I don’t play video games, but now I wish I did. Video game-like learning

May 21, 2019 9:36 AM

Hope me research how to improve the usability of a library discovery service! While we can create tutorials and provide library instruction, we have found that ‘just in time’ help is better than ‘just in case.’ No one wants to read a manual to do find sources for their paper. My ideal is to provide context-aware suggestions and tips analogous to how video games teach new users how to play. But, I also need to make these instructions easy to bypass for those that don’t need it and/or are annoyed by it.

MPOW, a university library IT department, has been working on a custom discovery service. It has a single search box and returns results from different sources in a ‘bento box’ style grouping. (Different sources include: the library online catalog, a knowledge base of research databases and journals, a repository of special collections finding aids, and a repository of digital images and electronic documents, both born digital and scanned in post.)

My goal is to improve the user experience. The good news about making a custom program is that the interface is more open to optimization. The bad news is, we are a small team stretched pretty thin.

I’ve gotten a lot of feedback from our users from usability testing. For example:
• They don’t understand what a ‘special collection’ is.
• They would look in one bento for are resource, when in actuality it would be found in another.
• They expect a certain result based on their search terms; our sources can’t provide that level of complexity. (This is likely related to expecting a library search box to behave the same as Google. Not surprising, but we aren’t at that level).

So, clearly, our users need guidance to make the research journey less confusing. While we want the service to feel as intuitive as possible, we know that there are some things that can’t just be gleaned automatically. We can’t put the burden of learning how library infrastructure works on our poor freshmen.

So, I want the guidance to be built in to the service. While we can create tutorials and provide library instruction, we have found that ‘just in time’ help is better than ‘just in case.’ No one wants to read a manual to do find sources for their paper. My ideal is to provide context-aware suggestions and tips analogous to how video games teach new users how to play. But, I also need to make these instructions easy to bypass for those that don’t need it and/or are annoyed by it.

I’ve been searching for articles using various terms: just-in-time online instruction, the gamification of learning, etc. but I’m not finding what I need. I'm clearly going about this the wrong way. What are some other keywords I can use to find some primers on designing a library tool that teaches you how to use it, as you use it? This is a thing that exists, right? I would appreciate any guidance you can provide.

posted by BeBoth to Education (9 answers total)

Best answer: I'm not deeply embedded in this area but I'm not totally sure this is a thing that exists. If it does, it might have been outlined once in Computers in Libraries magazine or at a CIL conference. The real trick you're dealing with is that the users who may NEED this option don't know they need it (and might ignore it if they were given the option to dismiss it) and placing some overlay on to everyone's experience is a total buzzkill. The big deal with gamification is that it's fun when it works, but often a tutorialis motivated by people wanting to play the game better. It's harder to incentivize people wanting to use the OPAC better because they don't know they're not using it well.

When I've seen tip-type situations with ILSes it's often something prompted by a blank set of results or a query "Didn't find what you were looking for? Want to try a tutorial?" and not placed at the beginning. It's tricky and good on you for trying to make it a better experience.
posted by jessamyn at 9:46 AM on May 21, 2019 [2 favorites]

As a web programmer type person as well as a recent-returner to libraries, I have thoughts.

• They don’t understand what a ‘special collection’ is.
• They would look in one bento for are resource, when in actuality it would be found in another.

If your system is simliar to the one I've been using, books (or media in general) are treated as parts of a library, collection, or other grouping. People search "George Washington" and get a page of boxes (bento sections) with balkanized results. "These three books and a filmstrip are at branch A." "Two of those books are also listed as part of the Heritage Collection on the third floor of branch C." "Branch B is having a celebration of famous Virginians!" and after a while it's just picking around to find the actual thing they're looking for and where it might be. I think it's more usable when the resource is primary, the actual thing they're searching for.

So, my preference is for the results to be paramount, to invert the bento-box model. Yes, like Google, but sometimes it's good to go with the flow. Search results are listed in some order, with all of the location and special topicalities listed as attributes of that result. You have 5 copies spread around? List the likely locations within a single result entry, if not a live record of availability, plus links to special collections or temporary programs where they would also be found. These sub-links to a result's membership in a group of books on a table with a sign can provide points of departure for a reader/user to find related info without having to hang on computer searches all day.

This is for me, and so I'm not sure about e.g. low-skills searchers.

• They expect a certain result based on their search terms; our sources can’t provide that level of complexity. (This is likely related to expecting a library search box to behave the same as Google. Not surprising, but we aren’t at that level).

I think you probably mean "specificity" where you have "complexity?"

It's not my favorite, but I feel this is usually handled by a link to "advanced search" that goes to an interface people are also used to: the many-blanks search form with every attribute of the database.

Beyond that I'd want to put work into Googleizing your single search field (and if it's a page with only a Google-like single search field with a button people are naturally going to expect Google functionality), but the good news is that a lot of that functionality tends to be packaged into libraries and plugins that will get you 80% there (+/-, quoted groupings, etc.), depending on what you use to provide the site and search (language, CMS, frameworks, etc.).
posted by rhizome at 10:31 AM on May 21, 2019 [2 favorites]

I'm with rhyzome. Most library search systems I can recall using have given me the full slate of results as a default, and then provided options for narrowing the search field to one or more of the 'bento box' categories you list. I consider myself a pretty savvy computer user, and I'd also find it frustrating to not be able to view all the results of a query in a single place.
posted by showbiz_liz at 10:36 AM on May 21, 2019

Do you offer chat reference currently? Proactive chat in your discovery layer might solve this problem if you do; it's synchronous, rather than the asynchronous tutorials you're referring to, but might be less work in the long run. It offers just-in-time help to students and depending on the skills of your development team might be something you could build yourself or find an open-source version to embed.
posted by stellaluna at 11:14 AM on May 21, 2019

I think your heart is in the right place, but it sounds like you might be re-inventing MS Clippy.
posted by doctor tough love at 11:44 AM on May 21, 2019

I would make sure you include an easy way to opt out of the 'bento box' and to search the catalog alone, since that's all I ever want to do with my library's search option. E.g. Harvard's main library page includes a prominent link to searching Hollis, their catalog.
posted by crazy with stars at 1:58 PM on May 21, 2019

Best answer: So this is a different implementation, but Meredith Farkas did a Library DIY project that makes the process of finding on-demand library instruction interactive.

It's an older project (2013), but I think researching Meredith's work would be a good starting point.

Back when I was following this kind of thing, I swear I remember a similar project that added an instructional sidebar to the library website, so that students could follow a tutorial and interact with the discovery layer at the same time. I think that might be a good way to go - a "training wheels" menu with instruction that you can turn on and off for new users.
posted by toastedcheese at 10:07 AM on May 22, 2019 [2 favorites]

Response by poster: Thanks, toastedcheese! You reminded me of Guide on the Side. I found the information about it here: https://ualibraries.github.io/Guide-on-the-Side/about.html

Also, thanks to jessamyn for pointing me to the CIL magazine. I've already found an article that looks promising, "Using Links and Widgets to Assist Users at the Point of Need" from the April 2019 issue.
posted by BeBoth at 6:42 PM on May 26, 2019 [1 favorite]

Response by poster: One more follow up: I found out that this is call onboarding in the user experience industry. Here is a good article about it: https://www.toptal.com/designers/product-design/guide-to-onboarding-ux
posted by BeBoth at 9:04 AM on June 14, 2019 [1 favorite]