Scirus Repository Search Service Growing

Elsevier offering improves content visibility on web

In recent years, Elsevier has been carving a niche in supplying targeted information to its mostly academic and special library users. Though its own content is voluminous, the company has developed software search devices that glean information from a plethora of sources beyond its vast holdings. Along with the Scopus search engine for science-specific resources, Elsevier is delving into the institutional repository realm with the recent launch of the Scirus Repository Search Service.

Sharon Mombru, Scirus senior product manager, told LJ that while the number of institutional repositories is growing, their content, nonetheless, can be difficult to find. Why? “A lot of the attention and focus in institutional repositories,” said Mombru, “has been on getting them set up and started, encouraging faculty submissions, getting them populated, managing them, etc. There hasn't been much talk about increasing their visibility.”

She contends that the underlying assumption is that merely being online guarantees your institutional repository can be found by everyone. “What our partnerships have showed us is that's not enough. What institutional repositories need to address is making sure that their content can be found by web search engines, that the content is easily accessible on the repository site itself, and that it also is sharable across other repositories, so you're looking at a cross-depository search facility there as well.”

Location is everything

A large factor at play here is the “invisible web.” Facilities can have top content, but if the popular search engines most commonly used (can you say Google?) don't read it, you're off the radar. “As a result,” says Mombru, “valuable content is lost or appears way down in the ranking during searches.” To combat that, the Repository Search Service takes a twofold approach: “We first index the content of a partner's repository, and that process involves looking in detail at how individual repositories are set up. They don't all do it the same way. We have a focused approach on a partner-by-partner basis.” Mombru said that paramount to Scirus's ability to increase a depository's visibility on the net is an indexing process the company developed that combines the metadata to the bibliographic information with the full text. “Very often you'll find that search engines do one or the other but [do] not combine them. This offers more powerful results.”

All indexed content is added to the scirus.com site, which also improves visibility—partners additionally can use the Scirus search engine on their sites—and all content is credited to the original source, which, Mombru said, is “valuable because it tells the user where the results are coming from, what data they're based on.”

Scirus's second wave is cross-repository searching, so each member's data set gets larger as the number of partners expands. Lastly, the service has a joint communications program, whereby Sciris tries to increase the repository's visibility through marketing efforts.

NDLTD and Caltech join

The service launched with the University of Toronto but has attracted two new partners: Networked Digital Library of Theses and Dissertations (NDLTD) and California Institute of Technology. The international, nonprofit NDLTD is an initiative to capture and centralize electronic theses and dissertations (ETDs) worldwide. It has roughly 100 member universities, which, combined, have roughly 200,000 ETDs in more than a dozen languages. At Caltech, Scirus is indexing the Collection of Digital Archives (CODA) of 4000 theses, technical reports, articles, oral histories, and conference papers produced by the school's 1200 staffers.