The hazards of beta testing (or...why isn't View It working?)

We hope you've been trying out the test version of MNCAT Discovery. So far, we're pleased with the variety of results you see with a keyword search, the facets on the left side (especially the new Include/Exclude functionality: Try clicking More Options under any list of facets and you'll see what we mean!), and the success rate for known item (aka title) searches.

But this is still a test version. Once you've found an item you want, the test version will correctly show you where to find a print copy when you click the Get It tab. But you will have less luck with getting online full text with the View It tab. You will be able to see sources for the item:

However, when you click one of the sources to view the item, you'll most likely see a very unfriendly error message:

While alarming, this error message is simply the result of working in a test version. When the final version is available later this month, you will be able to click a source and see full text, or the image, or the media file described by the item you chose.

If you see behavior you don't like in other areas (or if you love what you're seeing!), please do let us know using the feedback form.

Speedy website coming your way

Coming soon to your smartphone, tablet, and desktop computer: A lightning fast, Libraries website. The site is expected to launch in January as part of the roll out of MNCAT Discovery, the Libraries new search tool.

The new website will have a flexible, responsive design that is optimized first for smart phones. The site likely will include a single, blended search box with an option for advanced searches.

So, speed and a mobile-first design are the top goals for the new site, which is being designed by the Libraries web development team. But why?

Speed because web users are notoriously impatient. In its own research, Google has found that a half-second delay in delivering search results can reduce traffic by 20 percent. Amazon.com estimates that every 100 milliseconds of delay costs them 1 percent in sales.
Mobile-first is a goal because speed is even more important on a smart phone. And because the growth of mobile use is growing rapidly as people use their smart phones and other devices to search the Internet and conduct research. University Libraries has seen the growth of mobile usage on it site double in each of the last two years from 2 percent to 4 percent to 8 percent.

The new website and MNCAT Discovery

MNCAT Discovery, the Libraries new search tool, will greatly expand and enhance a user's ability to search print and digital resources. MNCAT Discovery is expected to go live in January 2014, at about the same time as the the new website. The site will include a single search box to search Libraries books and other holdings, in addition to articles and related materials.

In the Libraries' 2010 survey, the overwhelming majority who responded said that they were interested in a broader search that went beyond just our own physical collections. MNCAT Discovery will provide this and more, while still allowing users to search specific databases. The Libraries will continue user testing before going live in January and will continue to fine-tune this new search tool going forward.

A word about MNCAT Discovery

During this academic year, University Libraries will be implementing a new discovery tool called MNCAT Discovery. The tool will greatly expand and enhance your ability to search print and digital resources. We intend for MNCAT Discovery to be fully functional in January 2014.

This fall we’ve implemented the first phase: MNCAT Article Discovery, a new general-purpose database. MNCAT Article Discovery is meant to complement rather than replace our rich collection of licensed databases. Rather than focusing on a particular discipline, MNCAT Article Discovery allows you to search the overwhelming majority of articles and other materials currently licensed by the Libraries.

If you’ve used Google Scholar, you already have an idea of how this new database might work. But MNCAT Article Discovery offers some distinct advantages over Google Scholar:

• The ability to filter results after your initial search, similar to the product filtering you see on commercial websites like Amazon. Perhaps most important for undergraduates, users can quickly filter the results to only peer-reviewed literature.
• A direct connection to the full text licensed by University Libraries. While it is possible to set up a Find at University of Minnesota link in Google Scholar, the links in MNCAT Article Discovery don’t require any user setup or knowledge.

Other institutions have found that databases like MNCAT Article Discovery also help researchers with interdisciplinary work and re-introduce some of the serendipity that many miss from the days of shelf browsing.

Before spring semester, we plan to fully implement MNCAT Discovery, which will include the Libraries holdings as represented in the catalog. Details about MNCAT Discovery will be coming later this fall.

Knowing what pool you're swimming in

Last week I heard Susan Gangl, a librarian in our Arts & Humanities unit, describe the core message of a workshop covering a variety of library databases: Know what pool you're swimming in.

That message remains important for our newest, and arguably most visible, database, MNCAT Article Discovery. Unlike most databases, which are designed to focus on a particular discipline or audience, MNCAT Article Discovery is designed to include as many possibilities as is practical. Also unlike most databases, Libraries staff can choose which of the sources made available by our vendor, Ex Libris, to activate. The set of all sources available through Ex Libris is called the Primo Central Index (PCI).

As we work through the long list of available sources, we are discovering that selecting which to activate is a bit of an art. Our goal for MNCAT Article Discovery is that it provide a useful array of starting points for the user with minimal "clutter." If we activate everything, we run the risk of alienating the very users we want to help by providing too many choices. If we activate too little, too many users will miss out on useful starting points.

Among the guidelines we used to determine the initial configuration:

• If it's subscription-based and we subscribe, it's activated.
• If it's free references material, it's not activated. We've seen implementations at other institutions where the first hit is typically from Wikipedia. Our assumption is that our users will have already looked at Wikipedia before coming to the library if they needed the sort of background Wikipedia articles offer.
• If it's better discovered through specialized search tools, we leaned toward not activating. We still have a plethora of databases available that are designed to help people find materials in a particular discipline or of a particular type.
• Minimize multiple points of access to a title in a single search result.  For example, we could get e-book records from the PCI, but we also catalog those we have access to.  For now, such records will be available in both MNCAT and MNCAT Article Discovery, since the results are not blended.  This will be revisited when we move to Alma and blend it with MNCAT Article Discovery.
• Avoid clutter. For example, we chose not to activate HathiTrust Digital Library at this time. Currently the HathiTrust includes individual records for journal issues which we felt would provide too many returns.

You can find our current lists of what is active and what is not at http://z.umn.edu/pci. If you don't find a particular database on one list or another, that database isn't available through Ex Libris. For many sources, the lists include reason why the source was or was not activated. Note that the activated resources are on the main tab and deactivated resources are on the second tab.

Introducing MNCAT Article Discovery

The world of research contains a vast number of information sources: books, articles, conference papers, reference materials, and so on. Experienced researchers and librarians know that it's easier to find the sources you need by first choosing where to look. Libraries help facilitate the research process by providing access to a wide variety of databases. 

These databases typically index literature in a particular discipline or for a specific purpose. While helpful--even vital--for focused research, choosing among the hundreds of databases licensed by University Libraries is overwhelming for those less familiar with the process, and even for specialists in one field who need to branch into something new to them.

This fall, we are introducing yet another database to this mix: MNCAT Article Discovery. Rather than limiting a search to a predetermined scope, MNCAT Article Discovery indexes most of our licensed resources as well as a selection of the freely available content typically found through Google Scholar or an Internet search engine.

Of course, when it comes to results, more doesn't necessarily mean better. MNCAT Article Discovery uses a robust relevancy ranking algorithm combined with powerful facets to help the user focus on the most appropriate resources to meet their current research need. We'll provide more information on relevancy and facets in a future post.

In September, MNCAT Article Discovery will be the default search target from the Articles and Databases tab of the Libraries website. It will also be listed in the Databases A-Z list accessible from that page. For now, you can try out MNCAT Article Discovery in MNCAT.

Prior to spring semester, another resource will be added to MNCAT Article Discovery: MNCAT, our current "books and more" library catalog. This blended database will be called MNCAT Discovery, and it will be the default search from the Libraries main website. We'll provide more information about the scope, relevancy, filters, and features available in MNCAT Discovery, as well as a link to the test environment, during fall semester.