Assisting Underserved Libraries by Constructing Low-Cost Q/A Applications
Team: Adela Locsin, Carlos Eguiluz Rosas, Jared Payne & Kamrul Hossain
Our team’s design challenge was to create a public-interest technology (PIT) tool that would support librarians in their quotidien work through knowledge sharing. Through our initial conversations with librarians at the Queens Public Library , the Miami-Dade Public Library System , and other public libraries in the United States, we primarily considered how we might help librarians make exceptions to rules when needed for high-need groups. The scope of our PIT tool focuses on creating a tool that would redirect the day-to-day labor of librarians who are tasked with assisting patrons toward a more permanent resource that could continuously serve as a form of knowledge sharing, as opposed to fleeting and temporary interactions.
Landing page for patrons within the "Patron Help Center".
Research
Current Landscape
Public libraries are generally underfunded and often rely on outside resources to accomodate lack of literaty materials and services. Year after year, they scramble to respond to budget cuts and inconsistencies in funding.

(Source: ALA: 2010-2011 Public Library Funding Landscape ) Urban libraries reported decreases in each category (last row) for FY2010-2011, but anticipate some improvement in FY2011-2012 in all expenditure categories (3.9 percent salaries, 7.4 percent collections) except other expenditures (-4.4 percent).

We found that the public's perception on a library's success is typically based on statute numbers (ex. # of patrons, # of books lent) and not on the valuable (free) services that libraries provide to the public (ex. GED classes and Learn-to-Read programs). This perception influences how the public donates and how governing-officials allocate funds to their regional library systems. As a result, some libraries, like those in Oregon and New York , have begun an outcome-based approach to not only measure their impact, but also shift away from this traditional assessment of success. While a change in impact-surveying might prove useful for collecting more funding, it unfortunately doesn’t resolve uneven allocation of funding within library systems.


Two different branches within the Brooklyn Public Library system. Consider how the buildings' architecture, spatial arrangement, and density per square meter implies a difference in funding within the same system.
Pacific Street Library (Left) vs. Central Branch (Right)

Intrigued by how uneven allocation of funds can significantly impact the content and services provided by individual library branches, we decided to conduct interviews with the (primary) stakeholders of these public infrastucutres: Generalists, Specialist and Suuport Staff. Our objective was to develop a better understanding and language of this community while also considering the potential knowledge and technology gaps between different demographics and branches.

Generalists, Specialist, and Support Staff are not only those most affected by a change in funding, but they are also the main group of individuals who patrons interact on a daily-basis.
Synthesis
Data Collected
After conducting 9 interviews with various stakeholders and reviewing posts from multiple online forums (i.e. Library Twitter , Library Reddit , and blogs written by several librarians), we were able to extract 6 themes relating to pain-points in internal library communication. Throughout the analysis, we aimed to identify any larger systemic issue where our problem sits and if our to-be-created solutions would butt-up against larger societal issues.

Interpersonal Best Practices Communicating + Helping Patrons Resources Differ Communicating Value is Tough Some Library Processes Just Suck
Communication between departments often happens in person or through private messaging, but communication with admin is often one-way. Often communicated through face-to-face conversation. Technology divide is a huge issue. The structure of the library system dictates funding and resources allocated to a library. Standard library data focuses on numbers as they pertain to books and amounts of resources (circulation, collections, etc). Interlibrary Loans (ILL) are horrible to process + there is no way around it!
Tasks are long and there is often no incremental measure of progress. Range from extremely specific to extremely general. Patrons are often not aware of the full spectrum of resources. Larger libraries often have their own technology departments and funding allocations. No standard way of quantifying the value of library programming. Weeding (removing items from circulation) is tedious.
No paper trail for conversations. Often solicited through social media. (#librarytwitter). Patrons come in who need to use library services but are not members or cannot afford to. Paper-based processes abound!

Given all this, we narrowed our project's focus to:
  • Considering only libraries in urban locations.
  • Thinking about how libraries in low-income areas may struggle to serve their patrons fully.
  • Identifying technologies that ease these pain points.

How Might We's (HMW's)

With this new change in focus, we concluded our research + analysis and began moving towards developing ideas for how we might:
  • Support access to digital tools when a patron prefers analog mediums?
  • Help patrons access services when an exception is needed for high-need groups?
  • Help librarians make exceptions to rules of the library when needed for high-need groups?
V1 Prototypes + Testing
Ideation
After developing the 3 HMW's above, we generated 30 potential ideas for how we can benefit librarian communication. From these 30 different ideas, we unanimously decided on four to which we then combined two to create a "Hybrid Prototype" that generates the most value to the community we're serving. To provide some background, the two ideas we fused were the Searchable FAQ and the Forum/Polling Board.

Searchable FAQ
Documents when and how frequently exceptions are made as well as establishing answers or precedents to resolve queries.
Target Scenerio Requires
Librarians + support staff who are tasked with assisting patrons. A low-wealth patron asks a new staff member if they can check out a test-prep book, despite having incurred an amount of late fees that normally would prohibit the patron from checking out the book. The staff member doesn’t know if they can make a good-faith exception.
  • Manual entry/query of question by library staff.
  • Computer with data-entry capabilities

Walkthrough of a patron searching and posting a question. Users can flag questions for viewing later, and upvote/downvote to prioritize the questions most relevant to them.

Findings
  • Users found the prototype easy to use and they had a good understanding of it.
  • Users wanted to know if only those with library cards can access this service.
  • One user asked how they should request for an updated response to a question that was asked 5+ years ago.
Risks + Issues
  • Different libraries will have different rules regarding exceptions.
  • There is a possibility that librarians would either not be able to make any exceptions or apply them unequally.
  • The tool is exclusive – only works for patrons who know keywords.

Forum/Polling Board
Acts as a platform for all members to (anonymously) discuss and vote on certain exceptions.
Target Scenerio Requires
Librarians + support staff who regularly face patrons and deal with patron-related operations. A patron contacts a librarian regarding a service and asks if they can make an exception for them. The librarian inputs the exception onto the polling board and invites their fellow staff members to vote on what should be done. The other librarians (anonymously) cast their votes and discuss the decision underneath the poll. After some time, the librarian contacts the patron with the final (majority) decision.
  • Participation from library staff.
  • Time to reach decision.
  • (Optional) Anonymity feature.

Walkthrough of a librarian exploring questions via tags and viewing them via on a scale of urgency. Librarians are able to create their own questions and ask their fellow colleagues to vote.

Findings
  • A new employee would normally contact someone at the front-desk for help. If no one can be found, then they would resort to their supervisor or the person in change.
  • No digital tracking of questions; all tracked manually.
  • Patrons asking "strange questions" occurs quite often, in which case a staff member would refer them to the appropriate department.
    • For smaller branches, people have more knowledge than those in bigger branches who would call department instead.
    • Employees are advised to refer patrons to a department to not overstep boundaries even though they can faciliate the query.
Risks + Issues
  • Likelihood of employee participation in both discussions and polls.
  • Weighing of opinions.
    • Should all opinions be weighted the same?
    • Does the hierarchy-status or knowledgeability of the librarian matter?
  • Decisions are not immediate. Require time to collect votes and form meaningful discussions.
  • Idealogical clashes interferes with objective decision-making.

V2 Prototype + Testing
Following our user-testing of the V1 (low-fidelity) prototypes, we gathered our notes and went back to the drawing board to combine both ideas in order to create a low-cost database and training tool for public librarians that
  • Allows us to create impact in under-funded libraries.
  • Creates a common knowledge base for library systems.
  • Could potentially act as a form of librarian-centered mutual aid.
  • Doesn’t halt or obstruct the work of other employees.
  • Address the needs from both patrons and library staff in a feedback-loop design.
We took into consideration the risks that weren’t encompassed by our V1 prototypes in order to remove or lessen any blind spots in our application (ex. assisting patrons who are afraid to ask for help). With that in mind, we present our mid-fidelity hybrid prototype called the "Patron Help Center".

Patrons searchs and asks a question.
Librarian responds to question.
V3 Prototype + Testing
(Final Design)
Visual Analysis
Our low-fidelity and mid-fidelity prototypes were chiefly based on an early desire to tailor the tool around the needs of patrons who prefer analog mediums or may not feel comfortable with digital tools. Therefore, the early digital solutions focused on keeping them focused, short, and simple. We wanted the (final) design to be adaptable to many different types of users. Interfaces were "brought back to basics" by asking simple, direct questions. Reducing the visual impact of the interface allowed the functionality and straight-forwardness of the tool to be front and center. Therefore, white backgrounds with grey highlights were chosen as the base colors, brighter colors--such as blue or green-- were sparingly used for specific purposes such as ‘Submit’ buttons or update notifications. No iconography was implemented in the (final) prototype; however, custom icons could theoretically be implemented in order to differentiate the implementation of this system in different library systems. One analysis of note in the development of the prototypes was connecting the ideate phase back to new areas of exploration found during the ideation exercise; this allowed us to capitalize on the ‘invisible labor’ of librarian staff responding to questions, but not documenting those responses in a way that is useful for others that may have that same question. This was a key insight in developing a symbiotic tool of mutual aid for patrons and librarians.

The next four slideshows showcases how patrons would ask and follow questions, and how librarians would assign and respond to them.

Asking

Following

Assigning

Responding

For the final-round of testing, we looked back at our previous iterations and made sure that our final tests would continue capturing any relevant information regarding the usability and accessiblity of the prototype, while also capturing any information that wasn't previously attained like likelihood of usage.
Non-Target Users (Friends and Family)
  • Simplistic design with basic, yet essential functionalities.
  • Similar to other Q/A platforms (ex. Yahoo Answers and Stack Overflow).
  • The question-form isn’t your ordinary “enter question within the textbox and submit” form.

Target Users (Librarians and Patrons)
  • Easy to use and gets the job done.
    • Familiarity with similar platform.
    • Simple navigation can help patrons find the answers they need on their own without the library’s help.
  • Elderly individuals able to somewhat navigate on their own.
    • May need help learning how to use the platform.
  • Language support may be essential for non-native English speakers.
  • If well advertised, it can be something that patrons can use instead of contacting librarians directly.
Reflections on the PIT Design Process
The following are some of the challenges and opportunities that came about from following the PIT Double Diamond design process. We hope our experience serves as an example for future designers and developers who choose to use this approach.

  • Make note of all visible and invisible infrastructures present within the environment you’re observing. In our case, we documented both the technology infrastructure present in library systems (ex. computers/tablets, communication platforms, public and librarian-only websites), and the social infrastructure manifested from patron-librarian communications and the communities (growing within/living in/inhabiting/entering) the library.

  • The “PIT Double Diamond” process will make sure that your final design is a precise reflection of the actual needs requested by your target audience and not those which might’ve assumed during the initial phase. Repeating the cycle of “Research, Prototype, and Testing” phases may filter and/or conglomerate several ideas until you obtain the right one.

  • Figuring how NOT to design an overzealous application that infringes on the daily lives of your target user will be an interesting challenge. In some cases, like ours, your research and interviews will uncover a significant problem that can be solved with the simplest, yet most meaningful and beneficial application.
Acknowledgments
We would like to thank all the librarians and library staff we’ve interviewed throughout this semester for educating us on their day-to-day lives and for providing us with the insights necessary to create our tool.

In addition, we would like to thank the entire teaching staff, in particular, Anne-Laure for helping redirect our project in the right direction and providing meaningful feedback for all presentations. We could not have accomplished this without your help!