This guest post is the fifth in a six-part series from Jane Burke, Discovery and the User Experience, which focuses on the latest developments in discovery and their potential to impact academic library user communities.
Posts will continue weekly through the end of March. I hope that you will enjoy the series and look forward to your comments. Mike
I hope by now I have convinced you that libraries have a Problem. The Problem is that users have moved away from us, and that we need to take steps to go where they are. While users believe in the inherent goodness of the library, that goodness is often too hard to access. We need to offer easier ways to discover the richness and depth of our library collections.
The analogy is our physical buildings. Nearly every library I visit has invested in making the physical library space more accessible and more user-friendly. Furniture has been replaced with components that allow students to create collaborative ad hoc spaces, using the library as an environment that fosters information sharing.
We need to do the same thing for our digital environment. Just as we have removed barriers from the physical library space, we need to do the same for our digital library. We need a friendly front door into the collections, one which allows users to easily enter our information environment, on which we spend so much money.
So you’re convinced? Great!! What are your options for a Discovery service? Actually, there are quite a few. The one that is right for your library at this time depends on a number of factors:
- Amount of digital content that the library has
- Age and condition of your OPAC
- Readiness of the library staff to embrace and espouse a new approach
One other thing to remember is that you don’t have to live with your decision forever. Most discovery services are subscription-based today. You are not buying a solution that you have to live with for 10 years – and you probably won’t want to because these services are evolving rapidly. (Dare I say this is not like the ILS, whose average life span in a library averages 15 years.)
“Discovery” needs to meet end user expectations. That means it has to work like Google – simple, easy, fast. For me, an inherent part of Discovery is that all of the library’s collections, in all formats, need to be searched together and displayed in a single result set. The hackneyed phrase “break the silos” is about all the silos – the catalog as well as the e-journal databases. End users don’t think about format. All digital information is the same to them – and should be to your Discovery service. (I know there are still a lot of faculty who tell students to find 3 scholarly journal articles. But that’s one assignment. We shouldn’t build the format bias into students’ information- seeking experience.)
I see your options for Discovery as a continuum of services right now. On that continuum there are several categories of services, all of which employ a single search box. The user, who we know will use a search box, enters one of more keywords and gets results across multiple library collections.
- Federated Search - This is the earliest discovery technology. It searches across multiple collections and databases and delivers results. It is mature, affordable technology that is simple for the end user. But it is not fast, and it delivers limited metadata. Still, it is a valid first step. It’s not perfect, but it is a step for users. If you cannot afford anything else, at least embrace federated search. You can also embed federated search into some OPACs, which provides some journal articles in the familiar interface.
- Hybrid Search - This approach mixes metadata indexed together in a big index with federated search. This approach is widely used today, by many providers. It delivers a result set quickly for the data in the index and then uses federated search to display data not in the central index. There are some drawbacks with hybrid search because not all of the results are treated in the same way. Relevance ranking cannot be applied across all of the results because there isn’t a single result set. Some content is disadvantaged.
- Web-scale Discovery - This approach uses a Google-like architecture to index all metadata, from all collections, into a single index. The result set is unified, with all metadata in a single relevance-ranked result set. And it’s fast, because the index is pre-coordinated to deliver all of the applicable results together.
There are differences in these approaches. You will want to look at demonstrations of each in order to understand the end user experience. Try to compare that experience to the real competition for our users’ loyalty – Google.
Next week, Jane will complete the series with a discussion of Emerging Trends in Library Discovery Solutions.