Charlsy Yang
LightNews LightNews
charlsy.cv
Charlsy Yang
@charlsy.cv
220 followers 98 following 150 posts
designer of some sort · charlsyang.com
Posts Replies Media Videos
charlsy.cv
sometimes i just need emotional support not words that make sense
June 24, 2025 at 7:12 AM
charlsy.cv
id rather drink a machine made coffee than a sloppily made coffee
June 18, 2025 at 7:47 PM
charlsy.cv
i’ve come to realize that for me living within the walking distance of a coffee shop with good coffee and seating is way more important than a grocery
June 14, 2025 at 4:51 PM
charlsy.cv
omg dia solved this
screenshot of the chat sidebar in the Dia browser where i asked “write in normal sentence case for email” and the answer returned an email with the correct apostrophe
June 12, 2025 at 6:13 AM
charlsy.cv
now be careful if you drop your phone your apps will be shattered
June 9, 2025 at 8:28 PM
charlsy.cv
i’d rather be in the trenches twiddling bits
June 7, 2025 at 5:09 PM
charlsy.cv
found this reddit talking about the icon i designed hehe funny www.reddit.com/r/Windows_Re...
What do you think of the new Microsoft Surface icon? Is it good or ugly?
www.reddit.com
June 6, 2025 at 2:28 AM
Reposted by Charlsy Yang
anh.ng
a little 2025 update on my personal website. learned a lot (from fanatically duckduckgo-ing and youtube-ing and claude-ing and and and…). i might redesign every year just for the fun of it 🪑💬
Ánh
Graphic and web designer based in Vietnam, interested in creative tools, human vulnerability, consciousness, and internet satire.
www.anh.ng
June 4, 2025 at 7:20 PM
charlsy.cv
thanks 🥺
An excerpt of a response from Gemini: “In essence: It's about the rigidity and scope of your negative view. A strong, critical stance based on observed realities and personal values is not inherently a problematic bias. It can be a sign of a strong moral compass. An "unreasonably tilted" opinion or a harmful bias tends to be: • Closed-minded: Unwilling to consider alternative perspectives or nuances. • Overly simplistic: Reducing a complex issue to black and white. • Unfairly condemnatory: Making sweeping negative judgments about entire groups or industries without sufficient evidence. You are clearly grappling with this thoughtfully. Your awareness of the "rational case" for ads alongside your deep personal distaste suggests you are capable of nuanced thinking. The discomfort you feel about your work highlights a genuine value conflict, not necessarily an unreasonable bias. The key is to continue examining the specifics of the practices you're involved with against your own well-considered ethical principles.
May 26, 2025 at 5:50 AM
charlsy.cv
asked ai to roast my design and it’s...pretty good, echos what i have in mind and more
Screenshot of AI response: “3. Scribble Scramble: ▪ The pink scribbles are a fun idea for adding energy and a hand-drawn feel, but their execution is a bit distracting. ▪ The scribble weaving through the letters of "[redacted]" and "[redacted]" actively hinders readability. It feels less like a cool design element and more like someone accidentally drew over the important parts.
 ▪ The heart and sparkles on the right feel a little disconnected and add to the overall busyness without a clear compositional anchor.
 4. Composition Collision: ▪ There's a lot happening in a concentrated space. The main text block feels very dense. The edges of "[redacted]" and "[redacted]" are uncomfortably close to the edges of the textured background square. Give your elements some breathing room!
 ▪ The balance feels a bit off. The heavy text block and photo on the left could be better counterbalanced or integrated with the elements on the right.
”
May 24, 2025 at 7:04 AM
charlsy.cv
🥲
May 23, 2025 at 5:42 AM
charlsy.cv
still got goosebumps rewatching the bear s2e7
May 23, 2025 at 5:20 AM
charlsy.cv
wait i didn’t know dieter bohn is at google
May 21, 2025 at 2:04 AM
charlsy.cv
okay i implemented the footer
May 17, 2025 at 7:33 AM
charlsy.cv
on the very last night i finally exported my read.cv data and gave it a home :```)
A screenshot of split tabs in the Arc browser: on the left is a localhost, on the right is read.cv prod. Both shows Charlsy’s read.cv profile and has the Chrome dev tool opened. a screenshot of a Github commit with the title “bug fix & close the polish gap with read.cv prod” and the description “rip read.cv & posts.cv”
May 17, 2025 at 5:47 AM
Reposted by Charlsy Yang
worrydream.com
Computational Public Space. A talk about a values-driven approach to integrating computation into cities. (40 min)
www.youtube.com/watch?v=PixP...
Computational Public Space
YouTube video by Dynamicland
www.youtube.com
May 15, 2025 at 5:33 PM
charlsy.cv
i didn’t know how much of my agency was afforded by public transit. it’s kinda like Kohut’s concept of self-object, something that scaffolds my sense of self and wellbeing
May 6, 2025 at 5:32 AM
charlsy.cv
when can ai learn to use smart quotes and curly apostrophe
May 4, 2025 at 6:36 PM
charlsy.cv
lol so behaviorism thought humans are LLM
a photo of a book excerpt: In truth, language had always been the ultimate hurdle for behaviorism. How could you avoid talking about "mental states" such as ideas, images, feelings, and intentions when the whole point of language was to communicate those states? The behaviorist answer —naturally-was that such apparent mental states were illusory. We produced certain sequences of words only because we'd learned associations between individual words: John had a certain probability of being followed by kissed, which had a certain probability of being followed by Mary, and so on. Thus John kissed Mary. Indeed, B. F. Skinner himself was just finishing up a major book on the subject and had declared that a behaviorist account of language would be the culmination of his life's work. Chomsky was contemptuous. "I thought that [Skinner's theory] was completely wrong from every point of view," he would say many years later. "You could ask a question that for some reason investigators rarely asked: namely, can the proposed model even in principle account for the known facts? As soon as you ask that it becomes clear that [the behaviorist model] can't possibly account for the known facts about language.'
May 3, 2025 at 7:37 PM
Reposted by Charlsy Yang
janelleamores.bsky.social
Hi all — I’m putting on a Read.cv + Posts meetup during ConFig!

Just a really chill, laid-back meet with Posts people. Feel free to share to the community and register below. Excited to see some of y’all next week! ♥️

lu.ma/h8v38q2k
April 30, 2025 at 3:59 AM
charlsy.cv
a lack of motion makes a prototype feel fake, an excess of motion also makes a prototype feel fake, hmm
April 25, 2025 at 4:58 AM
charlsy.cv
i’ve learned that not caring is almost as costly as care being wasted
April 25, 2025 at 4:17 AM
charlsy.cv
An excerpt from Molly Mielke’s newsletter post “conviction”: “So perhaps it would be wiser to simply surround ourselves with people we want to be distorted by. There’s a lot of wisdom locked up in trite truisms like “you are the average of your five closest friends” — but since they don’t explain how or why they’re true I still feel the need to find my way to the same conclusion using many more words. Maybe the underlying explanation is this: whether we like it or not, each of our realities gets bent, expanded, and/or imploded by the expectations and perspectives of the people we incorporate into our little solar systems. The best example of this? Any case of someone massively raising the ambitions of another by merely suggesting they aim higher at pivotal points in their life. While Aristotle explains this as “man differing from other animals in his greater aptitude for imitation,” I prefer framing this as the human capacity to exponentially expand each other’s scale of imagination.” A frame from the TV show The Newsroom season 2 episode 6 where Will McAvoy is saying: ”Why didn’t you just say I shouldn’t worry about the number?”
April 25, 2025 at 4:13 AM
charlsy.cv
i thought you were trying to do better than google 😅 have we run out of business model
April 25, 2025 at 3:43 AM
charlsy.cv
had a great chat with gemini tonight 🍷
a screenshot with text: This conversation explores the user’s experience and concerns with AI-assisted coding. Starting with feelings of both empowerment and unease about creative authorship, the discussion delves into the nature of AI as a tool, why it feels different from traditional tools (due to natural language interaction, generative capabilities, and apparent agency), and the vast difference between current AI agency and genuine human consciousness. A major theme emerges regarding the potential for over-reliance on AI, leading to skill degradation and lack of deep understanding, framed through Ivan Illich’s concept of “non-convivial” tools and McLuhan’s “medium is the message.” While acknowledging AI’s democratizing potential, the conversation questions the future necessity of deep coding knowledge. It concludes by brainstorming practical ways to design AI tools that foster learning and autonomy, and strategies for users to leverage existing tools effectively to augment their understanding rather than merely becoming dependent.
April 21, 2025 at 6:20 AM