SRU/SRW

SRU/SRW
Name	SRU/SRW
Developer	Z39.50 community, Library of Congress, OCLC
Released	2002
Latest release	2008
Operating system	Unix, Linux, Windows
Platform	HTTP, XML
Genre	Information retrieval, Metadata
License	Various

SRU/SRW SRU/SRW are related protocols for search and retrieval over the Internet that use HTTP and XML to provide standardized query and response mechanisms for bibliographic and metadata resources. They were developed to bridge legacy systems such as Z39.50 with web technologies embraced by institutions like the Library of Congress and services such as OCLC and WorldCat. The specifications aim to support interoperability among repositories curated by organizations including British Library, Bibliothèque nationale de France, National Library of Australia, and consortia like Dublin Core Metadata Initiative.

Overview

SRU and SRW originated as companion profiles that define how clients submit structured queries and how servers return structured results using existing protocols and encodings. The protocols enable discovery across catalogues maintained by entities such as Library of Congress, British Library, German National Library, National Library of Scotland, and projects like Europeana. SRU focuses on a RESTful interaction model over HTTP commonly used by services from OCLC, JSTOR, HathiTrust, and Internet Archive, while SRW encapsulates the same semantics in a SOAP envelope appropriate for environments integrating Microsoft-based platforms, Oracle, and IBM services. The initiative involved standards bodies and projects including Z39.50, ANSI, NISO, and the Dublin Core Metadata Initiative.

History and Development

Development traces back to efforts to modernize the Z39.50 protocol developed in the late 20th century for networks such as OCLC and systems managed by institutions like Library of Congress and British Library. Stakeholders from OCLC, Library of Congress, National Library of Medicine, and regional bodies including CONSER and ISSN International Centre collaborated with standards organizations ANSI and NISO to produce the SRU/SRW specifications. Major milestones include the initial specification releases in the early 2000s, implementation trials by University of California campuses, integration tests with services like WorldCat, and adoption by national repositories including Bibliothèque nationale de France and National Library of Australia. The landscape also saw interactions with linked initiatives such as Dublin Core, MARC21, and namespaces used by MODS.

Protocol Architecture and Components

The architecture defines a client-server model where clients issue queries and servers return records and diagnostics. Core components include the query language layer that integrates Contextual Query Language (CQL), the transport mappings for HTTP (SRU) and SOAP (SRW), and XML schema definitions for response payloads leveraging vocabularies aligned with MARC21, Dublin Core, METS, and MODS. Implementations often incorporate registries or discovery services analogous to those run by OCLC and mirrors maintained by institutions like Stanford University and Harvard University. Security and authentication integrations reflect interoperability with identity providers such as Shibboleth, SAML, and service platforms from Microsoft Azure or Amazon Web Services when deployed by consortia including HathiTrust.

Query and Response Formats

Queries are commonly expressed in Contextual Query Language (CQL), an expressive syntax developed by contributors from the Z39.50 and NISO communities and used by services operated by British Library, National Library of Scotland, and research infrastructures like JSTOR. Response formats are XML-based, with schemas providing records in MARCXML, Dublin Core XML, MODS, or simple retrieval formats that mirror structures used by WorldCat and OCLC services. Diagnostic and explain responses follow conventions influenced by specifications from ANSI and NISO, and clients may negotiate record packing, pagination, and resultSet management comparable to behaviors seen in APIs from Google Books and Internet Archive catalogs.

Implementations and Software

Several library and repository platforms implemented SRU/SRW support, including integrated library systems from Ex Libris (e.g., Aleph), open-source services like Greenstone, DSpace, and EPrints, and search middleware such as Vufind and Blacklight when integrated by institutions like Princeton University and University of Cambridge. Commercial vendors including OCLC and ProQuest offered SRU endpoints for discovery, and software projects at Stanford University and University of Illinois provided toolkits and libraries in Java, Perl, PHP, and Python to facilitate adoption. Aggregators and service providers such as Europeana and HathiTrust used SRU/SRW alongside APIs from JSTOR and Internet Archive.

Use Cases and Applications

Common applications include cross-catalogue discovery deployed by national libraries like Library of Congress and Bibliothèque nationale de France, union catalog synchronization for OCLC-based networks, institutional repository harvesting at MIT and University of California campuses, and metadata-driven services powering discovery layers at universities such as Harvard University and Yale University. SRU/SRW supported subject gateways, union lists, and interlibrary loan systems interacting with infrastructures like WorldCat and protocols influenced by Z39.50 implementations used by library consortia such as CONSER.

Standards, Governance, and Interoperability

Governance and maintenance involved collaboration among NISO, ANSI, and contributors from institutions including Library of Congress, OCLC, British Library, and National Library of Australia. Interoperability testing and profiles were carried out in workshops attended by developers from Ex Libris, ProQuest, Stanford University, and Harvard University, with mappings to metadata standards such as MARC21, Dublin Core, METS, and MODS. While adoption varied across projects like Europeana and HathiTrust, SRU/SRW played a role in transitional architectures linking legacy protocols such as Z39.50 to web-native APIs used by major repositories including WorldCat and Internet Archive.

Category:Web service protocols