<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Jamie Chung on Medium]]></title>
        <description><![CDATA[Stories by Jamie Chung on Medium]]></description>
        <link>https://medium.com/@jamie_chung?source=rss-1765e734ba10------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*BulJHE4c9HXmHX043uYLBg.png</url>
            <title>Stories by Jamie Chung on Medium</title>
            <link>https://medium.com/@jamie_chung?source=rss-1765e734ba10------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 09:37:42 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@jamie_chung/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[[CD Studio] Project 2 : Advocating Equity: Representing Data to Aid Interrogation, Understanding…]]></title>
            <link>https://medium.com/@jamie_chung/cd-studio-project-2-advocating-equity-representing-data-to-aid-interrogation-understanding-4357af0b45ae?source=rss-1765e734ba10------2</link>
            <guid isPermaLink="false">https://medium.com/p/4357af0b45ae</guid>
            <dc:creator><![CDATA[Jamie Chung]]></dc:creator>
            <pubDate>Mon, 26 Oct 2026 18:11:55 GMT</pubDate>
            <atom:updated>2026-11-10T19:18:53.084Z</atom:updated>
            <content:encoded><![CDATA[<h3>[CD Studio] Project 2 : <strong>Advocating Equity: Representing Data to Aid Interrogation, Understanding, and Just Decision-Making</strong></h3><h3>WK9 Monday : Where Do We Start with Technology and AI?</h3><p><em>October 19, 2026</em></p><h4>Learning</h4><p>Our instructor suggested two ways to find a research direction. We could start with a question and search for data, or look through available datasets and develop questions from what we found. She asked us not to narrow the topic before checking whether the data would let us study it.</p><p>For example, we could search broadly within technology and AI rather than specify a subtopic in every search. This might introduce a dataset or issue we had not considered.</p><h4>Project progress</h4><p>Our team discussed business AI adoption, employment, data centers, education, and access to AI. The Business Trends and Outlook Survey was one possible starting point because it included business AI use by state, industry, and business size. We discussed connecting it to employment, but that was one possibility among several.</p><h4>Reflection</h4><p>Technology and AI felt almost too broad to begin with. Every idea brought up several more, and I kept wondering whether we should choose a direction first or keep searching. The instructor’s suggestion gave us something smaller to do: find a source, check what it measures, and see which of our questions it could answer.</p><h3>WK9 Wednesday : What Counts as a Flaw in Data?</h3><p>October 21</p><h4>Learning</h4><p>We asked our instructor whether finding a flaw meant finding two datasets with conflicting numbers. She explained that we also needed to examine how the data was collected.</p><p>She used a hypothetical patient survey as an example. If participation depended on opting in or having access to a particular device, some patients might never appear in the results. We needed to ask who was included, who was missing, and whether the method matched the population the findings were supposed to describe.</p><p>She recommended reading the documentation before trying to interpret the table: who collected the data, why they collected it, and how they collected it.</p><h4>Project progress</h4><p>We were still considering labor, infrastructure, and access. We discussed possible questions about who benefits from AI and who pays for its infrastructure, but we had not found a clear combination of question and data.</p><p>We also looked at the AI Incident Database and discussed how it collected reports of AI-related incidents. Another source was hosted on GitHub, which made us question whether it was an official source. Our instructor suggested tracing it back to the originating agency rather than judging it by the platform alone.</p><h4>Reflection</h4><p>I was frustrated because I could not tell whether we were choosing the wrong datasets or misunderstanding the assignment. We kept asking what made a source good enough to use.</p><p>The survey example helped me understand what to look for. A limitation could be in the collection method or in the way we interpreted the results. We did not have to find an obviously incorrect number. That made the assignment easier to approach, even though our topic was still unsettled.</p><h3>WK10 Monday : Reading the Dataset Before Drawing It</h3><p>October 26</p><h4>Learning</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*PlDvrKmhClknip2LkZvG1w.jpeg" /></figure><p>We discussed LATCH: Location, Alphabet, Time, Category, and Hierarchy. These are different ways to organize information.</p><p>One example was a comparison of countries’ populations. Although the entries were places, a map would not necessarily be the clearest representation. If the purpose was to compare population sizes, arranging countries by quantity could make the comparison easier.</p><p>We also discussed scales, ranges, and buckets. For our assignment, each of us needed to take a dataset and identify how its variables were measured and organized.</p><h4>Project progress</h4><p>Our research included business AI use, employment, and businesses without paid employees. We were interested in whether AI might make it easier to work independently, but the nonemployer data did not directly measure that connection.</p><p>We worked through the AI survey’s questions and reporting periods. Current AI use and expected use in the next six months were separate measures. We also had to interpret period codes and entries such as “S” before we could use the tables confidently.</p><p>During the check-in, we discussed comparing employment across years and occupations. Our instructor suggested that these trends might be worth investigating alongside AI use, while reminding us that a parallel trend would not establish causation.</p><h4>Reflection</h4><p>I had expected downloading the data to be the start of making charts. Instead, we spent time figuring out what the questions and codes meant.</p><p>That was slower than I wanted, but the distinction between current and expected use showed why it mattered. If we mixed those columns, we could make a clear-looking chart that answered a different question from the one we intended.</p><h3>WK10 Wednesday : Why Was Delaware So High?</h3><p><em>October 28</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yeRg76Y16jjBaMGfRB9nbA.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rPkB6dMM6rhvfe9pTk4HzA.jpeg" /></figure><h4>Learning</h4><p>When we described our research to the instructor, we explained that Delaware had ranked unexpectedly high in the business AI-use data. We had started asking whether funding, taxes, remote work, or industry composition could explain it.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*efs1eLqwOeL24vyuHbOmCQ.jpeg" /></figure><p>She asked how we had arrived at those questions and encouraged us to document the sequence: what we expected, what the data showed, and what we investigated next. She also asked us to look for evidence supporting our explanations and to keep checking for missing information.</p><p>Her spreadsheet suggestion was specific. Instead of naming tabs after team members, we could name them after their contents, such as AI use, employment, or funding. We could then copy related columns into a tab and reorder the tabs to try out a presentation sequence.</p><h4>Project progress</h4><p>We had expected California or New York to lead in business AI use. Delaware’s position made us look more closely at differences between states.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*6V2ImYn0CN2eHbbcq6nIQw.jpeg" /></figure><p>We discussed funding, tax rates, remote work, and business registration as possible explanations. I also looked at County Business Patterns data to compare Delaware’s industry composition with the national picture, focusing on information, finance, and professional and technical services.</p><p>Alongside this, we reviewed employment trends across occupations and nonemployer business growth. Office and administrative work, healthcare support, and transportation and warehousing came up in those comparisons. We still could not say which changes were related to AI.</p><p>By the end of the discussion, we had several explanations to test. We needed to follow article references back to their sources and look for data on funding and taxes.</p><h4>Reflection</h4><p>Finding Delaware was exciting because it gave us a specific question after days of broad searching. But we moved quickly from “this is unexpected” to “maybe this explains it.” Each new explanation sounded convincing until we tried to support it.</p><p>I was relieved when the instructor said that following these questions was part of the assignment. We still needed evidence, but not having a conclusion yet did not mean the investigation had failed.</p><h3>WK11 Monday : When More Data Made the Story Less Clear</h3><p><em>November 2</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*o_q9NE4c8RsCyv5Iygaehg.jpeg" /></figure><h4>Learning</h4><p>Our instructor questioned the connection between our business AI-use data and the resident population and migration data we had brought in. She asked us to either explain that connection more carefully or focus on business data.</p><p>She also asked us to return to the original survey description: why was the information collected, and what was it intended to show?</p><p>Later, we discussed visual variables: size, shape, color, value, texture, and position, as well as sound, timing, motion, and tactile qualities. In the instructor’s periodic-table example, size represented atomic mass, color represented electronegativity, and motion represented reactivity. Viewers could turn on different properties and compare the patterns.</p><h4>Project progress</h4><p>We used a whiteboard to reconstruct our research sequence. Business AI use had led us to Delaware. Employment was another research strand, rather than the reason we had selected Delaware. We needed to make that order clear.</p><p>We reviewed our funding, tax, and migration explanations, but had not established that they explained the AI-use pattern. Our discussion also revealed confusion between percentages and counts. A higher business adoption rate did not necessarily mean a larger number of businesses using AI, and a small resident population alone could not explain the business percentage.</p><p>We began discussing visual sketches. One used stars for states, color for AI use, and a possible three-dimensional change for growth. The star could suggest both U.S. states and AI, but we questioned whether the display would be easy to read across all states.</p><p>For Wednesday, we needed to organize our question, datasets, scales, ranges, buckets, coordinate systems, early visual ideas, and data limitations.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rM1FDarYyzd5bCWLR9uhhw.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2ve2KI55swzhyQKwH3xD1g.jpeg" /></figure><h4>Reflection</h4><p>It was frustrating to have so much material and still struggle to explain the project. We kept returning to the same questions because we had not agreed on how the datasets connected.</p><p>The feedback gave us a concrete problem to fix. We had started with businesses, then brought in information about people without explaining the step between them. Adding another dataset would not resolve that. We needed to decide which comparison we were making.</p><p>The visual-variable discussion was also a reminder to start sketching. We had spent so much time searching that we had barely tested what someone else would be able to understand from our work.</p><h3>WK11 Tuesday : Putting Our Research into Presentation Order</h3><p><em>November 3</em></p><p>We met to organize our research for Wednesday’s presentation. We had collected data on business AI use, industry composition, and employment, but we still needed to explain how these pieces connected.</p><p>One question was whether to follow the order of our investigation or arrange the material in the order that would be easiest for an audience to understand. Our research had involved several detours. Including all of them would make it difficult to follow, but leaving them out could hide why we changed direction.</p><h4>From business AI use to Delaware</h4><p>We considered beginning with a brief introduction to the emergence of generative AI, followed by business AI-use figures from 2023 and 2025. This would establish the broader change before we compared individual states.</p><p>While reviewing the survey, we returned to the distinction between recent AI use and expected use in the next six months. A plan to adopt AI could not be treated as evidence that adoption had happened. We also questioned how voluntary participation affected the results and what we needed to check before describing the estimates as representative of businesses.</p><p>The state comparison gave us a more specific question. We had expected California or New York to stand out, but Delaware caught our attention instead. Before explaining that result, we needed to confirm whether we were comparing adoption levels or increases over time. Some earlier Delaware values were unavailable or difficult to compare, which also limited what we could say about its growth.</p><h4>Testing our explanations</h4><p>We revisited several possible explanations for Delaware’s position: funding, taxes, business registration, and migration. Each seemed plausible, but we had not established that any of them explained the survey results.</p><p>The population comparison was particularly difficult to justify. The survey measured businesses’ AI use, while population and migration figures described residents. We could not use one to explain the other without evidence connecting them.</p><p>We shifted toward the number and types of businesses in each state. We compared industries reporting higher AI use with state industry-composition data. Information, professional and technical services, education, finance and insurance, and real estate were among the industries we examined.</p><p>Industry mix gave us another explanation to investigate. However, the datasets did not align perfectly in their years and definitions, so we still needed to decide which comparisons were suitable to show together.</p><h4>Returning to employment</h4><p>We then returned to whether increased business AI use was related to employment changes.</p><p>We tried comparing employment counts and growth rates, both across total employment and within selected industries. Several times, we stopped to check whether two lists were measuring comparable things. Missing or suppressed values also made some state rankings incomplete.</p><p>The comparisons did not produce a consistent pattern. Some states appeared in employment-growth lists, others in decline lists, and some were missing from the relevant comparisons. We could not support a claim that AI was increasing or decreasing jobs from what we had examined.</p><p>That was frustrating. After spending so much time collecting data, I wanted a clearer answer. But trying to explain the results to one another made it harder to ignore where our argument depended on assumptions.</p><h4>What I took from the session</h4><p>Preparing the presentation exposed problems that were easier to overlook while we were collecting sources. Whenever we tried to write a transition between charts, we had to explain why the next dataset belonged there. Sometimes we could not.</p><p>I began to see why we kept getting stuck. We were moving between businesses, residents, industries, and workers, but those shifts were not always supported by the data.</p><p>We left with a possible presentation structure and an unanswered employment question. I was still disappointed that we could not reach a stronger conclusion, but I felt clearer about what needed work: checking the comparisons and explaining their limits before developing the final visuals.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*pIpYOJSeQYCFD7vKW1OTAQ.jpeg" /></figure><h3>WK11 Wednesday : What the Share-Out Made Us Reconsider</h3><p><em>November 4</em></p><p>Today we presented our research on business AI use and employment. We had explored national trends, differences between states, and industry composition before comparing AI-use growth with employment changes. Preparing the presentation helped us organize that research, but the class discussion showed us which connections still needed explaining.</p><p>One classmate asked how we could separate AI’s effect from other factors affecting employment. We had found states where AI use and employment both increased, and others where they moved differently. That comparison did not give us a clear answer about AI’s effect on jobs. We needed to make that distinction explicit, because “we did not establish a relationship” could easily be heard as “there is no relationship.”</p><p>Another question concerned our definitions: what kind of AI, and what kinds of businesses? We had introduced the project through recent generative AI headlines, but we needed to check whether that framing matched the broader AI technologies covered by the survey. Otherwise, readers might interpret every figure as a measure of ChatGPT-like tools.</p><p>Stacie suggested that, in our case, the investigation could become part of the story. Showing what we expected, what we compared, and why the answer remained uncertain could help readers understand how we reached that point.</p><h3>What I wanted to borrow from other teams</h3><p>Pooja’s explanation in the CPI presentation stayed with me. She used a hypothetical set of neighborhoods and houses to walk us through the calculation. I could follow where the information came from and where an estimate entered. Stacie also highlighted how effective that step-by-step explanation was.</p><p>Our presentation had named survey limitations without giving the audience much opportunity to understand them. I wanted to try showing how businesses’ responses become the AI-use figures in our charts. If readers could follow that process, they could see why coverage and participation mattered rather than simply being told that the data had flaws.</p><p>The EV group gave me another useful comparison. They discussed why a count of charging ports could not fully describe charging access: working condition, speed, and home charging could change the picture. It reminded me to ask what our own measures actually represented. Employee totals could show a change in employment, but not identify whose work had been replaced by AI.</p><p>In the women’s health discussion, Stacie kept asking about the consequences of missing information and who needed to see it. That helped me think beyond listing limitations. For our project, the consequence was that someone worried about their career might read a broad trend as a prediction about their own job.</p><p>I left wanting to make our explanation more useful to that person. Our next version needed clearer definitions, a visible collection process, and more careful language around what the comparisons could establish.</p><h3>WK11 Thursday : Making Space for the Main Question</h3><p><em>November 5, 2026 · Team Meeting</em></p><p>We met to work through the presentation feedback and begin developing visual ideas. Before sketching separately, we tried to agree on what our symbols would mean. We wanted to avoid combining three different visual interpretations of AI into one project.</p><p>We discussed chips, stars, buildings, and a skyline with more lights turning on as AI use increased. The skyline could communicate growth at a glance, but we questioned how readers would compare the actual values. The star raised a different concern: would people read it as generative AI or even a particular product?</p><p>That discussion brought us back to the survey definition. We checked what technologies it included because the symbol needed to fit the measure. We also revisited the definition of businesses, which determined who the survey represented.</p><p>It was difficult to return to definitions when I wanted to start making. Still, discussing them through a sketch felt more concrete than reading them in isolation. We could ask whether an image would give someone the wrong idea before spending time developing it.</p><h3>Why we tried removing Delaware</h3><p>We listed the charts we had so we could see the whole sequence together. That made one problem clearer: the Delaware investigation led into funding, taxes, and industry composition, but did not give us a convincing transition into employment.</p><p>We had included it because it was an important part of our research. Yet a reader would not necessarily need every step we had taken to understand our main question. We decided to try removing the Delaware and industry-composition section and moving more directly from AI-use changes into employment comparisons.</p><p>I found this hard because we had put considerable work into investigating Delaware. But keeping it would require an explanation of why that detour mattered to the employment story. We had not established that connection, and including it risked creating an expectation that we would explain Delaware’s result.</p><h3>Applying the CPI group’s approach</h3><p>We returned to the CPI presentation because it offered a way to develop our story without adding another loosely connected dataset. Their explanation made the collection and calculation process understandable.</p><p>We discussed giving more space to how our survey worked: which businesses were approached, how responses were gathered, and which businesses were outside its coverage. An iceberg or zoom-out reveal was one idea. Readers would first see a chart, then discover the collection process and boundaries behind it.</p><p>The reason for trying that structure was to connect a limitation to something the audience had just seen. We did not want the methods section to feel like an unrelated warning at the end.</p><p>We finished with a shorter sequence to test and an agreement to develop sketches individually. We had not settled the visual style, but we had made a useful editing decision: our research history would help shape the project without determining everything that had to appear in it.</p><h3>WK12 Monday : Giving Our Story a Specific Reader</h3><p><em>November 9</em></p><p>Today Stacie asked us to identify possible audiences and write down what they might value, expect, and need. We began individually on paper, then compared our answers as a team.</p><p>We kept returning to people afraid that AI would replace their jobs, including workers and people entering the job market. We chose to focus on that concern because it connected our research question to a decision someone might actually be trying to make.</p><p>At first, our answers repeated the same idea: people wanted to know whether AI was taking their jobs. Stacie pushed us to explain why they needed that information. We discussed whether it could help them reconsider a career direction or decide which skills to develop.</p><p>That distinction gave us a better way to evaluate our content. A state ranking might be interesting, but we needed to explain how it helped readers assess the claims they were hearing about work.</p><h3>Why our own motivation needed checking</h3><p>The next part of the activity asked us to consider our purpose as communicators. We wanted to reduce some of the anxiety created by AI-and-job-loss headlines. We were also considering whether changes in job requirements could offer another way to investigate AI’s relationship with work.</p><p>During our check-in, Stacie asked whether this direction came from what we were discovering or from a conclusion we wanted to support.</p><p>I wanted the project to leave people feeling less helpless, so the question was uncomfortable. It made me consider how an optimistic intention could influence our reading of the evidence. We could not promise reassurance simply because we wanted to offer it.</p><p>We asked Stacie whether her feedback meant the entire story needed to become about bias. She clarified that she wanted us to integrate transparency, including using “might” when the evidence only suggested a possibility.</p><p>That conversation helped us move forward. We could investigate changing job requirements, but we needed to keep the claim at the level the source supported. We also needed to let the findings change our message.</p><h3>Applying the storytelling arc</h3><p>We then used a whiteboard to map a hook, rising points, an aha moment, and next steps. We tried placing our actual content into those stages because a list of charts had not helped us decide what the reader should understand at each point.</p><p>We considered opening with the fear that AI was stealing jobs, introducing AI-use and employment trends, then examining why those comparisons could not answer the question on their own. We began exploring job-posting sources because employment totals might miss changes in what employers were asking workers to do.</p><p>One of our early messages was about understanding trends and preparing for the future. Stacie pointed out that it could apply to many topics. We needed to say what was changing, which evidence we had examined, and why it mattered specifically to AI and employment.</p><p>The brainstorming also became more playful. We discussed dinosaurs watching an approaching meteor. We wanted the opening to make the fear of a large, unfamiliar change recognizable before asking readers to engage with the charts.</p><p>We also considered people entering and leaving buildings while AI adoption followed a separate path. That sketch was an attempt to show two trends without automatically drawing a causal link between them.</p><p>The dinosaur idea made me excited to develop the visuals. But our discussion showed that we still needed a shared meaning for each symbol. If the meteor represented AI in one scene and survey coverage in another, the audience would have to keep relearning the image. Defining those meanings became a task for our next meeting.</p><h3>WK12 Tuesday : Checking Whether the Storyboard Could Carry the Evidence</h3><p><em>November 10</em></p><p>We met in the morning to develop the storyboard and decide what its symbols represented. One rough opening showed dinosaurs having tea, spotting a meteor, and panicking. A later scene imagined them using its heat to roast marshmallows.</p><p>We liked the humor because it could make a topic associated with anxiety easier to approach. The change from fear to curiosity also connected to our wish to help readers examine AI more carefully.</p><p>But we still needed evidence for the message between those scenes. An amusing ending could not establish that AI was harmless or that everyone would benefit from adapting. We needed to work out where the data belonged before committing to the full narrative.</p><h3>Why we explored job-posting data</h3><p>We had started looking at Indeed because our employment comparison described totals, while our newer question concerned possible changes within jobs. Job postings might offer a view of the requirements employers were advertising.</p><p>We looked for the available data and methodology so we could check what the measure counted. During the discussion, we questioned whether an AI-related posting meant using an AI tool, developing AI, or another kind of AI-related work. We also distinguished an advertised position from a person actually being hired.</p><p>We discussed the denominator as well. A higher percentage of postings mentioning AI could reflect a change in the total number of postings, rather than a larger absolute number of AI-related postings.</p><p>These questions helped us make the proposed claim more specific. A trend in AI-related language could be relevant to our investigation, but would not by itself show how existing jobs had changed. We needed to explain the source’s coverage and keep those interpretations separate.</p><p>I appreciated that we questioned this source even though it seemed to support a direction we liked. After struggling to reach a conclusion, it would have been easy to accept it too quickly.</p><h3>Why we reopened the employment comparison</h3><p>Removing Delaware made us reconsider our selection of five industries. We had originally used them to investigate whether state industry composition might explain the AI-use pattern. Once that section was removed, we needed to ask whether the same selection still served the broader employment question.</p><p>We discussed comparing employment across industries more broadly and returned to the original files. In doing so, we found that cross-industry and industry-specific tables had become mixed in our discussion and AI-assisted analysis.</p><p>We stopped to locate and share the raw sources because we could not carry the earlier comparison into a revised story without confirming what it measured. We also needed to leave the result open. A broader comparison might not look like the earlier one.</p><p>The storyboard helped expose this problem. Writing “employment change” in a scene was easy. Choosing the file, years, industries, and measure for that scene required us to check the work again.</p><h3>Why the placement of limitations mattered</h3><p>We also debated whether to explain limitations beside each dataset or reveal them near the end. We wanted the story to be engaging, but each placement affected how readers would interpret what they had already seen.</p><p>A final reveal could be memorable. It could also feel as though we were withdrawing the story after asking readers to trust it. Explaining limitations alongside the charts could make the comparison easier to follow, although we worried about slowing the pace.</p><p>We had not settled the placement by the end of the meeting. We had identified what to test: could someone remember which limitation belonged to which chart, and would the explanation change their understanding at the right moment?</p><p>I felt more involved in the design as we worked through these decisions. We were no longer asking only whether a chart or scene looked interesting. We were asking why it belonged, what evidence it carried, and what someone might conclude from it.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4357af0b45ae" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[[CD Studio] Week 5 / 6 / 7]]></title>
            <link>https://medium.com/@jamie_chung/cd-studio-week-5-e991d7bf95a3?source=rss-1765e734ba10------2</link>
            <guid isPermaLink="false">https://medium.com/p/e991d7bf95a3</guid>
            <dc:creator><![CDATA[Jamie Chung]]></dc:creator>
            <pubDate>Fri, 09 Oct 2026 22:39:12 GMT</pubDate>
            <atom:updated>2026-10-09T23:37:49.082Z</atom:updated>
            <content:encoded><![CDATA[<h3>[Week5]</h3><h3><strong>Navigating ambiguity</strong></h3><p>What I really enjoy and find challenging at the same time in Communication Design Studio is navigating ambiguity.</p><p>Sometimes, I feel like I understand something intuitively, but when I try to explain it to others, I realize that my reasoning isn’t as clear as I thought.</p><p>Turning a vague understanding into a logical argument that others can follow is difficult. But there’s also something incredibly satisfying about the moment when everyone’s thoughts finally align and we reach a shared understanding.</p><p>I’m starting to realize that this process of working through ambiguity is one of the most challenging, yet rewarding, parts of design.</p><h3><strong>Challenging Our Initial Assumptions</strong></h3><p>As we started developing more detailed wireframes, we realized that each of us had been imagining the solution differently.</p><p>It wasn’t until we began turning our ideas into something concrete that these differences became visible. Through discussions and iterations, we gradually worked toward a shared understanding of what we were actually trying to design.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*oUqkNokjRX8AmkBf.png" /><figcaption>After 3 hours meeting, we all knocked down…</figcaption></figure><h3><strong>Questioning Our Own Solution</strong></h3><p>As we started developing more detailed wireframes, we realized that each of us had been imagining the solution differently, even in the smallest details.</p><p>This made us wonder whether we were too attached to the chatbot concept we had presented at our midterm review. Were we genuinely designing the best solution, or were we just trying to make everything fit our initial idea?</p><p>To challenge our assumptions, we went through another round of solution exploration.</p><p>We revisited several possibilities:</p><ul><li><strong>Comments on individual emails:</strong> Could create a feedback loop between students and the advisor like figma, but wouldn’t make it easy to search for or discover relevant information. Also, It could</li><li><strong>Summaries and Q&amp;A within each email:</strong> Could provide contextual support, but would keep the experience fragmented across individual emails. It could cause confusion to students if information is being added + workload would increase.</li><li><strong>A separate SoD information hub:</strong> Could centralize information, but would introduce another channel students would need to remember to visit. Also, adding extra workload to Academic advisor</li><li><strong>A chatbot extension:</strong> Could offer flexible access to information, but raised questions about where and when the assistant should appear.</li></ul><p>Eventually, we came back to the chatbot concept. But this time, we had a new question:<br><strong>How could a chatbot help students recognize which emails deserve their attention?</strong></p><p>This shifted our focus from simply answering students’ questions to helping them recognize relevant information before they even knew what to ask.</p><p>We also realized that we had different ideas about where the service should live.</p><p>Should it be a browser extension? Should it appear within each individual email? Or should we build a separate information hub?</p><p>Each option had its trade-offs. A browser extension could appear outside the email context, where it might feel out of place. An assistant embedded in each email would require students to open individual messages first. And a separate hub would add another communication channel, which didn’t align well with our original hypothesis.</p><p>Eventually, we decided to integrate the assistant into the Gmail workspace and began developing our wireframes around that direction.</p><p><strong>Takeaway : No exploration is wasted, even when it brings you back to where you started.</strong></p><p>At first, I wondered whether revisiting our solution was unnecessary. But going through the alternatives helped us understand why we were choosing a chatbot and revealed a question we hadn’t fully considered before.</p><p>I learned that questioning an idea doesn’t always mean abandoning it. Sometimes, it helps you discover what the idea actually needs to solve.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*5g9m8jPz9M6_YIqa.png" /></figure><h3><strong>When Details Reveal Different Ideas</strong></h3><p>As we moved into UI design, we realized that even though we had agreed on the overall concept, we still had very different ideas about how each function should work.</p><p><strong>What should we actually show students?</strong></p><p>We discussed whether the interface should have separate tabs for a calendar, important information, or selected information cards. We also debated how much detail each card should contain. Should students see just a title, an AI-generated summary, or the original email?</p><p>Even the way we categorized information raised questions. Should we distinguish between mandatory information and personal preferences? What about information that requires action versus information that is simply good to know?</p><p>We gradually defined our information categories and decided that some features, like sorting and filtering by preference, would be more useful during onboarding than as permanent controls in the interface.</p><p>But the biggest realization came when we questioned the relationship between the information cards and the chatbot.</p><p><strong>If the cards already provide AI-generated summaries, what exactly is the chatbot supposed to do?</strong></p><p>We realized that we were designing two separate features that partly served the same purpose.</p><p>That question led to one of our biggest breakthroughs: <br><strong>What if we merged the cards and the chatbot into a single interaction?</strong></p><p>Instead of treating them as separate functions, clicking an information card could open the AI assistant with that specific context already loaded. Students could immediately ask follow-up questions without having to explain what they were referring to.</p><p>It was a relatively small change in the interface, but it made the whole experience feel much more connected.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*gjulFskwD_j3ewa5s58tig.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Eu_o93o2aO0Hax6gILS8IQ.png" /></figure><h4>Takeaway</h4><p><strong>Sometimes, the most meaningful design progress comes from questioning why two features need to exist separately.</strong></p><p>I realized that designing more functions doesn’t necessarily create a better experience. By questioning the redundancy between the cards and chatbot, we found a simpler interaction that could serve both purposes.</p><p>It also reminded me that alignment doesn’t happen just because everyone agrees on a concept. We need to make our ideas concrete enough to discover where our assumptions differ.</p><h3>Revisiting Our Solution Hypothesis</h3><p>Our team felt that we needed to take a step back and reassess whether our initial solution would actually meet students’ needs.</p><p>To hear directly from students, we conducted a Google Forms survey over three days to better understand their needs and see whether our proposed solution was heading in the right direction.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*DrfNs5EpKGRYwMTx.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*IjDAs5XKVWq7xbG-.png" /></figure><p>After our midterm presentation, our team felt that we needed to take a step back and validate whether our initial solution — a chatbot — would actually meet students’ needs.</p><p>To hear directly from students, we conducted a Google Forms survey over three days to better understand their needs and see whether our proposed solution was heading in the right direction.</p><p>Through the survey, we identified a group of students who had questions but hadn’t reached out to ask them. We also gained a better understanding of how students currently find and access school-related information.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*lpPbUo8H7Wo2Ke6Y.png" /></figure><p>However, one finding challenged our initial assumption.</p><p>When we asked where students would naturally expect to find an AI assistant for School of Design information, they pointed to the SoD website.</p><p>This made us question our decision to integrate the assistant directly into email. Were we designing something that didn’t align with students’ expectations?</p><p>We shared this uncertainty with Stacie, and she offered a perspective that helped us rethink the finding:</p><p><strong>“People’s expectations are shaped by their previous experiences. They aren’t always the answer.”</strong></p><p>That helped me realize that what users expect isn’t necessarily the same as what would work best for them. Their expectations are valuable insights, but they shouldn’t automatically dictate our design decisions.</p><p>With that perspective, we felt more confident moving forward while continuing to question and refine our approach.</p><p><strong>Takeaway</strong></p><ul><li>Intuition alone isn’t enough to convince others. I need to articulate my reasoning and support it with evidence.</li><li>If I struggle to explain something clearly, I may not fully understand it myself yet. Explaining my thoughts to others helps me discover gaps in my own reasoning.</li></ul><h3>[Week 6]</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*nq-eSVbprnLB8uXM1BDG7w.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*X_cAVfsjBuyJK_SlyHVyqA.png" /></figure><h3>Naming Our Service</h3><p>We explored quite a few names for our service.</p><p>Chaewon is especially good at ideation, and one of her suggestions, <strong>Scotty</strong>, was a strong candidate. We also explored several names inspired by CMU’s Scottish heritage, but none of them quite clicked.</p><p>Then, we turned our attention to the <strong>Mellon</strong> in Carnegie Mellon.</p><p>Compared to Carnegie, Mellon seemed to get less attention, so we thought it would be fun to give it a little shout-out.</p><p>And that’s how we landed on <strong>Mellon!</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/796/1*tHmRwV0hTFRZoiX-VxE-qQ.png" /><figcaption>Chaewon explored different styles of logo</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/402/1*iy9tutCuoUMdgryBPst3wg.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/354/1*ZEwst3izIPumD5g2R2QpUg.png" /><figcaption>Reference we used</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/698/1*kAllq6gwudWXqfsmq7kG0g.png" /><figcaption>Mellon!</figcaption></figure><h3>Dividing Responsibilities</h3><p>We decided to divide the responsibilities among team members, and I took charge of the video, including scriptwriting, gathering visual assets, and editing.</p><p>To be honest, I felt a bit overwhelmed at first because I didn’t have much experience with video editing. However, with help from many people, I was able to enjoy the process and bring everything together.</p><h3>Developing the Video Scenario</h3><p>Since our product had a relatively clear user journey, developing the overall flow of the video was fairly straightforward.</p><p>However, the challenging part was figuring out <strong>how to visually communicate the problem we had identified.</strong></p><p>Initially, we came up with a scene where a student receives an endless stream of emails.</p><p>But we realized that simply showing too many emails wasn’t enough to communicate our actual problem: <strong>students struggle to navigate information, identify what matters to them, and take the necessary actions.</strong></p><p>So we refined the scenario to make the problem more specific. Instead of just showing an overwhelming inbox, we showed a student receiving multiple emails from different stakeholders and eventually missing an important action among all those messages.</p><p>This helped us shift the focus from simply showing <em>information overload</em> to illustrating <strong>how fragmented communication can lead to missed actions.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xIeS3Hv07jz0cpFM9HAh4A.png" /></figure><p>We also refined the hook of our final presentation to align with this direction. Instead of simply showing a student overwhelmed by too many emails, we made the situation more specific: a student struggling to find the information they need in an email from their academic advisor.</p><p>This helped us create a more consistent narrative from the beginning of our presentation through to the final video, keeping both focused on the core problem we were trying to address.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*71rrcnirijsc34sdjz7aDA.png" /></figure><h3>Week 7</h3><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e991d7bf95a3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[[CD Studio] Week 4]]></title>
            <link>https://medium.com/@jamie_chung/cd-studio-week-4-a94e1d9ef59a?source=rss-1765e734ba10------2</link>
            <guid isPermaLink="false">https://medium.com/p/a94e1d9ef59a</guid>
            <dc:creator><![CDATA[Jamie Chung]]></dc:creator>
            <pubDate>Fri, 09 Oct 2026 22:21:23 GMT</pubDate>
            <atom:updated>2026-10-09T22:21:23.948Z</atom:updated>
            <content:encoded><![CDATA[<p>This week, we had our checkpoint presentation.<br>Our project explores the <strong>School of Design’s communication system</strong>, and we narrowed our focus to the <strong>relationship between the graduate academic advisor and students.</strong> Since this relationship is closely connected to many other parts of the School of Design communication ecosystem, our team tried to clearly explain why we chose this relationship and how we narrowed down our scope.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*24LzBb4Njd3GFFhD.png" /></figure><p>The feedback from the presentation helped me understand which parts of our project still need more explanation and refinement. It was also useful to see other students’ presentations and observe how different teams structured and communicated their thought processes.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*BnIde-d4GyK8-Bjq.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*qd4zE0AkBdjwErQU.jpeg" /></figure><h3>Takeaways</h3><ul><li>When presenting an idea, always need to think about <strong>who the audience is</strong>.<br>: We need to provide enough context for them to follow our thinking, while also finding the right balance between high-level framing and detailed information.</li><li>Even if we have strong research, that effort can easily disappear if we do not translate it into something visually clear and presentable.<br>: Looking back, I think our group did not show enough evidence to support some of our decisions. We skipped too many parts of our research process, which made some of our conclusions feel less grounded than they actually were.</li><li>Importance of creating a strong hook at the beginning of a presentation :<br>We need to give people a reason to pay attention before explaining the details of the system.</li></ul><h3>Ideas From the Feedback</h3><p>The feedback also gave us several directions to explore.</p><p>One idea is to integrate tasks or reminders directly into communication channels so that messages become more actionable. For example, after receiving an important message, the system could automatically create or suggest a follow-up reminder.</p><p>Another important point was that the primary communication channel should also reflect Donna’s own workflow and preferences, rather than only focusing on what is convenient for students.</p><p>Currently, there is also a gap in the feedback loop. Donna and Stacie may send an email, but they do not necessarily know whether students read it, understood it, ignored it, or knew what action to take. At the same time, students may find it difficult to respond or ask follow-up questions with very little effort.</p><p>This made us think more carefully about the system as <strong>two-way communication</strong>, rather than simply improving how information is delivered from the school to students.</p><p>Instead of immediately framing the solution as a chatbot, we may need to ask a broader question:</p><p><strong>How might we reduce repetitive communication for staff while making it easier for students to understand, respond to, and act on information?</strong></p><p>An AI assistant could potentially help with recurring questions, summarize or clarify information, and support faster responses. However, AI can also make mistakes, so an important part of the system would be preserving Donna’s agency and allowing her to remain in control of what the system communicates on her behalf.</p><p>This feedback also suggested that our project may not need to focus only on events. The communication problem could potentially expand to broader School of Design information, similar to how an HR support system helps people navigate recurring organizational questions.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a94e1d9ef59a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[[CD Studio] Week 3]]></title>
            <link>https://medium.com/@jamie_chung/cd-studio-week-3-87980bfc8646?source=rss-1765e734ba10------2</link>
            <guid isPermaLink="false">https://medium.com/p/87980bfc8646</guid>
            <dc:creator><![CDATA[Jamie Chung]]></dc:creator>
            <pubDate>Fri, 09 Oct 2026 22:20:16 GMT</pubDate>
            <atom:updated>2026-10-09T22:20:16.666Z</atom:updated>
            <content:encoded><![CDATA[<p>I conducted two one-hour interviews: one with a new graduate student and another with Donna. Both gave me a lot of insight into how event information is communicated and experienced from different sides.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*knZ5-KFTFoo2NClh.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*IYIUu4lXWNP1aBEi.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*NDrHH85Ncc9fVyfl.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*btAztngeOdiKIFh4.jpeg" /></figure><p>During class, our team gathered around the table and shared what we had heard from each interview. Comparing the interviews was especially useful because we could start seeing gaps between how information is communicated and how students actually perceive and interpret it.</p><h3>Takeaway</h3><ul><li><strong>Good interviews depend on more than good prepared questions.</strong> I realized how important it is to ask the right follow-up question in the moment. Sometimes the first answer only reveals the surface, and one additional “why?” or “what happened next?” can uncover a much more useful insight.</li><li><strong>Think about whether an intervention should change a mental model or work with the one that already exists.</strong> Before jumping into solutions, I want to ask whether people’s existing understanding is actually the problem. Sometimes it may make more sense to design around an existing mental model rather than trying to change it.</li><li><strong>Synthesis becomes more valuable when perspectives are compared, not just collected.</strong> Looking at Donna’s and the student’s experiences side by side helped us see mismatches that would have been difficult to notice from either interview alone.</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=87980bfc8646" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[[CD Studio] Week2]]></title>
            <link>https://medium.com/@jamie_chung/cd-studio-week2-c44a56951e7c?source=rss-1765e734ba10------2</link>
            <guid isPermaLink="false">https://medium.com/p/c44a56951e7c</guid>
            <dc:creator><![CDATA[Jamie Chung]]></dc:creator>
            <pubDate>Fri, 09 Oct 2026 22:19:01 GMT</pubDate>
            <atom:updated>2026-10-09T22:19:01.419Z</atom:updated>
            <content:encoded><![CDATA[<p>Based on what we had explored so far, we started looking into the communication system at CMU as a group.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*kFNWzw_l30EGoNKG.jpeg" /></figure><p>But it took longer than we expected because, as we tried to put everything together, our approach became too general.</p><p>Another reason we struggled a bit was that each of us had approached the research at a different scale.</p><ul><li><strong>Chaewon:</strong> How to install an HP printer</li><li><strong>Pragya:</strong> How to communicate through email</li><li><strong>Jamie:</strong> How to use the shuttle bus</li></ul><p>The scope of our individual research was quite different, so it wasn’t easy to bring everything together at the same system level.</p><p>Press enter or click to view image in full size</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*7l_Tr8bIhhrYrhDa.png" /></figure><p>Thanks to Stacie and Jaseelyn’s guidance, we realized that we needed to focus on a more specific context. So, we decided to look into the School of Design’s communication system.</p><p>Interestingly, even within our team, we represented different types of audiences:</p><ul><li><strong>Chaewon:</strong> a new student who attended the School of Design orientation</li><li><strong>Pragya:</strong> a new student who didn’t attend the orientation</li><li><strong>Jamie:</strong> kind of a returning student who didn’t attend the orientation</li></ul><p>We noticed this through a casual conversation while sharing our orientation experiences.</p><p>Interestingly, the information each of us received was also different.<br>So, we decided to look into a SoD event.</p><p>Press enter or click to view image in full size</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*aqnUKCNlQaDMfjgm.png" /></figure><p>This approach was more of a general communication map showing how SoD event information is delivered to students. However, it still felt too broad to us, so we decided to zoom in further. We chose the orientation onboarding process to dig deeper.</p><p>Press enter or click to view image in full size</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*ZoZFSj2yUvKDgWbV.png" /></figure><p>Still, it didn’t show the complex communication happening behind the event. So we added a time layer and broke down the information in more detail.</p><p>Press enter or click to view image in full size</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*LO9AT-n0RE47jo85.png" /></figure><p>Finally, we found a good scale for the communication system and could clearly see the relationships within it.</p><h3>Here’s what we found</h3><ul><li><strong>Missing information creates knowledge gaps</strong><br>e.g. <em>Who is Ray? What does he look like?</em></li><li><strong>Information is scattered across different places and messages</strong><br>Students have to explore multiple sources, and the level of detail is inconsistent.</li><li><strong>Email is the key communication channel</strong><br>This creates redundancy and makes important information easy to miss.</li><li><strong>The amount of information increases abruptly over time</strong><br>Students can quickly become overwhelmed, especially around orientation.</li><li><strong>Senders don’t see the full communication load students receive</strong><br>Each sender knows what they send, but not necessarily how many other messages students are receiving.</li></ul><p>But we were concerned that our frame might be too narrow. Stacie told us that we could expand this model and apply it to event communication more broadly.</p><h3>Takeaway</h3><p><em>- To find opportunities in communication, we need to narrow down the scope enough to see the details and relationships.</em></p><p>Based on these observations, we wanted to break them down from a systems thinking perspective.</p><p>Press enter or click to view image in full size</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*w0N8caFWSxwOKcn7.png" /></figure><p>Reference : <a href="https://ecochallenge.org/iceberg-model/">https://ecochallenge.org/iceberg-model/</a></p><p>Individual events were easy to recognize, but when we started looking for patterns across them, we had to think more carefully.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*TWrqJwAH0DYgHXtu.png" /></figure><p>When we compared different events, we found several inconsistencies. For example, there were gaps in when students received information, how much information they received, and how easily they could access it. We also found differences in the importance of the information being communicated.</p><p>So we identified this as one of the key patterns. We also found that most communication happens through email, and in most cases, the information comes from a single sender.</p><p>For structure and mental models, the structure was more difficult to fill in because we didn’t really know the school system and the processes happening behind the scenes. So we decided to fill in the mental model first.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c44a56951e7c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[[CD Studio] Week 1]]></title>
            <link>https://medium.com/@jamie_chung/cd-studio-week-1-623f66a94281?source=rss-1765e734ba10------2</link>
            <guid isPermaLink="false">https://medium.com/p/623f66a94281</guid>
            <dc:creator><![CDATA[Jamie Chung]]></dc:creator>
            <pubDate>Fri, 09 Oct 2026 21:45:05 GMT</pubDate>
            <atom:updated>2026-10-09T21:45:05.454Z</atom:updated>
            <content:encoded><![CDATA[<p>We started with an activity observing toys on the floor.<br>The first question was:</p><blockquote><strong><em>What makes us think of these objects as toys?</em></strong></blockquote><p>Press enter or click to view image in full size</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Y9Y7sX2kKbSDqp7y.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*rmbEfriMayZFsmmA.jpeg" /></figure><p>As we observed them, a few questions stayed with me:</p><ul><li>When do you want people to be surprised?</li><li>How do people know where to grab?</li><li>Is it intended to look old? If so, where does that visual language come from?</li></ul><h3>Takeaways</h3><ul><li>Designers need to decode existing forms of communication.</li><li>Think about how things communicate with you, both physically and mentally.</li><li>Through design interventions, designers can help people think or act differently.</li></ul><blockquote><em>Card Sorting</em></blockquote><p>The next activity was about what we should prioritize in our learning environment throughout the semester.</p><p>There were cards with quotes related to interaction design, collaboration, and attitudes toward learning. Working in teams, we placed them along an axis, grouping and prioritizing them. It gave us a chance to reflect on what we want to value in our learning environment going forward.</p><p>My team, Pragya, Chaewon, and I, decided to use the amount of energy we would need to put into something as the axis.</p><p>Press enter or click to view image in full size</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*2BQUXqXvuObhfeJE.jpeg" /></figure><p>While we were going through the quotes, there was one that we initially struggled to understand:</p><blockquote><em>“Map relationships and non-obvious leverage points in systems.”</em></blockquote><p>At first, I understood this somewhat superficially as “look for points that are not immediately observable and see them as opportunities for change through design.”</p><p>But as we discussed it further, I started thinking more about what <em>non-obvious</em> could mean. What is easily visible in a system might not represent everyone within it. There can be people, relationships, or perspectives that are less visible because they are overshadowed by the majority.</p><p>This made me think about how easily I can approach a system based only on what I can immediately see, and what might be missed when I do that.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*XS7IKoFEVEheW0pR.jpeg" /></figure><p>My goal for the semester</p><h3>Takeaway</h3><ul><li>Think more carefully about how I communicate with others, especially when we have different opinions.</li><li>Instead of assuming that I understand something immediately, take time to understand how other people are interpreting the same thing.</li><li>When looking at a system, pay attention not only to what is obvious, but also to relationships, perspectives, or points of influence that might be less visible.</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=623f66a94281" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[[MDes] Week 5]]></title>
            <link>https://medium.com/@jamie_chung/mdes-week-5-8f729b1712d5?source=rss-1765e734ba10------2</link>
            <guid isPermaLink="false">https://medium.com/p/8f729b1712d5</guid>
            <dc:creator><![CDATA[Jamie Chung]]></dc:creator>
            <pubDate>Mon, 21 Sep 2026 21:59:42 GMT</pubDate>
            <atom:updated>2026-10-09T21:27:04.368Z</atom:updated>
            <content:encoded><![CDATA[<p>It’s already been a month since the semester started. Each day feels long, but the weeks pass by so quickly. It’s bittersweet. I wish time would slow down a little.</p><h3>Comminication Design Studio</h3><p>What I really enjoy and find challenging at the same time in Communication Design Studio is <strong>navigating ambiguity.</strong></p><p>Sometimes, I feel like I understand something intuitively, but when I try to explain it to others, I realize that my reasoning isn’t as clear as I thought.</p><p>Turning a vague understanding into a logical argument that others can follow is difficult. But there’s also something incredibly satisfying about the moment when everyone’s thoughts finally align and we reach a shared understanding.</p><p>I’m starting to realize that this process of working through ambiguity is one of the most challenging, yet rewarding, parts of design.</p><p>As we started developing more detailed wireframes, we realized that each of us had been imagining the solution differently.</p><p>It wasn’t until we began turning our ideas into something concrete that these differences became visible. Through discussions and iterations, we gradually worked toward a shared understanding of what we were actually trying to design.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NhR2INGKIfmX_wISG9y_vA.png" /><figcaption>After 3 hours meeting, we all knocked down…</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-t4CYNd4SBKry0Oo42cWdQ.png" /></figure><p>our team felt that we needed to take a step back and validate whether our initial solution (a chatbot ) would actually meet students’ needs.</p><p>To hear directly from students, we conducted a Google Forms survey over three days to better understand their needs and see whether our proposed solution was heading in the right direction.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hCxy8Xa2c-sS5-xF_mEE6w.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yLe3hnY3qY3WNNOdkeiocw.png" /></figure><p>After our midterm presentation, our team felt that we needed to take a step back and validate whether our initial solution — a chatbot — would actually meet students’ needs.</p><p>To hear directly from students, we conducted a Google Forms survey over three days to better understand their needs and see whether our proposed solution was heading in the right direction.</p><p>Through the survey, we identified a group of students who had questions but hadn’t reached out to ask them. We also gained a better understanding of how students currently find and access school-related information.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*thLy26j_g6McJQjxzS89hw.png" /></figure><p>However, one finding challenged our initial assumption.</p><p>When we asked where students would naturally expect to find an AI assistant for School of Design information, they pointed to the SoD website.</p><p>This made us question our decision to integrate the assistant directly into email. Were we designing something that didn’t align with students’ expectations?</p><p>We shared this uncertainty with Stacie, and she offered a perspective that helped us rethink the finding:</p><p><strong>“People’s expectations are shaped by their previous experiences. They aren’t always the answer.”</strong></p><p>That helped me realize that what users expect isn’t necessarily the same as what would work best for them. Their expectations are valuable insights, but they shouldn’t automatically dictate our design decisions.</p><p>With that perspective, we felt more confident moving forward while continuing to question and refine our approach.</p><p><strong>Takeaway</strong></p><ul><li>Intuition alone isn’t enough to convince others. I need to articulate my reasoning and support it with evidence.</li><li>If I struggle to explain something clearly, I may not fully understand it myself yet. Explaining my thoughts to others helps me discover gaps in my own reasoning.</li></ul><h3><strong>Seminar 1</strong></h3><p>On Monday, we read an excerpt from Friedman and Hendry’s <em>Value Sensitive Design</em>. The authors asked, “How can human values be systematically incorporated throughout the technology design process to support human flourishing?”</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/964/1*2ykNcTyiLEW0LNTGtmx91w.png" /></figure><p>To explore this question, we participated in an activity where we shared a value that mattered to us and developed it into a potential research question.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*C09qoOqxZlBX3NSriTod2Q.png" /></figure><p>The values I chose were:</p><ul><li>Support, trust, and authenticity in relationships</li><li>Belonging and supporting family</li><li>Gratitude in life, inspired by Eunbi</li><li>Sharing something meaningful with the community</li><li>Believing in yourself and taking action</li></ul><p>Based on the values we wrote down, another person on our team developed one of them into a research question.</p><p>This was inspiring because Jiyeong and I interpreted relationships differently. While I focused on support and trust between human, she brought a broader perspective to my value and proposed this compelling research question:</p><p><strong>In an era of AI, how can we preserve the authenticity that has traditionally existed in human relationships?</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8f729b1712d5" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[[MDes] Week 4 : Expanding the Perspective]]></title>
            <link>https://medium.com/@jamie_chung/mdes-week-4-98d301504683?source=rss-1765e734ba10------2</link>
            <guid isPermaLink="false">https://medium.com/p/98d301504683</guid>
            <dc:creator><![CDATA[Jamie Chung]]></dc:creator>
            <pubDate>Sat, 19 Sep 2026 15:13:24 GMT</pubDate>
            <atom:updated>2026-10-09T20:51:37.442Z</atom:updated>
            <content:encoded><![CDATA[<p>It’s already been a month since the semester started.<br>I’m still getting into the rhythm of the academic schedule, while also getting used to my cohort and courses.</p><p>This week felt especially meaningful for School of Design students because we had the <strong>Design Nexus Summit 2026</strong>.</p><p>The Design Nexus Summit is one of the School of Design’s signature events, bringing together students, alumni, and professionals. Through the event, students have the opportunity to hear from inspiring speakers, learn from their experiences, and connect with people across the design community.</p><h3>Communication Design Studio</h3><p>This week, we had our checkpoint presentation.<br>Our project explores the <strong>School of Design’s communication system</strong>, and we narrowed our focus to the <strong>relationship between the graduate academic advisor and students.</strong> Since this relationship is closely connected to many other parts of the School of Design communication ecosystem, our team tried to clearly explain why we chose this relationship and how we narrowed down our scope.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*mKP6QQTLDybYVy3tWzFQmg.png" /></figure><p>The feedback from the presentation helped me understand which parts of our project still need more explanation and refinement. It was also useful to see other students’ presentations and observe how different teams structured and communicated their thought processes.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*EaLUgPMtZ_5z-46fWwio0g.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*vnsZnjUIJ1fAHVT78FUXfA.jpeg" /></figure><h3>Takeaways</h3><ul><li>When presenting an idea, always need to think about <strong>who the audience is</strong>. <br>: We need to provide enough context for them to follow our thinking, while also finding the right balance between high-level framing and detailed information.</li><li>Even if we have strong research, that effort can easily disappear if we do not translate it into something visually clear and presentable. <br>: Looking back, I think our group did not show enough evidence to support some of our decisions. We skipped too many parts of our research process, which made some of our conclusions feel less grounded than they actually were.</li><li>Importance of creating a strong hook at the beginning of a presentation : <br>We need to give people a reason to pay attention before explaining the details of the system.</li></ul><h3>Ideas From the Feedback</h3><p>The feedback also gave us several directions to explore.</p><p>One idea is to integrate tasks or reminders directly into communication channels so that messages become more actionable. For example, after receiving an important message, the system could automatically create or suggest a follow-up reminder.</p><p>Another important point was that the primary communication channel should also reflect Donna’s own workflow and preferences, rather than only focusing on what is convenient for students.</p><p>Currently, there is also a gap in the feedback loop. Donna and Stacie may send an email, but they do not necessarily know whether students read it, understood it, ignored it, or knew what action to take. At the same time, students may find it difficult to respond or ask follow-up questions with very little effort.</p><p>This made us think more carefully about the system as <strong>two-way communication</strong>, rather than simply improving how information is delivered from the school to students.</p><p>Instead of immediately framing the solution as a chatbot, we may need to ask a broader question:</p><p><strong>How might we reduce repetitive communication for staff while making it easier for students to understand, respond to, and act on information?</strong></p><p>An AI assistant could potentially help with recurring questions, summarize or clarify information, and support faster responses. However, AI can also make mistakes, so an important part of the system would be preserving Donna’s agency and allowing her to remain in control of what the system communicates on her behalf.</p><p>This feedback also suggested that our project may not need to focus only on events. The communication problem could potentially expand to broader School of Design information, similar to how an HR support system helps people navigate recurring organizational questions.</p><h3>Seminar 1</h3><p>On Monday, we read about the relationship between humans, machines, and addiction through <em>Addiction by Design</em>. I found the reading fascinating because it connected some of the ideas we had discussed in previous weeks to a more tangible, everyday case.</p><p>According to the reading, addiction develops through repeated interactions between a subject and an object. What I found especially interesting was the idea that some objects are more likely than others to trigger or accelerate addictive behaviors.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*q3KyZbATxaswkOWAeIR8EA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OCZsUPGDKvJTVbIrq3HJtw.png" /></figure><p>After the discussion, I asked about the difference between what I want to believe and what I actually believe. One response that stayed with me was that our actions and behaviors can reveal this difference. We may describe ourselves as believing one thing, but the patterns in our behavior can sometimes tell a different story.</p><p>Also, for this week’s seminar, our assignment was also to listen to an AI-related podcast. My assigned episode was Esther Perel’s discussion on <strong>Artificial Intimacy</strong>.</p><p>I really enjoyed listening to the episode because it made me think about how quickly our relationship with AI has changed. Nowadays, people are becoming increasingly comfortable not only sharing their emotions with AI, but also developing a sense of relationship with it.</p><p>These human–AI relationships can feel genuine, comforting, and even helpful. At the same time, they may weaken our capacity to tolerate the discomfort, friction, ambiguity, and imperfection that are unavoidable in human relationships. If we become too accustomed to interactions that are always responsive, supportive, and relatively frictionless, we may become less patient with the complexity of being with other people.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*gEPMrfe9n1qgLbXi3Wsebg.jpeg" /></figure><p>This tendency also made me think about how our expectations of relationships might gradually change. If AI can provide emotional support without asking much from us in return, we may begin to see a “good relationship” as one that minimizes discomfort rather than one that requires us to work through it.</p><p>This reminded me of a conversation I had around 2022 in an AI-related book club I was part of. Someone said that she felt deeply touched by the way an AI supported her, and people laughed at the idea at the time. Looking back, it is striking how quickly this kind of interaction has become normalized.</p><p>At the time, what struck me most was the realization that conversational AI allowed me to share embarrassing moments, mistakes, or vulnerable thoughts without first building a deep relationship with another person. I did not have to worry as much about judgment, awkwardness, or the emotional investment that usually comes with opening up to someone.</p><p>However, Perel argues that these uncomfortable feelings, tensions, and ambiguities between people are actually what help us mature and learn how to live with conflict. By encountering different perspectives and contradictions, we are pushed to negotiate, reflect, and genuinely try to understand one another. In that sense, the friction we often want to avoid may actually be an important part of building meaningful human relationships.</p><h3>Studio</h3><p>For our studio project, we had a check-in presentation this week.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*952RVzjqkbBMGSQRLREGmg.png" /></figure><p>It was a great opportunity for us to <strong>refine our story and overall flow</strong>. Pragya created an excellent storyboard for the team, which helped us organize our ideas and make our story much stronger visually.</p><p>Through the presentation, we were also able to see which parts of our narrative were clear and which parts still needed further development.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*BQdjRINoyAbcFNJeztaQ3g.png" /></figure><p>As I also realized in Communication Design Studio, we spent time thinking about how to create a strong hook at the beginning of the presentation.</p><p>One of the main pieces of feedback we received was that while the bio-based textile itself was compelling, it was still unclear how it would actually be used by the end user.</p><p>That feedback hit a little close to home because, to be honest, this was also an area our team had not focused on deeply. We were more interested in developing the material concept and the speculative system around it, so we had somewhat avoided the practical question of how people would actually interact with and use the textile.</p><p>Because of that, the feedback was especially useful. It reminded us that even if the concept is visually or conceptually strong, we still need to clearly explain how it becomes meaningful and usable in people’s everyday lives.</p><p>During Design Nexus, our guests circulated around the studio and gave feedback on our projects. We received several ideas that significantly expanded the way we were thinking about our concept.</p><p><strong>Patrick Hubner,</strong> a designer and creative technologist, was particularly interested in the kinds of patterns our bio-based textile could generate in response to different environmental conditions. His feedback pushed us to think more deeply about how the material itself could visually reflect the environment in which it grows.</p><p><strong>Gaetan DeSimone </strong>also introduced us to ideas about how we might showcase the material and helped us think about bio-plastics as another possible direction for presenting or developing the concept.</p><p>I did not get a chance to speak with Julienne Devita, who works in creative higher education and design futures at Adobe, during the studio session itself. However, when I later had an opportunity to get her feedback, she suggested that the project could be pushed much further in a speculative direction. That was especially useful because it encouraged us not to limit the concept too quickly based on what seems immediately feasible.</p><p>Receiving feedback from guests with very different backgrounds made the project much richer. Each person noticed a different aspect of the idea.</p><h3><strong>This Week’s Little Fun</strong></h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*aTJYlOEzH0VZ6KxwEsmCmw.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hoie7oPw_W1aphAyGqPdAQ.jpeg" /></figure><p>On September 19, Scott Fahlman visited campus. He is known for proposing the :-) emoticon as a way to reduce misunderstandings on text-only online message boards.</p><p>He signed shirts for students and gave out T-shirts, and he was incredibly kind and friendly.</p><p>One small thing that caught my attention was his signature. It was so neat that it almost looked printed. I was curious, so I asked him about it, and he told me that he is also good at writing in other handwriting styles!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*N7kbXbV_N9w2hNgpsslhGw.jpeg" /></figure><p>The campus police were also grilling hot dogs and burgers and handing them out to students, which made the day feel even more lively.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NeKEEnD8WhW9yX3CQK47_w.jpeg" /></figure><p>I also walked around the career fair this week. It was interesting to look around and see what kinds of companies were there, though I wish there had been a few more booths related to design.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*z7onSzxq4u-BcyTGT_JmOQ.jpeg" /></figure><p><em>Will come back with Design Nexus experience..</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=98d301504683" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[[MDes] Week3 : Looking One Layer Deeper]]></title>
            <link>https://medium.com/@jamie_chung/mdes-week3-0584e2a508d8?source=rss-1765e734ba10------2</link>
            <guid isPermaLink="false">https://medium.com/p/0584e2a508d8</guid>
            <dc:creator><![CDATA[Jamie Chung]]></dc:creator>
            <pubDate>Sat, 12 Sep 2026 00:47:46 GMT</pubDate>
            <atom:updated>2026-10-09T20:52:17.602Z</atom:updated>
            <content:encoded><![CDATA[<p>Monday was Labor Day, so the week seemed to fly by.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/974/1*fa2atlpBK7EOFqmlxBiV4Q.png" /><figcaption>The view from my studio always makes me feel refreshed.</figcaption></figure><h3>Communication Design Studio</h3><p>For CD Studio, I conducted two one-hour interviews: one with a new graduate student and another with Donna. Both gave me a lot of insight into how event information is communicated and experienced from different sides.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-dS4DV6W3UV1fsZs5a8AzA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*QWQI4Yyh26-RdKgodeS7Ug.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Q551XZLVW9uZXgejk6kbvA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*cD4QjiURTriOfMQYS6xCsw.jpeg" /></figure><p>During class, our team gathered around the table and shared what we had heard from each interview. Comparing the interviews was especially useful because we could start seeing gaps between how information is communicated and how students actually perceive and interpret it.</p><h4>Takeaway</h4><ul><li><strong>Good interviews depend on more than good prepared questions.</strong> I realized how important it is to ask the right follow-up question in the moment. Sometimes the first answer only reveals the surface, and one additional “why?” or “what happened next?” can uncover a much more useful insight.</li><li><strong>Think about whether an intervention should change a mental model or work with the one that already exists.</strong> Before jumping into solutions, I want to ask whether people’s existing understanding is actually the problem. Sometimes it may make more sense to design around an existing mental model rather than trying to change it.</li><li><strong>Synthesis becomes more valuable when perspectives are compared, not just collected.</strong> Looking at Donna’s and the student’s experiences side by side helped us see mismatches that would have been difficult to notice from either interview alone.</li></ul><h3>Design Lab</h3><p>In Design Lab, Daphne invited Sanju Sunny, a Principal Design Technologist at AWS, to speak with us.</p><p>Sanju comes from a software engineering background and has moved back and forth between UX and engineering throughout his career before becoming a design technologist.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*0iybt6OrCIZDwSG1PluukA.jpeg" /></figure><p>What made his talk especially interesting was that, before he visited, we had already been discussing in class what prototypes are actually for. Sanju described prototyping as a way to move the discussion forward.</p><p>He also showed us how he incorporates AI tools into his design process and how the nature of his work has changed because of them.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3JhmMQ3hKjvKBi7Pbyhwdg.png" /></figure><p>Recently, I’ve been trying to bring AI more actively into my own design process, so seeing his actual workflow was particularly inspiring. His work broadened my perspective on what I could potentially do with AI.</p><p>One example that stayed with me was his personal AI agent. He had created an agent that could act almost as an intermediary between himself and another person, answering questions about his work, perspectives, and ideas.</p><p>Of course, that immediately raises another question: How much agency should we actually give an agent? But seeing someone experiment with that boundary rather than only talking about it was fascinating.</p><h4>Takeaways</h4><ul><li><strong>The way designers work is evolving.</strong> AI isn’t only changing the tools we use; it may change what we consider part of a designer’s practice.</li><li><strong>Push technology further.</strong> Instead of stopping at what I already know how to make, I want to see how far I can push an idea with the technology available to me.</li></ul><h3><strong>Design Studio</strong></h3><p>As we continued exploring the future through a speculative design lens, our team chose to imagine what the textile industry might look like 50 years from now.</p><p>Pragya and I (aka PJ Sandwich) brought personas and scenarios into our exploration.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Pq0EFzGTpMHZYtHymYd2LQ.png" /></figure><p>I developed three different personas and scenarios to imagine how people might experience and interact with textiles in that future.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OPSkl-ByRy0jCFpLhOjhiw.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zdUPrxBTfu11y3VrzXMpDg.jpeg" /></figure><p>At first, I approached the topic quite superficially. Since I didn’t know much about the textile industry itself, I naturally started from the part I was most familiar with: the customer experience. Two of my scenarios imagined how people might live with smart textiles in the future.</p><p>But something felt off. The scenarios were easy to imagine, and some of them felt like they could become reality much sooner than 50 years from now. I realized I was mostly imagining futuristic products rather than really questioning what the future of the textile industry could be.</p><p>So I went back and started researching how the textile industry actually works today.</p><p>That’s when I learned more about the environmental problems embedded in the system, including the massive amount of water pollution caused by textile dyeing. Since our team is focusing on <strong>Restore</strong> within the TTFFF framework, this gave me a completely different place to start. Instead of simply imagining more advanced textiles, I started thinking about how an existing problem in the industry could be transformed into an opportunity for a radically different future.</p><p>What changed wasn’t just my idea. It was where I was looking.</p><p>Initially, I thought the textile industry offered only a few interesting intervention points for speculative design. After researching it more deeply, I realized that I had been looking at the industry too flatly as a customer, which is the perspective I’m most used to working from as a designer.</p><p>Once I looked beyond the customer-facing layer and into the system behind it, I found much more space to speculate.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DXvyxbSVNxpSMCF7LIaR_w.jpeg" /></figure><h4>Takeaway</h4><p>Not every industry looks exciting at first. But that doesn’t necessarily mean there is little room for design. Sometimes it means I haven’t looked deeply enough yet.</p><p>Speculative design requires me to move beyond the most obvious user experience and question the systems, processes, and assumptions behind it. When imagining a future 50 years away, I also need to give myself more permission to be imaginative. If the future I imagine already feels easily achievable today, I probably haven’t pushed it far enough.</p><p>This also made me think again about what kind of designer I want to become — not only someone who improves what people already interact with, but someone who can find possibilities in systems that initially seem difficult to change.</p><h3>Seminar</h3><p>This week, we discussed what “interaction” actually means through <em>Things That Keep Us Busy: The Elements of Interaction</em> by Janlert and Stolterman (2017).</p><p>It was a difficult article to digest, but also a surprisingly meaningful one because it gave me a way to think about a question I had from our last class.</p><p>In the previous session, I wondered: <strong>Is a poster interactive?</strong></p><p>My instinct was to say no. But I couldn’t clearly explain why. A person sees a poster, reads it, interprets it, and might even change their behavior because of it. Something is happening between the person and the artifact. So why wouldn’t we call that interaction?</p><p>This week’s reading gave me the vocabulary to explain what I had been sensing but couldn’t articulate.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/886/1*IuFPCuyZi99ZGN295gMHJg.png" /></figure><p>The authors describe interaction as something that happens within a “window of interaction,” shaped by eight dimensions: agency, independence, receptivity, predictability, freedom of action, pace and time, attention demand, and awareness level.</p><p>Thinking about these dimensions helped me return to my poster question. A poster may influence what a person thinks or does, but it has very little agency of its own. Its behavior is not guided by an internal design toward a goal, and it cannot respond differently depending on what I do. I can act on the poster, but the poster cannot really make a move in response to me.</p><p>That distinction helped me understand why having an effect on someone is not necessarily the same as interacting with them.</p><h4>Takeaway</h4><p>We use words like <em>interaction</em> so casually that they can feel self-explanatory. But when we break them down, they are often made up of multiple dimensions and conditions.</p><p>This reading reminded me to be more careful about using broad design terms as if their meanings were obvious. Sometimes, breaking down a familiar concept gives me a much more precise way to understand and talk about what I am actually designing.</p><h3>Thesis Prep</h3><p>Today in Thesis Prep, I spent some time thinking about what I can leverage from my past experiences and, more importantly, what I am actually curious about as a designer.</p><p>I’ve had this vague feeling that I’m not quite ready for a thesis yet. I don’t have one clear topic or question that I know I want to pursue.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*iE7pg6IKDTGEz_iy82R-5Q.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jePz2gUWPRUJdBQYe0wlrA.png" /></figure><p>But looking back at my previous projects today made me realize that maybe I have more direction than I thought. Instead of trying to come up with a thesis topic from scratch, I started looking for the things I keep coming back to.</p><p>I wrote down telepresence, asynchronous togetherness, spatial experiences, traces of people, and mental health. At first, they seemed like separate interests. But I started to see a connection between them.</p><p>I’m interested in the traces people leave behind and how those traces can create a sense of connection across time, distance, or even absence. I’m also interested in how design can support people.</p><p>I don’t know yet what form this will take as a thesis, and I don’t think I need to decide that now. But I feel like I’m beginning to identify a question underneath several of my past projects: How can design create meaningful connections between people when they cannot simply be together?</p><h3>Takeaway</h3><p>Maybe preparing for a thesis isn’t only about finding a new topic. It can also be about looking backward and noticing what I have already been repeatedly drawn to.</p><p>I want to keep exploring how traces, presence, space, and technology can be used to support human connection and see where that takes me.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WO7XstgBn29MhNGbQKHTFQ.jpeg" /><figcaption>In remembrance of 9/11.</figcaption></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=0584e2a508d8" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[[MDes] Week 2 : Learning to Navigate Uncertainty]]></title>
            <link>https://medium.com/@jamie_chung/mdes-week-2-learning-to-navigate-uncertainty-94214dbbdb8a?source=rss-1765e734ba10------2</link>
            <guid isPermaLink="false">https://medium.com/p/94214dbbdb8a</guid>
            <dc:creator><![CDATA[Jamie Chung]]></dc:creator>
            <pubDate>Fri, 04 Sep 2026 18:11:47 GMT</pubDate>
            <atom:updated>2026-10-09T20:53:14.226Z</atom:updated>
            <content:encoded><![CDATA[<p>Time flew by, and somehow the weekend disappeared before I knew it.<br>Anyhow, a new week started.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_i58S8UisGRdjKeEN4bGtA.png" /></figure><h3>Communication Design Studio</h3><p>Based on what we had explored so far, we started looking into the communication system at CMU as a group.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*s0dQV5o5aC2py7kTRaImMA.jpeg" /></figure><p>But it took longer than we expected because, as we tried to put everything together, our approach became too general.</p><p>Another reason we struggled a bit was that each of us had approached the research at a different scale.</p><ul><li><strong>Chaewon:</strong> How to install an HP printer</li><li><strong>Pragya:</strong> How to communicate through email</li><li><strong>Jamie:</strong> How to use the shuttle bus</li></ul><p>The scope of our individual research was quite different, so it wasn’t easy to bring everything together at the same system level.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2wAjijCp6MJFd94siNzdZQ.png" /></figure><p>Thanks to Stacie and Jocelyn’s guidance, we realized that we needed to focus on a more specific context. So, we decided to look into the School of Design’s communication system.</p><p>Interestingly, even within our team, we represented different types of audiences:</p><ul><li><strong>Chaewon:</strong> a new student who attended the School of Design orientation</li><li><strong>Pragya:</strong> a new student who didn’t attend the orientation</li><li><strong>Jamie:</strong> kind of a returning student who didn’t attend the orientation</li></ul><p>We noticed this through a casual conversation while sharing our orientation experiences.</p><p>Interestingly, the information each of us received was also different.<br>So, we decided to look into a SoD event.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZCj1APB5vkubMLZ3ytBlqA.png" /></figure><p>This approach was more of a general communication map showing how SoD event information is delivered to students. However, it still felt too broad to us, so we decided to zoom in further. We chose the orientation onboarding process to dig deeper.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ODI7Hhg4GXJXKyVFPxUblA.png" /></figure><p>Still, it didn’t show the complex communication happening behind the event. So we added a time layer and broke down the information in more detail.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*H_AXN58Jnx6phWotOwQbag.png" /></figure><p>Finally, we found a good scale for the communication system and could clearly see the relationships within it.</p><h4>Here’s what we found</h4><ul><li><strong>Missing information creates knowledge gaps</strong><br> e.g. <em>Who is Ray? What does he look like?</em></li><li><strong>Information is scattered across different places and messages</strong><br> Students have to explore multiple sources, and the level of detail is inconsistent.</li><li><strong>Email is the key communication channel</strong><br> This creates redundancy and makes important information easy to miss.</li><li><strong>The amount of information increases abruptly over time</strong><br> Students can quickly become overwhelmed, especially around orientation.</li><li><strong>Senders don’t see the full communication load students receive</strong><br> Each sender knows what they send, but not necessarily how many other messages students are receiving.</li></ul><p>But we were concerned that our frame might be too narrow. Stacie told us that we could expand this model and apply it to event communication more broadly.</p><h3>Takeaway</h3><p><em>- To find opportunities in communication, we need to narrow down the scope enough to see the details and relationships.</em></p><p>Based on these observations, we wanted to break them down from a systems thinking perspective.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hWKcwJrgiQSyf4UKtAt8nQ.png" /><figcaption>Reference : <a href="https://ecochallenge.org/iceberg-model/">https://ecochallenge.org/iceberg-model/</a></figcaption></figure><p>Individual events were easy to recognize, but when we started looking for patterns across them, we had to think more carefully.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4O3TxahotcOA9KlxvJYyGQ.png" /></figure><p>When we compared different events, we found several inconsistencies. For example, there were gaps in when students received information, how much information they received, and how easily they could access it. We also found differences in the importance of the information being communicated.</p><p>So we identified this as one of the key patterns. We also found that most communication happens through email, and in most cases, the information comes from a single sender.</p><p>For structure and mental models, the structure was more difficult to fill in because we didn’t really know the school system and the processes happening behind the scenes. So we decided to fill in the mental model first.</p><h3>Seminar 1</h3><p>We had to read part of Schön’s writing. It took me quite a while to understand it while I was reading, but after finishing it, I found it really rewarding.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Y9YU8PA-Ntnc4y2BjSp3eg.jpeg" /></figure><p>Schön was a professor at MIT, and he argued that professional education should be redesigned to combine the teaching of applied science with coaching in the artistry of reflection-in-action.</p><p>Through this reading, I started to understand more fundamentally how studio-based learning works in an academic setting.</p><p>I come from a non-design background, so when I was doing my MA last year, I didn’t really know how to approach a studio class or how to adapt to this way of learning.</p><p>But throughout the program, studio helped me develop not only self-discipline, but also the ability to navigate ambiguity and uncertainty. These were some of the biggest lessons I took away from the program.</p><p>Schön’s writing gave me concrete words for thoughts that had remained vague in my mind.</p><p>We also read Herbert Simon’s “The Science of Design: Creating the Artificial.”</p><p>To be honest, this reading was quite difficult, and I wasn’t fully confident that I understood everything correctly.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RstUwkv4a6yIYSFfUO64yQ.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*AeT2NeMzB2knqD2iZdgFog.jpeg" /></figure><p>But through the activity, I started to understand the differences between the two authors more clearly.</p><p>If someone asked me which approach I prefer, I would say that, if there were no time constraints, I would naturally see Schön’s approach as an ideal environment for designers. I also think this kind of reflective process is necessary for individual growth because it allows designers to continuously question and reframe their own ways of thinking.</p><p>Still, I think Simon’s argument is meaningful because it helps us understand constraints and communicate design in a language that people from other fields can understand. In a collaborative environment with limited time and relatively established mental models, his approach can be highly efficient. It gives designers a practical way to navigate existing structures without having to explore every possibility.</p><p>Simon’s approach can help us work efficiently within existing constraints. But if we want to make a big leap beyond existing ways of thinking, I think we also need Schön’s approach. Reflection and reframing give us a chance not only to find a better solution within an existing frame, but also to question and reshape the frame itself.</p><h3>Takeaway</h3><ul><li>Motivation: What bothers me? What motivates me?<strong><br></strong>I wonder where frameworks like the Double Diamond belong between Simon and Schön.</li><li>Simon’s approach feels more like working efficiently within an existing frame: understanding constraints, exploring possible alternatives, and finding a satisfactory solution.</li><li>If efficiency is the goal, Simon’s approach makes sense. But if I want to make a big leap beyond existing ways of thinking, I think Schön’s approach becomes more important.</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=94214dbbdb8a" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>