Graduate Studio: Technology B [Class Notebook]
(DES-720B-04)

Week 2

How does a system turn the world into something it can read and act on? What is gained or lost when meaning becomes symbols, categories, and rules?

This week, we will trace systems of representation from divination, logic, and notation to digital computation and symbolic AI. Through a hands-on reconstruction, we will test how one rule or relationship behaves across different inputs and examine the assumptions and limits built into every representation.

Class Notebook

Andrea Cheng

-

Binjia Li

This week, I learned that reconstructing a system means making its assumptions explicit. My model represents skin, hair, and eyes as three color values, then turns their lightness difference into a category using designer-chosen thresholds. Changing only one input produced three different classifications. I also discovered that the middle lightness value does not affect the result, even though it appears as an input. Numerical precision does not necessarily mean the system understands a person. My next question is: how stable would its classification remain if the same person were photographed under different lighting conditions?

Jiayi Li

What kinds of information must be made explicit before a system can act intelligently?

Li Zhang

There are many different basic algorithms about wayfinding. How can the real map software be optimized? Apart from efficiency, what other standards exist?

Lulah Balde

This week I learned more about technical system development as I was understanding the important of considering details you wouldn't otherwise think about when building an entirely new prototype built on a new system of choice. Questions that I'm still exploring are how to make my prototype more distinct to my design style and unique to my initial project concept.

Lulah Balde

refer to previous submission

Mingfu Yang

If a system only needs to recognize a few gaming keywords and provide context-aware responses, can it be perceived as a "teammate who understands the game"?

Rebecca Davidson

This week, I learned how to experiment with my simulation to learn more about my system.

Samridhi Jain

How does fear influence people’s decisions? Fear of losing influenced my final situation. Does this happen in daily life? Are there moments where fear outweighs the odds?

Seulgi Choei

This week, I learned that recognition is not the same as understanding. I also started questioning how much a machine’s answer depends on the rules and thresholds we define for it.

Wendy Wang

I learnt how this waggle dance system exactly works with each other.

Xiran Wang

I learned something really basic about coding this week from the workshop.

Zoe Liu

I'm not against other people believing gods really exist in the world, but asking a god this way feels like it comes down to luck. What if someone actually had the skill to guarantee they'd get the answer they wanted from the deity every single time? Of course, when you're out of options, using this method to find a direction a choice without other choice.

Week 2: Systems of Representation

An installation of books, wall panels, and hanging scrolls covered in thousands of invented characters that look readable but carry no meaning. Xu Bing, [Book from the Sky](https://blantonmuseum.org/exhibition/xu-bing-book-from-the-sky/), ca. 1987–91. Installation view at the Blanton Museum of Art, 2016. Courtesy of Xu Bing Studio.
An installation of books, wall panels, and hanging scrolls covered in thousands of invented characters that look readable but carry no meaning. Xu Bing, Book from the Sky, ca. 1987–91. Installation view at the Blanton Museum of Art, 2016. Courtesy of Xu Bing Studio.

Lectures

Readings

Resources

Assignment 2

This assignment is part 2 of the project Epistemological Research of Technology

Submit Assignment

For this second assignment, return to the subject you researched last week and choose one specific part of its underlying logic to reconstruct. Rather than trying to reproduce the entire system, focus on one rule, decision, relationship, sequence, classification, transformation, or feedback process that seems important to how the system operates. The goal is to move from describing what the system does to understanding what has to be represented, defined, and made operational in order for that behavior to occur.

Spend the week building a small working model of this mechanism. Your reconstruction might take the form of code, cards, diagrams, objects, a spreadsheet, a decision tree, a conversational script, written instructions, a performed procedure, or another simple prototype. Keep the system small enough that you can clearly explain how one input leads to another state, decision, or output. As you work, pay attention to the choices that have to be made explicit: What information does the system need? What categories or symbols does it use? What rules determine what happens next? What possibilities are excluded because they cannot be represented within the system?

Test your reconstruction using at least three different inputs, situations, or conditions, and document what happens each time. From these experiments, identify at least five observations about how the system behaves and one question that you would like to investigate further. Pay particular attention to moments when the system behaves unexpectedly, produces an ambiguous result, fails to account for something, or appears more convincing than you expected. What do these moments reveal about the assumptions built into its representation of intelligence?

Bring your working reconstruction, documentation of your tests, observations, and question to class next week for a short informal presentation. This experiment does not need to become your final Project 1 yet; use it as a way to discover what part of the system you may want to investigate, challenge, or develop further.