Special Interest Group Archive: Information Technology 11/21/14

The SENYLRC hospital libraries recently implemented a federated search tool - used to be offered by Swets - technically a federated search

What is the distinction between discovery and federated search?

KL - A true discovery layer has a unified index, all the contents compile the metadata into one index. You need to have a standard to pool the metadata for the vendors. You need to normalize data, but that could depend on agreements with vendors.

(ProQuest and EBSCO don't play nicely together - sometimes they change small things and searches stop working.)

Another thing is the link resolver - how you are going to get things back to the user. Federated search goes inside the EDS, but you don't see it behind the scenes.

(Elsivier - they want more money for all the SUNY schools, but negotiations broke down. A task force is looking for other options for Science Direct.)

SUNY is also looking into open access - 64 campuses with research work - library as open access publisher. - another SIG topic? MSMC does a per-article system. In SUNY, only the academic centers subscribe to the entire package, and open access seems like the solution.

HLSP has content through deep web, the follow-up to next gen, a consortial model works for shared content among the hospitals, some is just individual libraries for their own licensed content. The system is still in the trial phase, and one system still needs to come online. It requires training, because the results for librarians are different than those that come in from just a PubMed search. NIH funding requires an open access copy of articles after one year. Compared to other systems, it was inexpensive.

AH - Has EDS with the SUNY deal on linksource - they are transitioning to article finder, EBSCO's new feature. SUNY's contract is with EBSCO, although individual campuses can opt out.

DG - no discovery layers, too expensive for the mount. They are waiting for discovery layers to get less expensive. III is even more expensive.

TO - Has Encore and is getting EDS through them. However, it has failed, when EBSCO and III don't play well together. Some libraries have purchased their own systems, but most are using the system's resources. This is more OK for a public library.

RA - They also don't have anything. She is familiar with EBSCO and Encore from other libraries, but there were issues on those campuses again, and the popular solution became open access. Then again, they have fewer than 100 students, and not a lot of original research. But looking ahead, they could have more students and might need solutions.

What prompted you to delve into this? What is the feedback you got from students?

DG - they redesigned their website to look discovery-ish, without actually having it?

KL - had to redesign the whole website to accommodate the search box.

AH - we needed to "simulate Google." - Students want information from Google, even when they get books in their hands.

DG - their students want physical books over ebooks (TO, KL agree - they want to photocopy books.)

MHLS discovery layer - relies on the relevance from III.

There is a misunderstanding about what is full text and what is not.

The difference between indexed searching and federated is that indexed searching is faster, because it's going off of metadata, not looking at the entire database. HLSP has StatRef results, and a nice-looking result. MULTISearch is what SENYLRC calls the system.

Discovery layers don't always work well for everyone, especially people with specific needs like nursing students or history scholars. It's better to send them to the direct resource.

For those of you with discovery layers, do you have special collections that aren't in the layer?

KL - if it's in the catalog, it's in there. But things like the digitization project are not in the catalog, so they're not in the discovery layers. The catalog is exported once a week.

TK - if you want to, you can bring in HRVH as an API and have it show up.

AH & KL have placards that show up with information that people put in like library hours.

KL - it works well when you don't know what you are looking for, but librarians who know the best place to go might get annoyed by realizing that they aren't finding the best resource.

AH - not a place to look for specific articles, there is too much

Training? Did you train your students or anyone else?

Generally - no, because they just want it to be Google.

AH - he can give more specific training when students come to the reference desk. When students are looking for something specific, he'll show them specific books. They do library instruction

KL - the librarians are not comfortable teaching with discovery layers because they know the databases so well. They did a full-day EBSCO training, but mostly they saw things that were not available yet.

Were you part of the decision making process?

KL - pricing was attractive for EDS, they had a task force among the SUNYs that she was part of. They also looked at WorldCat, ProQuest Serial Solution, and Primo from ExLibris. They had it turned on for a while before she started to publicize it. The librarians didn't know it was there, and it has switched between catalog and discovery layer for a while.

Other choices are Encore from III, Aquabrowser, You Find (You View?)

TO - wasn't there for the big switch, but was there for Encore, they have purchased EDS through III. EDS is for journals, Encore is for the catalog.

Link Source can turn some information sources up higher in algorithms, so that you can have control over what shows up first in a search.

AH - They followed the hard work of KL and other SUNY libraries. New Paltz and others had done some of the trouble-shooting.

EDS can give you search term reports, and you can also use Google Analytics to figure out what terms students type in and then turn away. They can aslo tell you what users find in search but can't access because it's not in the collection.

KL - Google books and Google scholar will show up with your library's information that is pushed out through EBSCO. They use the library's metadata as well as IP addresses based on location.

Words of wisdom from Andy...

"Don't Let Perfect be the enemy of the good."

You can change it - you don't have to tell them. We like to survey things about everything, but no one else cares as much as you do.

RA - Google and Apple change things on us all the time, there's no reason that we can't do it anymore.

TO - not so much if you are a public library. Everything has to go through a board or a public decision. They are a cooperative. III also is the only entity that can change links on the search page.

In attendance:

• Rhonda Altonen (convener) - Barrytown Unification Theological Seminary
• Andrew Heiz - SUNY - Orange Community College
• Thomas O'Connell - MHLS
• Denise Garofalo - Mount Saint Mary College
• Kristy Lee - SUNY New Paltz
• SENYLRC - Carolyn Bennett, Tessa Killian, Zack Spalding, Patricia Carroll-Mathes