![\"\"](\"https:\/\/sites.temple.edu\/assessment\/files\/2022\/12\/screenshot-1-1024x320.png\")
<\/a>With this configuration, item records with long title statements or multiple subfields (like the one above) can become lengthy and hard to decipher, making search results hard to scan and distinguish from one another. It\u2019s not ideal for how users read online<\/a>, scanning quickly for the information most relevant to them. We know from our own past user research that users tend to scan through results in Library Search quickly, in search of key bibliographic information like author, title, or date, that serve as sign posts to help them find what they need and discard what they don\u2019t. We\u2019re also aware of how challenging it can be to distinguish one item from another when looking through dozens of results. <\/p>\n\n\n\n With Library of Congress moving towards omitting ISBD punctuation from MARC records, we needed to rethink how we are indexing and delimiting between different MARC subfields for relevant Library Search displays, such as title statement, imprint, etc. With the physical space of a 3×5 card catalog record no longer a limiting factor in how we display information about works, we can explore alternative approaches that do not rely on prescribed punctuation, such as using the infrastructure of our discovery systems to create a human-readable display. <\/p>\n\n\n\n The Discovery Oversight Group formed a working group to investigate these opportunities and challenges. Led by Emily Toner (Library Technology Development), the group included Leanne Finnegan (Metadata and Digitization Services), Jackie Sipes, (user experience) Holly Tomren (Metadata and Digitization Services), and Joi Waller (graphic design). Ahead of the fall Library Search development sprint, the working group held a week-long design sprint to explore how we might display MARC data without the use of ISBD punctuation. The design sprint was held ahead of the development sprint to allow ample time for collecting user feedback before moving forward with significant design changes. <\/p>\n\n\n\n We completed the design sprint across 5 days with group members spending anywhere from a 1-3hrs per day on meetings or other sprint activities. Focusing on highly visible data, such as title, author, and publication information that provides essential information to our users when they are reviewing search results and record pages, we proposed three “How might we<\/a>” questions to guide us as we brainstormed solutions: <\/p>\n\n\n\n A few of our ideas included: <\/p>\n\n\n\n Based on these ideas, Joi created 6 different prototypes<\/a>, each with different configurations of the title, subtitle, and statement of responsibility. <\/p>\n\n\n\n For the prototypes, we used the search Hidden Figures which provides more than one example of the problems we wanted to solve \u2013 helping users disambiguate between different items by the same title in multiple formats as well as containing items with longer title statements where ISBD punctuation is used to separate MARC data. <\/p>\n\n\n\n To gather feedback, Joi and I (with lots of help from John Pyle) set up a table and rolling whiteboard in the Charles atrium. Noticing that traffic in the building felt unusually low that day, we hauled our setup outside and positioned ourselves under the overhang near the 13th street entrance on a perfect, sunny, fall day. <\/p>\n\n\n\n We had 8 participants in total, including 3 undergraduates, 3 graduate students, 1 faculty, and 1 staff. All but two had used Library Search previously, and those who hadn\u2019t had experience using library catalogs at their previous institutions. Participants came from a range of TU schools and colleges including Biology, English, French, Music, and Public Health. <\/p>\n\n\n\n To get feedback we used a combination of visual preference testing, informal in interviews, and a brief survey. <\/p>\n\n\n\n For the preference testing, we showed participants the 6 prototypes printed on large 11×17 paper and taped to the white board for easy comparison. Participants were given the following prompt: <\/p>\n\n\n\n Here\u2019s a scenario. You\u2019ve just searched for the book Hidden Figures,<\/em> and these are the first three results you see<\/strong>. Spend a minute or so examining the prototypes. There are subtle differences; don\u2019t get too caught up trying to look for those. Just think about would be most helpful to you if you were looking at these results and trying to evaluate these results.<\/p>\n<\/blockquote>\n\n\n\n Of our prototypes, most participants preferred E, with F a close second. Though design differences were subtle throughout the prototypes, participants agreed that the blue background in E and F helped draw them to the item title right away. The response to the updated displays was generally very positive, confirming our decision to separate the statement of responsibility, which contains author information, as well we any other persons, corporate bodies, etc. that contributed to the content of the source, out from the title section.<\/p>\n\n\n\n To implement a separated the title and statement of responsibility in Library Search, we created separate fields for indexing the title and the statement of responsibility and configured the display of those fields through design elements and page structure. Shifting the statement of responsibility below the title suggests a hierarchy to the information presented, while additional styling elements, such as the high-contrast font and royal-blue background color and reconfigured (larger!) font sizes in the title area, help users hone-in on the details needed to identify and disambiguate sources as they browse search results. <\/p>\n\n\n\n There was less consensus from users around the display of subtitles. Some participants found prototype E (with the subtitle on a separate line) to be less cluttered and easier to read. A participant who preferred F, commented that she liked seeing title and subtitle together on a single line as it provided context for how the item was different than the titles above and below.<\/p>\n\n\n\n We initially planned to implement the separated subtitle (prototype E), but once the development sprint started, we realized the need for a careful and more thorough investigation before proceeding with additional metadata mapping updates. So, that change is on hold for now. <\/p>\n\n\n\n The near consensus from users on the visual design\u202fwas interesting, but not unsurprising. Breaking the statement of responsibility out of the title statement was a clear improvement for readability. Talking with users also helped us clarify what item information they find most useful in the search results display. We asked participants to review a list of 10 items and indicate which 5 they rely on most when scanning search results. <\/p>\n\n\n\nDesign sprint<\/h2>\n\n\n\n
\n
\n
User feedback<\/h2>\n\n\n\n
![\"\"](\"https:\/\/sites.temple.edu\/assessment\/files\/2022\/12\/charles-exterior-1024x833.jpeg\")
<\/a>\n
Title statements<\/h2>\n\n\n\n
![\"\"](\"https:\/\/sites.temple.edu\/assessment\/files\/2022\/12\/1-2-1024x320.png\")
<\/a>![\"\"](\"https:\/\/sites.temple.edu\/assessment\/files\/2022\/12\/prototype-EF-hidden-figures-1024x320.png\")
<\/a>![\"\"](\"https:\/\/sites.temple.edu\/assessment\/files\/2022\/12\/prototypeEbook.png\")
<\/a>![\"\"](\"https:\/\/sites.temple.edu\/assessment\/files\/2022\/12\/prototype-F-1024x404.png\")
<\/a>Summary previews and other improvements <\/h2>\n\n\n\n
![\"\"](\"https:\/\/sites.temple.edu\/assessment\/files\/2022\/12\/final-design-Edited-1024x576.png\")
<\/a>