lispybanana
Oatmeal Spice: Interesting Procedurally Generated Output
Selling My House with a Chatbot
Basics of Film Financing
Time-Locked Cryptography
Re:
on Sunday November 30, 2025 @10:57PM (#46103337)
Open-access article
Re:Tragic Algebra of Stock-Based Compensation
on Sunday November 30, 2025 @10:55PM (#46103324)
See here instead (open article): https://news.ycombinator.com/item?id=46103320
Tragic Algebra of Stock-Based Compensation
Re:Transducer: Composition, Abstraction, Performance
on Saturday November 08, 2025 @01:09AM (#45854541)
Nicely done.
Re:The Case That A.I. Is Thinking
on Tuesday November 04, 2025 @04:31AM (#45809025)
Would they have diagnosed an issue if you hadn't presented it to them?
Life solves problems itself poses or collides with. Tools solve problems only when applied.
Re:A History of Large Language Models
on Wednesday October 15, 2025 @01:00AM (#45588237)
How does NN create a concept not in its training data? (Does it explore negative idea-space?) What if a concept uses a word not yet invented? How does LLM produce that word and what cosine similarity would such a word have if it's never appeared next to others? How would we know if such a word is useful?
Re:A History of Large Language Models
on Saturday October 11, 2025 @03:08AM (#45547257)
Humans do this but this is not all they do. How do we explain humans who invent new concepts, new words, new numerical systems, new financial structures, new legal theories. These are not exactly predictions (since they don't exist in a training set) but they may be composed from such sets.
All Revenue Is Not Created Equal (2011)
Lawlessness, Small and Large (2014)
McPhee Method
How Differential Steering Works (1937) [video]
Bitcoin: The New TradFi
Best Way to Advertise a Programming Language
Re:Ask HN: Quiz or test shared on HN you only take once
on Sunday April 20, 2025 @02:22PM (#43745486)
I believe I found it: https://nicholas.carlini.com/writing/llm-forecast/question/C...
Re:Ask HN: Quiz or test shared on HN you only take once
on Sunday April 20, 2025 @01:08PM (#43745015)
Questions like mine ("help me find it") occasionally posted here. I believe "Ask HN" is the appropriate section.
Ask HN: Quiz or test shared on HN you only take once
Quiz or challenge shared on HN. User allowed to take test only once, or rather something about the test made subsequent attempts not valid. (You learn something about the test as you take it.) At the end you receive a percentile score relative to other users. I don't exactly remember the content or what was being tested. Possibly it had something to do with LLMs.
Official U.S. Time
Stripe's payment API: The first 10 years (2020)
Re:How to Use Em Dashes (–), En Dashes (–), and Hyphens (-)
on Friday March 28, 2025 @01:18AM (#43501807)
Super- or subhuman intelligence can be identified in the pre-Mason–Dixon line era.
I Built a Mechanical Calculator [video]
Re:Ask HN: Substack article about 50 surprising mathematical observati...
on Tuesday February 11, 2025 @04:07PM (#43018370)
found it, hn link https://news.ycombinator.com/item?id=28764416 and source URL https://axisofordinary.substack.com/p/the-most-counterintuit...
Ask HN: Substack article about 50 surprising mathematical observations?
Shared here was something of a listicle (brief 1 or 2 sentences) compiling 50 or so surprising observations of math or statistics. E.g., Simpson's paradox. I believe article was hosted on Substack.