The Who, What, When, Where and Why of Library Discovery (text as intended for presentation)

Posted on     10 minute read

× This article was imported from this blog's previous content management system (WordPress), and may have errors in formatting and functionality. If you find these errors are a significant barrier to understanding the article, please let me know.

[caption id="attachment_KkOCeAtKHIc" align="alignright" width="210" caption="Me and my Jester's Cap"]

Last week I was at the NISO Forum: The Future of Library Resource Discovery with a great group of colleagues as we challenged ourselves to think about the role of discovery services in the information-seeking habits of our patrons. In the closing keynote, I was projecting what library resource discovery interface might look like five years from now, and I was weaving in comments and ideas that had bubbled up in the in-person conversation and the Twitter channel. And yes, I did wear a jester's cap for the presentation.

Included below is the text of the presentation as intended to be given on October 6, 2015, at the Mt. Washington Conference Center in Baltimore, Maryland. (I did stray from the text in a few places, but not in any significant way.) At the bottom is a postscript based on a conversation I had afterwards about the role of mobile devices in library resource discovery.


First off, the hat. I started blogging as the Disruptive Library Technology Jester about 10 years ago. Just after I started the blog a colleague, Greg German, gave me this hat, and I thought this forum on library discovery services was a good time to give it its professional debut. A jester's role in medieval courts was thought to be able to speak truth to power. I don't know if it will so empower me today, but it does have the added bonus of embarrassing my 14-year-old daughter as I wore it in the car on the way to the airport on Sunday.

Secondly, [mock-Shakespearean-actor-tone] and with the gracious blessing bestowed by the NISO education committee for whom we are grateful for organizing this event, the jester will not prognosticate in a long monologue upon this stage for the next hour, but instead will enlist accomplices from among those present here and those watching from afar through glowing windows connected to us by a mystical ether.[/mock-Shakespearean-actor-tone] (Not a day of Shakespearean training, can you tell?) Instead I want to provoke a few thoughts and prick a few comfortable conceptions in this last session before we all leave the Forum exhausted but fulfilled. And if I make you laugh along the way, well, that's just a bonus.

To do this, I want to lead us through an exploration of this topic via the five double-u’s of journalism: who, what, where, when, and why. The who, the what, the where, the when, and the why of a discovery service. And since the end of a meeting like this is a time for looking ahead, let's speculate what this might look like just five years from now. As a foil I want to use this gadget, the Amazon Echo. Have you heard about it yet? Let's see what Amazon wants us to think about it.

[caption id="attachment_KkOCeAtKHIc" align="aligncenter" width="570" caption="'Introducing Amazon Echo' -- the first one minute and 25 seconds were played in the presentation"][/caption]

There is also a cute, not-safe-for-work parody of this commercial if you search YouTube for “Amazon Echo Early Beta Version”. But I think this gets the point across for one vision of a discovery service -- ask a question into the air and get an instance, accurate, and concise answer back. Is what many library users are looking for when they come to our one-search-box discovery service? How do we want our discovery service to look like this and in what ways do we want to be distinctly different? Let's run through the questions.


Close your eyes. I want you to imagine the most challenging person to support. For librarians that work on reference desks or for user experience facilitators, this probably not a hard challenge. For the rest of us, perhaps if you work in tech support think of someone that you’ve had to help with a computer problem. Or just think of someone who walked up to you in the hardware store and asked you which screws you think they should buy. Or someone at a grocery store asking you how to know when a melon is ripe. Got that person in mind? Okay, keep your eyes closed and picture that person.

Do they know how to navigate the web, operate a mouse, and understand the user interface cues that are now ingrained in your experience? Do they have a speech, mobility or visual impairment? Do they have the life experience needed to even form the question they are asking, or are they a child or a budding scholar in an unfamiliar research area?

Our discovery services need to take this range of “whoness” into account. They need to work for a wide variety of skill, abilities and knowledge. The Amazon Echo probably has some kind of built-in discriminator filter that learns the voice of each person, and it might start to tailor answers based on its knowledge of that individual. Our discovery layers don’t have that context of the person asking the question.


Speaking of context -- in that Amazon Echo ad, a simple factual question met with a simple factual answer. Is this what we envision library discovery to be? Personally, and as much as I love libraries, the library's discovery service is not the first, second, or third place I going to go to look for the answer to that question. No, instead the “what” of our discovery service should be deeply rooted in the tradition of the reference interview. An interaction with the questioner that clarifies the intention of the question and the form of the reply: a factual answer or instruction on how to arrive at the answer; whether an encyclopedia article is enough or an in-depth review article is sought; whether a single reputable source is all that is needed, or a comprehensive literature review is required.

The art of the reference interview carefully guides the user through the maze of possibilities. The user might not even know they’ve been lead through one of a thousand paths that could have been taken when the question was first formed. Do our discovery layers account for that complexity as they lead users to their end goal? Or are our discovery layers attempting to mimic the Amazon Echo, or the single Google Search box, to present users with the single, one-size-fits-all answer to their question.


If we go back to the Amazon Echo, that was an innocuous black cylinder in various rooms of our house. Do we envision black cylinders from our library's discovery provider in our homes, offices and dorm rooms answering esoteric questions? Do we expect these cylinders to be placed on reference desks across the country where patrons can ask their questions?

Let's not stop at Amazon's black cylinders -- we also have Apple’s Siri on a watch and Google Now on a phone. For me, “where” is clearly not a place where we need to keep up with the jones’s. While our library discovery interface should have all of the responsive design techniques that make it scale from phone size to wall size, we should not lose sight of where users will conduct their research -- on their tablets and desktops. [I got push-back on this idea after the talk; see the postscript below.] Let’s test this, of course, with anthropological study and user experience testing, but nothing I have read so far leads me to believe that we need to go down this path.

Within our own community though,i think we are looking for the ubiquity of our discovery service in all the places the user is. They may not speak their question aloud to the room, but we do want to meet them where they are.


When do undergraduates do their research? At 10pm on Sunday night, of course. In fact, I wonder if one of the contextual clues that the discovery service could use to tune its reference interview algorithm is the time of day, day of the week, and week of the year. The Google web search engineers would call this a “signal” -- one of more than 200 clues it says it uses to tailor search results to a user’s specific needs. The reference interview’s artful probing questions present more signals.


There is one signal that requires very special handling. Why is the patron asking this question? What more can I infer from the series of questions he or she has asked the discovery service over the course of lifetime. I don't want to know. If fact I'm actively going to take steps to be sure I don't know. This here, I'll argue, is what distinguishes us from Echo, Siri, Cortana, Google Now, and the rest off the emerging personal assistants. If I were to ask my discovery service the height of Mount Everest, then ask the distance between Baltimore and Mount Everest, then ask how warm Mount Everest is in October, I don’t want advertisements for books on mountains or tourist packages to Nepal to follow me around the web for the next six months.

The third principle of the American Library Association Code of Ethics says, "We protect each library user's right to privacy and confidentiality with respect to information sought or received and resources consulted, borrowed, acquired or transmitted." As a librarian, there are few things that make me more anxious than to show up as a professional quoted in an article like this.

I know this is going to be controversial. In a world that is being given customization, complete contextual awareness, and near prescient capabilities for meeting its information needs, it seems important that the library profession should chase after those capabilities. I don’t know. There is a large working group within NISO right now trying to put meaning that code of ethics principle in a meaningful way in our highly networked, distributed service model. Records of patron activities cross many organizational, contractual, and network layer boundaries. I, too, have been amazed at the capabilities of systems that are arguably not bounded by professional ethics like ours. I just don't think our libraries should go there.

Jester’s Cap

I don’t know. I’m just a guy with a funny cap on his head teaching himself how to juggle. Maybe our five-years-older selves will have figured all of this out. We’ll look back and wonder why it took so much thought to create this thing we called a discovery service. Maybe we’ll wonder why we put so much thought into something that was such a flash in the pan. Maybe in five years Googlezon will give each of us our Evolving Personalized Information Construct and we don’t have to go to the library any more for our information needs. Or maybe we will have found that a gathering in Baltimore, Maryland back in 2015 generated some really good ideas -- ideas that sparked a wave of innovation between publishers, service providers, and libraries with far reaching impacts.


That’s what I think will happen. Now comes the audience participation part, but not with the juggling. I want us to cast our imaginations forward five years and see how we’ve addressed some of the things we’ve talked about over the past two days. I’ve written down some key thoughts and ideas expressed here in the room, on twitter, in Marshall’s document and other documents. If this is something you said or wrote, and are willing to fess up to it, I’d like to give you a chance in this closing session to project yourself into that future place. If you didn’t say what was written on the card, but you want to add your own ideas, jump in as well.



Based on a conversation, I would present the "Where" section differently. Users may come to the library resource discovery service from a mobile device with a very small screen, but they will likely do so with a much different intent than if they were on a desktop or tablet device. The smaller devices are more likely to be used for "known item" searches -- such as when the patron wants to send a citation to a collaborator in the course of a conversation. (Such known item searches are also useful for more traditional reasons such as finding an item in the stacks.) In those cases, the discovery service should take the mobile nature of the inquiry as another signal to use to tailor the user experience to that small device. I will still argue that very, very few are going to conduct a full-blown literature search -- with all of the bells and whistles -- on a small mobile device.