Discovering Systems for Discoverability

My Role
Research, Wireframing, UI (Sketch), Conversation Design, Prototyping (Principle)

Collaborators
Lipika, Mahit & Niranjan

Duration
5 weeks (Apr -> May 18’)

The Problem
For libraries, across their physical and digital collections, people only ever access a fraction of them. Items aren’t visible to people, both onsite and online.

The Solution
We built a system which allows people to conveniently access items. This includes an app, a voice companion, and an in-person experience.

Background

Many institutions such as libraries, museums and galleries have growing volumes of digital collections. Items which were ‘born digital’, such as digital art, as well as items which have been digitised.

Like many cultural institutions, libraries wants to make it easier for people to discover items in its collection. At any point, the library’s shelves have only 5% of the library’s entire collection and the users do not have a convenient method to access the digital content.

As a part of the Student Design Challenge in OzCHI 2018, State Library Victoria (in Australia) wanted us to help their users explore, discover and use digital items.

Defining Success

Based on our existing understanding of the problem space, and the brief — we defined what success would look like for us.

Increase the metric of total number of physical visits to the library

Make discoverability easier while maintaining serendipity

Accommodate for knowledge in different types of media (digitised art, videos etc)

Design inclusively for people across all ages, genders and disabilities

Libraries and Learning

We conducted a thorough literature review spanning research papers, articles and existing solutions for discoverability.

We wanted to see what “a library” and “learning” really meant to our target audience and where the existing models were falling short.

Some key takeaways:

Libraries are more than providing access to knowledge. They have the potential to be platforms for storage and circulation of local community knowledge.
‍
There exist a lot of skills or topics (eg. pottery) which are best learnt through human interaction. Learning isn’t only limited to books.
‍
The ‘environment’ and ‘experience’ of a traditional library is very important especially for those who visit the library as a sanctuary to work and learn.
‍
People don’t go to libraries anymore. They either find no value in it or are unable to access it due to a lack of accessibility to people with disabilities.

Who goes to a Library?

Libraries are for everyone. Whatever we design, would need to be universal. To cover as much of the demographic as possible, we looked at two extreme personas.

Personas to help us cater to two ends of the target demographic

Opportunity Area

We identified potential challenges for our spectrum of users. Reframing the challenges turned them into opportunities for design.

How might we design for the curious who want to learn something new?

We came up with multiple ideas and then mapped all the potential solutions in a 2x2 matrix to downselect.

The result of a 2 hour brainstorming session

At the end of this, we settled on Victoria. A personal assistant who accompanies you inside the library and helps you find resources to learn from. Outside, she can send you media, from books to tutorials and aid in finding items which align with your interests.

Concept Testing

Onsite Communication
‍A similar implementation was possible with a smarter version of an assistant like Google Assistant. However, in our earlier research we had found that the experience of the library (silence, being away from screens) was something people visiting the library valued a lot.

Not only this, we realised typing or even speaking as the only input could cause inconvenience and wasn’t very inclusive to differently abled users.

Developed by the MIT Media Lab, Alter Ego met all our design goals and works as a great addition to our solution by facilitating silent communication as well as being inclusive to all. You can read about it here.

Alter Ego's Form

We bodystormed the concept to test it’s usefulness.

Once the User enters the library premises, they are handed the headset and can connect to Victoria from their phone.

If the user wants to learn anything new, Victoria can send relevant material based on recommendation algorithms.

User saves content to their digital collection on their phone

1. The FAB opens up a modal with a handlebar. Pulling it up reveals the full window with the conversation history.

2. Victoria can suggest material directly from the chat window itself. Clicking on the tile directly opens the book’s summary and details.

1. New additions are given priority, where the user can scroll through newly downloaded material.

2. The library contains all the books downloaded.

3. Each book displays how much it has been read. The books are ordered in a chronologically descending order, with the most recently opened book at the top.

Designing Personality

After concept testing, we had a slightly more fleshed out version of Victoria. Now we needed to figure out the personality that Victoria would have as an AI (colour, form and microcopy).

Personality and Tone
We asked ourselves what a personal assistant with a focus on discoverability of resources in the library would sound like.
‍
After a lot of iteration, we arrived at these traits — Authoritative, Reassuring, Calm and Knowledgeable.

By conversational modelling I identified the user’s intentions and mapped a range of possible utterances to responses from the system.

A happy path

A complex conversation showing the pre-attentive and attentive states for the voice interface

Once I had mapped out some conversation flows, I had a better idea of the different interaction states I'll have to account for. I used Victoria's personality traits to inform the form factor of the visuals to convey interaction states.

A newspaper article talking about why dating apps make you feel awful

A newspaper article talking about why dating apps are no way to find true love

Six fundamentals interactions which I kept in mind while designing Victoria

Look and Feel
‍Because the State Library of Victoria didn’t have a colour palette of its own, we had to come up with a visual language from scratch.

A moodboard created based on Victoria’s personality traits

Visual guidelines for GUI

Meet Victoria

Archives contain all the materials and resources in the State Library of Victoria. Allowing for a more immersive experience of looking for more media based materials.

Speak with Victoria by simply addressing her through the headset. Through the app, with the click of a button, Victoria is summoned and a conversation history is also available.

Explore the various new arrivals in the State Library of Victoria. These shows the latest additions and new arrivals along with session updates.

Concept Video

A low fidelity video demonstrating what the experience of being in the library (onsite) might look like.

Measuring Success

Along with directly satisfying most of the conditions we had set for ourselves, our solution also had second order effects which aligned with our goals.

Since Alter-Ego is a niche technological innovation, and is only available onsite, it would drive more people to visit the library more often, increasing our ‘physical visits to library’ metric.

We were also awarded the SDC Chairs Commendation for our work in tackling this brief.

Our team went on to publish and subsequently present ‘Victoria’ in the Student Design Consortium at India’s largest HCI conference as well!

The team presenting Victoria to the conference

Takeaways

The unique brief helped me look outside of just the digital interface and look at the experience of visiting a library from a service design lens.
‍
Along with getting extremely lucky by having an amazing team, I was also fortunate to have found the scope to work on Voice as a medium, and Inclusive design. Two areas I’m really passionate about.

Things I'd do differently

If I had more time, I would have tried to conduct primary interviews with people who fall into our target audience.
‍
I would also have liked to build out multiple conversations on DialogFlow, from beginning to end, and try them out with said users.
‍

Want to talk about Voice, Inclusive Design & more? Say hi!

Copied to clipboard!

Jump to Section