Hands-on Program Design and Operations
From Digital Literacy to an AI-ready World — for Paju Local Education Resource Activists
Foreword
On 7 October 2026, we meet local education resource activists at Paju Citizen Hall. This is Session 3, following Sessions 1 and 2. Today is less about theory than about what we can use in schools and villages tomorrow. Whether you live on a phone or a laptop, both are fine. Raise your hand whenever you have a question.
This piece expands the day’s talk in order. It is not a copy of the slides; it is written to be read like a book. I have left a little of the room’s voice in the lines, so you may read aloud or return to it later.
There is one question for today: Is the person who runs experiences well the same person who uses tools well? I hope the answer opens naturally by the end of the text.
A short map of the years we have walked: before 2000 we learned to fix and reuse in front of machines; through the 2000s into the 2010s we moved from desk to hand, from laptop to smartphone; through the 2010s into the 2020s digital became ordinary; from 2021 toward 2030 we are in the age of AI, records, and collaboration. This is not a history lesson. We only ask what remains for activists in each era’s cases.
Chapter 1. Tool Learning — AI Starts from Devices, Not Magic
What we casually call AI begins with digitally grounded devices. Literacy is less about “how to use that app” than about imagining and handling the device itself.
Tool learning unfolds in six moves: imagination (what could be made), grasp (what parts and rules exist), design (order and structure), assembly (connect parts until they work), disassembly (why it is built this way), and reassembly (repair, reuse, re-purpose).
In experiential education these translate into a simple aim: not “how to use the machine we bought,” but “become someone who can take apart and put back together.” One more app fades; the power to imagine, grasp, assemble, and reassemble lasts.
Devices expand the person, then quietly put them inside a rule
New devices keep arriving as times change. If we read the shift from “machines for people” to “people fitting the machine’s logic,” activists can see how to treat tools.
Human-centered devices follow the body—hands, eyes, voice. Pencil, typewriter, telephone, camera. People run the machines. The main question is “What shall I do with this tool?”
Computer-centered devices follow machine rules—process, data, computation. PC operating systems, algorithmic feeds, AI agents. People slot into the flow and respond. The question becomes “What is this device asking of me?”
The flow looks like this. First, tools that extend the body: writing, sound, sight. Then tools that train us to the rules on the screen: keyboard, icons, app logic. Now tools that arrange life around us: recommendations, alerts, automation touch schedule and information first.
When activists borrow a tool, two questions help. Which of my body and time does this expand? Which of my behaviors is it trying to rule and automate? Experiential design grows the first and names the second so it can be handed back to the learner.
Teachers—devices enlarge us, then one day they place us inside rules. Spotting that day is today’s strength.
Chapter 2. Design from Joy and Happiness
Experiential programs do not start with explanation; they start with what someone likes. When opening conversation with children, teens, and students at school, “Let’s start from what you like” opens the door faster than “You must learn this.”
The old explanatory intro lists features and steps. “Do it like this,” and the child only follows. Feedback is a finished product. A joy-based intro starts from likes, play, and interest. “Shall we try something we have not done together?” Participation, faces, and laughter become feedback; the process becomes memory.
Three design principles. First, open from what they like—games, music, drawing, sports, photos. Second, do something untried together—side-by-side experience over lecture, one more try over presentation. Third, treat happiness as achievement—expressions, stories, and “I want to do this again” over the right answer.
Prompts you can use on site: “What are you into these days—shall we open with that?” “This is new for me too—okay if we get a little lost together?” “Thirty seconds to touch it before I explain?”
Let them define, explain, and name happiness
Use a game, device, or AI the student already loves, and walk three steps. The activist listens and follows.
Define: “X is Y to me” in one or two sentences. Explain: as if to a partner or someone you just met. Happiness: “Where is the most fun? Tell me that moment.”
Any game, everyday device, or AI app works. Output: one definition, one minute of explanation, one happy point. Then one bridge: “If we move that happy point into today’s experience, what would it be?”
Education succeeds not with “What did you learn today?” but with “I want to do this again.”
Chapter 3. Happiness Makes Repetition; Repetition Makes Literacy
When joy appears across products and devices, children echo and repeat. But behavior grows only when fun is joined by “What I did made a result.”
Happiness begins from what they like. Shared action creates echo; pair and group experience is the base. Repetition happens when there is one more try and a natural next round. Confirming results means keeping “that was fun, and then this happened” in photos, sound, objects, or words. When they choose the next mission themselves, motivation is complete.
Fun is the first spark. Result-checking is fuel. To carry joy into the next action, show “I did this, and then X happened” with eyes and words. Repetition is growth. Motivated behavior expands on its own—the point where the activist no longer has to force practice.
On site: open with a liked game, device, or AI; make once; check the result; let them choose the next difficulty. “What changed from yesterday? What shall we try next?” Output: a motivation card with one fun point, one result, one next goal.
Repetition is not homework. It grows when fun and result connect.
Repetition grows literacy
Repetition born of happiness is not mere habit. Reusing the same tool builds fluency and frees attention for thinking. Trying, failing, and fixing grows debugging; errors become material. Explaining and showing grow expression. Reviewing results grows judgment. Choosing the next level grows self-regulation.
Several light repeats with short explanations beat a single full session. Six short times beat one long day. The activist’s role is not to force repetition but to build a structure where repetition happens.
Literacy can be taught, but it also grows from hands and heads that repeated for fun.
Chapter 4. AI Is Designed to Manage the User’s Memory
AI is not only an answering machine. It favors document forms that people and AI can both read, and it is designed to manage memory around those documents.
Markdown is the clearest example: friendly to humans, easy for machines. An .md file is both a human record and raw material for AI memory. Documents stacked in folders become operating principles, experience, and context. AI is especially good at catching spoken language—audio, conversation, interview—and keeping it as written text. What is said in the field does not scatter; it becomes record.
Pair AI’s memory layers with human records. Rules and judgments become operating principles: “We do it this way”. Status summaries capture where we are today so work can continue. Notes catch what must not be lost. Task files accumulate materials and evidence per activity or program.
Agreements for shared documents: one context per file (humans read, AI summarizes and searches); title and date (keys for search and reuse); newest on top (flow up, history kept); share the same folder with the AI. The .md you leave becomes AI memory.
Leave today’s definitions, explanations, and pledges as .md and they become shared memory you can call tomorrow.
Memory beyond the person—class, group, family
Memory is not only personal. When class, group, and family learn to keep records together, individual memory becomes community asset.
Personal memory: my definition, explanation, pledge in notes-like memos. Class memory: lesson flow, presentations, one-line reflections in dated files and a shared index. Group memory: team decisions, roles, outputs in a folder named for the group. Family memory: shared activities, photo captions, anniversaries in a family folder and short conversation logs.
Agree who leaves what and memory stops scattering. One person need not do all. Share roles; use the same form. The learning point is the agreement to keep records together, not the technique of writing.
Good memory management is not “I remember everything” but “each of us leaves a little.”
Chapter 5. Technical Diversity — an Eye That Can See More Than One Way
The more tools we have, the more we need the power not to be trapped in one way of working. Not taking information at face value—verifying, filtering, and re-seeing from another value standard—is technical diversity.
Verification: is it true, where is the source? Cross-check the same fact in two places. Filtering: keep what is needed now, drop the rest. Other value standards: look from another stance, era, or culture—“How does this look from the other side?” Diverse tools: handle one theme through writing, photos, speech, hands, and games, not one app.
Diversity is the attitude that does not believe in a single answer. Verify and filter without harming relationships. Doubt while living together. Different values show different pictures. The same holds for devices, games, and AI.
Tool switching begins with judgment
Start by checking how you judge information. Once judgment is steady, you can move across tools.
Check judgment: “Why do I believe or doubt this?” Set criteria: source, date, emotion, common sense. Switch tools: same goal, other app, device, or method. Use flexibly: choose, do not be trapped.
With criteria, tools can change without shaking you. Freedom to switch tools comes from your judgment.
On-site prompts: “How might this sound to someone else?” “What if we think the opposite?” “Could we do this another way than our app?” Technical diversity is not using many tools; it is the eye that can see more than one way.
Chapter 6. The Easiest Practice — Interview, Question and Answer
There is one method that moves all the axes at once: interview. Ask, answer, ask again.
Interviews touch joy design because they speak about likes and experience. They grow motivation and repetition by talking again and refining. They lead to memory and documents because speech becomes text. Asking “why” checks judgment; hearing the other’s criteria practices verification and other value standards.
Three steps you can use immediately. Open questions: “When was this most fun?” “Why do you like it?” Specifics: “What happened in that moment?” “For example?” Then record and confirm: leave what you heard as .md and ask, “Is this how you understood it?”
Set direction in speech, and carry one theme to the end
The power of interview is setting direction in the spoken word and finishing one theme. Starting in speech sets direction without weight. Gathering scattered talk onto one theme deepens it. Flow refined in speech becomes writing, becomes record.
A spoken interview: set direction (“Today is about X”), bind scattered words to one line, deepen with why/how/example, then leave the record.
An interview card for activists and students. Define: “What is X to you?” Happiness: “Where are you most happy?” Result: “What happens when you do that?” Other criteria: “I see it this way—what do you think?” Record: “Of what we heard today, what is the one line to keep?”
Without heavy tools, one question-and-answer can move literacy, memory, and motivation together. Unpacking one theme in speech with students is the easiest door to all of today’s axes.
Chapter 7. Background — Open Source, the Quiet Twenty Years, and the Door Opening 2025–2026
7.1 What we call open source
Open source means technology and code published so anyone can view, modify, and share again. Opposite are closed products, proprietary software, and secret recipes you may only use. The core of open source is not free price; it is freedom to inspect, modify, and redistribute. Looking under the hood, adapting to our needs, and putting changes back into the world is the culture and right we call open source.
The same happens in documents. Open text anyone can open, wikis anyone can edit, open data and open standards anyone can reuse—“open source that is not code.” This text is in Markdown on purpose: a public document format humans read, machines parse, and AI can hold as memory.
7.2 Why open source stayed quiet in Korea for 20–30 years
From the 2000s to 2026, open and public technology in Korea was rarely celebrated or centered in education. Layers of reasons stack.
First, closed ecosystems of portals, apps, and platforms were strong. Naver, Kakao, and commercial services owned daily entry points; users learned only “how to use.” Second, schools and public institutions preferred commercial products for convenience and support—understandable under budget, security, and maintenance. The cost: “open and fix” experience stayed last. Third, certificates, contests, and curricula targeted tool use, not structure, licenses, and remixing others’ code and materials. Fourth, open source felt like “for developers” and “hackers outside the company,” distancing everyday users and activists.
This is where tool learning locks in. Using only closed tools, we cannot practice imagine, grasp, disassemble, reassemble. Public technology—open source, open documents, open data—are tools you can open. The six moves of tool learning are gestures open-source culture has practiced for decades.
7.3 Visible product, quiet foundation — Samsung and Android
Look at Samsung’s digital stack. Phones, TVs, appliances, services rest on open source such as Android, with product and experience layered on top. AOSP is public code anyone can fetch; each firm adds UI, service, brand. Still, what gets attention in Korea is the finished product and brand, not the open foundation.
Galaxy and apps are visible; AOSP is quiet. Commercial edtech is visible; open standards, open data, and open source are not. “How to use the product” is taught strongly; structure, license, and reuse are weak. Much of the world runs on open source—browser engines, Linux kernel, Python, open map data, Wikipedia markup—yet textbooks and workshops rarely show it.
Products are emphasized; open understanding and reuse stay quiet. That is why tool learning is needed. Using a well-made app matters too. But when technology anyone can open and remix is nearby, students and we move from “receiver” to “maker and sharer.”
7.4 2025–2026 — the door opens because AI was born on open source
After those quiet decades, 2025–2026 shift the ground. The reason is clear: AI itself rose on an open-source base.
AI and machine learning grew as a chain of public research, datasets, code, and papers. Knowledge from universities and labs was already unlocked in repositories, papers, and data—so generative AI was possible. Around 2025, downloadable weights and tools anyone can run spread; “AI = an expensive company product” began to move toward “AI = technology you can open, use, and fix.”
The flow: research communities shared models and training methods in papers and code; large language models, image generation, and speech recognition stacked on top. Open weights and OSS inference tools made “run it yourself” possible at universities, schools, libraries, and villages. Review, reproduction, and improvement happen in public; prompts, project settings, and result logs cycle faster. Above all, public document formats like Markdown sit on the surface of AI memory and collaboration. The .md you leave becomes AI work context.
Generative AI, open models, and public tools enter daily life, education, and work. Public technology once “only for developers” becomes the talk of users and activists. Open documents, reusable templates, and open data are recognized as material for memory and collaboration. The door is open—why this training is “now.”
AI was born on open source. Only in 2025–2026 does that quiet technology start to become everyone’s language.
7.5 On the activist’s site — how to keep public technology close
Start with one class using one public tool or material.
Write handouts in Markdown anyone can open instead of a commercial word processor. Cross-check a public map and open statistics at least once. When using images and audio, record source and license. Ask AI for a draft and keep the result, prompt, and edit reasons in .md. These plant the open-source spirit of inspect·modify·redistribute inside program operations, not only tool use.
With students, be concrete: “Let’s open this app’s code or help once.” “Who made this map data?” “How will the next class reuse what we made?” One cycle of imagine, grasp, fix, and remix lasts longer than one expensive device.
7.6 Closing — from product to ecosystem
Korea’s 20–30 years put product and brand in front and quiet public technology behind. In 2025–2026, as AI was born on open source and enters daily life, the center of gravity moves. The point is not knowing more of someone else’s models. The point is the habit of asking: Who is this tool open to? Can it be fixed? Can it be shared again?
When activists hold that question, experiential programs become entrances to ecosystems, not product manuals. Students become the next contributors, not only users. We spoke of open source at such length not to boast about technology—but because borrowing tools and making experiences should last and stay shared.
Chapter 8. What Remains from the Eras
From cases before 2000 we learned repair and reuse in front of machines: Logo for designing paths in language, word processors for drafts and reading aloud, CD-ROM encyclopedias for questions, search terms, and summaries. One line: not expensive gear, but sequence and revision make literacy.
From the early 2000s to early 2010s, desk to hand. Laptops meant desk and classroom, keyboard input, time online. After iPhone and Android: place became the field and transit; input became touch and camera; time became always connected. The activist’s role widened from writing notices to shooting the field and reporting at once.
Wikipedia was an encyclopedia you write and reading that verifies; blogs turned submission into publication. The iPhone shift moved the desk web into the palm. Carry Session 1’s discussion: name one thing in our program that needs a laptop, and one thing that works standing on site.
From the 2010s into the 2020s, the smartphone became work: KakaoTalk for notices and coordination, Instagram and short-form for outreach, smart learning and Classroom for class structure, digital citizenship and fake news for trust. As convenience grows, the three-second pause and check becomes operational baseline.
From 2021 toward 2030: AI, records, collaboration. Hybrid participation that includes those who cannot come; generative AI that drafts while humans own facts and responsibility; Markdown memory that hands records to the next person and to AI.
Chapter 9. Operations — One-Pager and Capabilities
Session 2 practice: your team designs one experience. The frame is a one-pager.
Define: one sentence on who and what change. Structure: opening, immersion, sharing, connection. Operations: time, headcount, safety, access. Record plan: photos, copy, feedback to keep. Digital: announcement channel, photo consent, three-second check.
Filling the frame surfaces three capabilities. Question and design—what to make people experience. On-site operations—safety, inclusion, time, care. Record and hand-off—keep it so the next person can take it.
Today’s pledge in three lines: When I build an experience, what do I start from? When I keep records, what form do I use? Filling the blanks is the pledge.
For what comes next: within 24 hours keep three photos and three lines of reflection. Make one team folder and start with an index file. At the next meeting, review each other’s materials and add cases.
Closing Words
We borrow tools; we make experience. From machines before 2000 to AI in 2025–2026, tools keep changing and will change again. Whether an experience succeeds is still decided by how we meet people.
From laptop to mobile to AI, the activist’s work remains: ask questions, run the field, keep records. When records pass to the next participant, the next activist, and to AI, they become operational know-how.
May the skills learned today help one more person join in school and village tomorrow.
Appendix. What to Open Again on Site
Lines to emphasize
AI starts from devices, not magic—imagine, grasp, assemble, reassemble. Devices expand people, then put them inside rules; spotting that day is today’s strength. Untried together over explanation—joy is design. AI is designed to manage memory—the .md you leave is tomorrow’s context. Technical diversity is an eye that can see more than one way. The easiest practice is interview—questions and answers move every axis at once. Not expensive gear, but sequence and revision make literacy. As convenience grows, pause-and-check three seconds is operational baseline. AI may draft; facts and responsibility are human. We borrow tools; we make experience.
Case material folders
Era cases live in project folders. Pre-2000 in 2000/, laptop-to-mobile transition in 2010/, smartphone daily work in 2020/, AI and records in 2030/. In the session, open Logo, word processor, and CD-ROM encyclopedias from pre-2000; iPhone shift, Wikipedia, and blogs from the transition; KakaoTalk and digital citizenship from the 2010s; generative AI and Markdown memory from recent years first.