One of the biggest changes in academic libraries over the past few years has been the development of “discovery layers”: collections of paid, unique data that are pre-indexed and then easily searched by specialized search engines.
For those readers not in the library industry, keep in mind that Google, Bing, Yahoo!, and other search engines cannot crawl through data that is in siloed, subscription database collections. That data is limited only to people and institutions that have paid for access to it. So, a big benefit that libraries have held over Google is offering the content inside these databases. Such databases range from big vendors who gather together (or “aggregate”) content from many different sources – examples include ProQuest, Gale, EBSCO, Project MUSE, and a variety of others – to smaller publishers or content providers who generate unique content that they believe they can offer for sale to individuals or institutions. The drawback, however, has been that it wasn’t easy to find all that data – you had to go to each different silo and search that database to see if there was anything of interest there. And of course, first you had to know that each database (or silo) existed.
For a while librarians used “federated searching”, but it wasn’t a great solution. With a federated search tool (and they are still certainly in use in many libraries), the computer takes your search terms and goes out to search each of the many different databases that you’ve selected, waits for all the search results to come back, and then compiles the results together. In most cases, it’s not a very elegant solution, and it’s easy to see why the speed and simplicity of a Google search became so popular – even when the content wasn’t as good.
Google, of course, doesn’t go out and do a search the moment you type words into its search box; it has already reviewed and ‘indexed’ all of that content, and whatever it has indexed is what will be in the search results it provides to you.
So, library database vendors tried to create solutions that allow libraries to compete with Google in this area. Their strong differentiator is that the data they’re indexing is the subscription-based content, rather than data on the free web, which Google indexes. Examples of these are “EBSCO Discovery Service” from EBSCO, “Primo Central” from Ex Libris, “Summon” from Serials Solutions, a division of ProQuest, “Encore” from Innovative Interfaces, and a few others. (As an aside, I was a co-founder of Serials Solutions; I was involved in the sale of Serials Solutions to ProQuest, and remained with the company for a while after the sale; and was slightly involved with the development of Summon. While I will always have a soft spot in my heart for Summon – to the extent one can have a soft spot for a discovery layer, I suppose – I am today very interested in making sure that ALL institution patrons have access to the ShipIndex.org data, through ALL discovery layers.)
Content from ShipIndex.org is now indexed in Summon and EDS, and I’m looking to get it into other discovery layers, as well. Here’s an example of what search results look like at a library that subscribes to both ShipIndex.org and Summon, from Serials Solutions:
When a student does a search for a ship — in this case the Elizabeth Davidson — they find a citation for that vessel in ShipIndex.org, and a link to take them directly to the page for that ship. They didn’t even need to know that ShipIndex.org exists. They search in Summon (or EDS, Primo Central, or another discovery layer) and they find content that they wouldn’t have otherwise found.
To be clear, a library must subscribe to both a discovery layer and the underlying databases, for the databases’ contents to appear in the discovery layer. Discovery layers are definitely not cheap, but they do make a huge difference in improving how library patrons discover the resources that the library already subscribes to.
I’m pleased that ShipIndex.org’s data is in Summon and EDS, and I look forward to doing whatever I can to make it available to users of other discovery layers, as well.