Fundamental services for scholarly reference

This page is obsolete: it is superseded by CITE architecture.

This replaces the earlier overview of technical projects at CHS: phase 1, services for using classical texts

Background: an architecture for digital publication

Digital libraries are more than just collections of digitally published works: a digital library allows digitally published works to interoperate in terms of their fundamental semantics. (See a fuller discussion in "Digital publications for digital libraries.")

The architecture for these operations is a group of network services. Each service works with a particular kind of object, and provides operations appropriate for that object type. Simple objects serve as building blocks for constructing composite objects.

The most fundamental services in a digital library enable identifying, citing and retrieving scholarly citations: until an object can be cited, it cannot be used to create more complex objects.

We have defined and implemented services for two fundamentally different forms of citation: citation of canonical texts, and citation of objects in a collection. We have also defined and implemented a service for indexing information referring to either kind of reference. An enormous range of digital scholarly work can be constructed on the basis of these three fundamental services (supported by a group of ancillary services and standards).

Canonical Text Services: identifying, citing and retrieving citations of texts

The CTS protocol uses a hierarchical notion of a textual work (similar to the works, expressions and items of FRBR model), and joins to this a continuous canonical citation scheme. The service supports reference to and retrieval of passages from any version of a work. The CTS URN is a notation for expressing the semantics of this traditional form of scholarly reference.

(See more information on CTS.)

Collections Services: querying, citing and retrieving objects in a collection

In contrast to the continuous citation used to refer to passages of texts, scholars often refer to discrete objects using some unique identifier to pick an individual object out from a group. Examples include anything from a collection of entries in a bibliography, to artifacts in a museum. In addition to retrieving an object by its identifier, a Collections Service provides information about the structure of the object, and allows querying based on that structure.

(See more information on Collections Services.)

Reference Indexing Services: associating information with references

Reference Indexing Services associate information with scholarly citations. References may be in the form of CTS URNs, unique identifiers from a Collection Service, or standardized identifiers from a Registry Service.

(See more information on Reference Indexing Services.)

Links