en
Feedback
Continuous Learning_Startup & Investment

Continuous Learning_Startup & Investment

Open in Telegram

We journey together through the captivating realms of entrepreneurship, investment, life, and technology. This is my chronicle of exploration, where I capture and share the lessons that shape our world. Join us and let's never stop learning!

Show more
2 296
Subscribers
-124 hours
-137 days
-4130 days
Posts Archive
"Selling work opens up markets that weren't attractive for software businesses" "It’s not about selling "virtual employees". It's about unbundling the specific tasks or outcomes that employees have traditionally performed, and selling those." https://www.sarahtavel.com/p/a-few-sell-work-not-software-updated#%C2%A7its-not-about-selling-virtual-employees-its-about-unbundling-the-specific-tasks-or-outcomes-that-employees-have-traditionally-performed-and-selling-those

It feels good in the moment for hotheads to call for retaliation but we have to think through the second-order consequences. The U.S. only has about 4,000 Tomahawk missiles in its inventory. We can do about a week of Shock & Awe. Then what? Are we going to send in ground troops? Iran’s territory is vast, its forces are decentralized, and enough of its infrastructure is underground by now. It will also be supplied by China and Russia. In retaliation, our highly exposed bases in Syria and Iraq (which Neocons refused to remove, leading to this tragedy) will likely come under heavy fire; in fact our troops there are sitting ducks. Our carrier groups could even be targets. America will take meaningful casualties. Meanwhile, the Strait of Hormuz could be closed, compounding the situation in the Red Sea, causing a global oil shock. Assuming Iran wants to continue the war, we will have no exit strategy. Regime change will fail. We will be stuck in a new forever war. In fact, there’s a decent chance that we will get run out of large parts of the Middle East. https://x.com/davidsacks/status/1751707656262455559?s=46&t=h5Byg6Wosg8MJb4pbPSDow

GUI ->? μŒμ„±κΈ°λ°˜μ˜ λŒ€ν™”ν•˜λŠ” λ“―ν•œ UI: ν…μŠ€νŠΈ 쀑심이 μ•„λ‹Œ μŒμ„±κ³Ό μ‹œκ°μ΄ 합쳐진 μƒˆλ‘œμš΄ UI. μΉ νŒμ„ 놓고 ν˜Ήμ€ μ—¬λŸ¬ 이미지, λΉ„λ””μ˜€λ₯Ό 보며 μ‚¬λžŒκ³Ό λŒ€ν™”ν•˜λŠ” λ“―ν•œ μƒˆλ‘œμš΄ μΈν„°νŽ˜μ΄μŠ€. GUIμ‹œμ ˆμ΄ λŒ€ν˜•λ§ˆνŠΈ λ§€λŒ€μ— μžˆλŠ” λ‹€μ–‘ν•œ μ• ν”Œλ¦¬μΌ€μ΄μ…˜λ“€μ„ 비ꡐ해가며 κ΅¬λ§€ν–ˆλ‹€λ©΄, AI기반 μ œν’ˆλ“€μ΄ λ§Žμ•„μ§€λ©΄ 백화점에 κ°€μ„œ μƒλ‹΄λ°›μœΌλ©΄μ„œ μ œν’ˆμ„ κ΅¬λ§€ν•˜λŠ” κ²ƒμ²˜λŸΌ μ• ν”Œλ¦¬μΌ€μ΄μ…˜ μ‚¬μš©μž κ²½ν—˜μ΄ λ³€κ²½λ˜μ§€ μ•Šμ„κΉŒ. 일방적으둜 μ‚¬μš©ν•˜κΈ° μœ„ν•œ μš©λ„μ—μ„œ λŒ€ν™”κ°€ κ°€λŠ₯ν•œ(μ‚¬λžŒμ΄λΌκ³  λŠλ‚„λ§ŒνΌμ˜ λ§₯락을 μ΄ν•΄ν•˜κ³  μ•½κ°„μ˜ μΉœμ ˆν•¨μ΄ κ°€λ―Έλœ) μ„œλΉ„μŠ€λ“€μ΄ 더 λ‚˜μ˜€μ§€ μ•Šμ„κΉŒ.

Generative AI created a renewed interest in text-based interfaces. For the first time in history, computers can truly "understand" natural language and respond to open-ended questions and poorly structured commands. For many, this was a callback to the UNIX philosophy as best articulated by @tszzl in his famous "Text is the Universal Interface" blog post that came out just a few months before ChatGPT. While tools like DALLE and Midjourney are entirely driven by text, many apps like RunwayML augmented traditional UIs with prompt-based generative AI. But actually, looking a little closer, you can start to see that Midjourney is growing its own JIT UI for controls inside the chat window. Some more recent examples of JIT UIs: - @perplexity_ai copilot where the AI can generate forms on the fly to get more information from the user to tune the search. - ChatGPT Code Interpreter and products like @JuliusAI_ that can show charts or, in some cases, entire UI. In computing, nothing ever is a silver bullet. Every new system creates new problems to solve and this one is no different. As we're all finding out, Chatbots present novel UX challenges, which were best articulated by @Wattenberger in "Why Chatbots Are Not the Future". And JIT UI introduces a new set of challenges, such as how to think about state and persistence. It's unclear that JIT UIs holds the answer, but it's a thread worth exploring. It's also unclear if this is the kind of thing that startups can innovate on, as it's predicated on having a platform, and UI innovations get copied super quickly. However, clearly, it's time to try and reimagine UI and solve some issues with GUIs that have plagued computing for decades! https://x.com/amasad/status/1749993684580593991?s=46&t=h5Byg6Wosg8MJb4pbPSDow

1. 백쒅원 씨가 μ°½μ—…ν•œμ§€ 무렀 30λ…„ λ§Œμ— 상μž₯(IPO)을 μ€€λΉ„ 쀑이라고 ν•œλ‹€. 2. μ’€ 더 μ •ν™•ν•˜κ²Œ λ§ν•˜λ©΄, 백쒅원 씨가 μ°½μ—…ν•œ 더본코리아가 과거에도 상μž₯을 μ‹œλ„ν•œ λ°” μžˆμœΌλ‚˜, μ½”λ‘œλ‚˜ 팬데믹 λ“±μœΌλ‘œ 인해 상μž₯이 μŠ€νƒ‘λ˜μ—ˆλ‹€κ°€, μ΄λ²ˆμ— λ‹€μ‹œ λ„μ „ν•œλ‹€κ³ . 3. 사업을 μ‹œμž‘ν•œ ν›„λΆ€ν„°, λ†λ‹΄μ²˜λŸΌ β€˜λ‹¨κΈ°μ μœΌλ‘œ 빨리 μ„±μž₯ν•˜λŠ” κ²ƒλ³΄λ‹€λŠ”, 30년을 끈질기게 μ‚΄μ•„λ‚¨μ•„μ„œ, κ·Έλ•Œμ―€ 상μž₯ν•  수 μžˆλŠ” νšŒμ‚¬λ₯Ό λ§Œλ“œλŠ” 것이 λ°”λž¨μ΄λΌλ©΄ λ°”λž¨"μ΄λΌλŠ” 말을 μ’…μ’… ν–ˆλŠ”λ°.. 4. 백쒅원 씨쑰차도 상μž₯κΉŒμ§€ κ°€λŠ” 데 30년이 κ±Έλ Έλ‹€λ‹ˆ 괜히 마음이 κ²Έν—ˆν•΄μ‘Œλ‹€. μŠ€νƒ€νŠΈμ—… 씬에 μžˆλ‹€ 보면, 마치 λΉ λ₯Έ 게 μ§„λ¦¬μ΄μž 선인 κ²ƒμ²˜λŸΌ λŠκ»΄μ§€μ§€λ§Œ, μ•ˆμ •μ μΈ 기업을 λ§Œλ“œλŠ” λ°λŠ” 생각보닀 훨씬 더 였랜 μ‹œκ°„μ΄ ν•„μš”ν•œμ§€λ„ λͺ¨λ₯Έλ‹€.