🌿 Leahhha

Politics of Code NotePad

Apr 12, 202514 min read

Politics of Code 1.27 - W01

Politics ←> Economics 历史: People originally lives in small groups. Survival → Politics (ethics)

Politics of Code - W01 02 03

Politics of Code - W04 - Social Media

Historical, geographical influence: institutions fear of revolution, and chaotic society.

(Compare France and United State: France as long as you obey the law, there is free speech. United State not concern that much. )

Traditional Media: printed matter 影刷物; address to mass audience (be able to reach more people)

Newspaper, Poster (on construction site), Pamphlet, Books, Radio: reach people easily; share words

Pamphlet definition: Pamphlets may consist of a single sheet of paper that is printed on both sides and folded in half, in thirds, or in fourths, called a leaflet or it may consist of a few pages that are folded in half and saddle stapled at the crease to make a simple book.

smart phone: spread images seamlessly (unpresidented); the emergence of social media drive this trend.

S-curve of technology engagement: probably never reach 100%. Start quickly, then slowly, because mosts people have it.

Reading and writing is a technology.

Company success: publish new product to replace the old one; economical success, have sticky and loyal consumer.

vast technology that change the culture!

How technologies shape us and how our invention of technologies shape our culture?

Phone company: IOS (only apple), Android (Huawei, LG…).

The mechanics for using social media across culture is the same

!! culture is inherent in the technology !!

visually imagery post similar things - images = visual language

Socialization and globalization is supported by technology

Technology of social media accelerate the globalization, cross cultural communication.

Design: giving shape to things

Polarization: either black or right; hard to maintain no middle position; No feature of advertising.

Social media pervasive. How do you think it is going to affect

Why can’t people being neutral in daily life, but more polarized argument in social media platforms.

connect people, stay connect with people in a casual way.

Server support the content, so that you can view; privacy analyzer.

Politics of Code - W05 - Surveillance and Control(Ethic of Technology)

Exam - 起源于 do you fit? in military - measurement > classification Normal Distribution 正态分布 : bell curve Eugenics is a set of beliefs and practices that aim to improve the genetic quality of a human population. 一套旨在提高人类遗传质量的信仰和实践 the originator of eugenics: Francis Galton

Progression means moving forward, while regression means moving backward. Example: Correlation of Weight and Height is arbitrary Different country uses different chart and reference standard. Reference standard is judgement. = I judge according to these rules Every opinion is a judgement! Everything is redefined all the time. Labeling people is a judgement. !! There is no disability, only diversity. !! Advertising on website > making assumptions > guess you like (people who buy this may probably buy that) Analysis( cut things into small pieces and analyze them), analytic approach, coming up analytical description of things Example: understanding of TIME: put people into the same schedule and on the same page. functional, efficient culture: keep time Different approach to things. In the society, things are culturally specific, there is no right or wrong way of things happening.

gyroscope: a device used for measuring or maintaining orientation and angular velocity

手机里有类似的检测手机状态的App(像gyroscope), 例如当人们把手机拿出来, 图表上会显示独特的图标状态, 所以 app 可以检测出使用者的状态

Fingerprint: only I use my laptop in this way. little bits of information stored in your browser; it could only plausibly match one person or a very small group of people.

Under oppressive force by 设身处地地思考 Under oppression - being giving the perspective of 减少 消除 oppression seeing the world of a perspective change

People being the content of technology

Politics of Code - W06 - Big Data

Current technology: APP, AI

Epistemology is the study of knowledge, including how we know things, what we know, and the limits of knowledge.

How the trained data is not just outputting what is inputting and knowing that the system understand? How do we know if the system is not a parrot?

  1. knowledge: consciousness, the way we know about the world
  2. Information: giving your mind a form of something
  3. Communication: the transmission of ideas or data from one entity to another.

How human have the structure to obtain knowledge?

Human have structure of knowledge is like a rectangle, therefore lots of data is being stored and present this way.

Facts: 如何获得Experiments > Observation > Measurements

e.g.: Poem is just reshuffle of letter; letters frequency is different.

Two ways of looking at data:

  1. Hypothesis-Driven (Deductive Approach) Starts with a theory or hypothesis and tests it using data.
  2. Pattern Recognition (Exploratory Data Analysis - EDA, Inductive Approach) Starts with data and looks for patterns, trends, or relationships without a predefined hypothesis.

Data visualization: Deriving structure from data (hypothesis bring into the data already) (e.g: getting people’s location from photos and form a map of a city)

Json = key value structure

Hyperscale refers to the ability of a system, architecture, or infrastructure to scale efficiently as demand increases. It’s commonly used in computing, data centers, and cloud services but can also apply to broader concepts like business growth or network expansion.

Hyperscale data centers are massive facilities designed to efficiently support high-volume computing, storage, and networking demands. They power cloud services, AI, big data analytics, and large-scale applications.

Each major topic subject: have their own structure of thinking. (how a new subject 出现)

Extract, Transform, Load (ETL) is a process used in data integration, analytics, and warehousing to move and refine data from various sources into a centralized system.

Parsing is the process of analyzing and converting data, code, or text into a structured format that can be understood by a computer. It’s widely used in programming, data processing, and natural language understanding.

Statistics - the company doesn’t know people’s preferences, but people who click on this link, have a higher possibility of clicking the other link.

e.g.: Google is thinking through the inputed data searches, by analyzing the corelation between things

From ethniticy of part: people who search for racists words, and also get recommend to click on similar racist words. Therefore, does the computer engineer 促使 more racist people appear.

Decision people made may lead to sth good or bad!

Reading Note 07 Global and Local

Digitally disadvantages: 1. Access integrity 2. surveillance

  1. Access integrity: wrong tools connections make things worse.

Why are some languages digitally thriving?

Series of advance decision that lead us today, English is the dominant language.

History: Keyboard designed for English and it is now used worldwide. Subtly reinforcing the dominancy of English in the digital world.

In the base level, english is baked in the system.

At the edge of loosing language diversity, due to the lack of digital support!

  1. People have the ability to create the visual representation, their own font of their language to join in the visual representation!

  2. Create custom keyboard layout: building digital infrastructure from the ground up!

Aim: Create a seamless experience for diverse languages.

Solution of solving the digital disadvantages in languages: changing the mind set and approach. We want the other language to thrive digitally.

What is the ultimate goal of solving the digital disadvantages in languages: create a digital ecosystem that values diversity.

Notice the importance: linguistic diversity is just the important as biodiversity.

Based on this weeks reading, I notice the fact that the dominancy of english in digital realm. The ultimate goal is to create a digital ecosystem that values diversity. Historically, Keyboard designed for English and it is now used worldwide. Subtly reinforcing the dominancy of English in the digital world. In the base level, english is baked in the system. I highly agree on linguistic diversity is just the important as biodiversity. Recalling from previous in-class discussion, the digital world magnify the embedded society problems. It is also great for noticing the problem and then actively put out rules and restrictions to prevent it being bad.

The reading above showcase the fact that nowadays people are also having trust-crisis among the authenticity of the information(news) we get around the world.

Shutting down the internet is also a hot topic in China, as it blocks major social media and search engine, such as instagram and google. Those software failed to get connection when users location is in China. Government seems to use this as an effective solution when chaotic things happened that might threaten government’s reliability.

Politics of Code - W04 - Global and Local

Racist: Culture: Ones culture cannot be changed. Not our choices. Believe things can be changed or not. People are born unformed, and can be educated to both extreme.

bites = 8 bit

Leet:  a way of writing that uses numbers and symbols to replace letters Leah: 134|-| Jiaxin Yu: .]|4 }{|// ’/ [_]

Artificial Language: a language consciously devised for a specific purpose, rather than evolving naturally

Art Nouveau: meaning “new art” in French, is an international style of art, architecture, and applied art, especially the decorative arts, that flourished from the late 1880s to the First World War, characterized by organic, flowing lines and nature-inspired motifs.

Code Point → Glyph(abstractions of numbers, characters represented by numbers) → Font

Politics of Code - Labor and Exploitation

Private space has become accessible to everyone online. (It is pricy todo that)

The cost of using social media: People’s privacy being exploited and exposed to unknown people.

Exploitation: 三角贸易triangular trade - ship is never empty, each sail is lucrative to someone

早期England: Mills 工厂, 制造厂 monopoly 了 market, 使货价飙升,使大家不满, 放火烧了Mills

Origination of Romanticism: Romand language: write about love

Orderly, Things can be in control: Nature and Culture: Nature, represents the raw, untouched beauty of the world around us; Culture, inside inertia(dark side in your mind).

Before 19th century, there is no creativity. (only god is creative)

Capitalism cultural production work: there is no good way out.

AI companies another type of publishers

The emergence of recoding music: Publish music without paying a publisher. Encourage people to be more creative.

Political technology: AI

Create things: Give shape to things.

Why is the high technology targeted creative field?

  1. Is it going to be a larger market, and larger profit from the field.
  2. Economic side, creative field provides a sandbox for everything.

How AI company is profiting from?

Contradictions happens both sides.

ReadingNotes:

  1. Newsletter (blog), 1 November 2022. https://c0nverse.substack.com/p/ai-artists-and-the-future-of-images . historical ladder images: major shifts:
  2. traditional images (visual symbols): pure sensorial experience, without the layer of interpretation added on to it
  3. linear text (abstract systems of writings, like newsletters): actively engaging the world and shaping it. Newsletters: Human imagination become to surf. (Time is being set as linear)
  4. technical images(photography): mythical consciousness become dominent.

AI affect our way of process information and how we make connections. Traditional media: tangible objects. (touch, smell… with it) Technical images: They look complete, but they can always be deconstruct (Illusory wholeness: Illusions created by tiny particles) Programmed particles into informative situations.

AI are made of data. Through prompt, reassemble them into new image. AI give everyone the ability to be creative to be an artists.

How does this challenges artists? Artists dedicated years to mastering their crafts. (tortured genius, starving artist) Questions early artist seem their role. Simple text to complex visuals. Traditional artists goal:Artists want to connect somethings larger to themselves. Decoding the myth giving everyone to discover their creative potential. Decentralizing the centralization of contents, like newsletters to youtube tiktok videos. We are communicating visually in an interactive way. Technology shaping the way people communicate.

What does it means being creative in this whole new world? Ways of our experience is evolving. All arts involves in a degree of curation. Artists are curator. Selection of curating. Traditional media: Transform of mental effort into physical form. Prompt engineering: prompt through the words you choose. Curating: Making choices, guiding the AI, shaping the final output. Narrowing the possibilities through the words you choose. Lower the barrier of being creative, being visually impressive. Higher the bar of depth of expressing something meaningful. You have the tool doesn’t mean you automatically have the vision. Required: intention, thought and unique perspective to create something genuinely compelling Working with AI: A collaborative process. AI not replacing artist but changing how artists work. Integrating AI as a powerful tool. AI accelerate how art is produced and shared. AI makes art endlessly modifiable. AI = Fun (Immediate satisfaction, playful) Distopian world: we get stuck and only reusing old ideas. Constantly revisiting the familiar, instead of forging new creative territory. Use the AI to bridge the gap between people foster communication , create new forms of connection. Using AI in a conscious and thoughtful way.

Politics of Code - AI

Large Language Model (LLM): make predictions (no human soul: LLMs has its own way)

Being creative: pushing the boundary of existing rules

AI is controlled by people who trained them. Design decision

  1. Neural network: like frog eye through selections to detect what is the item.

  2. “Connection theory,” or connectionism, is a learning theory, developed by Edward Thorndike, that posits learning occurs through forming associations between stimuli and responses, with these associations strengthening or weakening based on frequency and nature of pairings.

  3. Backpropagation is the process of adjusting the weights of a neural network by analyzing the error rate from the previous iteration.

  4. A regression model: provides a function that describes the relationship between one or more independent variables and a response, dependent, or target variable.

    1. 分析两个毫不相关的数组之间的关系. 所以有些relation is being set up in a bad way. (Assumptions) → making the next prediction.
    2. models 作用: what is most likely to be next. → make predictions
    3. Every culture have their own way (that’s why there is culture shock) → animals have soul but no mind. Every species have their own way of communication (probably human cannot understand)
    4. Input data machine become conscious 机器有了人类般的感知, become ?alive? with trainers? Is there a soul in AI? Input data: texts are syntax(grammatical rules with language: have structure for better understanding) and brain don’t work syntactically.
    5. Being creative: pushing the boundary of existing rules
    6. Training data: books from years ago, probably comes from public domain.
    7. There is no perfect translation: Eg. Translation is its own culture and has its own value. It is only passing through the superficial idea, not the soul.

  5. A perceptron is a foundational model in machine learning, serving as a simple artificial neuron capable of binary classification, meaning it can decide if an input belongs to one class or another. 感知器是机器学习的基础模型，可作为二元分类的简单人工神经元，这意味着它可以决定输入是属于某个类别还是另一个类别。Human comes in to tell the model what is good or bad, it is necessary for the diversity of people who train the model eg. Delvish(Nigerian: cheap, who trained the data) comes in.

  6. How to think texts as vectors

  7. Chat GP T: transformer vector.

  8. Logic: Culture shapes data; data shapes culture.

  9. Trained the model with data 数组: model gives Prediction: eg. what is likely to be the next word.

  10. Neural network assigning probability. The model trying to understand the logic

  11. Beautiful Soup is a library that makes it easy to scrape information from web pages.

  12. Training data of thousand of people walking: it is like a unique way people walk, to recognize people.

Graph View

Backlinks

  • No backlinks found