Review: SRU, Open Data and the Future of Metasearch

[ Editor's note: Once upon a time, when I was doing much more software development than I am doing today, I found myself with a project where I had to deal with the differences between SRU and SRW. The world of web services and their related standards was quite mysterious to me at the time. Carl Grant, President of CARE Affiliates, reviews an essay in Christopher Cox's book about federated search that discusses these two standards in the context of a number of standards of interest to the federated search and library community. While you'll need to read the actual essay to understand the standards, Carl Grant does a fine job of reviewing how the chapter treats SRU, in particular.

Given the quality of the essays in Mr. Cox's book plus the severe lack of any books related to federated search, I highly recommend the book. You can purchase a copy of Mr. Cox’s book of essays from the publisher, Taylor & Francis, who donated the review copies, by calling their Customer Service department, Monday-Friday 9 A.M. – 5 P.M. EDT, at (800) 634-7064.

Some months ago, as part of my content access basics series, I wrote about SRU, SRW, and Z39.50.

You can find other reviews of essays from Mr. Cox's book in the "Cox essay review" category.]

~~~~~~
As a long time supporter of standards in libraries through my work in NISO, I’m always extremely happy when I find a publication that makes standards understandable and accessible to non-technical people. Such is the case with this chapter on SRU in the book Federated Search edited by Christopher Cox. Put simply, this is a readable, thorough work providing overview, details and context.

As Reiss states in the chapter, “SRU supports the standardized, machine-readable transmission of textual queries via a URL and return of search results in XML to any type of information source on the Internet.”. Using this capability, libraries (and others) can deal with the ever diversifying array of resources on the web and apply search services to those resources. He is absolutely correct in saying that users expect easy to use search interfaces that provide them with useful results linked directly to the full-text resources and that SRU is capable of meeting that need.

Reiss outlines the history and basics of SRU and SRW beginning with a description of what the difference is between the two. Frequently this is not understood, but he clearly lays out the difference by stating that SRU is used where Representational State Transfer (REST) occurs using HTTP GET or POST requests and that SRW is used where Simple Object Access Protocol (SOAP) is utilized and therefore can support transmission protocols like email and SSH. He goes on to cite research done by OCLC Research showing that SRU performs two to three times better than SRW.

With those basics in place, he then gives an overview of SRU used with both software and metadata. He explores the SRU server and client software and basic features that might be expected from each. He examines metadata and the infrastructure needed in the future in that area. Citing Roy Tennant’s 2004 document outlining the nine factors needed in infrastructure he concludes “SRU is an excellent candidate technology for widespread adoption as the protocol for distributed search and retrieval within this infrastructure.”

Next, SRU is explored within the larger context of open standards and the data landscape, examining how SRU works with METS, OAI and RSS. The explanation of how SRU and OAI work together to provide user driven services is equally clear and informative. When coupled with the open access movement, he correctly concludes that scholars will benefit from this process because their research will be more visible as a result of multiple access points being made available.

Reiss then moves to the specific subject of metasearch and the important role that SRU can fill there in searching library catalogs, repositories and websites. He correctly notes that “as more libraries acquire metasearch tools the number of content providers that open their data via SRU is likely to increase”, although as he then notes the implementations of SRU are still young.

A review of some of the existing SRU projects notes major initiatives that have occurred, including OCLC’s portal, the OCKHAM initiative, the European Library, and the NISO MXG Initiative. He also discusses Amazon’s OpenSearch initiative, which while not yet SRU compliant, does show potential for integration and co-existence with SRU.

The chapter concludes with a discussion of the future of SRU. He concludes “SRU can be an important component that libraries and their vendors can employ to attack this problem by creating effective metasearch tools. It is well supported by the open source library community and relatively easy to implement.”

This conclusion is absolutely right on the money. SRU services are increasingly being enabled and offered to the marketplace. Key to adoption is the requirement for librarians to take the time to understand this technology. This well written chapter by Kevin Reiss will serve as an excellent way to achieve that goal. I highly recommend it.

If you enjoyed this post, make sure you subscribe to the RSS feed!

Tags: CARE Affiliates, carl grant, federated search