of the Linked Data Platform (LDP) they are developing at W3C. They avoid describing the details of the protocols or the ontologies that might be used which is meant to enable easier digestion of the use cases provided. Personally I was a little disappointed not to see some elaboration of how these processes might work in the context of these stories. Perhaps these will be fleshed out in the future.
More information on the Linked Data Platform including discussion of the HTTP protocol is available as part of the Working Group’s Working Draft. And the Working Group welcomes comments on this Use Case document which can be sent to public-ldp@w3.org.
Comments are closed.