Last spring dh+lib published the special issue “Making Research Tactile: Critical Making and Data Physicalization in Digital Humanities,” which featured seven case studies on ways critical making could be integrated into a digital humanities (DH) research practice. This follow-up special issue features concrete ways we can integrate critical making into our (library) instruction. Given the library’s role in technology instruction and data literacy–as well as its function as a space that embraces creative methods–we wanted to provide some explicit examples of teaching methods that can work in everything from a technological one-shot to a more embedded instructional model. If you want to know more about what data physicalization is and why it is a great tool for DH, please see the introduction to that previous issue. 

Technological instruction without computers 

Teaching data, or other technological instruction, through crafting can strengthen our pedagogy in many ways. Russell and Hensley warn of the dangers of “buttonology” and focusing our instruction too much on the interface of specific tools (2017). When we do this, we risk not teaching students how to think computationally, and instead how to navigate specific software, which could become immediately outdated when something new is released or when that software is no longer available. And as Cammeron et al. show us, a minimalist and low-tech pedagogy can more quickly get students thinking about technological concepts (2024). This can also help address technological anxiety and frame assignments as more “fun” for students who are less drawn to doing humanities work digitally, which can foster students’ curiosity to pursue the topics further. 

Over-emphasizing programming languages or computer software perpetuates limited notions of what counts as technical skill. Adopting a more expansive view of technology to include things such as weaving, sewing, paper arts, and more, enables us to honor and value the broad range of skills, experiences, and interests our students bring with them into our classrooms. By situating these technical and creative skills in larger contexts beyond the digital, our digital or data literacy instruction can appeal to a more diverse range of learners. Put another way, de-emphasizing the coding aspects of technological instruction opens up learning spaces to a wider range of people, including women, BIPOC, LGBTQIA, and economically-disadvantaged students, who, because of past negative experiences or stereotypes, may not gravitate toward programming-oriented topics or classes. In this way, we see this approach as one of many interventions, such as Girls Who Code, to expand and make more accessible technological instruction. We hope this issue inspires you to imagine more accessible and inclusive approaches to your digital humanities instruction that are rooted in this expansive view of technology.

In this issue

This issue consists of eleven data physicalization case studies, specifically situated in instructional contexts. Like our previous issue, these case studies take inspiration from cooking blogs: each piece includes a narrative describing the authors’ experience doing the work and a set of instructions to try doing it yourself. We’ve also collected these instructions into their own zine, which you are free to print, share, and remix. 

Within the issue, you will find projects that use a wide range of crafting methods, from beading to circuit building. We encourage you to take inspiration from these examples, but do not feel limited by them. If there is a type of craft or making you are most comfortable using, there is probably a way to use it for this type of work too! We hope this special issue can serve as an invitation to join us in this work. Please try any of the projects in the issue, post pictures of what you make using #DHMakes—a hashtag many folks in the community have been using—and if you find more readings on this topic, or write one yourself, add it to our Zotero group for resources on data physicalization.

Citations

Cammeron, Malcom, Carter, Caroline, Pérez Martínez, Winnie E., Stephen, Samantha, and Brandon Walsh. 2024. “A Way In: Digital Pedagogy Training with Speculative, Low-Tech Workshops” Journal of Interactive Technology and Pedagogy, vol 24.

Russell, John E., and Merinda Kaye Hensley. 2017. “Beyond Buttonology: Digital Humanities, Digital Pedagogy, and the ACRL Framework” College & Research Libraries News, December. https://doi.org/10.5860/crln.78.11.588.

Introduction

Data visualizations—and, by extension, data physicalizations—often make data more accessible visually. Colour-coded graphs and flow charts with graphics and arrows can be easier and quicker to read at a glance than a long table of data or numbers and percentages hidden within long prose. Wearing that data as a scarf is also visually appealing, and it makes for a great conversation starter about the work. Although these graphs and charts can make data more accessible for sighted audiences, they often make data less accessible for visually impaired audiences, as the details of some data visualizations can be difficult to describe in alt text, and alt text often is not used or difficult to access in many publications. (For information on how to make data visualizations more accessible, consult Chartability.) Due to their inherent tactile nature, there is an opportunity to make data physicalizations more accessible than their flat, digital counterparts.

To take examples from the previous special edition featuring data physicalizations, the woven emotional bookmarks, the data ball, and the porcelain Me & You project all resulted in physical, tactile objects that one could hold or display on a shelf or their self (in the case of wearable projects). Though these projects are tactile, they still largely rely on the sense of sight to be interpreted. As I read the 2024 special issue, I began to think about how we could take data physicalizations a step further to be accessible to people who are blind or visually impaired by using the advantages of the tactile nature of the crafted data. Though I have never considered myself an artist, I did pick up a craft during the lockdowns of 2020 (as many other people did), and I began to crochet. As I became more comfortable with my craft, I began to accumulate what fibre arts folks describe as a yarn stash. My stash takes up the entire closet in my home office, and as I studied my stash, I contemplated how I could use the variety of yarn I had available to create something beautiful, informative, and accessible tactilely and visually. Yarn is a wonderful material because it comes in so many different colours, weights (thickness), and textures, and, because of its versatility, it can be used for a wide variety of crafts and projects.

The goal of this woven tactile pie chart project is to make data accessible to sighted and visually impaired audiences. By using a combination of yarn colour and weight, the pie charts are completely legible by sight and by touch. Adding braille data labels makes the data even more accessible to people with visual impairments. For the samples used in the pattern tutorial below, I used data from the 2023 Statistics Canada report on Print Material Accessibility. According to this report, 5.2 million people in Canada report difficulty reading print materials. The weave that illustrates the tutorial below uses the statistic from this report that 77% of those 5.2 million people have difficulty seeing words in print. I used additional data from this report for more woven pieces, which I then sewed together to make a larger tapestry of data.

The pattern is specifically designed for percentage charts with two values. This pattern will not work as effectively for pie charts that include more than two values, as it would become more difficult to feel the difference between weights of yarn if there is a more gradual change (from 4-weight to 5-weight to 6-weight, for example) rather than the stark difference used in the sample (from 4-weight to 6-weight). Multiple data points may work better if additional textures of yarn are used, rather than simply relying on the thickness, but it is possible that more data points would still make it more difficult to read tactilely than the speed of visually seeing the different colours used. Therefore, the pattern is most effective for charts with two labels.

For the data labels, I am using unified English braille (UEB, which is the most commonly used variation of English braille worldwide), and I chose to use uncontracted braille, or grade 1 UEB. Using uncontracted braille means that every braille character stands for a letter in the visual alphabet (with exceptions for some punctuation marks that require two characters); contracted braille has short hands—or, more simply, contractions—that help to shorten the text. Contractions can be a single character for a whole word (the ⠮), a single character for a letter formation (ar ⠜), or multiple characters for a single word (braille ⠃⠗⠇). Typically, braille readers will learn the uncontracted alphabet before learning the contractions, so using uncontracted braille for the data labels makes it more accessible to a wider braille-reading audience—it also makes it more accessible for people who do not read braille to add the braille labels, as contracted braille has a lot of complicated rules.

The tutorial below discusses how this project was possible, so I now want to discuss why this project is important—why making data accessible is important. One of my favourite books about data is Data Feminism by Catherine D’Ignazio and Lauren F. Klein (2020). In their chapter called “What Gets Counted Counts,” D’Ignazio and Klein discuss how being counted in data—or, conversely, not being counted—matters significantly in terms of political, cultural, and economic representation. Though the chapter focuses on gender representation in data (such as census data, among others), the intersectional feminist perspective with which the entire book is written concludes that all types of diverse data are important to ensure accurate, ethical data (and data representation). My research overall focuses on accessibility and representations of disability, so ethical data collection and representation are core focuses for my work. As an extension to that, accessible reproductions—including visualizations and physicalizations, as well as the inclusion of alt text and captions—of data are also vitally important to ensure research is being communicated ethically to diverse audiences. It is not enough to count diverse data if that data is not then disseminated and represented accessibly and ethically.

In this same chapter, D’Ignazio and Klein argue that “the absence of data becomes an important takeaway, as meaningful as the data themselves” (D’Ignazio and Klein 2020, 113). In this quote, they are referring to nonexistent data, such as that of nonbinary and genderqueer people when the only gender choices on a form are “male” and “female,” but their argument stands for the lack of accessibility to data as well. The absence of ethically, accessibly disseminated data has meaning for the audiences who are unable to access that information. Data physicalizations are, of course, only accessible by the people who are physically able to access them in the space where they exist, but they should be fully accessible to all audiences who are present in that space. As D’Ignazio and Klein say, what gets counted counts; being able to access what has been counted counts, too.

The pattern below will guide you in creating your own tactile pie charts, including brief instructions on adding braille data labels. It is okay if you do not know how to read braille; you can still add braille data labels. Figure 1 has the basic UEB alphabet and numbers. To use a capital letter, use the capital letter indicator before the letter you want capitalized. To capitalize a whole word, use the indicator twice, and to capitalize a whole passage, use it three times and then put the capital terminator indicator at the end of the affected passage. To use a number, put the numerical indicator before the braille character you want to read as a number (note that this is not the same as the character #, which has its own braille character). The numerical indicator persists until a space. If you want a combination of letters and numbers, use the grade 1 indicator or the capital letter indicator before the characters you want to read as letters (52e = ⠼⠑⠃⠰⠑ or 52E = ⠼⠑⠃⠠⠑).

Figure 1: The unified English braille alphabet, including some punctuation marks.

Materials

• A handheld loom—any shape will work as long as it is a notched loom, not a pin loom, but I used a hexiloom
• A loom comb
• A pick-up stick
• A darning needle and a long weaving needle—you could just use one or the other, but I find having two is often useful
• Yarn scissors
• A 4-weight yarn and a 6-weight yarn in different colours
• Cardstock business card size paper
• A braille slate and stylus—I used a business card size slate that fits the paper, but any size slate will work with some adjustments

Figure 2: Materials required for the weave, excluding the braille slates for labelling.

Pattern

If you would prefer a visual tutorial on how to use this type of loom, the maker of the loom I used made a video tutorial on YouTube.

1. Tie your 4-weight yarn to one side of the loom. Make sure the knot is snug, and leave yourself enough yarn on the end to easily sew it in at the end. (Leave more than I did in Figure 3, as I found mine too short to sew in easily.)
   

   Figure 3: How to tie your warp layer onto the loom.

2. Create the warp layer on the front of your loom. Wrap the yarn around the notches at the top and bottom of the loom. The strands should only be on one side of the loom, as shown in Figures 4 and 5.
   

   Figure 4: The front of the loom with the warp layer completed.

   

   Figure 5: The back of the loom with the warp layer completed.

3. When you finish the warp layer, tie the end of your yarn in the same way you did at the beginning. Make sure it is snug, as having the end loose can create problems as you weave. (I had to retie mine partway through making this sample, as it was too loose and it started to fall off the last two notches as I was weaving.)
4. Now using the 6-weight yarn, cut the amount you need from the skein. You need to weave the whole amount through the loom on every row, so you cannot use a full skein. Roughly layer your yarn over the loom to estimate how much you will need. If you have too much, you can cut off the remainder at the end. If you do not have enough, you can tie another piece on to continue weaving. (I recommend the magic knot. I used this method, and it’s impossible to tell that there is a knot in my work after it was completed.)
5. Thread the yarn into the darning needle and begin your weave. For the hexiloom, start with the bottom five warp strands, as shown in Figure 6. Weave by moving your needle below a warp strand, then above the next one, then below the next one, and so on. Leave yourself a long enough tail to sew in at the end. (Again, I recommend leaving a longer tail than shown in Figure X, as mine was too short to easily sew in at the end.)
   

   Figure 6: The beginning of the weave with the 6-weight yarn and a tail to sew in.

6. Weave the next layer by alternating which warp strands you are going above/below, adding a strand on either side to start working up the sides of the hexiloom. (If you are using a square loom, you will start by working the full width of the loom.)
7. At this point, you may want to start using the pick-up stick. Place the stick under the warp layer along the full width of the loom. This helps to pick up the work so it is easier to weave. If your warp is tight and it is difficult to move the pick-up stick into place, using the long weaving needle to lift the warp.
   

   Figure 7: There are a few rows completed now, and the pick-up stick is underneath the warp.

8. Continue this pattern, working back and forth along the loom and increasing the number of strands you work with until you are using the full width of the loom. Use the comb after each layer to tighten the new layer against the previous layers so there are no gaps in your work. If you wish, you can thread your yarn onto the longer weaving needle so you can weave longer sections on each layer. I used my darning needle for the entire weave, as the eye of my weaving needle was not big enough for the 6-weight yarn.
9. The rest of the pattern will vary depending on the numerical value of the data you are weaving. I am weaving 77%, so I am using the 6-weight yarn to take up approximately 77% of the loom surface. Adjust accordingly to the value of your data.
10. Figure 8 shows approximately 50% of the loom woven with the 6-weight yarn. Since I am creating a pie chart, I will no longer weave the full length of the loom after this point. To create the effect of 27% on the top half of the weave, my next layers will be slightly longer than halfway across the width of the loom. Figure 9 shows the length of the remaining layers. Continue the back and forth weave pattern until you reach the top of the loom, then begin to reduce the number of warp strands you weave with to get to the top tip of the hexiloom.
    

    Figure 8: Half of the weave is completed with the 6-weight yarn.

    

    Figure 9: Half of the weave is completed, plus a few rows to slightly past halfway across the loom to simulate the remaining 27%.

11. When you finish the 6-weight section, leave yourself a tail to sew in. Figure 10 shows the completed 77% portion of the work.
    

    Figure 10: The 77% section of the weave with 6-weight yarn is complete.

12. Now thread your needle with the 4-weight yarn again to weave the “negative” data portion.
13. Start your weave at the inner corner to meet the 6-weight yarn. Put your first stitch into the 6-weight yarn to connect the sections, as Figure 11 shows. Weave towards the outside using the same above/below pattern.
    

    Figure 11: Begin your 4-weight section by weaving into the 6-weight section to connect them.

14. For your second layer, stitch the 4-weight into the 6-weight as you work towards the centre of the loom. Weave your needle above and below the first two strands, then stitch below into the 6-weight yarn. Then weave above and below the next two strands, then stitch into the 6-weight again. Repeat until you finish this layer.
15. Work your third layer back to the outside of the loom as normal. Then follow the pattern of weaving above and below as normal up the remaining section of the loom. Approximately every second or third layer towards the centre of the loom should stitch into the 6-weight yarn to keep it connected. If you do not stitch into the 6-weight yarn like this, the result will be disconnected and not as strong. Figure X shows approximately half of the 4-weight section completed with stitching into the 6-weight section.
16. When you finish the rest of the 4-weight section, cut off your remaining yarn and leave a tail long enough to sew in. You can now remove the project from the loom. Carefully remove the warp from the notches around the bottom of the loom, and then the project should easily pop off of the top. Untie or snip the warp that is tied to the loom. You can then sew in all of your ends.
    

    Figure 12: The completed project on the loom.

    

    Figure 13: The completed project off the loom.

17. You now have a tactile pie chart! Using different weights of yarn allows you to feel the different sections on the chart, and using different colours allows you to quickly see them as well. The next steps are optional, if you want to add braille data labels to your work.
    

    Figure 14: The completed project with the ends sewn in.

Adding braille labels

1. See Figure 1 for a guide on braille characters.
2. Plan where the braille characters need to be stitched to align properly when finished. If desired, you can use stitch markers to mark where the characters should be to avoid error.
3. Thread the 4-weight yarn on your needle to cross-stitch the braille characters. I used the same colour since it is meant to be read tactilely, but you can use a different colour if you want to.
4. Braille characters are made up of dots, so each dot in a braille character should be one stitch, and the collection of stitches in a particular pattern creates the braille character. If you are proficient in embroidery, you can use the French knot stitch to create each dot in the braille character. However, I found that my weave was not tight enough to hold the French knot stitch in place, so I did a modified cross-stitch instead. To do the cross-stitch, stitch two lines to make the x and then go over the first line again, so you stitch: \ then / then \ again. The third stitch makes it thicker and easier to read tactilely. For the sample weave, I stitched “77” (⠼⠛⠛—a numerical indicator then “gg” for “77”). There was not enough space to stitch the two-character % sign (⠨⠴), which I would include on labels where I have more space.
   

   Figure 15: The completed weave with the braille label cross-stitched on.

5. To create the paper labels, I used cardstock-thick paper in business card size. This is specifically braille paper, but any form of cardstock should work. I wrote the data label with a pen: 77.4% have difficulty seeing words in print. I then used a braille slate and stylus to write the same label in braille. I chose to use uncontracted braille to make it legible to a larger number of braille readers (explanation in the introduction). Remember to write with your pen before brailling, because it will be difficult to write over the embossed braille. Also note that you need to braille with the stylus backwards and on the backside of the paper so that it embosses to be read on the front, the same side as the writing.
   

   Figure 16: The paper label on the back of the completed weave.

   

   Figure 17: Multiple completed pie charts sewn together to create a tapestry of data.

   

   Figure 18: The back view of multiple completed pie charts sewn together to create a tapestry of data.

Ellen Forget

Ellen Forget is a PhD candidate at University of Toronto in the Faculty of Information and the Book History and Print Culture specialization. They are a graduate of the Master of Publishing and Editing Certificate programs at Simon Fraser University, and they are certified in braille transcription from the Canadian National Institute for the Blind.

• More Posts

Brandon Walsh describes this as the “three-speed problem” of DH pedagogy: in any given room, instruction will always be too fast for some, too slow for others, and just right for a few (2023). To address this, Walsh proposes low-tech approaches that foreground inquiry over learning to press the right buttons or “buttonology,” per Russell and Hensley (2017). Similarly, Jentery Sayers explores hands-on methods using paper, scissors, glue, and cardboard, shifting the focus from tools to broader discussions about how data are “produced, processed, and expressed” (2019). Inspired by these approaches, I found new possibilities for DH pedagogy at the #DHMakes mini-conference at DH2024. Surrounded by paper, fabric, yarn, and a range of creative engagements with technology, I saw an opportunity to rethink instruction about data to center critical conversations over technical tools.

This piece reflects on my attempt at designing a data physicalization workshop. I share lessons learned, practical pointers, and workshop materials for those interested in integrating crafting into DH pedagogy.

Tracking Changes with Tapestry

In early 2025, the Trump administration began scrubbing federal websites related to gender and diversity, climate science, and healthcare issues. Some websites were temporarily unavailable, such as Census.gov. Among pages deleted on the Center for Disease Control website were health disparities in LGBTQ youth and information on HIV including stigma, living with HIV, and preventing HIV. Other sites were significantly altered, such as the removal of transgender and queer people from the National Park Service Stonewall National Monument.

I have been trying to comprehend these changes and to begin understanding their scope and magnitude. At the time of writing this, there is no comprehensive dataset of these edits. Even if all the links were captured, a dataset alone would not capture details such as choices in design, emphasis, language, and linked resources.

For example, the Stonewall page’s banner once featured a drag performance but was replaced with a rear view of George Segal’s Gay Liberation sculpture. Using the Wayback Changes tool, I compared the January 19th, 2025 version with its revised state, visualizing insertions and deletions side by side.

Two versions of the Stonewall National Monument website viewed through the Internet Archive Wayback Changes tool. Some items are highlighted due to the underlying code being changed, though the content remains the same.

My attempts to grapple with this reminded me of the “data visceralizations” of COVID deaths, such as Jacqueline Wernimont’s Memory Weaving (2022) or the National Covid Memorial Wall, both leveraging embodiment for an affective understanding of incomprehensible change (D’Ignazio & Klein 2020). The tactile representation in the tapestry, the experience of walking along the wall, and the visibility of the labor that went into these pieces is what makes them so effective.

I had an idea for turning these website revisions into a woven tapestry with students in a workshop setting to create an opportunity for discussion, exploration, and processing the erasure we are seeing. Turning the revisions into tapestry offers a way to represent changes beyond a binary diff—distinguishing between design, media, interactive elements, and text while also revealing major losses, such as deleted videos and resources.

As a teaching tool, this approach encourages close reading across text, media, and design. Students translate their analysis into structured data, make choices for physicalizing that data, and then craft a tapestry representing a portion of the site. While the Stonewall homepage presented rich material, I chose a different example for the workshop, being mindful of participants’ relationship to the material. We worked with the CDC’s About HIV page, which underwent significant revision before being partially restored after February 14th, 2025.

Lessons Learned

This workshop was my second attempt at teaching data physicalization. With both workshops, I was taken in by the deceptive simplicity of the tools—cardboard, yarn, scissors, pen, paper. What can go wrong?

From the generous feedback of colleagues who attended the first workshop, I learned that you cannot skip over discussing what data are and how they are made, as Miriam Posner discusses in “Humanities Data: A Necessary Contradiction” (2015). To grasp physicalization, it is helpful if participants practice and create a full piece, not merely translate a few data points, such as abstractly representing a couple lines of data in yarn. Working toward a completed piece helps to concretize the motivations behind data physicalization, understand what this kind of representation is hoping to achieve, and allow individuals to develop an argument around their own craft. Focusing on creating a piece also brings the artistry of data physicalization into the room which can get lost if you focus merely on the process of learning a craft and converting the data.

I am still seeking ways to simplify the craft and allow more space for discussion. This time, as I was standing in front of the room, attempting to weave with the board in front of me (you should practice this), I instructed people to do a twining stitch to secure their tapestry. This was a mistake. I had flashbacks to teaching python—the flash of errors across the room and frustrations. Weaving was easy for some people, but challenging for others who lacked experience and dexterity. It was a reminder that crafts, like coding, are also complex technologies. Even in an analog setting, workshops can replicate the same pedagogical challenges as digital tools—varying skill levels, uneven comfort with technology, and the three-speed problem. I had turned to craft to avoid buttonology, aiming for a low-tech approach as Walsh proposed, but crafting workshops, too, risk becoming rote technical instruction rather than spaces for critical engagement. The more intricate the technique, the more likely the instructor-student dynamic mirrors Freire’s banking model of education (2000).

This is not to say that the workshop was a failure or that crafting workshops should be avoided, but that they are not inherently “low-tech” in the same way that Walsh and Sayers are working, simply by the virtue of being non-digital. This workshop was more successful at teaching data physicalization as a decision-making process, but successfully completing a craft does take time away from critical discussion. Moving forward, I hope to refine the balance between making and meaning—ensuring that technical instruction supports, rather than overshadows, the inquiry at the heart of the workshop.

Workshop Design and Materials

This workshop seeks to meet the needs of a mixed classroom, where some learners may have advanced knowledge of how to work with the data or the craft and some may have no experience at all. Ideally, schedule 90 minutes to allow for an introduction to frame the workshop followed by time to do the craft. Leave 10-15 minutes at the end for people to share their process and reflect.

This workshop focuses on the affective impact of physicalization on data representation and the process of translating data into representation in a tapestry, which may be more interpretive than other visualization processes familiar to learners. The concept of affective data is introduced at the beginning of the workshop with a brief presentation and discussion on data feminism (D’Ignazio & Klein 2020) and examples of woven data.

Learning Outcomes

• Examine the impact of physicalization on data representation
• Read and construct tabular data
• Perform a close reading of two versions of a text
• Develop a method for representing changes in text in tapestry

Supplies and Preparation

• Cardboard
• Cardstock, cut in strips
• Scissors
• Warp string
• Yarn or fabric, variety of textures, weight; pre-cut for students
• Popsicle sticks
• Yarn needle
• Darning needles (optional)

You don’t need a lot of supplies to run this workshop, including looms—DIY cardboard looms work well. When selecting yarn, vary texture and weight rather than relying solely on color. Instead of offering a full-color spectrum, provide a curated selection of complementary shades, including neutrals.

Any yarn weight will work, but note that while thicker weight yarns will build a project faster, very chunky yarn will make a project somewhat messy. In the example in these instructions, which comes from the workshop, the large contrast in yarn weight, the very heavy yarn, and the narrowness of the tapestry led to the piece being slightly unstable and wonky. Correcting for these would make a more even tapestry.

If working with a large group, pre-cutting materials helps streamline the process.

Tapestry Step-by-step

Step 1: Choosing a Document

To make this tapestry, begin by choosing a document that has two versions. This example uses the Internet Archive Wayback Changes tool, but it could also be a manuscript revision (e.g., The “Yellow Wallpaper” digital edition), edits in a GitHub repository, or changes to a Wikipedia page.

I recommend making physical, printed copies of the documents you are comparing. For efficiency, print versions with diffs visible.

Two versions of the CDC webpage about HIV viewed through the Internet Archive Wayback Changes tool. Some changes highlighted are differences in code at the back end rather than changes to the content, but starting at the navigation bar, it is evident that some content has been removed, such as pages for “How it Spreads,” “Stigma,” and “HIV Awareness.”

Step 2: Annotating Changes

Mark insertions, deletions, and modifications in the text. Consider changes in different media types—text, links, images, or design elements. With the CDC HIV example, a missing link often signals the removal of an entire page of information.

The diff checker missed a few changes and failed to capture some of the features I wanted to represent. I annotated deletions, insertions, and modifications, noting media, links, alert boxes, and alterations to design. In this example, I tracked line numbers by hand. In the workshop, I provided a table of the diffs with empty spaces for filling in the types and notes to model a method for converting these observations into structured data.

The table, which was printed out, had two completed rows to provide an example for attendees. In the short timeframe of a workshop, this provides clarity and helps people build their tabular dataset.

Step 3: Organize your data

Create a table with columns for:

• The changed content
• The type of change (insertion, deletion, modification)
• Any qualities you want to represent in the tapestry

Each row will correspond to a section of woven data.

Step 4: Planning the Physicalization

Decide how you would like to represent these changes on a tapestry and select your materials for the physicalization.

Will you show only changes, or also the unchanged content for context and contrast? If most of the document remains the same, representing unchanged content may not be useful.

For the workshop, I created two patterns as options: one focusing solely on changes, another including the full diff. I used word count, but line count may provide a clearer structure and help learners draw the connection from method to physicalization.

Color choices: I selected cream, blush pink, and sage green, incorporating varied textures like sage cord and undyed wool roving. I inverted the common “green for new, red for old” convention, avoiding the implication that “green” always signifies positive change. Thick sage cord represented removed links and media, emphasizing their weight.

Step 5: Making the Loom

Set up the loom. DIY cardboard looms work well, but should be pre-set for a workshop. Small, even cuts along the cardboard create space nestle the warp string. Some loom tutorials cut and glue horizontal strips of cardboard which make the weaving easier, but are more time consuming to make for a whole class.

Nestle the warp string into the first notch in the cardboard, leaving a 3-4” tail. Pull the string all the way down the loom and nestle into the corresponding notch at the bottom.

The string is nestled into the first notch. Make the cuts even as possible both in terms of their distance from one another and depth in the cardboard.

To make your way back up the loom, do not take the warp string up the length of the reverse side. Instead, from behind the loom, pull the string sideways and through the next notch to the front. Now you can pull the string back up to the top.

The warp string pulled tight across the back and through the next notch to the front.

Repeat this process of pulling the string down, behind, through to the front, and back to the top until you have enough warp to work with. For the tapestry, I have made it a little tall and narrow. The start and end of the loom should be on the same side and oriented to the bottom of your tapestry.

Completed view of the front.

Completed view from the back. When you start working, you should flip it so that the tail ends are at the bottom.

Step 6: Weaving

Create a base. Weave a strip of cardstock or flexible cardboard (such as a cereal box) at the bottom for stability, following a simple over-under weave—going over the warp string and under the next, repeating all the way across. This base provides a straight edge for weaving and also ensures there is enough string to tie off the tapestry when finished.

Optionally, add a twining stitch to maintain stability when the tapestry comes off the loom. A twining stitch creates a strong edge finish by twisting around the warp strings. It is best understood by watching a tutorial, such as the Spruce & Linen guide to starting a weaving.

Cardstock is used to create the base.

Weave from the bottom up using plain weave—simple over and under, back and forth across the warp. Pulling tightly across will start to build tension and bring the sides closer and closer together. Avoid this by making an arch shape as you pull the yarn across the warp and then push it down.

The weft is pulled up and then across in the shape of an arch, helping prevent building too much tension.

Tapestry weaving typically starts from the bottom and moves to the top. I wanted my tapestry to somewhat represent the webpage visually, so I started at the bottom of my data table and worked in reverse order, rendering the diffs in thread: unchanged text, deletions, insertions, modifications.

Experiment with stitches. You may stick to plain weave for simplicity or introduce new weaves for modifications, looser weaves for deletions, or loops for errors. However, a selection of yarn weights and textures can add a lot of variety to a tapestry without teaching new knots. When working with roving or cord, try pulling a little out of the warp threads to build some texture. Roving is unspun wool and its fluffy texture provides a nice contrast.

When you begin and end a new piece of yarn, leave a few inches hanging off the side to be tucked in later.

The long ragged ends will be woven into the back.

The majority of this piece is completed using plain weave. The introduction of different materials provides visual interest without increasing technical difficulty for beginners.

Step 7: Removing the weaving

Before reaching the top, optionally secure the tapestry with a twining stitch, the same as at the bottom if you did that. Cut the warp threads at the bottom, leaving enough length to tie off in pairs. These can be left hanging or woven into the piece.

The tapestry is still on the cardboard loom, but the bottom threads have been cut at the longest possible length.

After tying off the bottom strings in pairs, release the warp loops from the top and leave them as they are.

The tapestry is off the loom. The bottom ends have been tied off in pairs and top loops are available to slide over a dowel or popsicle stick for hanging.

For a detailed walkthrough of creating a mini tapestry, I recommend the Spruce & Linen 3-part tutorial by Jenell Flynn, which was how I learned.

Step 8: Finishing touches

Flip the tapestry over and weave in loose ends using a darning needle. The back side of the tapestry may not be clean, especially if you are using a variety of textures.

The ends are tucked into the back of the piece. Make sure to leave the ends long enough to do this. The wide contrast in yarn weight, narrowness of the piece, and frequent shifts between colors have made this piece a bit uneven.

Thread a dowel, stick, or popsicle stick through the warp loops for hanging.

These are two interpretations of the CDC HIV About page. The tapestry on the right uses rya knots for tassels, representing the unchanged features in the footer of the page. Greater emphasis is placed on the modifications to create larger bands of pink.

Works Cited

Bradley, Duncan. “Taking Time Is an Effective Way To Process Large Numbers.” Nightingale (blog), May 31, 2023. https://nightingaledvs.com/time-dimension-in-data-visualization/.

Center for Control and Prevention. “Wayback Machine: About HIV.” Internet Archive Wayback Machine, January 16, 2025. https://web.archive.org/web/diff/20250116033727/20250213010210/https:/www.cdc.gov/hiv/about/index.html.

D’Ignazio, Catherine, and Lauren Klein. “On Rational, Scientific, Objective Viewpoints from Mythical, Imaginary, Impossible Standpoints.” In Data Feminism. MIT Press, 2020. https://doi.org/10.7551/mitpress/11805.001.0001.

Freire, Paulo. Pedagogy of the Oppressed. Translated by Myra Bergman Ramos. New York: Bloomsbury Academic, 2000.

Posner, Miriam. “Humanities Data: A Necessary Contradiction.” Miriam Posner’s BLog (blog), June 25, 2015. https://miriamposner.com/blog/humanities-data-a-necessary-contradiction/.

Russell, John E., and Merinda Kaye Hensley. “Beyond Buttonology: Digital Humanities, Digital Pedagogy, and the ACRL Framework.” College & Research Libraries News 78, no. 11 (December 4, 2017): 588. https://doi.org/10.5860/crln.78.11.588.

Sayers, Jentry. “Low-Tech Approaches to Digital Research.” Presented at the NEH Office of Digital Humanities Institute on Textual Data and Digital Texts in the Undergraduate Classroom, February 26, 2019. https://jentery.github.io/lowtech/#/title.

U.S. National Park Service. “Wayback Machine: Stonewall National Monument.” Internet Archive Wayback Machine, January 19, 2025. https://web.archive.org/web/diff/20250119000200/20250215002842/https://www.nps.gov/ston/index.htm.

Walsh, Brandon. “‘The Three-Speed Problem in Digital Humanities Pedagogy’ in ‘What We Teach When We Teach DH.’” In What We Teach When We Teach DH, edited by Brian Croxall and Diane K. Jakacki. Debates in the Digital Humanities 10. University of Minnesota Press, 2023. https://doi.org/10.5749/9781452969558.

Wernimont, Jacqueline. “A Gather: Documenting Death.” Jacqueline Wernimont: Researching Data from Bodies and Data Flows (blog), March 24, 2022. https://jwernimont.com/a-gather-documenting-death/.

Kaylen Dwyer

Kaylen Dwyer is the Digital Humanities Librarian at Tufts University where she provides support for a wide range of digital humanities projects, including digital exhibits, digital editions, mapping, and multimodal storytelling in research and pedagogy. Her research interests are critical making, creative coding, and disability and access in DH.

• More Posts

Introduction

Dear Data Binghamton started as an interdisciplinary discussion between the Digital Scholarship (DS) team and a professor from the Department of Teaching, Learning and Educational Leadership (TLEL). During their initial conversation they realized they had a shared admiration for Giorgia Lupi and Stefanie Posavec’s Dear Data project. Through Dear Data professional data illustrators Lupi and Posavec challenged each other to send each other weekly postcards over the course of a year that contained a thematic data visualization on one side and a key on the other (Dear Data – The Project n.d).  The result is a blend of art, data, and self-reflection (Carpenter et al. 2024). The DS librarians and TLEL faculty member at Binghamton believed Dear Data might be used to structure a similar initiative at Binghamton focused on engaging data-intimidated or data-detached undergraduate students in data literacy activities. The conversation led to the two joining forces by bringing in a cohort of faculty and graduate students from across campus to create Dear Data Binghamton.

Figure 1

Examples of student-created postcards based on screen usage. 

The first iteration was based on creating postcards similar to Lupi and Posavec’s original project around the theme “A week of social media,” which asked participants to keep track of their social media use over the course of a week. Throughout the semester, events focused on walking participants through a series of creative exercises to teach some of the basic principles of data visualization and design and engage students in best practices around data collection and analysis. Participants were welcome to create and drop off postcards in the Libraries or during events with more than 30 postcards submitted during the semester. 

Now in its third year, the project has continued to evolve to encompass a wide range of different personal data projects. Last spring, a new round of postcards were created, encouraging participants to track the number of times they expressed gratitude through text messages. This iteration allowed participants to easily collect and analyze their data in one session, enhancing the accessibility of the activity. The project has also expanded to encompass a number of new formats, including a postcard exhibit at the library, an in-class workshop for an undergraduate Data Visualization course, and multiple presentations for a campus Data Visualization interest group. Members of the Dear Data team have also designed a course on Data Storytelling for first-year students to bring the team’s passion for data literacy into the classroom more formally.

Figure 2

Participants at a Dear Data kickoff event  

From Postcards to Physicalization 

From our experiences running Dear Data events focused solely on postcards we found that participants often struggled to move beyond visualizations such as pie charts, line, and bar graphs. Inspired by projects from dh+lib‘s “Making Research Tactile: Critical Making and Data Physicalization in Digital Humanities” and the collection of art within “Making with Data” the Dear Data Binghamton team considered if pivoting to using physical materials would help students think beyond the bar graph a little further (Huron et al. 2022).  Links between data physicalization and data literacy have been explored by many, especially for younger students, showing increased understanding of basic data literacy concepts when utilizing physical objects, such as through crafting physicalizations, to create and explore data visualizations (Bae et al. 2023; Zhu et al. 2023, 2023). Eager to explore this hypothesis, the Dear Data Binghamton team created and integrated a data physicalization activity into the Data Storytelling course taught in Fall 2024. The goal of the assignment was to push the bounds of 2D visualization of their personal data (i.e. Dear Data) to a 3D model that reflected a more creative practice and approach, while also considering the tactile accessibility of representing data.

Process (How to…)

We provide guidance towards creating and implementing a similar assignment as part of a class or workshop. This can be done as an individual or as a small group of students. 

• Data: Students collect 5-7 days of personal data (e.g., music, bad habit, emotions). From our experience, students should submit their ideas for approval as some data ideas do not lend themselves well to telling a story through data physicalization such as “how many times a day do I study.” We asked students the following: (a) What data will you collect?; (b) What three variables will you collect?; (c) How will you collect the data?; and (d) How will you remind yourself to collect the data? Other approaches to data include utilizing an online dataset of interest or asking students to collect data through observations, surveys, or informal interviews.
• Data Building Kits: Kits should include a range of objects to create physicalizations including distinct textures, forms, and functions of materials. Our kits included low-cost materials such as cotton balls, pom-poms, pipe cleaners, felt pieces, macaroni, ribbon, and popsicle sticks (see Figure 3).

Figure 3

Data physicalization kits

• Create: Students create their data physicalization. They should consider things such as how each variable can be represented using the different materials (e.g., ribbon or pipe cleaners may represent length of time, cotton balls for intensity or size); the scale, layout, and interaction of their physicalization as a whole; and how viewers will experience and interpret their final visualizations. We further suggest providing examples of data physicalizations, such as Pierre Dragicevic and Yvonne Jansen’s Gallery of Physical Visualizations and Related Artifacts, and discuss what they notice and what they wonder about those examples (Dragicevic & Jansen, 2021). This step will take at least one hour to complete.
• Peer Evaluation: Students should be provided time to evaluate one another’s data physicalization. Questions may include the following: (a) How do the design elements (colors, shapes, layout) contribute to the overall impact of the physicalization? (b) What story does this physicalization tell? (c) What works well? What could be improved? Based on the feedback, students should consider changes to their data physicalization.
• Data Storytelling: Students craft a 250- to 300-word story – setting, climax, and closing – using the data physicalization as a point of reference. An alternative approach is to create a gallery or an exhibit in a library. Instead of a story, students write an artwork label. It could also be an interactive showing in which the visitors take an active approach, such as answering a question about the data physicalizations.

Insights

In examining the assignments, there were several patterns in how students created their data physicalization. One pattern across all students was the use of color to represent one of the variables from their data set. As illustrated in the first image in Figure 4, color was utilized to represent time of day — orange being morning, yellow being daytime, and purple being nighttime.

Shapes and/or icons were also employed by students to represent variables. One student used shapes to represent emotions — circle for bored, square for happy, triangle for tired, and a half circle for annoyed — while another created symbols to display what they were doing when they were biting/picking at their nails (e.g., star at work, circle on the phone).

Figure 4

Use of color in data physicalizations

In addition, there was variation in how students represented rating scores. For instance, one student recorded the meals they ate for breakfast, lunch, and dinner. Under each meal, noted by the different colored pom-poms, the students included a satisfaction rating by the number of orange pipe cleaners on a scale of 1-5 (see Figure 5). The second image  (see Figure 5) is another example in which a student created spikes on various crowns to represent the number of stars they rated each music genre given a scale of 1-5.

Figure 5

Representation of rating scores 

In addition, students represented frequency of a variable by utilizing one of a material, such as one pom-pom or one sequenced shape, to represent one data point. This is illustrated in the examples provided in Figure 4 and the first image in Figure 5. In Figure 7, each pom-pom was a data point, with different colored pom poms representing a different movie genre.

Length of time was another consideration by students. They employed popsicle sticks or pipe cleaners as observed in Figure 6. In the first image, the students used the height of pipe cleaners to denote the amount of time they were on the phone during different locations – class (pink), outside (purple), and dorm room (white). The second image displays the amount of sleep a student slept over the course of four days. The first stick shows the student slept for four hours between 12:30 am and 4:30 am.

Figure 6

Representation of frequency through height and length

While students were creative and intentional about how they represented each of the three variables (i.e., data points) through the various materials and their visual properties, they did not seem to consider the tactile nature or properties (e.g., fuzzy, sphere-shaped) when choosing the material for their data physicalization. Some students did seem to engage more creatively with data and created visualizations outside of the typical bar chart, pie chart, and line graphs foundational to middle and high school curriculum. Additionally, a few of the students found that creating the physicalization inspired how they represented the same data as a Dear Data postcard visualization. As stated by one student, “Because these were all very visual traits, it made the translation to the postcard a simpler task.” Figure 5 further displays the similarities between the two representations with the data physicalization showing “deeper depth by using more shapes and colors.” 

Figure 7

Data as represented by postcard and data physicalization

Feedback and Themes

The 3D visualization assignment pushed students to think beyond conventional data representation, offering an opportunity for hands-on creativity and abstract storytelling. One student noted, “When doing the 3D-Visualization, I truly had to think outside of the box at how I could depict a certain data set that I had collected, yet show it in an abstract way. I had to do my best to tell this story still without it getting lost in translation.” This freedom to explore personal datasets and represent unique experiences was a highlight for many, as it allowed them to go beyond typical graphs and infuse personal creativity into their work.

However, the challenge of this medium was significant. Many students expressed difficulty in transitioning from more familiar formats like drawing to constructing a 3D model. As one participant reflected, “I enjoyed doing the postcard, yet found the 3D part extremely hard. I like that I tried something out of my comfort zone, because now I know I would much rather draw my data.” Others echoed this sentiment, appreciating the hands-on nature of the project but finding the effort required to gather materials and create a model to be a barrier. One student commented, “The postcard was fun to make, but pretty time-consuming compared to creating a simple graph like the one above.”

Despite these challenges, the tactile and multi-sensory nature of the project was a key advantage. One student remarked, “This method of data storytelling is more tactile, allowing it to be more accessible, and also texturally different… I was able to utilize so many mediums like color, texture, and shape, to tell a story.” Another student highlighted the specificity offered by 3D, explaining, “I utilized the third dimension to increase the specificity of my data representation, using the length of pipe cleaners to show the amount I slept instead of using a color to represent a range of hours slept.”

Ultimately, the 3D assignment offered a complex, yet rewarding, approach to data visualization that allowed students to explore new dimensions of creativity and storytelling.

Future Plans

We believe that our success in the classroom can be adapted for library programming. As such, we have planned a library data physicalization event, using the same methods, during the upcoming 2025 Love Data Week. The planned event will be a two-hour workshop in the Digital Scholarship Center wherein participants can drop in anytime. Dear Data Binghamton will supply the crafting materials, and the group members will lead them through the physicalization activity. Depending on its success, it may become an annual event, an asynchronous tabling option, or another step in Dear Data Binghamton’s evolution. We also have begun building collaborations with other groups on campus, including the art museum and fiber art groups, to form additional modalities for data physicalization projects. 

With our success in all Dear Data Binghamton’s iterations, we believe that librarians can adapt the project’s concepts in library programming for their audience. Hopefully, this physicalization activity is one example of how to do so. We believe no one should be too intimidated by data or creativity, and creating a connection with one’s own personal data helps to bridge the gap for students on either side of the spectrum. 

Citations

Bae, S. Sandra, Rishi Vanukuru, Ruhan Yang, Peter Gyory, Ran Zhou, Ellen Yi-Luen Do, and Danielle Albers Szafir. “Cultivating Visualization Literacy for Children Through Curiosity and Play.” IEEE Transactions on Visualization and Computer Graphics 29, no. 1 (January 2023): 257–67. https://doi.org/10.1109/TVCG.2022.3209442.

Carpenter, Ruth, Amber Simpson, Melissa Haller; Mamen Rodriguez Galindo; Mary Tuttle; Hannah Jones; Amanda Ortiz. “Data + You: Promoting Data Literacy at Binghamton University.” In Data culture in academic libraries : a practical guide to building communities, partnerships, and collaborations, edited by Marcela Y. Isuster and Alisa B. Rod. Chicago, Illinois: Association of College and Research Libraries, 2025.  

Dear Data. “THE PROJECT.” Accessed February 6, 2025. http://www.dear-data.com/theproject.

 Dragicevic, Pierre, and Yvonne Jansen. “Gallery of Physical Visualizations and Related Artifacts.” List of Physical Visualizations, April 15, 2021. https://dataphys.org/list/gallery/.

Huron, Samuel, Till Nagel, Lora Oehlberg, and Wesley Willett, eds. Making with Data: Physical Design and Craft in a Data-Driven World. New York: A K Peters/CRC Press, 2022. https://doi.org/10.1201/9781003264903.

 Zhu, Caiwei, Chloe Oi-Ying Leung, Eleni Lagoudaki, Mariana Velho, Natalia Segura-Caballero, Dietsje Jolles, Gavin Duffy, Günter Maresch, Marianna Pagkratidou, and Remke Klapwijk. “Fostering Spatial Ability Development in and for Authentic STEM Learning.” Frontiers in Education 8 (April 17, 2023). https://doi.org/10.3389/feduc.2023.1138607.

 Zhu, Caiwei, Remke Klapwijk, Miroslava Silva-Ordaz, Jeroen Spandaw, and Marc J. de Vries. “Cognitive and Embodied Mapping of Data: An Examination of Children’s Spatial Thinking in Data Physicalization.” Frontiers in Education 8 (December 21, 2023). https://doi.org/10.3389/feduc.2023.1308117.

Halie Kerns, Ruth Carpenter, Kirsten Pagan, Amber Simpson, and Melissa Haller

Halie Kerns, Ruth Carpenter, Kirsten Pagan, Amber Simpson, and Melissa Haller are all actively engaged in advancing data literacy at Binghamton University. As members of the Dear Data Binghamton initiative (recently renamed Data & You) they continue to explore innovative strategies for making data approachable, creative, and personally meaningful. Their collaborative work centers on integrating data storytelling, visualization, and reflection into classrooms and activities across campus.

• More Posts

Since 2020, Mary Elizabeth Leighton and Kalea Furmanek-Raposo have been crafting encounters with humanities data together as literary scholars and in collaboration with university librarians. First, we worked together in a pandemic-era online undergraduate classroom as instructor and student, combining archival research in 19th-century digital databases with 19th-century hands-on scrapbooking practices. Second, we collaborated as part of the Crafting Communities team, researching and creating online DIY craft tutorials (with step-by-step written instructions, photos, and videos) on 19th-century craft practices. And third, we planned and piloted a fall 2024 English majors capstone seminar, teaching students about critical crafting methods–that is, how to use crafting as a way of deepening their understanding of literature. 

This capstone seminar, titled “ENSH 482: Community-Engaged Learning and Public Humanities: Crafting Literacies,” was structured around three modules: paper, textiles, and printing. For each module, we invited community and scholarly experts into the classroom to teach students a range of crafting methods: for the paper module, students learned lino carving and block printing, paper quilling, and décollage; for the textile module, rag rugging, embroidery, and sewing; and for printing, typesetting by hand and hand press printing on our university library’s 1953 Vandercook press. We also invited librarian experts to teach us about creativity and book arts, how to edit audio with Audacity, and how to curate digital exhibits using Omeka. The seminar concluded with pop-up exhibits of students’ final crafted projects at our university library as well as a local public library; a month later, we launched Crafting Readers, complementary in situ library and digital exhibitions of students’ final crafted projects, which responded to the literary texts they had chosen, and their written and audio reflections on crafting as a mode of literary analysis.

English capstone students at the “Crafting Readers” pop-up exhibit at the Sidney-North Saanich Branch of the Vancouver Island Regional Library on 7 December 2024

English capstone students’ final projects and rag rugging workshop at the “Crafting Readers” pop-up exhibit at the Sidney-North Saanich Branch of the Vancouver Island Regional Library on 7 December 2024

We recommend paper quilling as a productive classroom and library activity with rich potential for data visualization. We provide some background about the history of paper quilling before sharing our experiences of quilling in the literature classroom as well as some examples of capstone students’ quilling and reflections. Since that first capstone quilling workshop, we have adapted the workshop for a second-year foundations literature course, and we also include some examples of second-year students’ quilling. The pattern that we provide here has thus been honed for general use in the literature classroom through several iterations. We conclude by making recommendations about how quilling might prove fruitful for students working with library materials regardless of their academic discipline. 

Paper Quilling: A Brief History       

Paper quilling (also called paper filigree, paper mosaic, mosaicon, or rolled paperwork) exists in various forms across time and space. For example, Jianzhi, or Chinese paper cutting, dates back to the second century and informs the style of Chinese paper quilling, which usually focuses on contrast and blank space (Ding 21-22). The earliest evidence in Western traditions comes from churches, nunneries, and monasteries scattered across Europe (Williams), where fifteenth-century nuns and monks used quilling in place of metal filigree (that is, precious metalwork). However, the exact origins of paper quilling are unknown, and the craft may be as old as the creation of paper itself. 

When paper quilling, the crafter manipulates long strips of paper into spirals with a thin tool and then applies an adhesive to the paper coils to stick them onto a surface. Long strips of paper can also be left uncoiled and glued along their edges for structure as the crafter creates a three-dimensional design. Changing the color of the paper strips and the size of the tool used to roll the paper gives the design shape and texture. Paper quilling requires few materials or special equipment and lends itself well to upcycling, making it a popular craft throughout time.

English foundations students’ paper quilling

Extant examples of paper quilling appear on everything from mirror frames to tea caddies now housed in museums (Battison). Quilling became a fashionable pastime for women in the eighteenth century. A 1786 article in the New Lady’s Magazine speaks to the versatility of this craft: “[paper quilling] affords an amusement to the female mind capable of the most pleasing and extensive variety” (530). Today, paper quilling most frequently appears on greeting cards but has great potential as a method of data visualization.

Paper Quilling in the Classroom

We have experimented twice with paper quilling in undergraduate literature classrooms: first, in the capstone seminar for English Majors that we have already described and that focused on what crafting methods can teach students about literary texts; and second, in an introductory foundations course for English Majors focused on building students’ literary-critical vocabulary, strengthening their literary analysis, and honing their prose. After our first classroom workshop in the capstone seminar  with upper-year Majors, we adapted the workshop for first- and second-year Majors in the foundations course, integrating lessons we had learned about workshop planning and delivery as well as implementing a clearer focus on quilling as a data visualization exercise.    

By the time we piloted our first paper-quilling workshop, Kalea had already developed a paper quilling tutorial for the Crafting Communities website as part of her graduating M.A. project. Drawing on historical and contemporary sources, she compiled a list of steps and materials—including a range of paper and tool options to make the tutorial accessible to a range of makers in a range of settings. To test this pattern and demonstrate how paper quilling can be incorporated into literary analysis, Kalea created a large paper eye to represent the concept of narrative perspective in Charlotte Brontë’s Villette, which is a novel thematically preoccupied with what is seen and what remains unseen. She noted how many times specific eye colors (blue, green, brown, and violet) were mentioned in the text, using the Project Gutenberg digital edition to search for terms relating to eyes and tracking their frequency in a spreadsheet. The creation of searchable editions of novels enables efficient data-based analysis, and we encourage workshop participants to seek out such resources. 

Per each mention of an eye color, Kalea quilled three strips of paper in the corresponding color. These paper rolls filling in the iris of her eye act as data that visualize the protagonist’s attention to detail. Scholars have theorized about the function of Villette’s secretive and unreliable first-person narrator (Lucy Snowe) since the novel’s publication. Despite the pseudo-autobiographical mode, the protagonist often withholds information about the plot and other characters from the reader. Kalea demonstrated how Lucy’s attention to detail—specifically, whose eyes she notices, how often she describes their eyes, and whether or not the adjective that she uses to describe them changes as her relationship to that character evolves—betrays the very information about her personal life that she strives to conceal from the reader. Through her paper quilling work, Kalea concluded that Villette’s unique combination of detail and concealment cleverly resists the nineteenth-century gender roles prohibiting explicit expressions of female feeling. 

Kalea’s paper-quilling project on Charlotte Brontë’s Villette

Paper Quilling in the Classroom 1: Capstone Seminar

We used Kalea’s tutorial to devise our paper quilling workshop for the capstone class. For this workshop, we provided and prepared the materials: pre-cutting strips of colored paper using a paper-cutter, providing a heavy paper base, and pouring liquid glue into small recycled containers. However, each student chose and brought their own design connected to the literary text that they worked on throughout the course. During the 80-minute workshop, students transferred their design to a paper base, outlined their design by building sturdy paper walls, rolled paper coils, and quilled their design by filling in the paper walls with rolled paper coils.

We presented paper quilling as a potential mode of data visualization, emphasizing how Kalea’s quilled eye incorporated data about narrative perspective that she gathered from Brontë’s novel, yet students engaged with this craft in mainly aesthetic and thematic rather than data-driven ways. Encouraged to choose their own design, students tended to pick objects that were mimetically represented in their literary texts (such as water flowing over rocks at the shoreline in a novella set on California’s coast) or metonymically related to their literary texts (such as a lamp in a poem about home and domesticity). Despite this design focus on representing objects that appeared in their literary texts, the following students’ reflections suggest how the quilled design became a way of visually representing complex relations in the literary text (among characters, settings, themes) and how the act of paper quilling became an analogy for—and, indeed, facilitated—the act of criticism by drawing attention to literary aspects of their text.

English Capstone Students’ Paper Quilling and Reflections

Carly Goodman on Christine de Pizan’s The Book of the City of Ladies (1405)

Carly Goodman on Christine de Pizan’s The Book of the City of Ladies (1405)

“In The Book of the City of Ladies, Pizan writes of a ray of light that comes to her in a moment of devastation, appearing first as the sun but revealing itself to be “three crowned ladies” coming to bring her enlightenment. Deceptively difficult, I found paper quilling to be a rewarding task. Though the slightest deviation in size, color, or placement of the paper coils threatened to ruin the illusion of the image, I realized that this papercraft merely demands patience and a willingness to adapt, much like Pizan’s perception of her coming struggles and the role the women would play in her overcoming them. Through perception, the sun becomes a symbol of resistance, change, and female empowerment.”

Kyra Hetherington on T.J. Klune’s Under the Whispering Door (2021) 

Kyra Hetherington on T.J. Klune’s Under the Whispering Door (2021)

“I chose to quill a dandelion while reflecting on T.J. Klune’s Under the Whispering Door because the flower represents hope, healing, and resilience and marks the start of the main character Wallace’s journey in character development. Quilling relies on creating firm foundation walls while leaving space for intricate details and patterns that rely on tension and delicacy. The same can be said for character development in Klune’s novel. Wallace’s initial description takes the place of the stem and leaf walls, giving the reader an idea of the character. Closed loose coils in my dandelion serve as representations of the many relationships and events that Wallace experiences. This internal shaping gives the final quilled flower more depth and complexity, just as Wallace changes from a flat, mean character into a person who craves the possibilities of what life can offer.”

Faith Lapointe on Virginia Woolf’s To the Lighthouse (1927)

Faith Lapointe on Virginia Woolf’s To the Lighthouse (1927)

“Like Virginia Woolf’s novel To the Lighthouse, paper quilling relies on small details coming together to form a larger cohesive image. The process of paper quilling–which involves choosing, cutting, rolling, and arranging strips of paper to create a design– illuminates Woolf’s prose style. She builds the novel by intertwining fragmented streams-of-consciousness to layer characters’ perceptions and reflections, a strategy that draws attention to the insight that can be found in small, seemingly insignificant moments while also demonstrating how life itself is composed of and defined by such moments.”

Tamara Laing on Isaac Asimov’s I, Robot (1950)

Tamara Laing on Isaac Asimov’s I, Robot (1950)

“The colors of paper provided an opportunity for color-play…. While the coiled spring in the center of the gears is made of a neutral black-and-white print stock and thus reads as more ‘mechanical’, the outer gear has the appearance of a yellow flower that speaks to the way the robots most often engage with the humans: they try to work with them rather than in opposition. There are three pink open-topped hearts, which gradually increase or decrease in size depending on the viewer’s perspective and represent the Three Laws of Robotics in the novel that exist to keep humans safe from robots but also distinctly ‘other’ them…. Connecting the gears and the hearts is a gear heart made of neutral text strips. It is engaged with the gears as well as following the shape of the heart and is meant to connect the two elements that are forever at odds in the book. I opted not to follow the traditional technique we were shown, which involved building frame walls and filling them with quilled curlicues and the like. The end result is a piece that rejects rigid rule structures while inviting touch and evoking a sense of play.”

M.L. Prenger on Elizabeth Smart’s By Grand Central Station I Sat Down and Wept (1945)

M.L. Prenger on Elizabeth Smart’s By Grand Central Station I Sat Down and Wept (1945)

“There is purposeful chaos to Smart’s writing: the fragmented prose reflects and amplifies themes in the novel. I was unsure how to approach the text, but one line stuck: “The sea booms. The stream rushes loudly … the kelp in amorous coils appear to pin down the Pacific.” The contrast of delicate curling and unfolding (wavering) kelp beds holding down something as vast and uncontainable as the Pacific is what I wanted to capture in the coils of my paper quilling. I chose to be unprepared, without form to trace. I scratched some squiggles in pencil and felt ashamed of how childlike they appeared. I wanted to sit and spin paper, thinking about the ocean itself (its proximity, my feelings while observing it). The coiling kelp beds suggest a delicate balance between control and surrender, reflected in Smart’s writing but also within myself. Apprehensive to attempt something new, I hoped my intention would move through my fingers, winding the paper strips up with my desires, my entanglement (and history) with the text…. The novella affords Smart emotional release but never resolves the issue of love, loss, and longing. Likewise, my coiled paper forms are holding down the rising swell of opposition within, allowing me to touch that tension point and use a dab of glue to hold it in place.”

Ria Sheoran on Hanya Yanagihara’s A Little Life (2015)

Ria Sheoran on Hanya Yanagihara’s A Little Life (2015)

“Through paper quilling, I was able to conceptualize the intricate coils of each relationship that the main character Jude in Hanya Yanagihara’s A Little Life nurtured. Not every coil of paper was easy to create for this project, and at times the coils fit uncomfortably against the other pieces of paper. This, in turn, enabled me to think that even though these relationships were what shaped Jude’s life, they were not always providing what was best for him despite being necessary to his story.”

Melanie Turunen on Antoine de Saint-Exupéry’s The Little Prince (1943)

Melanie Turunen on Antoine de Saint-Exupéry’s The Little Prince (1943)

“Reflecting on my experience of paper quilling a baobab tree from The Little Prince by Antoine de Saint-Exupéry, I found that the craft mirrored the novella’s deeper themes of care and growth. The baobab trees in the story are a powerful symbol of the consequences of neglect. The Prince’s task on his asteroid is to prevent the baobabs from growing too large and overwhelming, much like how we must care for the things and people we love. The tightly wound strips of paper in my baobab tree, requiring close attention and care to prevent unraveling, became a visual metaphor for the need to nurture what we cherish before it grows burdensome. Just as the paper could easily tear or loosen if not carefully applied, so too can the delicate aspects of our lives unravel if not given the attention they need. The Prince’s love for the Rose, his companion, only grows once he understands that he is responsible for her and nurturing their emotional relationship. Each twist and turn of the paper was a gentle reminder of how important it is to tend to the small details—whether in art or our lives—so that they can flourish without becoming overwhelming.”

Paper Quilling in the Classroom 2: Foundations Class

After the first workshop, we realized that if we wanted students to use paper quilling as a data visualization method, we needed to prompt them explicitly to prepare a data set ahead of time. For our second workshop, with junior students in the second-year foundations class, we focused students on data visualization by 1) providing everyone with the same literary text, Elizabeth Bishop’s “One Art” (1976),) and the same design, an image of an open book; and 2) requiring students to assemble a data set before they started quilling.

To model paper quilling with a data set from a literary text, we shared not only Kalea’s quilled eye based on her analysis of Villette but also samples that she had prepared based on her analyses of two poems: Eavan Boland’s “The Pomegranate” (1994) and Theodore Roethke’s “My Papa’s Waltz” (1942), which students had discussed in a previous class.

Kalea Furmanek-Raposo on Eavan Boland’s “The Pomegranate”

Kalea Furmanek-Raposo on Theodore Roethke’s “My Papa’s Waltz”

Kalea’s samples (seen above) focus on meter. In her paper quilling of “The Pomegranate,” each colour represents a number of syllables in a line (e.g., there are five lines with eleven syllables represented by the five green coils). Kalea divided the stanzas in “My Papa’s Waltz” into each section of the pattern and then used pink coils to represent stressed syllables and blue coils to represent unstressed syllables. She glued them in rows that loosely replicated the poem’s lines and marked the ending syllable of each line with a triangle.

For this second workshop, we built in time for students to gather data in small groups before launching into quilling. We gave each group a few minutes to assemble their data set. As the class had discussed in a previous class on poetry and closed form, “One Art” takes the form of a villanelle, with five tercets and a quatrain as well as a tight rhyme scheme and pattern of repetition. (For example, in a villanelle, the first line of the first tercet repeats as the third line of the second tercet.) The poem thus provided students with useful options for data collection.

Kalea’s samples focused on meter, but some students chose to count words relating to the poem’s key themes. They then rolled the corresponding number of paper coils and used different shapes to identify them, as in the following example.

Data chart for paper quilling by Cerine Chittattu and Fermín Roja

English Foundations Students’ Paper Quilling and Reflections

English foundations students’ paper quilling samples by (top to bottom, L to R) Julia Apap, Cerine Chittattu, Olivia O’Leary, Wesley Oertel-Sproule, Dallia Knight, Isabelle Porter, Sage Stobbe, Abi Heinrichs, Sienna Sabina, Jake Rees, Fermín Roja, Ruby Lautermilch, Cynthia Guo

We built in time for students to talk briefly in their small groups and then as a class–at the beginning of the workshop, about their data, then at the end of the workshop, about how the paper quilling process helped them visualize these data. For example, students who focused on diction and repetition noticed the number of times that references to (and variants on) loss appear in the poem; they noted not only repetition of the word losing but repeated variants on this word, including lost, lose, and loss. Other students counted caesurae, enjambed lines, and rhyme. Would students have noticed such poetic elements without paper quilling? Hopefully–but the acts of preparing data and paper quilling these data underscored the relationship of poetic form to theme, and the different visualizations within the same open-book design suggested the poem’s formal complexity.

Recommendations:

1. Data Assembly: Invite students ahead of time or during the workshop to assemble the data that they will visualize in the form of paper quilling. A short poem provides ample data for a single workshop. Longer fiction and non-fiction would also provide useful data, but students might need more time to assemble such data. Project Gutenberg is a useful resource for students working on literary texts that are available in HTML format because of such texts’ searchability using keywords. For example, Kalea searched the Project Gutenberg version of Villette using keywords to determine the number of occurrences of eye color in the text. Paper quilling could be adapted for non-literary forms of data as well.

2. Paper Quilling Materials: Assemble and prepare quilling materials ahead of time: paper strips (in different colors) that measure the same length; cardboard or stiff paper bases; copies of the same pattern; stick glue (for gluing the paper pattern onto the base) and liquid glue (for attaching paper coils to the pattern on the base).

3. Reflection Time: Build in time for reflection and take-aways, either as part of a workshop discussion or as a check-in after the workshop.

Pattern

Supplies:

• Paper (colored paper, newspaper, magazine pages)
• Cardboard or heavy paper for the base
• Scissors, knife, or metal ruler
• Large paper clip
• Glue stick
• Liquid glue
• Small brush or cotton swab to apply the liquid glue
• Container for liquid glue (optional)

Paper quilling supplies

Step 1: Preparing the materials

Cut the paper into strips along the long edge so that each strip measures approximately 1 centimeter (1/3 inch) in width. If you have access to a paper cutter, we recommend using it to cut the paper into uniform strips.

Print copies of a common design to be glued onto the base (or bring a copy of your own design). We recommend keeping designs for classroom use compact because of time constraints. We chose this image of an open book, which Kalea drew and which measured approximately 6 centimeters (2.5 inches) wide and 4 centimeters (1.5 inches) tall.

Kalea and Mary Elizabeth’s paper-quilling pattern for workshops

Pour liquid glue into containers for sharing. We made small trays out of aluminum foil.

Unfold your paper clip to create a makeshift quilling tool around which you will coil your paper strips. (We found it easier to keep the coils taut when rolling them around such a tool instead of using our fingers.)  

Step 2: Determine your data set

Choose something to visualize with your paper quilling. We encouraged students to plan their paper quilling using data-informed decisions. 

Step 3: Creating the design 

Glue the printed design on your base with a glue stick (or, in the alternative, draw your design on the base with a pencil). Do not use the liquid glue for this step because it will make the base soggy. 

Create an outline of your design by placing some of your strips of paper on the base (they should be standing upright like short walls) and gluing them down with the liquid glue. You will fill in these borders with your curled strips later.

 Step 4: Begin your quilling!

To form the paper coils, place a strip of paper at the base of your paper clip and roll it onto the tool (or just use your fingers). Once you have rolled the whole strip, press your thumb to the loose end of the strip and slide it off your tool. Roll the paper between your fingers to tighten it or let it spring open for a loose coil. You now have a circular paper roll. You can shape this roll into another shape by pinching the edges or glue down the loose end to keep it circular. 

Glue the paper rolls within your pattern’s walls until you have completed your design.

You can explore Kalea’s short videos on coiling techniques to learn more. 

Works Cited

Battisson, Clair. “‘Natural Born Quillers’ – Conservation of Paper Quills on the Sarah Siddons Plaque Frames” Conservation Journal, no. 27, April 1998. V&A. 

Ding, Eva. “Jianzhi: A Silhouette of Tradition, Art, and Feminism Through Chinese Paper Cutting.” Scope: Contemporary Research Topics (Art and Design), no. 26, 2024, pp. 21–33.

“Historical Account of the Origin and Progress of Paper Filigreework, with an Accurate Description of the Present Practice of That Ornamental Art.” New Lady’s Magazine, vol. I, no. 10, Nov. 1786, pp. 529–30.

Williams, T. C. “Paper Gold: The History and Art of Paper Quilling.” Medium, 15 Feb 2022.

Dr. Mary Elizabeth Leighton and Kalea Raposo

Mary Elizabeth Leighton teaches Victorian literature and culture at the University of Victoria. With Lisa Surridge, she co-wrote The Plot Thickens: Illustrated Victorian Serial Fiction from Dickens to Du Maurier (2019) and is currently researching pregnancy in Victorian fiction and culture. With Andrea Korda and Vanessa Warne, she co-founded Crafting Communities, a resource hub about 19th-century craft and material culture. Kalea Furmanek-Raposo holds an MA in English from the University of Victoria. Her research interests include nineteenth-century girls’ print culture and women’s writing as well as hands-on learning and digital humanities. She has been a part of the Crafting Communities team since 2021, she was a research assistant on the Great Expectations Pregnancy Project, and she was a 2023 UVic Lowens Fellow.

• More Posts

Introduction

I (Theresa Quill) have been teaching with maps in a library context for over a decade; supporting a wide range of disciplines, levels, and topics. Regardless of the discipline, I often use print maps to illustrate concepts such as visual literacy, authority, bias, accuracy, and the information creation process. I also lead library instruction sessions on digital mapping and GIS tools, and have found that if students do not have an introduction to foundational spatial literacy concepts it is easy to get bogged down in the software and lose the thread of why we are mapping. Leanne and I piloted a data physicalization assignment using paper circuit maps in an attempt to talk about mapping and spatial literacy without the distraction of software.

Teaching with Maps

Cartography, even maps that we view as “scientific” or “objective” offers a particular view of place, born from the data and technology available at the time, and reflective of the motivation and viewpoint of the cartographer. Harley wrote that “[maps are] refracted images contributing to a dialog in a socially constructed world” (2001, 53). Viewing maps that were made to be intentionally persuasive and have an obvious point of view (tourist maps, propaganda maps, etc) can be an accessible entry point for talking about how all maps are abstractions and simplifications of the world.

“The Awakening” from Henry Mayer, an example of a persuasive map from the PJ Mode Collection at Cornell University.

Map-making is an iterative process that involves several steps: conception, data gathering, feature plotting, labeling, design, quality control, and dissemination. This is true of historic paper maps, modern maps made with Geographic Information Systems (GIS), and hand-drawn maps. When I teach GIS, I explain how vector data is made up of points, lines, and polygons. A common example is that a road network is likely to be a line feature, country borders are polygons, and points of interest are, well… points. I typically use static images to help explain how these different data types are organized into layers, but I wanted to find a more effective example.

A typical representation of points, lines, and polygons.

Last year I attended a data physicalization workshop that GIS & Research Visualization Librarian Meg Miller (University of Manitoba) led at the Western Association of Map Libraries (WAML) conference where she used embroidery to illustrate these data types. I thought this was such a clear and innovative approach to talking about spatial data without the complication of software, and I have been thinking about ways to incorporate data physicalization into my own work ever since. 

Hand-drawn maps are another way to get at these core concepts of cartography and map-making. Through the process of drawing a map, students make decisions about what they will include, but also about scale, relative distance, representation, and abstraction. Through the process of creating maps themselves, students are asked to think about the many ways that features from the real world are abstracted, distorted, and left off of maps. Students as cartographers engage in place-making in a very concrete way that can invite conversations about objectivity, “truth,” and the iterative nature of information creation. This can of course be accomplished by working with GIS and computer cartography as well, but there is something about the physical act of drawing, erasing, and making that gets to the core of these concepts. In their guide to creating hand-drawn maps, Helen Cann remarks, “In all its beautiful fallibility, hand-crafted cartography offers something digital maps could never do; it reveals something richly fascinating about a given landscape, its culture and history, and the unique character of the cartographers themselves” (2018, 9).

Teaching with Paper Circuits

The way that paper circuits bridge the divide between STEM and the arts has always fascinated me (Leanne Nay). It’s a sneaky way to engage artists in STEM and engage STEM folks in the arts. As e-textile designer and educator Leah Buechley explains, “By situating technology in a creative, artistic (feminine) context, e-textiles invite new audiences to the table to tinker and play” (2014, x).  I’ve taught paper circuits to high school students at summer camp, families at outreach events, instructors at workshops, and college students in the classroom. The materials are affordable, the process is collaborative, and the end result is endlessly satisfying. Regardless of the context, the novelty of making things light up never wears off.

When I teach paper circuits, I typically lead a demonstration or provide step-by-step instructions, but there is still ample opportunity for trial and error. As with all maker endeavors, the response to “failure” is often the most meaningful. Creating a paper circuit is naturally product-driven: does your LED light up or not? When the LED does not light up, things get interesting. At this moment, participants have to decide how they will proceed. Will they ask for help from the instructor? Will a peer join their cause? Will they stop trying altogether? There are no wrong answers.

Examples of a paper circuit greeting card and light up wristband.

Geography Class Instruction Sessions

We partnered with faculty member Olga Kalentzidou to incorporate paper circuits and digital mapping into her undergraduate course, Human Geography in a Changing World. During our first class visit, Olga asked her students to “Map YOUR Bloomington.” In 20 minutes, they sketched maps from scratch and did not rely on base maps or templates. After drawing their maps, students walked around the room to see what their classmates created. Their maps illustrated their dorms, study spots, restaurants, and other campus landmarks. As the only non-students in the room, we immediately noticed that their maps almost exclusively represented the university campus, with little attention to the surrounding city.

We joined the class later that week to show them how to incorporate paper circuits into their maps. We started with a brief overview of the supplies and concepts, followed by an open-ended period where they worked together to create their circuits. Our instruction focused on the technical aspects and we did not dictate what the lights should represent. Many students chose to highlight key points on their maps and some constructed paths. We asked students how they decided what to highlight on their maps and heard stories about football games, favorite restaurants, where they live and work, and how they get around town. We also drew maps of our city that were entirely different. Our identities as parents/librarians/adults undoubtedly influenced our choices to map libraries, swimming pools, and other toddler-focused locations around town far from campus.

Analog vs Digital

During that second visit to the class, we talked about what type of spatial analysis we could do with their maps in a visual, and then also a digital/GIS format. Did they notice common themes from the collection of student maps? Were there certain locations that appeared on multiple student maps? Where was the geographic center of their maps, and how did that differ from official city maps?

We planned a second activity, though we ran out of time before we could implement it. We envisioned asking the students to recreate their hand-drawn maps using the GIS software Felt. Students would translate their map locations into points, lines, and polygons with the locations or route of their paper circuit LEDs highlighted as its own “points of interest” layer. Then we could incorporate all of their highlighted data layers into a single map to perform analysis using GIS that would give us quantitative data about which locations or parts of town students in this class frequented or otherwise found significant. Data from a project like this could be used to inform city planners decision making about the placement of infrastructure such as increased sidewalks, bus routes, or public toilet access. This process is referred to as Community Mapping or Asset Mapping.

What started out as a highly individual hand-drawn map about a specific student’s experience and view of Bloomington can be combined with others to give us an idea of the student community. These are themes that Dr. Kalentzidou reinforced in her Human Geography class lectures and readings.

Conclusion

By blending analog and digital methods, students learn mapping concepts and experience the process of information creation. Crafting hand-drawn maps encourages critical engagement with place and paper circuits allow them to continue developing and examining their narrative. Teaching multiple approaches side by side broadens students’ understanding of digital humanities methods.

Pattern: How to Make a Paper Circuit Map

Supplies Needed

• Chibitronics LED stickers*
• Conductive nylon tape
• Coin cell battery CR2032
• Scissors
• Binder clip
• 2 sheets of copy paper
• Colored pencils, crayons, or markers

Figure 1. Paper circuit supplies can be purchased from Sparkfun Electronics, Brown Dog Gadgets, Adafruit, or Chibitronics. Chibitronics LED stickers are great for folks who are new to tinkering with electronics, but they are significantly more expensive than regular LEDs. If you decide to use regular LEDs: Use clear tape to hold the legs in place and poke holes through your drawing so they come through your map.

Circuit Fundamentals

A little bit of knowledge about electricity goes a long way towards creating paper circuits! 

• LED
  • LEDs, or light-emitting diodes, have a positive (+) and a negative (-) side. The orientation of an LED is important for creating a successful circuit. If you’re working with standard LEDs instead of LED stickers, the longer leg is the positive side (+) and the shorter leg is the negative side (-). 
• Circuit
  • In order to light up an LED, we will use a battery and nylon tape. Current will flow from the battery, through the tape, to our LED. This circular path is known as a circuit.
• Trace
  • A trace is the path that connects components in our circuit. We will use nylon tape traces because nylon tape is conductive, meaning it allows electricity to pass through. 

Parallel Circuit Example

In this tutorial, we’ll show you how to make a parallel circuit so that you can incorporate multiple LEDs into your design. The actual path of the circuit will depend on the location of your LEDs, but a basic circuit will need a positive trace and a negative trace to allow power to flow from the battery through the LED.

Figure 2. This example illustrates the necessary components of a basic parallel circuit with one LED. We can use this setup to create a larger path with more LEDs.

Step One: Draw Your Map

Start by creating a hand drawn map. If you’re starting from scratch, we recommend choosing a simple prompt and time limit. For example: Map your city in 20 minutes.

Figure 3. Most of my personal and professional life in Bloomington is spent in various libraries. I mapped three libraries in relation to my house and included the time it takes to get to each library from home.

Consider the following before you begin your circuit:

• Are you working alone or collaborating with others?

• What story do you want to tell?

• How can you emphasize parts of your story with light?

• Will all of your lights turn on together (one circuit) or do you want multiple different sets of lights (multiple circuits)?

• How many LEDs do you want to use?

• Do you want to use different colors?

Step Two: Plot Your Points, Traces, and Battery

We’re going to create a circuit on a separate piece of paper that will sit underneath our map drawing. Next, we’ll plot out the location of our components.

• Place the circuit paper on top of your map to identify the exact spots where you want to have an LED and make a pencil mark.
• The battery will be placed in a corner of the map to create a switch to turn the lights on and off. Sketch out where your battery will be located.
• Finally, sketch out two traces: one that will connect the battery to the positive leads of your LEDs and one that will connect to the negative leads of your LEDs.

Figure 4. Sketching out the traces ahead of time will save you time and frustration. Sketch your traces close enough that an LED sticker can fit between them. Don’t let the traces overlap or touch or you will get a short circuit (which sounds dramatic, but it simply means your LED won’t light up).

Step Three: Create Traces

Now we’re ready to place the nylon tape on the paper.

• Remove the adhesive backing and stick the tape along the positive and negative paths that you sketched on your paper. 
• If you need to round a corner, you can cut the tape and start a new piece. Make sure both pieces of tape make full contact.  

Step Four: Add LED stickers

Once your traces are complete, add your LEDs. 

• Place your LED stickers so that the positive side is touching the positive trace, and the negative side is touching the negative trace.

Step Five: Add Battery

It’s time to add power to light up our LEDs.

• Place your battery in the corner with the negative side facing down.
• Fold the corner of your paper so the positive trace is resting on the positive side of your battery. 
• At this point, you should see light! 
• Add a binder clip to hold your battery in place.

Figure 6. Your completed circuit will rest underneath your hand-drawn map. You can join them together with paper clips or glue to keep everything in place.

Figure 7. You map is complete!

Troubleshooting

If your circuit doesn’t light up, have no fear! Consider the following troubleshooting tips.

• Check the connections
  • Make sure your LED stickers are making full contact with the nylon tape.
• Check the orientation
  • Check the positive and negative orientation of the LED stickers.
• Check for short circuits
  • Confirm that the positive and negative tape traces do not overlap or touch, this will prevent the current from flowing.
• Check the battery
  • Ensure that the battery is sandwiched firmly between the top and bottom tape lines, making full contact with the tape.

References

Buechley, Leah. 2014. Foreword to Soft Circuits : Crafting E-Fashion with DIY Electronics, by Kylie Peppler, Melissa Gresalfi, Katie Salen Tekinbas, and Rafi Santo, x. Cambridge, Massachusetts: The MIT Press.

Cann, Helen. 2018. How to Make Hand-Drawn Maps : A Creative Guide with Tips, Tricks, and Projects. La Vergne: Chronicle Books LLC.

Harley, J. B. 2001. The New Nature of Maps: Essays in the History of Cartography. Edited by Paul Laxton. Baltimore, Maryland: Johns Hopkins University Press.

Theresa Quill

Theresa Quill is the Map and Spatial Data Librarian at Indiana University Bloomington where she curates map collections, liaises with the Department of Geography, and supports faculty and students in spatial data inquiries across disciplines. Her research interests include humanitarian mapping, critical cartography, spatial humanities, and teaching with maps. Theresa is a current co-editor of the Journal of Map & Geography Libraries and she teaches a course on Map and GIS Librarianship in the Luddy School of Informatics, Computing, and Engineering at Indiana University Bloomington.

Leanne Nay

Leanne Nay is the Creative Technologies Librarian at Indiana University Bloomington where she manages the Wells Library Makerspace. Her research explores gender imbalance and belonging in library makerspaces. Leanne is an adjunct instructor in the Luddy School of Informatics, Computing, and Engineering and a faculty champion for the Center of Excellence for Women and Technology 3D Printing and Design Team.

• More Posts

Introduction + Backstory

This workshop is designed to provide librarians / library workers / information professionals with an alternative mode to help data learners create their own data visceralization exploration that goes beyond the traditional data visualization methods. By engaging the body, senses, and physical objects in their environment, workshop participants can better relate to seemingly abstract concepts in data, the data processing, and its outputs.

This workshop builds off of my individual data visceralization experiment that was first conceived during a collective research project in 2022 by Carol Choi, Jessika Davis, John Decker, Ava Kaplan, and Lubov McKone, “Making the Desert Island Discs Dataset: Data Visceralization and How We Don’t Know What We Know”(Choi et al. 2022). This project explored alternative modes of representing data, such as materialization and sonification, as a means of connecting critical making and feminized labor to a more visceral understanding of the data itself.

In determining what dataset to use in our project, we focused on a topic that wouldn’t interfere with, nor directly reinforce, the concepts mentioned above. We also wanted to choose a dataset that had already been through the Digital Humanities research cycle (in promotion of the concept of reproducibility in the field). In the Knowledge Commons (aka Humanities Commons) we found our dataset, Desert Island Disks (Finnegan, Rangel Smith, and Van Stroud 2021), which built upon Desert Island Discs 1942-2020(Gustar 2020), which in turn builds on BBC Desert Island Discs Dataset v 1.0 (Morgan 2018). You can read more about it on our “Data Choice” project page.

This dataset represents data from the long running BBC radio show Desert Island Discs, where the hosts ask their invited guests (“castaways”) which 8 songs they would like to have accompany them if they were marooned on a desert island. Over the years captured in the dataset, 1942-2021, the top 8 songs the guests chose ended up being by classical composers:

• Wolfgang Amadeus Mozart
• Ludwig Van Beethoven
• Johann Sebastian Bach
• Franz Schubert
• Giuseppe Verdi
• Edward Elgar
• Pyotr Ilyich Tchaikovsky
• Giacomo Puccini

Top 8 Journey

The visceralization experiment served as an exploratory exercise to transform data from a one-dimensional realm into a physical (visual / tactile) and aural experience. Aptly titled Top 8, my individual experiment focused on the relationship between the program guests and the artists of their 8 song choices. To translate the data to a final output, I was confronted with many unexpected considerations that needed to be negotiated in terms of how we think about, utilize, and communicate our data process, specifically the ideas of “choice” and “absence.” Cat Hicks discussed the absence when processing data for an intended outcome, “Sometimes I like to call this ‘when we miss missingness,’ a quip to remind myself that we need to understand data not just in terms of datasets but in terms of the context around it. We need to ask what data was collected and, perhaps more importantly, what data was left out.”(Hicks 2022)

Working through those considerations, the experiment has evolved from the first iteration, a table-top proof of concept (Figure 1), to a second iteration, a life-sized version that was installed as an exhibition (Figure 2).

Figure 1. The first physicalized / sonified iteration for the individual “visceralization”, Top 8, from the “Making the Desert Island Discs Dataset: Data Visceralization and How We Don’t Know What We Know” Project in 2022. This output consisted of poster and foam board layers, crochet thread, screws, tape, sonification of labels via mini-recordable sound devices, penned dots on paper (when audiences ran their fingers against the underside of the board, they felt the raised texture of each dot, which signified the data not used in the network physicalization).

Figure 2. The second physicalized / sonified iteration for the individual “visceralization”, Top 8, from the “Making the Desert Island Discs Dataset: Data Visceralization and How We Don’t Know What We Know” project as presented at HASTAC 2023. This output consisted of metal coat racks, passive infrared (PIR) motion sensor-activated sound players, thread, tape, zip ties, and an “absence rug” consisting of not used choices/data points – 1:1 count of matpe beans – between plastic sheets.

My past iterations and this workshop builds from previous explorations of how engaging our senses can enrich understanding, particularly when data moves from flat visualizations into three-dimensional space (Jansen et al. 2015). I’m particularly interested in how our physical bodies, working in concert with all our senses, can create new layers of understanding. This approach draws on embodied cognition—the idea that “our cognitive activities are anchored by our sensorimotor interactions with our surroundings” (Wilson 2002; Noë 2006; Barsalou 2008; Zhu et al. 2023). Wilson’s framework identifies six dimensions of embodied cognition, including how cognition is situated in physical contexts, how we off-load cognitive work onto our environment, and how even our abstract thinking is grounded in bodily experience (Wilson 2002). In this third iteration of my work, I wanted to experiment specifically with how physical engagement affects sense-making and data interpretation. When we process information, our minds aren’t working in isolation—we’re constantly navigating between abstract data and concrete experience, encoding and decoding meaning through multiple dimensions simultaneously. Physical objects act as bridges in this process, grounding abstract concepts and encouraging exploration.

Definitions

Before we dive in, here’s an overview of some of the key terms and concepts that will be used during the workshop:

The output of this workshop and the previous iterations assume the form of a network graph, which can be defined as a map of concepts or things (nodes) that are connected by some sort of shared relationship (edges). Network graphs can help understand and analyze commonalities, differences, missingness (or absence), and the complex patterns in communities existing within these concepts and things.

It is important to note that my aim was to visceralize a network graph, not just visualize one. What does that even mean? In the early 2000s, the concept of  “data visceralization” was thought to be introduced through Kelly Dobson’s work on the Data Visceralization Project at Rhode Island School of Design (RISD). Her project called into question the over-reliance of data visualization as the singular pipeline to understanding what the data represented. Data visualization tends to make audiences feel like they are in control, while also being detached observers of (rather than participants with) the information. The term was “later popularized by Luke Stark” (WERNIMONT 2021) (Greenhalgh 2022). Stark defined data visceralizations as the “representations of information [that] rely on multiple senses including touch, smell, and even taste, [which] work together to stimulate our feelings as well as our thoughts.”(Stark 2014) 

To me, data visceralization connects and uses our physical body to map understanding from abstract concepts to the tangible world around us – even weaving in feelings and emotions.

Data Transparency

In the spirit of communicating data processing actions clearly and transparently, I share a bit about the steps I took and considerations I made to accommodate potential limitations around time, space, and what is realistically feasible with a human body.

The workshop picks up from the already processed dataset utilized for “Top 8” iterations mentioned above, which centered around the top 8 artist occurrences (total = 4478) which the BBC guests chose from 1942-2021. The pared-down dataset will just consider the time period of 1990-1999 (total = 678). This will still allow for experiencing the data in the original project configuration, but at a smaller scale to allow time for everyone to participate in the full workshop activity. The new pared-down dataset is available for download and can be reviewed at during the workshop. There is another step of data processing that is needed to transform the data into a network graph.

To transform raw data into a network graph, a nodes and edge table needs to be constructed to prepare the data according to the types of relationships you want to show in that network graph format. To build that table, I used the same Python script utilized in the previous iteration (changing out the file directories as needed). This updated Python script generates a nodes and edge table (nodes pair and the connecting edges with weight). The artists are represented as the nodes and the edges are the co-occurrence of any 2 choices. The weight is the frequency of the occurrence.

For instance, when a “castaway” chose Mozart and Bach in their top 8, that was considered one edge weight which connected those 2 nodes. The more times the guests chose that same pattern, the weight of the edge increased. For the sake of transparency, and to ease workshop facilitation, I have included all the “already processed” files in the GitHub repository. 

Another important calculation to call out is that of the Absence: the data not selected from the dataset to construct the network graph. So, for the first and second iterations, the calculation was to take the total number of rows (all 8 choices through the years of 1942-2021) 26260 and subtract from that the 4478 rows I used in the network graph (only the occurrences of the “top 8” artists from 1942-2021) which yielded 21,782 rows not “used”, i.e. the absence.

For the pared down workshop version of the dataset, the calculation is as follows: 4478 (only the occurrences of the “top 8” artists from 1942-2021) minus the 678 rows of the “top 8” artist occurrences from the date range of 1990-1999 which yields 3,800 data points not considered in the construction of the human network graph visceralization, i.e. the absence.

TL;DR

The first 2 iterations called out an absence (data not used in the final network graph output) that totaled 21,782 data points. This workshop iteration follows the same data processing steps, but considers a smaller subset of that data, which decreases the total absence to 3,800 data points. This is important to highlight in keeping with the spirit of the themes of the project – transparency and choice.

It took extensive trial and error to determine the most manageable population data for the all the physicalization efforts. The original project team processed the data in OpenRefine to combine similar values through text facets, clustering, and the removal of special character encodings and transformation into their proper diacritics. Additionally, manual visual checks were performed to group different values representing the same concept, including variations in name spelling (anglicization vs original), capitalization, spacing, and punctuation. I welcome you to view the entire data processing steps (all 6+ trials…) from the start of the original project “Making the Desert Island Discs Dataset: Data Visceralization and How We Don’t Know What We Know,”(Choi et al. 2022). You can follow the full data processing for the “Top 8” under the Technical Process header.

Framework

The workshop / activity employs the framework of data literacy and is inspired by the design principles set forth by Rahul Bhargava and Catherine D’Ignazio in their 2015 paper, “Designing Tools and Activities for Data Literacy Learners,” in which they describe data literacy as including “the ability to read, work with, analyze and argue with data”(Bhargava and D’Ignazio 2015).

Reading data asks the person to connect their understanding of the data to real-life concepts. Working with data involves the activities and actions taken with data directly (processing, generating, maintaining it, etc.). Analyzing the data involves re-ordering the data in such a way to compare/contrast and perform other evaluations. Arguing with the data demands the construction of a larger story outlining what the data is actually saying to a targeted group of people(Bhargava and D’Ignazio 2015). I directly connected this framework to the workshop activity steps below.

Bhargava and D’Ignazo also highlight the importance of transparency throughout the steps mentioned above, which help data learners connect the concepts to the domain-specific language utilized. They propose “activities that support learners must be focused, guided, inviting, and expandable.”(Bhargava and D’Ignazio 2015)

Building off their 2015 research to better support data literacy learners, Catherine D’Ignazio introduces the concept of creative data literacy and argues that non-technical types of learners “may need pathways towards data which do not come from technical fields” (D’Ignazio 2017). She emphasizes that multiple actors, such as “data creators, data publishers, tool developers, tool and visualization designers, tutorial authors, government, community organizers and artists” (D’Ignazio 2017), should be charged with facilitating creative data literacy.

Goals and Hopeful Outcomes

I hope that by engaging a constellation of senses (aural, touch, and visual) not typically used in conceptualizing data, participants will come away with increased data literacy, and in turn, a new awareness of how data can be transformed, processed, or removed. By employing tactics of embodied cognition (using the body to help form new understandings between the external world and internal thought), transparency of process, and the data literacy principles of Rahul Bhargava and Catherine D’Ignazio, I hope participants will better understand what that data means after experiencing it with their bodies. Maybe this will spark your own curiosity to transform data into the physical and aural realm. So, I leave you with this question:

How will you consider the power of the senses, choice, and absence in your own data processing after experiencing this project?

Workshop Activity

Note before starting: This workshop will require the participants to sit or stand for a period of time and it is recommended that the space have chairs available if people need to alternate between standing and sitting. The facilitator can also check in with the participants throughout the workshop to see if they need to take short breaks or switch positions. Additionally, it is recommended for the participants to wear comfortable clothes.

Potential trigger warning: the thread will be wound around the arms, legs, and mid section of the participants – connecting one participant to another. Participants will be in close proximity to each other, but if there are any feelings of discomfort, they can engage the facilitator. The participants should feel safe and empowered to remove themselves from the activity (i.e. the thread) at any time. The facilitator should also be on the lookout for signs of discomfort with the participants and intervene as necessary.

Time Frames

• Preparation for the workshop:
  • Collecting recycled materials (for the absence) over a couple days/weeks prior to the workshop.
  • Cutting up the absence: For one person, it takes approximately 7 hours to cut 3,800 pieces of the recycled materials. If this task can be split up between the participants, it will diminish the total time (for example, with 8 people, it would take a total of ~55 minutes to cut 3,800 pieces).
• Workshop duration: 
  • Activity: ~3 hours for main activity
  • Post activity discussion: ~30 minutes
  • Clean up: ~20 minutes
    

What is needed

• An open space/room that can accommodate at least 10 people situated at about an arm’s length from each other
• Laptop (optional: projector) arranged so that participants can see the screen (by either taking turns or passing around the device)
  • Another use for the laptop is to play music during the workshop activity. In the trial run, one of the participants put on a streaming music platform and played a mix of classical composers (many of them were the nodes in this dataset). The participants reported how it helped them get transported into a more focused state.
• At least one facilitator
• 8 Data Players (participants enacting the network graph- the nodes)*
• At least 2 Data Makers (participants stringing the crochet thread – the edges – from Data Player to Data Player)**
• Thread (for example, Aunt Lydia’s Crochet Thread Classic 10 or 3 ) in 8 different colors

Figure 3. Example of the crochet thread brand, Aunt Lydia’s Crochet Thread Classic 10.

• Absence Material– material that you can get in high volume and can be walked on (like paper crumpled up into balls, cut up cardboard such as soda can cases, cut up toilet paper rolls, etc)
  
  

  Figure 4. Examples of cut up recycled material, like cardboard soda can case or toilet paper roll. Depending on the Absence Material quantity needed and amount of resources that can be collected for the workshop, you will need to make decisions about how large or small to cut the pieces to get the amount required.

• Scissors to cut the thread (ideally one pair per “Data Maker” role)
• Potential other materials needed:
  • Masking tape or binder clips of assorted sizes (to secure the thread loops on the body / clothing if need be, although tension between to people should be enough to keep the thread stay put)
  • Printed out copies of the nodes and edges table to pass around to all the participants. It is helpful for the facilitator to also have a copy to keep track of the counts of thread wrapped between the nodes. 
  

  Figure 5. If a laptop can’t be made available for the workshop, a print out copy of the nodes and edge table file can be used as the “blueprint” to wrap the thread between nodes. Here the rows are highlighted in the color of the corresponding thread and checked off after the wrapping was completed.

Activity

1. Data Processing/ Preparation ((read)) and ((work with))
   
   1. The facilitator will introduce the project by discussing these points:
      1. Share the trigger warnings* / accessibility notes** above.
      2. Restate the information above in the Introduction and Backstory. This workshop is based on an existing project with multiple iterations that build from each other. That body of that work focuses on themes of feminization of labor, critical making as a means to knowledge-making, challenging the over reliance on creating visualizations to understand data (and the concepts it would represent), and the importance of communicating about choices and absence during the processing of data.
      3. Talk through the key definitions mentioned above: network graph and visceralize / visceralization.
      4. Data transparency: This dataset has been worked with by different people at different times, those people made choices along the way that impact what is seen or not seen in the data output, and the data has been processed to optimize grouping like concepts and to mitigate encoding errors at time of data extraction (see “Individual Data Processing” section in the original project for more detail).
      5. Share / show the data that will used to support this workshop: This is the parsed down data set (1990-1999) and the Python script that generated the nodes and edges table, which serves as the blueprint to determine how many times the thread needs to be wrapped around the 8 Data Players’ arms, legs, and torsos (connecting one Data Player to another Data Player).
   2. The participants can collectively decide which color will be assigned to which Top 8 artist nodes: Wolfgang Amadeus Mozart, Ludwig Van Beethoven, Johann Sebastian Bach, Franz Schubert, Giuseppe Verdi, Edward Elgar, Pyotr Ilyich Tchaikovsky, and Giacomo Puccini.
2. “Absence” Creation ((analyze))
   
   1. Depending on the amount of Absence Material that has been collected and the time allowed for the workshop session, decide what the ratio will be used to represent the “Absence”. For instance, if there has been a lot of material collected and the participants are able to dedicate time to cutting up the material, you can attempt a 1:1 ratio of material piece to data point not used. If time and materials are more limited, try using a different ratio like 1:2 or 1:5 (1 material piece represents  2 or 5 data points not used).
   2. Examples:
      1. 3,800 material pieces needed if using the ratio 1:1
      2. 1,200 material pieces needed if using the ratio 1:2 
      3. 760 material pieces needed if using the ratio 1:5
   3. Save this material for distribution for the very end of the workshop (post completion of the human network graph construction.
3. Constructing the Human Network Graph ((analyze))
   1. The 8 Data Players place themselves in a spread-out formation:
      

      Figure 6. Overview sketch of the staggered formation of the participant nodes in the activity space.

       

      

      Figure 7. Chairs and a mic stand serve as a placeholder for the participants in preparation to build the human network graph.

       

   2. Each Data Player adopts one of the 8 Top artist nodes.
   3. The Data Makers will then follow the node and edges table to connect the Data Players to each other with the string.
      
      

      Figure 8. Detail of the node and edges table in Google sheets that displays the key components of the table, source, target, weight, and an example of a recursive pair occurrence.

       

   4. Source = the node to start the thread wrapping, this also denotes the color thread to use (reference step 1b).
      1. Tip for the wrapping: first tie the end of the string to the body of the source node (not too tight, but tight enough so it should stay at the level securely). You might want to use a binder clip to the participant’s clothes or masking tape if  skin is bare to secure this first attachment.
   5. Target = the node that the thread will be wrapped around after the Source node, then returning back to the Source node or terminating there, depending on the Weight.
   6. Weight = the number of times the thread will pass between the 2 nodes.
   7. This first connection from Source node to Target node counts as a Weight of 1. When the thread is brought back to the Source node, that increases the Weight to a count of 2. Continue this path until the final count is met according to the  node and edges table.
   8. You may encounter a Source and Target pair that is the same. This means the artist existed multiple times in a guest’s top 8 choice or just by itself. This scenario can be referred to as a recursive relationship, meaning it evokes or references itself in the prescribed relationship (“Recursion” 2025).
   9. The Data Makers will construct the recursive loop as either a necklace, crown, or headband (or Data Player’s choice) for each Data Player.
      
      

      Figure 9. An example of a recursive loop (aka crown) – which is comprised of multiple lengths of string, cut to the same length, and joined together by masking tape.

       

      

      Figure 10. The participants chose how they wanted to wear their “recursive crowns” or loops that they created during the workshop. Some chose to wear around their heads, necks, and even as a loop attached to their clothing. Pictured from left to right: Scarlett Moldovan, Sara Zucker, Calista Donohoe, Kelsey Kiantoro, and Veneranda Aquirre.

       

   10. The thread should be anchored to the body in 1-2 spots on each arm (upper arm and wrist), 1-2 on each leg (thigh and calf), the waist, and the hips. Each Data Player will have, at most, 7 places of contact (not including their recursive crown).
       1. Note: some Data Players will not have a source part of the path and others might not have a target part of the path. This is due to the way the Python script matched the pairs of the chosen “top 8” artists within the same episode date as it was ordered in the dataset (and any subsequent sorting / filtering activities performed in the data processing steps). The weight of the connections is driving the “popularity” more than the color frequency.
       2. *If there are less than 8 Data Players: During the trial run of the workshop, we ran into this situation. We utilized elements in the classroom, like chairs, a mic stand, and a structural pole as stand-ins for the missing Data Players. We later had more participants arrive, so we cut away the thread wrapped around those elements and rewrapped around the human Data Player.
          

          Figure 11. *Since we didn’t have enough “Data Players” and “Data Makers” at the start of the workshop trial, we used an existing pillar in the classroom to serve as one of our “nodes”. We all worked together to wrap the thread around our “Data Player” nodes.

           

          

          Figure 12. *This is the pillar (playing the role of “Data Player” representing the node of Mozart).

           

       3. **If there are less than 2 Data Makers: Similar to being short on Data Players, we only had one participant that could play the dedicated role of the Data Maker.  In this case, each participant played both parts (until they were too limited in their motility from being connected with thread in multiple anchor points of the body to others).
          

          Figure 13. **A “Data Player” passes the thread to one of the other “Data Players” to assist the singular “Data Maker” in the wrapping of the thread. Photograph taken by Calista Donohoe, 2025.

           

   11. The Data Makers will continue the process of first securing, wrapping, and then tying off the thread for each of the line items in the node and edges table. It is recommended to symmetrically match the place on the body for any particular source-target pathway. For example, if the Data Maker starts on the right wrist of the source node Data Player and they are facing the target node Data Player, the Data Maker should attach to the target node Data Player’s left wrist. In other words, crossing the axis of the body should be avoided when connecting a related source-target pair. 
       1. Don’t forget to check in with the participants to see if they need to take a break, sit down, or even change roles. Changing roles mid-way, involves removing the attached thread from one participant to the other.
       

       Figure 14. Preliminary sketch depicting plan to connect the participants to each other by wrapping crochet thread from different points of one participant’s body to another participant’s body.

        

       

       Figure 15. A “Data Maker” ties the thread at one of the anchor points, the upper thigh, of a “Data Player”. Photograph taken by Calista Donohoe, 2025.

        

       

       Figure 16. Close up view of a “Data Maker” tying the thread at one of the anchor points, right above the knee, of a “Data Player”. Photograph taken by Calista Donohoe, 2025.

        

       

       Figure 17. “Data Players” stand in formation with thread connected at the key anchor points. Photograph taken by Calista Donohoe, 2025.

        

       Figure 18. A 360 view of the “Data Players” in the final human network graph formation. Since we only had 8 total participants for the trial run, I filled the “Data Maker” role. To hold my place in the final formation, we used a mic stand to connect the threads. After the video was taken, I slipped into the loops that were around the mic stand, completing the human network graph.

   12. The “Data Makers” will then distribute the absence on the floor surrounding each of the “Data Players” and in the center.
       
       

       Figure 19. A “Data Maker” distributes the “absence” – paper packing material – in the center and at the feet of the “Data Players” in the human network graph formation. Photograph taken by Calista Donohoe, 2025.

        

       

       Figure 21. Figure 22. A “Data Maker” distributes the “absence” – cut up pieces of cardboard, and packing material – in the center and at the feet of the “Data Players” in the human network graph formation. This represents the 3,800 data points not used in the nodes and edges above. Photograph taken by Calista Donohoe, 2025.

        

       

       Figure 20. The workshop participants take a moment and check in after the wrapping has concluded.

        

       

       Figure 22. The workshop participants shot from above. As the only “Data Maker”, I had to complete the wrapping. Not pictured is the node -holder (the mic stand) that held all the loops I would need to slip into to complete the human network graph.

        

       

       Figure 23. I finally made it into the network graph! Photograph taken by
       Kelsey Kiantoro, 2025.

        

4. Interaction and Response (Tactile and aural engagement, visual evaluation)
   1. The Data Makers walk around the Human Network Graph and “activate” the nodes by gently pulling or pushing on the arm, leg, or torso of a particular Data Player where the loops of thread are attached. The Data Players that make up that “source” and “target” configuration will then vocalize the name of the artist they are representing as the node.
5. Post Activity Discussion (( argue with))
   1. After the activity concludes and all the participants have extracted themselves from the network graph, the facilitator will encourage everyone to gather around for a discussion.
      
   2. Suggested investigative and prompting questions to ask during this time:
      1. Which artists were chosen together the most? Which pair were the least chosen together?
      2. Do you feel like you understand the data better before we made the Human Network Graph? In what ways did you or didn’t you?
      3. What did you learn from each sense (hearing, seeing, touching)?
      4. What do you find most challenging about the activity?
      5. What did you find most fun or interesting?
      6. Is there anything that you have encountered in your real life that this reminds you of?
      7. What do you think happens to the data that doesn’t make it into the charts and graphs that you see presented in your daily life (online, advertisements, media)? What do you think could be missing from those presentations based on what you experienced today?
      8. Were you surprised by anything you experienced today?
6. Clean Up Time!
   1. Everyone pitches in to collect the “absence” pieces for recycling
   2. The string can be kept by the participants, reused, or recycled

Figure 24. The aftermath – close up of the “absence” and thread on the floor after the workshop wrapped.

Figure 25. The aftermath – overview of the “absence” and thread on the floor after the workshop wrapped. This was then collected and recycled.

Share your Feedback

If you have facilitated or participated in the workshop activity above, please fill out this feedback form. I would love to hear about your experience. The hope is to use your feedback to improve the workshop in future iterations.

Acknowledgements

I would like to extend a special thanks to the following folks for participating in the experimental trial run of “Top 8 Workshop: Exploring Embodied Cognition Through Data Visceralization”. Their collective openness, willingness to play, and trust facilitated such a meaningful experience that brought to life an idea on paper. Thank you Veneranda Aguirre, Calista Donohoe, Kelsey Kiantoro, Scarlett Moldovan, Chris Moscardi, Nene Villalobos, and Sara Zucker!

Additionally, thank you to Anthony Cocciolo, Dean of the Pratt Institute School of Information, for providing the access/space on campus to run the workshop. I appreciate your continued support of alumni’s data/digital humanities research.

I also wanted to recognize and highlight the contribution of the initial project group, Carol Choi, Ava Kaplan, and Lubov McKone. Their collaboration, ingenuity, and research not only help shaped the trajectory of this workshop, but challenged the bounds of the Digital Humanities domain. I am honored to have collaborated with such thoughtful people.

Figure 26. Workshop selfie!

Figure 27. The trial workshop crew upon completing the activity. Pictured from top left going clockwise: Kelsey Kiantoro, Veneranda Aguirre, Chris Moscardi, Jessika Davis, Scarlett Moldovan, Calista Donohoe, Nene Villalobos, and Sara Zucker.

Bibliography

Barsalou, Lawrence W. 2008. “Grounded Cognition.” Annual Review of Psychology 59 (Volume 59, 2008): 617–45. https://doi.org/10.1146/annurev.psych.59.103006.093639.

Bhargava, Rahul, and Catherine D’Ignazio. 2015. “Designing Tools and Activities for Data Literacy Learners.” MIT Media Lab. 2015. https://www.media.mit.edu/publications/designing-tools-and-activities-for-data-literacy-learners/.

Choi, Carol, Jessika Davis, Ava Kaplan, Lubov Mckone, and John Decker. 2022. “Making the Desert Island Discs Dataset: Data Visceralization and How We Don’t Know What We Know.” 2022. https://studentwork.prattsi.org/data-visceralization/.

D’Ignazio, Catherine. 2017. “Creative Data Literacy: Bridging the Gap between the Data-Haves and Data-Have Nots.” Information Design Journal 23 (1): 6–18. https://doi.org/10.1075/idj.23.1.03dig.

Finnegan, Bill, Camila Rangel Smith, and Sam Van Stroud. 2021. “Desert Island Disks.” https://doi.org/10.17613/2gg4-hw68.

Greenhalgh, S. P. 2022. “Subjectivity in Data Visualization.” Subjectivity in Data Visualization, 131–41.

Gustar, Sndrew. 2020. “Desert Island Discs – Statistics in Historical Musicology.” April 18, 2020. https://musichistorystats.com/desert-island-discs/.

Hicks, Cat. 2022. “When We Miss Missingness.” Data Science by Design. July 27, 2022. https://datasciencebydesign.org/blog/when-we-miss-missingness.

Jansen, Yvonne, Pierre Dragicevic, Petra Isenberg, Jason Alexander, Abhijit Karnik, Johan Kildal, Sriram Subramanian, and Kasper Hornbæk. 2015. “Opportunities and Challenges for Data Physicalization.” In Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems, 3227–36. Seoul Republic of Korea: ACM. https://doi.org/10.1145/2702123.2702180.

Morgan, Paige. 2018. “BBC Desert Island Discs Dataset v 1.0.” https://doi.org/10.17613/M6RJ48T8X.

Noë, Alva. 2006. Action in Perception. The MIT Press.

“Recursion.” 2025. In Wikipedia. https://en.wikipedia.org/w/index.php?title=Recursion&oldid=1272345034.

Stark, Luke. 2014. “Come on Feel the Data (and Smell It).” The Atlantic. May 19, 2014. https://www.theatlantic.com/technology/archive/2014/05/data-visceralization/370899/.

WERNIMONT, JACQUELINE. 2021. “Visualizing Energy Data or Visceralizing Energy Transitions – JACQUELINE WERNIMONT.” April 17, 2021. https://jwernimont.com/visualizing-energy-data-or-visceralizing-energy-transitions/.

Wilson, Margaret. 2002. “Six Views of Embodied Cognition.” Psychonomic Bulletin & Review 9 (4): 625–36. https://doi.org/10.3758/BF03196322.

Zhu, Caiwei, Remke Klapwijk, Miroslava Silva-Ordaz, Jeroen Spandaw, and Marc J. de Vries. 2023. “Cognitive and Embodied Mapping of Data: An Examination of Children’s Spatial Thinking in Data Physicalization.” Frontiers in Education 8 (December). https://doi.org/10.3389/feduc.2023.1308117.

Jessika Davis

Jessika Davis is a New York City by way of Rhode Island based researcher, data analyst, knowledge organizer, photographer, performer, dancer, and multimedia artist. She is an avid pattern seeker, whether it be via data, light, space, sound, or the human condition. She received her BFA in Photography at Lesley University College of Art and Design and MS in Museums and Digital Culture at Pratt Institute, with an Advanced Certificate in Digital Humanities. Research areas include: Experiential Data | Multi-sensory Knowledge Making | Information Technology | Linked Open Data | Knowledge Organization | Digital Humanities.

• More Posts

Introduction

Quilting can appear a complicated task at the outset, and rightly so. It uses specialized equipment, patterns, and involves many phases to complete a project. But when broken down into discrete steps, it becomes manageable. The same is true of data visualization: it is a science and also an art, taking sets of data and presenting them in a way that is visually understandable, inviting, and hopefully, aesthetically pleasing.

It may seem like the two have nothing in common, but quilting and data visualization can be a happy marriage. Quilting joins together multiple layers of fabric with a middle layer of insulation using stitching. Quilting can produce functional objects, such as blankets, coats, and desk pads, or decorative objects like wall hangings. For these reasons, quilting can be a good medium for beginning data crafters. Quilting provides rich opportunities for conveying information, from patterns in the fabric, to the size of cut pieces, to the pattern of the stitching ultimately holding together the layers.¹

As a longtime figure skating fan, I have a lot of feelings about the sport! I only began figure skating as an adult in my early thirties, but my fandom stretches well back into my pre-teen and early teenage years. (Fellow fans may recognize this piece’s title as a reference to the 1992 cult film classic, The Cutting Edge). This project uses a dataset from the sport of figure skating. While many data crafting projects rely on datasets gathered by the crafter themself, this can be a barrier to less experienced sewists and data visualizers.² Using sports datasets can be fruitful for a variety of reasons:

• These sets are often meticulously curated; makers don’t have to collect data of their own, over time, as with quantified self or personal informatics data crafting projects. This provides an immediate gratification factor.
• There is so much data collected for sports and there are so many different statistics even within a single sport.
• While it might seem “lightweight” on the surface, sports can still be very personal. Think about childhood or family fandoms, sports that a maker played or participated in, and obscure sports that have very specialized fandoms. Curling fans, I know you’re out there!

The project demonstrated here requires cutting and sewing in straight lines only, making it accessible for beginning sewists; it is also one that could be produced in other media. Many public and academic libraries already include sewing machines in makerspaces, but makers could alternatively use cut paper, paints, or colored pencils on graph paper.³ It can also be adapted to other sports data sets fairly easily. Following Vladis, et al’s example, it could be expanded into a day-long or multi-session series.⁴

Data gathering

Before we can develop a pattern, we need to have our dataset. I developed a different quilt in late 2022 for a university library data crafting workshop hosted in our digital humanities research center, so I returned to the idea for this project.⁵ That project visualized individual medal winners, by country, at the World Figure Skating Championships from 2013-2022. This time, I returned to World Figure Skating Championship data, but shifted my focus to the percentage of medals won, by country, for the period 2014-2023. How would this visualization change my view of this data? Would it reveal different interpretations and insights? This design mimics a word cloud graphic produced by text analysis, where more frequent words or phrases appear larger in the cloud and less frequent ones smaller.

I used a Google spreadsheet to gather my data from a fan-run website, SkatingScores.com. SkatingScores offers a queryable database, but also tidy overview pages for competitions.⁶ My spreadsheet had 3 sheets for gold, silver and bronze medal data. I placed the years on the vertical axis and countries on the horizontal axis and began counting medals by country. 

Here’s the resulting gold medal table:

Figure 1. World Figure Skating Championship gold medal winners by country, 2014-23

I used simple SUM functions to add up the column and row totals. Four medals are awarded each year for each placement, one for men’s individual, women’s individual, pairs, and ice dance. The horizontal sums in the final column were a double-check for me that I was collecting data for all disciplines for each medal type in a given year.

From these tables, I derived the percentage won by country by dividing the country medal totals by 36 (the total number of medals awarded in the period; no Championship was held in 2020 due to COVID-19):

Figure 2. Percentage of gold medals won by country (2014-23)

I ordered them by hand in descending order of percentage in a new column, then rounded to whole numbers for ease of cutting.

Developing a pattern

Now, the real fun begins when thinking of how to represent each nation visually! I relied heavily on my previous project, recycling many of my design choices that were based on country flag colors and features. I began by collecting data on colors used in flags, then moved to features to help differentiate as so many nations share flag colors. I still had pieces of most of these fabrics on hand from my previous project, so I only ended up changing South Korea’s fabric from a solid white to a white and navy polka dot.

Figure 3. Fabric key for countries winning World Figure Skating Championship medals, 2014-23

Next, I decided how large this piece would be when completed. I landed on 30” wide (three columns each measuring 10” wide finished) by 40” long, which allows about 4” of fabric per year. This makes the finished object rectangular and a little smaller than 3 by 4 feet– a good size for a small wall hanging or blanket.

I used a standard quarter inch (0.25”) seam allowance, so the cut pieces will all have a standard width of 10.5” (allowing 0.25” on each edge). To figure out the lengths to cut, I returned to my spreadsheet and the column of percentage of medals won by country in descending order. Using spreadsheet functions, I multiplied the percentages by 40 (our desired final length). I then added 0.5” to each value in a new column for the seam allowances, and rounded to whole numbers for ease of cutting.

Figure 4. Gold medal winner fabric calculations

I repeated this process for each of the medals. This also helps us see how much fabric we really need to purchase. Fabric is sold by the yard (36” long by width of fabric, typically 40-44” for quilting cottons), so for smaller countries a quarter or even an eighth of a yard might suffice.

Assembly

There is a dirty little secret to sewing: there is very little time spent actually sewing! Most of our time is spent in the preparation and finishing phases, but we’re finally there. 

I gathered and cut my fabrics according to my calculations, then began sewing the columns together. I laid out the cut pieces in the order they should be sewn:

Figure 5. Gold medal fabrics laid out in order on cutting mat, ready to be sewn

When working with patterned fabrics, sewists need to take care to match the printed sides together when sewing so that the patterns all appear on the top (see Figure 7 in the Recipe). 

I repeated this process for each of the medal columns. With this design, sewists can make aesthetic choices when countries have equally sized pieces, as in the case of silver medals where five nations each earned 3% of the medals (see Figure 6). I chose to order mine in a color pattern I liked, but this is up to the sewist. 

Once the columns were assembled, I laid them out side-by-side and discovered an issue.

Figure 6. Gold, silver, and bronze columns laid out side by side. The column lengths are not equal!

My rounding calculations led to the columns being unequal lengths! Now I was left to decide: did I want to sew this as is, and leave the object as a wall hanging? Should I try to equalize the lengths by taking a tiny bit off the silver and bronze columns at the top? I could also fill in the gaps at the bottom using a different fabric. Or a hybrid of both?

While trimming a bit off the top seemed most attractive at first, it went against the visualization’s logic and risked reducing the visual impact of the top medal-winning nations’ accomplishments. In the end, I decided to fill in the blank spaces at the bottom with a neutral color not used by any of the countries’ flags, in favor of completing a small quilt as I’d originally intended. 

With the columns now of a more equal length, I sewed them together to complete the quilt’s top. I won’t detail my finishing steps here, but Melanie Ham’s Your First Quilt video series is a very clear resource for beginning quilters to learn the process of putting the quilt fully together. I’ve begun using a no-show binding method, which takes your design right to the edge of the front, but sewists may also use a traditional binding to provide a frame around the top’s design.

Conclusions

This project’s design is much less granular than my previous skating data visualization. As far as the design and assembly of this quilt, it was easier because it simply used fewer pieces of fabric. This quilt has just 31 cut pieces, versus 120 in the previous design, and yet it lead to some unique observations:

• Twelve nations won World medals in this period.
• Japan earned the most golds and silvers, followed by Russia with the most golds and bronzes, then the USA with the most silvers and bronzes.
• In every category, one of the “Big 6” countries (Canada, USA, Japan, China, France, Russia) claimed the first five spots. These are host nations of the major international senior competitions leading to Worlds.
• Silver and bronze have slightly more variety with Kazakhstan, Belgium, and South Korea winning at least one medal.
• Success at one step of the medal podium doesn’t guarantee it in another; for example, France won many golds, but relatively few silvers and bronzes. However, many of the countries have won at least one medal in each category.
• The northern hemisphere continues to dominate figure skating.

Although I’ve done two projects now using this data, this design could be reworked again to give weightings to the different medals. Imagining the way the physical medal podium has the gold medalist standing tallest, a sewist could multiply the gold or silver medal sizes by a factor to increase their sizes. Another possibility is to calculate the total percentages of all medals won by each country and build an abstract visualization more like a word cloud. 

Supplies and Recipe

The following items are commonly used in sewing and quilting projects and you may find them useful:

Sewing supplies

• Cutting mat – to protect your work surface and help measure
• Rotary cutter – this type of cutter produces clean, straight edges better than scissors
• Ruler/Cutting guides
• Pins/clips
• Iron + Ironing board – pressing the seams as you go helps the finished piece look neat
• Sewing machine – this pattern can be completed on a regular sewing machine 
• Quilt-piecing/patchwork foot – makes sewing 0.25” seam allowances much easier!
• Walking foot – this specialized foot is helpful for putting together the layers of fabric without bunching or gathers
• Scissors – small size for trimming threads 
• Seam ripper – we all make mistakes at some point and need this tool to take out stitching!
• Hand-sewing needle – most quilts are finished with hand stitching the edge binding to the back
• Painters tape – can be helpful in tacking down layers as you assemble your quilt

Pattern: Making a sport dataset quilt

1. Following the steps detailed above, begin your work by choosing a sport you feel passionate about and choose a dataset from it. The dataset can be large or small, but keep in mind that quilts usually repeat design elements (known as blocks) across rows and columns, so having a dataset that employs a range (ex years) may be helpful. 
2. Next, refine the dataset to the data you plan to visualize. Sports collect all kinds of figures and statistics, so hone in to a specific statistic, team, player, competition, or country.
3. With your chosen subset of data, organize them using a method of your choice. This could be done by hand, or using digital tools like spreadsheets. Since quilting usually produces rectangular objects, tabular tools help you begin to visualize what a pattern could look like. 
4. Developing a pattern is where the real fun begins! This is where you will decide the form your quilt will take. Will it be a traditional rectangle? How big do you want the finished piece to be? How will you represent your statistics with color, fabric pattern, size or texture? Will your quilting stitches (the stitching that holds all three layers together) convey information as well? It can be helpful to draw your vision out on graph paper at this stage to help scale it.
5. With your design worked out, it’s time to source your fabric and other materials. Etsy is a great source for small cuts of fabric (less than a half yard) as well as obscure novelty prints. For example, I found my maple leaf fabric that represents Canada in my projects via Etsy.
6. Using your cutting mat, rotary cutter or scissors, and rulers, cut your fabric to the planned lengths and widths to form your blocks. It is very helpful to lay the fabrics out as they will be sewn at this step, then sew them piece by piece together, making sure your design is coming out as planned. If you’re using patterned fabrics, be sure to place the right sides of the fabrics together and sew from the wrong side so that your patterns come out on top: 
   

   Figure 7. Make sure to sew patterned fabrics with right sides together, from the back.   

    

7. Sew together all your blocks into a single top piece to form the front of the quilt. This is typically done by sewing either rows or columns first, then sewing those together. There may be uneven edges– this is completely normal! As you put together the layers of your quilt (the batting and backing), you will trim the edges to be even with the other layers. Again, YouTuber Melanie Ham’s Your First Quilt series provides a clear overview of finishing steps for your quilt.
8. Use, display, and enjoy your quilt! I like to use my skating data quilts as lap blankets in my library office in chilly months.

Notes

<sup>1. I don’t use the term “data quilting” as this term has a distinct meaning used by data analysis methodology: “a three-layer approach and drawing on the comparable and parallel process of quilting, we introduce and describe each layer: backing, batting and top. The backing of the data quilt is the research problem and method, which supports the upper layers. The batting of the data quilt is the data and data analysis, which adds depth and dimension to the data quilt. Finally, the top layer of the data quilt is the presentation, visualization and storytelling, which pieces together the results into a single, cohesive deliverable.” I lean instead towards data crafting or data physicalization. (Murugan Anandarajan and Chelsey Hill, “Data Quilting: Art and Science of Analyzing Disparate Data,” Cogent Business & Management., June 1, 2019, https://doi.org/10.1080/23311975.2019.1629095). ↩
2. Jordan Cunliffe, Record, Map & Capture in Textile Art: Data Vizualization in Cloth and Stitch (United Kingdom: B. T. Batsford Ltd, 2022); Pierre Dragicevic, Yvonne Jansen, and Andrew Vande Moere, “Data Physicalization,” in Handbook of Human Computer Interaction, ed. Jean Vanderdonckt, Philippe Palanque, and Marco Winckler (Cham: Springer International Publishing, 2021), 1–51, https://doi.org/10.1007/978-3-319-27648-9_94-1 ↩
3. Elizabeth Hartnett, “Why Make? An Exploration of User-Perceived Benefits of Makerspaces - Public Libraries Online,” Public Libraries Online - A Publication of the Public Library Association (blog), November 28, 2016, https://publiclibrariesonline.org/2016/11/why-make-an-exploration-of-user-perceived-benefits-of-makerspaces/, accessed January 21, 2025 ↩
4. Nathalie Vladis, Aspen Hopkins, and Arvind Satyanarayan, “Data Crafting: Exploring Data through Craft and Play,” in IEEE VIS Workshop on Data Vis Activities to Facilitate Learning, Reflecting, Discussing, and Designing, 2020, http://vis.csail.mit.edu/pubs/data-crafting/↩
5. Teresa Hebron, “Why Me, or, How I Learned to Stop Worrying and Tried to Data Craft,” Nightingale, April 19, 2023, https://nightingaledvs.com/data-visualization-quilting-craft/↩
6. “World Championships,” Skating Scores, accessed December 3, 2024, https://www.skatingscores.com/wc.↩</sup>

Teresa Hebron is Associate Librarian and Program Director for Mountain West Digital Library, headquartered at University of Utah's J. Willard Marriott Library. Her research focuses on digital collections metadata in large-scale aggregations, librarian employment patterns, and data crafting.

• More Posts

Introduction

Academic libraries love data. Reference transaction data is essential to how academic libraries function, we’re told; the aggregated number of questions answered helps justify the institutional budget. The material reality of this means that librarians spend time logging what they do—the questions they answer must be entered and collected. 

With these collected data points, I made a quilt. I was drawn to the idea of using a traditionally feminine handicraft to represent traditionally feminine labor. I say this with full awareness that non-feminine individuals are also quilters and librarians—but I am aware of the demographics of both disciplines and aware of vague stigmas surrounding them. 

Beginning this undertaking, I knew the quilt would not be a perfect product. I am new to quilting, and I knew there would be mistakes. Stitches, which are supposed to be hidden, would definitely be visible. But this flawed product would highlight my goal: I wanted to make the messy, yet invisible labor of answering reference queries visible again. 

I am one of several Research & Instruction Librarians at Fairleigh Dickinson University (FDU). We are a small team in a mid-size academic library, and we take turns sitting at the reference desk and responding to online chats. My colleagues and I are urged to fill out the transaction form for every reference interaction we have, from “How do I print?” to “Can you help me find open educational resources about HR practices from the last five years with a focus on North America?” (Both are real questions that appear in the quilt.) We collect data for every question we receive, but what happens to that data afterward? It gets anonymized and sanitized and reported. We don’t look back at individual reference transaction data from semesters past. With this project, I am arguing that maybe we should—to help us understand more about the work we do.

Methods & Patterns

At FDU, we classify questions as “directional/technical assistance,” “transaction,” or “consultation.” We include other data with each of these interactions, such as the type of patron, the length of the interaction, and the academic area to which the question pertained. This was great data for the library, but for my purposes, I needed to split it into further categories to be able to create an interesting quilt. About a week and a half into the semester, I devised the following nine categories for my data:

1. Faculty or staff request resources 
2. Requests for “stuff”
3. Asking about directions
4. Asking about hours or access
5. Printing, scanning, and copying
6. Requests for books
7. Basic “how to search” help
8. Website or technology issues
9. Students request resources

To illustrate this a little better, here’s an example of each category.

1. Can you get me access to all issues of this journal? 
2. Can I borrow a pen?
3. Where is the athletics advising center?
4. How late is the library open?
5. How do I print?
6. Do you have any copies of The Handmaid’s Tale?
7. How do I look something up in the catalog?
8. How do I access the software on the computers?
9. Can you help me find information on mental health in college?

With my categories set, it was time to pick out fabric. I chose vibrant and geometric patterns. Stripes, splatters, triangles, diamonds—all in pinks and purples and blues. Would it have been simpler to pick solid colors, ones that easily stood out from one another? Sure. However, the messy nature of this data is something I wanted to embrace. I didn’t want the quilt to be easy on the eyes. Colorful, brash patterns complicate reading the quilt, make it harder to deduce patterns. I wanted anyone who viewed the quilt to spend time with it. The wild fabric patterns facilitate that.

Fabric patterns 1 through 9, in order.

I learned English paper piecing (EPP) as a quilting method a few weeks before the semester started at the #dhmakes mini-conference at DH2024. I was enamored by how easy it was—the fabric and paper came together at a surprising pace, and I found myself in a flow state while working. The process was meditative. I looked forward to integrating it into my research. 

For those unfamiliar with EPP, it involves a paper shape—in my case, a hexagon. Fabric is wrapped around the hexagon and sewn into place. The thread never goes through the paper, though—you sew the fabric to itself until it forms a hexagonal shape. Once you have multiple fabric/paper hexagons, you sew them together and remove the paper from the backside of the quilt.

A single hexagon from the quilt, showing the paper piece inside.

I decided that I wanted to write the reference questions on the paper hexagons themselves, creating a handwritten, hand-quilted record of the semester. On each paper hexagon, there’s the week #, pattern #, and question itself. I consider the paper hexagons, even removed from the quilt, as a part of this project. When the quilt is complete, they’ll be strung together on a cord and placed next to the quilt as a way of “browsing” it. Viewers should be able to place the paper hexagon on a corresponding fabric one.

A pile of paper hexagons with reference questions.

Week 9’s reference transactions, ready to be stitched together.

Results & Lessons

Over the course of the Fall 2024 semester, from August 26 to December 20, I recorded 200 reference questions. I had to buy a second pack of paper hexagons a few weeks into the semester. Then a third. I was not wholly prepared for the number of questions; I started my job over the summer. 

I feel the need to disclaim that, due to the volume of reference questions, my librarian duties, and the rest of my non-work life, the quilt is not yet complete. As of my first draft, I have quilted 9 of 17 weeks and begun work on the 10th. As of publication, I hope to have my questions quilted through week 12 of the Fall 2024 semester. One could question if an incomplete quilt is a quilt at all—this one currently has no backing, for example, as I’m waiting until I complete the top to add backing fabric. I argue it’s still a quilt, even if it’s a quilt in potentia. It is, at the very least, a quilted handicraft, and one I’m quite proud of.

The partly-completed quilt, showing weeks 1 through 8. Week 1 is at the bottom of the image and week 8 is at the top.

A close-up of the quilt, showing the small stitches the hold it together.

Let’s explore the trends of the quilt thus far. We answer a lot of questions as reference librarians, but reference transactions ebb and flow over the course of a semester. Week 2 was my busiest week for questions—first-year students found their way to the library and had questions about how it worked. Most of the questions I got that week were about printing (14 of 33 queries), which makes sense—it’s not intuitive how to print at our library. 

Later in the semester, the questions shifted. Students had questions about how to find articles for their assignments. Many of those questions required research consultations. Weeks 6, 11, and 13 were especially busy, likely because the nursing students, our library’s biggest users among the student body, had major assignments due around those times of the semester.

Surprisingly, the last few weeks of the fall semester were not busy at all. Weeks 15 through 17 were quiet. I theorize that this is because research for final assignments was already done; students were now focused on writing, something librarians at FDU do not assist with. Questions I got the last weeks of the Fall 2024 semester were about more material things (how late are you open? can I have a pen? can you unjam the printer?) rather than complex research queries.

For an interactive version of this chart, click here.

However, a line graph like this one doesn’t do it justice. That’s why I am quilting. The specificity of the queries is lost in a graph. A quilt (one that’s wildly colorful and a little rough around the edges) allows you to zero in on a single data point, on one hexagon, on one question. The paper hexagons, which I am saving, serve as a complement to the fabric object. 

And the fabric itself—simple cotton material—provides a tactile sensation that I can’t get from a graph. The quilt is tangible. It’s warm when I lay it across my lap to add another piece. I hope it’s something people can connect to when they see it. Something they can touch. Something that brings them clarity on the work librarians do.

Conclusion

I am a proud parent to this quilt. Even though my hands ache, it’s been such a good learning experience to create something chaotically beautiful. I mentioned earlier that my first time trying EPP put me into a meditative flow state. Everything left my head except fabric, paper, and thread. As I work through my reference quilt, I find myself focusing on fabric, paper, and thread again. But now I think of the question on the paper as I cover it with a corresponding fabric. I find myself remembering the person who asked, the time I took to answer it. It’s become a reflective process—one that I would not have if I added a reference transaction through our library’s system and never looked at it again. 

The general consensus from viewers is that the quilt is a cool project, but opinions vary on whether or not it’s research. As librarians, we are all aware that research is a process as well as a result. I focus on the process here. Further, I argue that this quilt is research—that reflection on work done, that creative visualization of mindlessly-collected data, can too be research. I am connecting to the work I do in a way I had not envisioned. I am hoping that showing the quilt to other librarians evokes a similar response—one of reflection, connection, and understanding.

The project’s most effusive supporter was my partner’s grandmother, whom I showed the partially-done quilt in December 2024. She was delighted that it was hand-quilted. “That’s real work,” she said. 

I will be taking the finally-completed quilt to the American Library Association Annual Conference in June 2025 as a non-traditional research poster. I am hoping it’s received well there. Above all, I hope the quilt does justice to the work we do as academic librarians—the work that we usually don’t see.

Erica Leslie Weidner

Erica Leslie Weidner is a Research & Instruction Librarian and Adjunct Instructor at Fairleigh Dickinson University. She is an early career librarian, longtime writer, and newbie quilter. Find more of her work at www.erica-leslie.com.

• More Posts

At Princeton University, there is a two-week period every January in which all members of the institution are encouraged to teach and learn outside of the formal classroom contexts. This time between the Fall and Spring semester and all of its events and offerings is known as “Wintersession.”  As part of a Wintersession offering facilitated by Ariel Ackerly, Ofira Schartz-Soicher, and myself, I created a “friendship bracelet” that represented trees I routinely encounter when walking between two libraries on campus.

Nearly every day, I walk a path between Firestone, the main library on campus, and Lewis Science Library. This is a common thoroughfare for many of us on campus, and it serves as the central artery for traffic of all kinds. The trees along this route set the scene throughout the year, blooming in the spring and turning fiery hues in the autumn. We step on their fallen leaves and sneeze when they release their pollen. However, trees and other plants are all too easy to ignore, which we as humans do at our own peril.¹ The trees along this path are central characters of university life that warrant appreciation and attention. Indeed, there has been a push to increase the labeling of plant species generally on campus, which is why I believed there would be a dataset of campus trees that we might use for our Wintersession – a way to connect via shared experience to the topic of data physicalization. As part of this event, I created a friendship bracelet that represents the trees I encounter as I walk between our library locations. The form lends itself to linearity, like my path, and it also evokes the kind of warmth I feel towards our “true and silent friend,” the tree.²

Librarians might consider using friendship bracelets as part of an effort to find and engage with local datasets, particularly those well suited to linear forms of representation. In our case, creating this data physicalization connected us with the groundskeeping team, an otherwise distant university department. In the end, we found that this new connection benefited both parties. The groundskeeping team had started to track the types of trees as part of a project to label species of plants on campus and better understand their own work through the tracking of data. The data originators in groundskeeping were enthusiastic about how their data were engaged with in unexpected ways. Their tracking key information about trees on campus opened the door for students to notice not only the trees but also the groundskeepers’ efforts in planting, tracking, and supporting plant health across campus.

Planning:

Identify your source of data for trees along a given path.

Option A: Get your data from an existing source. This may be published online or may be informally shared.

I was able to acquire a mapping of trees from our University Facilities, Groundskeeping Team. I believed this information would exist in part because of the labels that have been placed on some of the trees on our campus. If there is a labeling initiative, then I thought there would probably be a list with locations and tree types being kept somewhere. I reached out and spoke to a number of people in facilities before being directed to the person who managed that dataset. They were happy to share the information and were enthusiastic about seeing what students did with that information.

Option B: Alternatively, tree data can be manually created data, by taking a photo on your phone of each tree in question and using an application or built-in feature to make an identification of the species.

For example, on iPhones, a small leaf icon will appear in the bottom-center of the screen of the Photos application if a plant is “recognized” by the Visual Lookup feature. Alternatively, Android users might use an application like iNaturalist to identify species they are not yet familiar with. Regardless of which tool or system being used, the computer can only make suggestions about species identification, suggestions which need to be verified by a human. In practice, this can look like taking the species suggestion from the algorithm or artificial intelligence, and comparing that with information about that species which is available from a reputable source. Sometimes it may take multiple photographs of a single specimen to find an angle that is readily identifiable through computational methods.

List your trees.

Once you have an idea about the species represented by various trees, those data will need to be organized linearly. List out the trees in the order they appear along your chosen path.

This process works best for me when I narrow down the data in a couple of stages.

Prepare your data.

Isolate out the data you will be using to make your data physicalization. By the end of data preparation, you should have a simple, color-coded table that is able to be printed on paper.

Make a simple table.

First, I use an Excel or Google Sheet to list out the key pieces of information that I am going to use in my physicalization. For example, I removed the extra information about side and age, since I did not intend to represent those pieces of information in my object.

To find out which trees fell along the path I selected, I uploaded the data table (which included location information) into a mapping tool online, in this case, Google’s MyMaps. At this stage, I found it helpful to explore the data using the faceting provided by the mapping tool.

Mapped trees on campus

Color code that table.

I like to color-code the information, so it will be easier to follow when I’m tying knots later. If you were using Google to map the location of trees along a path, you can take their arbitrary color coding as a starting point, but do feel free to edit these automated selections. If you were in a spreadsheet, you can use conditional formatting to color code values in the table.

If you can print in color, then highlighting the rows and shading them in different colors (or setting up conditional formatting rules in Excel) can be the most efficient way to color-code at this stage. Depending on your comfort level in different tools, you can even put your table in Word and color code it there, but keep in mind that Word’s format is entirely focused on display and there won’t be an easy way to manipulate the data in that program. If color printing is not an option, I print out the table and add in color with a highlighter, colored pencil, or other similar material.

Use paper when possible.

I find referencing information on paper beneficial for making. Paper can be folded up, crammed in a bag with other materials, and balanced on a knee without having to worry about damaging a screen or running out of battery. Paper also limits the amount of information I have access to, which is helpful in keeping me focused on the act of making.

Preparing Materials:

Line up your embroidery thread.

Gather together as many colors of embroidery thread as there are species of trees along the path.

If there are too many species for each to have its own color of embroidery thread, see if you can group them together by any classification (i.e. all birch tree types are in one family group so they can be represented by the same color, regardless of subspecies).

You may need to iterate over your coding schema based on the materials you have available to work with. Color coding your data table is one way to anticipate that the number of colors required might be too great. However, do be ready to iterate on your coding schema in order to find the best balance between individual examples and categories.

Decide on a knot pattern that represents each tree.

For example, I used 5 knots to represent each tree, but you may wish to use 10 knots for a longer bracelet or smaller number of trees. Once you choose a number of knots for each data point, stick to that number.

Cut thread to length.

Cut all the colors of embroidery floss to be approximately the same length. In general, the length of the thread should be two to three times as long as the finished bracelet. If you run out of length for a color unexpectedly, you can use a fisherman’s knot  to lengthen threads. Keep in mind that you’ll want to “bury” these knots in the central group of threads to avoid seeing the added material. If you know that you will be using more of one color than another, I recommend adding a second string to the group rather than lengthening that color, which more easily becomes tangled.

Making:

Make a loop.

Start your bracelet with an overhand loop knot.

Tie half-hitch knots.

Pull out two threads that represent your first data point and tie them around the other threads using a basic knot (also known as a half-hitch) for the number of times you selected. I chose to tie 5 knots for each tree, so after the first group of 5 knots, I switched the threads to tie the next group of knots with a new color that matched the data it represented. As you tie the half-hitch knots, the knot will begin to twist around the central bundle of threads, which will create a spiral effect (like in this instructional content).

Make a Bracelet Closure:

Option A: Tie a bead to the end and secure it using one final knot.

Once you have tied your last group of knots, finish your bracelet with a large bead for fastening through the loop on the other end. You may have to be creative about how you tie your bead on and knot it securely. If your bead is large enough, you can keep the loop the same. If the bead slips out of the loop too easily, reduce the size of the original overhand loop-knot to be just slightly larger than the size of that ending bead.

Option B: If you prefer not using a bead, you can tie the two ends together for a permanent closure. To do this, use the remaining thread to attach one end of the bracelet to the originating loop knot at the other end.

Bonus: For a more refined look, use another piece of embroidery floss to wrap the loop of the bracelet with a spiral of knots. Start by tying the thread to one end of the loop and then repeat the half-hitch knot until all of the exposed threads from your starting overhand-loop-knot are covered.

Voila! Your data are ready to wear.

This process can also be used to physicalize other kinds of data that are readily rendered in linear form and are able to be grouped by around ten categories. This format is suited to minimally complex data with limited facets.

Humanities Data Example:

I used the same technique to create a wearable bracelet representing the objects associated with miraculous images of the Virgin Mary located in the Germanic South in the late medieval and early modern period. That friendship bracelet symbolizes the objects I investigated in my art history dissertation.

Each object is represented by 7 knots of a particular hue, indicating which of the three sites of worship it was made for: Blue for Regensburg, Red for Altötting, or Green for Mariazell.

The shade of the color further distinguishes the type of object along a spectrum from three dimensional and entirely unique to examples of two-dimensional and reproducible objects. The lightest tones signify three-dimensional sculptures, dark tones represent prints, and middle tones denote paintings and cast or punched metal pilgrims badges. Additionally, I added golden embroidery thread to the knots that represented objects that were understood to be capable of performing miracles. The length of floss needed corresponded to the number of examples that were going to be represented in knots. The more knots appearing in one color meant there needed to be longer floss to pull from in order to create those knots. For example, I needed much more dark blue floss because most of the objects I wanted to represent were prints.

The objects are arranged in chronological order, from the 12th century to the 16th century, showcasing historical progression and emphasizing the massive influx of printed material at the turn of the sixteenth century. The overhand loop served as a starting point, and the first knots tied represented the oldest objects in my data. The farther away from the starting loop, the later in date the object represented was created.

This bracelet looks similar to the tree bracelet I made, yet it serves a different purpose – a visual and tactile representation of my research that blends craft and scholarship and tangentially engages in the wearable and meditative practice entangled in the objects of both devotion and study that it represents. Not wholly unlike rosaries used in practices of prayer at Marian worship sites through the centuries, this bracelet is a symbolic and contemplative tool, weaving together the historical and spiritual significance of the objects in my dissertation.

^{1. Ro, Christine. 2022. “Why Plant Blindness Matters.” February 24, 2022. https://web.archive.org/web/20240714083447/https://www.bbc.com/future/article/20190425-plant-blindness-what-we-lose-with-nature-deficit-disorder Archived July 14, 2024 by the Internet Archive.}

^{2. Quote from the poem “The Friendly Tree.” Wynne, Annette. For Days and Days: A Year-round Treasury of Child Verse. United States: Frederick A. Stokes Company, 1919, p. 180.}

Sarah Reiff Conell, PhD

Sarah earned her Ph.D. in the History of Art and Architecture from the University of Pittsburgh in 2022. Her research involved collecting and exploring data from stories about miracles that were attributed to the Virgin Mary in 16th Century Bavaria and Austria. Now, Sarah is a Research Data Management Specialist at Princeton University, where she curates data and serves as an editor for C19 Data Collective, a humanities data publication.

• More Posts

I. Introduction

As introduced by Knight in “Black Ribbon for Mourning: Affective Solidarity and Feeling Very Difficult Data” [1], the quantified self movement (also commonly referred to as self-tracking or lifelogging) [2] has been growing and sweeping the nation for over a decade [3]. It has primarily brought attention to user-centered tracking of health data, where users commonly monitor such clinical and health metrics as diet, weight, blood pressure, and physical activity [4]. Self-tracking is often facilitated through popular mobile applications including Apple Health or Fitbit, the latter of which pairs with an external Fitbit device [5].

Despite the popularity of tracking health vitals, self-tracking of data has also extended beyond common clinical metrics like diet or blood pressure. Users increasingly engage with the quantified self by tracking key aspects of their daily lives [6], such as goals, mood, mental health, and finances. For instance, during the COVID-19 pandemic, the popularity of the mobile app Calm rose significantly as users sought to manage their mental health and mindfulness [7]. In quantifying aspects of themselves and their lives, users reference their self-collected data to plan and achieve goals, visualize personal changes and trends over time, and organize or share information about themselves.

Given these goals, users of self-tracked data may benefit in their practices from incorporating mnemonic devices to reinforce important thoughts, activities, or goals. Mnemonic devices have long been touted as tools to improve learning and memory [8], where classic examples like the phrase “Every Good Boy Does Fine” have been used in music education to remember the five lines of the treble clef staff (E, G, B, D, and F). Studies also indicate the utility of visual and tactile techniques to strengthen memory and learning, with evidence suggesting these approaches may be more effective than auditory methods [9]. The act of crafting presents a viable tool integrating both visual and tactile stimulation to enhance memory, especially since crafts convey meaning to both creators and users who experience their multisensory qualities [10]. The act of making activates both learning and memory by facilitating the intake, processing, and storage of new information [11]. In this process, new knowledge is reinforced and committed to memory, making learning through creation an embodied cognitive experience [12].

I thereby propose the “mnemonic bracelet,” a physicalization of quantified-self data that users can create to remind themselves of their goals, emotions, achievements, and thoughts, with a combined tactile and visual component. The mnemonic bracelet is a self-created bracelet made via knotting, weaving, beading, and/or other techniques—much like a “friendship bracelet”—that offers a tactile and visual physicalization of quantified-self data. It may be used to represent and encourage mindfulness of one’s emotions, goals, and thoughts over a period of time and through all stages of its creation and use, whether the end product is worn or displayed. The mnemonic bracelet reinforces learning and memory not only through the making process [13], but also through its role as a mnemonic tool for cognitive offloading, or reducing the mental effort required for processing a task, particularly within the constraints of short-term working memory [14]. In serving as a physical reminder of goals and memories that might otherwise occupy the mind, the mnemonic bracelet may help alleviate the cognitive burden of holding and recalling these ideas.

II. Overview of Methodology

The mnemonic bracelet may be used for several purposes, including to increase mindfulness of emotions, to remind oneself of key goals and achievements or milestones throughout the week, and to visualize changes over a period of time. Users can and should adapt the focus, timeline, and specifications of their mnemonic bracelet to their individual needs.

To complete a data physicalization project like that outlined in part III, participants will engage in a simple, week-long data collection process. Each day, participants will spend five to ten minutes tracking their emotions, thoughts, and goals in a physical format [15], and this quantified-self data will then be analyzed and codified with colors, beads, and knot sequences.

After collecting their data, participants will use it to create a mnemonic bracelet that will physically represent their codified data and help them practice mindfulness surrounding their emotions, goals, and achievements. The activity below involves creation of one bracelet for the entire week, whereby a large knot will separate each day of the week. Users may alternatively opt to create a more “detailed” bracelet for each day of the week, but the focus will be on the former—a week-long bracelet—where a sample schema might include the following:

1. Emotions as colors of the woven threads: Blue for calmness, red for frustration, orange for joy, yellow for focus, purple for inattention, black for sadness, etc. [16];
2. Goals as beads of different shapes, sizes, and colors: A small blue glass bead for a particular long-term goal, a small red glass bead for a second long-term goal, a large red plastic bead for a particular goal with a fast-approaching deadline, etc.
3. Knot colors and sequences as emotions: Alternate the order and frequency in which a thread is used to create left-hand knots so that a dominant color and/or pattern emerges.

Please note that this is a sample schema and that users can revise the project creatively to suit their preferences and needs.

III. Instructions with Sample Project

Figure 1. Mnemonic bracelet completed according to part III specifications.

For the purposes of the following project, the sample user and their data is loosely based on the author, who identifies as neurodivergent.

Suppose the sample user seeks to physicalize their data with three mnemonic goals: to increase mindfulness of their emotions, to remind themselves of key achievements and work goals throughout the week, and to visualize daily changes in these dimensions over a week-long period. However, users can and should adapt the focus, timeline, and specifications of their mnemonic bracelet to their individual needs.

A. Materials

The following materials will be needed for this particular project. Substitutions may be made depending upon a user’s project and preferences.

• Five colors of embroidery floss
• A clipboard, or a flat surface and object for securing thread (e.g., tape, a binder clip)
• A notebook or journal (to track data) [17]
• Seven writing utensils of assorted colors (to denote recurring moods and goals of significance) [18]
• Four beads of assorted colors, shapes, and/or sizes
• A ruler
• Scissors

B. Procedure and Sample Project

1. The first step involves the tracking of quantified-self data by engaging in a simple, week-long data collection process: Each day, spend five to ten minutes tracking your emotions, thoughts, and goals in a designated journal. This stage may result in raw data with a significant amount of detail, but this data will be simplified, analyzed, and codified in the steps that follow. For the sake of clarity and brevity, Figure 2. below represents an abbreviated sample weekly log.
   

   Figure 2. Step 1: Sample data (abbreviated for clarity) for one-week data collection process.

2. Once you have collected your data for each day of the week, the next step is to review your data and note common patterns, then codify your data through annotation, including using colors, shapes, text, and other notes or marginalia that will best support your abilities and needs. The focus will be to identify dominant emotions or moods that recur on a given day or across days, as well as to identify goals from the week. For the sample project below, a system of color-coding, shapes, numbers, and symbols were used to denote recurring moods and goals of significance. Consider that the color green is assigned to the mood designated “neutral,” and a light blue is designated for the mood “calm/refreshed” (please note that the light blue shade appears darker in Figures 3. and 4. below).
   

   Figure 3. Step 2: Analyzed and codified sample data (abbreviated for clarity) with annotations.

3. Now that the data has been codified, assign a unique embroidery thread color to each mood identified, and assign a unique bead to each goal identified. This step represents the assignment of physical components to each data point for the mnemonic data physicalization process.
   

   Figure 4. Step 3: Assigning each data point a corresponding physical component for mnemonic physicalization.

4. Measure your wrist size using one of the embroidery threads, wrapping a thread around your wrist and then measuring the noted length with a ruler. (The author’s wrist is roughly 7 inches in circumference.) Then, multiply this measurement by 3 or 4 to achieve the length to which you will cut your embroidery threads. (Note: The author used 28-inch lengths of thread, or 4 times their wrist size. After finishing their project, 13 inches of thread remained on the shortest thread. However, your length may vary depending on the tightness of your knots, the size of your chosen beads, and other factors.)
5. Using the length you determined in step 4, cut one thread of each of the five embroidery thread colors, which will result in a thin and delicate bracelet like the author’s. (Note: If you would like to opt for a chunkier looking bracelet, you may cut two or three threads for each color, but you will need to extend the length of each thread by roughly 6-12 inches or more per additional thread per color. Please experiment with the length, as this will depend on your wrist size as well.)
   

   Figure 5. Step 5: Cut embroidery floss threads, one per color (five total).

6. Now, knot your five threads together 2 inches from the top; this 2-inch “margin” will be used to tie the bracelet ends once it is completed.
   

   Figure 6. Step 6: Knot your threads about 2 inches from the top.

7. Secure the threads by using your clipboard to clamp the section right above the knot. If you are not using a clipboard, secure your threads with tape or a binder clip to a flat surface, as appropriate.
   

   Figure 7. Step 7: Secure your threads.

8. Determine how long each “day section” of your bracelet will need to be, given your wrist length and the number of days of data. The author’s wrist is approximately 7 inches in circumference, and they collected 7 days of data, so they determined that each section of their bracelet will need to be at least

   7 inches ÷ 7 days = 1 inch per day.

   To create a looser bracelet, the approximate length of each section will be extended to slightly longer than 1 inch.

9. Plan how much variation each “day section” will require in terms of patterning, beading, and required length. For example, the author noted that for their sample user’s Wednesday data, three equally dominant moods emerged, particularly in the pattern ‘stressed – happy – stressed.’ Therefore, the bracelet section representing Wednesday will be divided into three distinct mood phases, which will result in over 0.33 inches per phase. The author also determined that having many “day sections” will require four different beads for four different goals over the week-long period, and the length of these beads will need to be taken into account when physicalizing each day.
10. In terms of knotting, this project will be created using a series of tight left-hand knots that will accumulate to result in a spiral-staircase look. (The creation of a left-hand knot is detailed in Figure 8. below.) To determine which thread to knot first, begin by looking at the first day of your codified data and identifying which moods are recorded. For the sample user, Sunday is entirely “neutral,” which corresponds to the color green. Thus, we will begin knotting with the green thread and sustain this color choice throughout the physicalized length for Sunday. Begin by creating your first left-hand knot, as shown in Figure 8. Make sure that the knot and all knots that follow are relatively tight.
    

    Figure 8. Step 10: Make a green left-hand knot.

    

    Figure 9. Step 10: Completed green left-hand knot.

11. Continue to make 9 more left-hand knots for a total of 10 left-hand knots. Note that the spiral-staircase pattern is emerging. In the next steps, we will follow these knots with goal beads for day 1, then return to 10 more left-handed knots to finish up the first “day section.”
    

    Figure 10. Step 11: Ten completed green left-hand knots.

12. Now we will proceed to add our first goal bead. To determine which bead to add first, begin by looking at the first day of your data and identifying which major goals are recorded. For the sample user, Sunday consists of three goals: “submit research proposal,” “work on publication 1,” and “drink water.” These goals correspond, respectively, to a brown polymer clay bead, a peachy polymer clay bead, and a small, pearly blue glass bead. We can choose to begin with any goal we prefer, in the order that is most suitable or logical for us. The author chooses to begin with the “submit research proposal” goal, represented by a brown bead. Since the hole of this bead is large, we will thread the bead onto the four non-green threads, then secure the bead with a left-hand knot made by the green thread around the remaining bundle of threads.
    

    Figure 11. Step 12: Add the first goal bead, then secure it with a green left-hand knot.

13. Repeat the beading process for the remaining goals tracked in the first day.
    

    Figure 12. Step 13: Completed goal beads for day one.

14. Now that we have attached all our goal beads, complete 10 more green left-hand knots to complete the Sunday section. As shown in Figure 13., the resulting section will be just over one inch long. Please note that if the user had recorded two emotions for Sunday, they would use a different color for the next 10 knots to correspond to the second emotion.
    

    Figure 13. Step 14: Completed green left-hand knots after goal beads.

15. (Optional) Before moving onto day two, complete day one by creating a large left-hand knot with all five threads. This is the same kind of knot you used to tie all the threads together in step 6. By creating a knot here, it will be easier to visually separate each day from one another.
    

    Figure 14: Step 15: Tie a large knot to represent the end of day one.

    

    Figure 15: Completed day one, after final knot.

16. Repeat steps 11-16 for each remaining day within the collected data, being sure to use the appropriate colors, beads, and knots in the process. Be mindful of representing multiple moods in one day. For example, for Wednesday, the user experienced three dominant moods (alternating stressed-happy-stressed), so the author made 6 left-hand black knots, then 4 left-hand orange knots, followed by goal beads, then the 4 remaining left-hand orange knots, and a final 6 left-hand black knots. This sequence and number of knots was carefully chosen to maintain the length of 20 knots that was established for Sunday.
    

    Figure 16: Step 16: Completed day one and day two.

    

    Figure 17: Step 16: Completed all seven days.

17. Now that all seven days have been completed and knotted off, cut the remaining length of unused thread so that 2 inches remain for tying the bracelet together, end to end.
    

    Figure 18. Step 17: Cut off the remaining thread.

18. Knot the bracelet together securely so that you have a complete bracelet! If you would like the bracelet to be tighter around your wrist, have another person tie the bracelet for you. Please note that the author ended up with a very loose bracelet that may be taken on and off. Your mnemonic bracelet is now complete! As you wear it or display it, note what questions and thoughts arise in your head. How did the creation process help you with your overarching goals? How does the bracelet work for you? What questions did the physicalization process and the bracelet answer for you? What could you do next time to better serve your needs?
    

    Figure 19. The finished mnemonic bracelet!

    

    Figure 20. The finished mnemonic bracelet, worn by the author.

    

    Figure 21. Visualizing early data and the mnemonic bracelet together.

Footnotes

[1] Kim Brillante Knight, “Black Ribbon for Mourning: Affective Solidarity and Feeling Very Difficult Data,” dh+lib: 2024 Special Issue (2024), https://dhandlib.org/black-ribbon-for-mourning-affective-solidarity-and-feeling-very-difficult-data/.

[2] Other names may also exist. See: Brent I. Fox and Bill G. Felkey, “The Quantified Self,” Hospital Pharmacy 51, no. 2 (2016), 189, https://doi.org/10.1310/hpj5102-189.

[3] Melanie Swan, “Quantified Self,” in Encyclopedia of Behavioral Medicine, ed. Marc D. Gellman (Springer, New York, NY, 2018), https://doi.org/10.1007/978-1-4614-6439-6_101945-1.

[4] Brent I. Fox and Bill G. Felkey, “The Quantified Self”; Huong Ly Tong et al., “The use of Mobile Apps and Fitness Trackers to Promote Healthy Behaviors during COVID-19: A Cross-Sectional Survey,” PLOS Digital Health 1, no. 8 (2022), https://doi.org/10.1371/journal.pdig.0000087.

[5] Tong, “The use of Mobile Apps and Fitness Trackers to Promote Healthy Behaviors during COVID-19: A Cross-Sectional Survey.”

[6] Shan Feng et al., “How Self-Tracking and the Quantified Self Promote Health and Well-being: Systematic Review,” Journal of Medical Internet Research 23, no. 9 (September 21, 2021), https://doi.org/10.2196/25171.

[7] Susannah Fox and Maeve Duggan, Tracking for Health, Pew Research Center, (2013), 1–40; Tong, “The use of Mobile Apps and Fitness Trackers to Promote Healthy Behaviors during COVID-19: A Cross-Sectional Survey.”

[8] Małgorzata Krzeczkowska et al., “Comprehensive Review of Mnemonic Devices and their Applications: State of the Art,” International E-Journal of Science Medicine & Education 9, no. 3 (December, 2015), 4–9, https://doi.org/10.56026/imu.9.3.4; Martin Dresler et al., “Mnemonic Training Reshapes Brain Networks to Support Superior Memory,” Neuron 93, no. 5 (March 8, 2018), 1227–1235, https://doi.org/10.1016/j.neuron.2017.02.003.

[9] James Bigelow and Amy Poremba, “Achilles’ Ear? Inferior Human Short-Term and Recognition Memory in the Auditory Modality,” PLoS ONE 9, no. 2 (February 26, 2014), 1–8, https://doi.org/10.1371/journal.pone.0089914; Ling Leh Qing et al., “Visual, Tactile and Auditory Memory: Any Difference between Students?” Journal of Psychology and Social Health 3 (December 1, 2019), 24–29.

[10] Nigel Morgan and Annette Pritchard, “On Souvenirs and Metonymy,” Tourist Studies 5, no. 1 (April 1, 2005), 29–53, https://doi.org/10.1177/1468797605062714; Sherry Schofield-Tomschin and Mary A. Littrell, “Textile Handcraft Guild Participation: A Conduit to Successful Aging,” Clothing and Textiles Research Journal 19, no. 2 (March, 2001), 41–51, https://doi.org/10.1177/0887302X0101900201; Sinikka Hannele Pöllänen and Reetta Marja Hirsimäki, “Crafts as Memory Triggers in Reminiscence: A Case Study of Older Women with Dementia,” Occupational Therapy in Health Care 28, no. 4 (September 2, 2014), 410–430, https://doi.org/10.3109/07380577.2014.941052.

[11] Marte S. Gulliksen, “Embodied Making, Creative Cognition and Memory,” FormAkademisk: Forskningstidsskrift for Design og Designdidaktikk 9, no. 1 (June 21, 2016), 1–19, https://doi.org/10.7577/formakademisk.1487.

[12] Gulliksen, “Embodied Making, Creative Cognition and Memory.”

[13] Gulliksen, “Embodied Making, Creative Cognition and Memory.”

[14] Evan F. Risko and Sam J. Gilbert, “Cognitive Offloading,” Trends in Cognitive Sciences 20, no. 9 (September, 2016), 676–688, https://doi.org/10.1016/j.tics.2016.07.002. For more information on cognitive offloading, see also: Lauren L. Richmond and Ryan G. Taylor, “The Benefits and Potential Costs of Cognitive Offloading for Retrospective Information,” Nature Reviews Psychology 4, no. 5 (March 17, 2025), 312–321, https://doi.org/10.1038/s44159-025-00432-2.

[15] Users may choose to track these metrics digitally, if preferred. The author chose physical tracking methods for the project delineated in this paper.

[16] A user’s color choices are personal preferences but may be informed by color psychology and symbolism. For more information on color psychology and correlations between colors and emotions, check out sources such as the following: Christof Kuhbandner and Reinhard Pekrun, “Joint Effects of Emotion and Color on Memory,” Emotion 13, no. 3 (March 25, 2013), 375, https://doi.org/10.1037/a0031821; Laura Dilloway, An Exploration into Colour Symbolism as Used by Different Cultures and Religions, National Centre for Computer Animation, (2006); Mitsuhiko Hanada, “Correspondence Analysis of Color–emotion Associations,” Color Research & Application 43, no. 2 (August 30, 2017), 224, https://doi.org/10.1002/col.22171.

[17] Users may choose to track these metrics digitally, if preferred.

[18] For users choosing to track metrics digitally, 7 different colors should be chosen and incorporated in a digital format.

Bibliography

Bigelow, James and Amy Poremba. “Achilles’ Ear? Inferior Human Short-Term and Recognition Memory in the Auditory Modality.” PLoS ONE 9, no. 2 (February 26, 2014): 1–8. https://doi.org/10.1371/journal.pone.0089914.

Dilloway, Laura. An Exploration into Colour Symbolism as used by Different Cultures and Religions: National Centre for Computer Animation, 2006.

Dresler, Martin, William R. Shirer, Boris N. Konrad, et al. “Mnemonic Training Reshapes Brain Networks to Support Superior Memory.” Neuron 93, no. 5 (March 8, 2018): 1227–1235. https://doi.org/10.1016/j.neuron.2017.02.003.

Feng, Shan, Matti Mäntymäki, Amandeep Dhir, and Hannu Salmela. “How Self-Tracking and the Quantified Self Promote Health and Well-being: Systematic Review.” Journal of Medical Internet Research 23, no. 9 (September 21, 2021). https://doi.org/10.2196/25171.

Fox, Brent I. and Bill G. Felkey. “The Quantified Self.” Hospital Pharmacy 51, no. 2 (February, 2016): 189. https://doi.org/10.1310/hpj5102-189.

Fox, Susannah and Maeve Duggan. Tracking for Health: Pew Research Center, 2013.

Gulliksen, Marte S. “Embodied Making, Creative Cognition and Memory.” FormAkademisk: Forskningstidsskrift for Design og Designdidaktikk 9, no. 1 (June 21, 2016): 1–19. https://doi.org/10.7577/formakademisk.1487.

Hanada, Mitsuhiko. “Correspondence Analysis of Color–Emotion Associations.” Color Research and Application 43, no. 2 (August 30, 2017): 224. https://doi.org/10.1002/col.22171.

Knight, Kim Brillante. “Black Ribbon for Mourning: Affective Solidarity and Feeling Very Difficult Data.” dh+lib: 2024 Special Issue (April 29, 2024). https://dhandlib.org/black-ribbon-for-mourning-affective-solidarity-and-feeling-very-difficult-data/.

Krzeczkowska, Małgorzata, Kamil Jurowski, and Anna Jurowska. “Comprehensive Review of Mnemonic Devices and their Applications: State of the Art.” International E-Journal of Science Medicine & Education 9, no. 3 (December, 2015): 4–9. https://doi.org/10.56026/imu.9.3.4.

Kuhbandner, Christof and Reinhard Pekrun. “Joint Effects of Emotion and Color on Memory.” Emotion 13, no. 3 (March 25, 2013): 375. https://doi.org/10.1037/a0031821.

Leh Qing, Ling, Kellin Wong Kai Xin, Lim Chin Fen, Treshia Tapui, Toh Bing Zhao, and Chua Bee Seok. “Visual, Tactile and Auditory Memory: Any Difference between Students?” Journal of Psychology and Social Health 3, (December 1, 2019): 24–29.

Morgan, Nigel and Annette Pritchard. “On Souvenirs and Metonymy.” Tourist Studies 5, no. 1 (April 1, 2005): 29–53. https://doi.org/10.1177/1468797605062714.

Pöllänen, Sinikka Hannele and Reetta Marja Hirsimäki. “Crafts as Memory Triggers in Reminiscence: A Case Study of Older Women with Dementia.” Occupational Therapy in Health Care 28, no. 4 (September 2, 2014): 410–430. https://doi.org/10.3109/07380577.2014.941052.

Richmond, Lauren L. and Ryan G. Taylor. “The Benefits and Potential Costs of Cognitive Offloading for Retrospective Information.” Nature Reviews Psychology 4, no. 5 (March 17, 2025): 312–321. https://doi.org/10.1038/s44159-025-00432-2.

Risko, Evan F. and Sam J. Gilbert. “Cognitive Offloading.” Trends in Cognitive Sciences 20, no. 9 (September, 2016): 676–688. https://doi.org/10.1016/j.tics.2016.07.002.

Schofield-Tomschin, Sherry and Mary A. Littrell. “Textile Handcraft Guild Participation: A Conduit to Successful Aging.” Clothing and Textiles Research Journal 19, no. 2 (March, 2001): 41–51. https://doi.org/10.1177/0887302X0101900201.

Swan, Melanie. “Quantified Self.” In Encyclopedia of Behavioral Medicine, edited by Gellman, Marc D. Springer, New York, NY, 2018. https://doi.org/10.1007/978-1-4614-6439-6_101945-1.

Tong, Huong Ly, Carol Maher, Kate Parker, et al. “The use of Mobile Apps and Fitness Trackers to Promote Healthy Behaviors during COVID-19: A Cross-Sectional Survey.” PLOS Digital Health 1, no. 8 (August 18, 2022). https://doi.org/10.1371/journal.pdig.0000087.

Ruth S. Xing

Ruth S. Xing is a Research & Instruction Librarian at UConn Library - Stamford. Ruth holds an MS in Library and Information Science from Syracuse University and dual BA degrees in Psychology and English from Cornell University.

• More Posts