on a learning arc | alt: @springofmaya | 日本語: @haru_no_maya | 华语: @chuntiandemaya | also on 🦋
Singapore (sometimes London)
Joined June 2024
- Tweets7.5K
- Following607
- Followers967
- Likes49.9K
Pinned Tweet
I am travelling around the world next month! if you're in one of the locations below let's meet up!
San Diego 5 Nov
Seattle 6 Nov
San Francisco 7-8 Nov
NYC 9-19 Nov
London 20 Nov - ~2 Dec
Simplified Chinese in theory: reading and writing is going to be so much simpler now that characters take fewer strokes
Simplified Chinese in practice:
"If you had an exponential amount of compute" we don't even have an exponential amount of the universe 😭
If you had an exponential amount of compute you could solve simulate biology exactly and wouldn't need to run any experiments.
Some of you will think "of course, but that's an unrealistic assumption" and others will think "no, you need experiments even with that much compute. That's just How It's Done".
You would of course need evaluations once you have a solution, but you don't in principle need experiments to find it: Experiments are a way to save compute, dating from a time when it was much more expensive.
And I think the presentation is exponential but real systems aren't irreducibly exponential. AlphaFold worked, after all.
It is still actually useful. For instance, I was at a café earlier today, and they had a 'teh susu' and I wondered "what is susu" but then I saw 鲜奶茶 below and thought "right" and ordered without wasting an extra conversation turn 😇
When I'm waking up, my sleepy brain sometimes convinces me to not get up just yet as whatever it is that I'm dreamily thinking about is actually Important and Valuable
Then I finally get up half an hour later and the illusion evaporates in a second
What is that about
so now what I wonder about is: is Lean now locked in forever as the language for mathematics? or could you have interoperability, where the two proof assistant systems formalise each other to trust each other's claims
"We believe torment is the most natural way to interact with the Nexus" – every tech startup nowadays for some reason
We believe voice is the most natural way to interact with intelligence, so we created Dessa, the most authentic full-duplex speech model ever built. Talking with it feels like talking with a person…full of emotion, nuance, pauses, and a natural cadence.
block.xyz/dessa
Turns out that the word 'sidereal' (like 'sidereal day' for a single rotation of the Earth not with respect to the Sun but to the rest of the universe, i.e. ~23h56m) comes from Latin and means "relating to the stars"
All this time I parsed it as side-real. It made sense!
I ended up just now re-reading the essay on Jane Jacobs's economics written by @etiennefd for the 2023 ACX Book Review contest, and one point struck me a lot more than previously when put in the context of the heated debates of Q4 2026
To Jane Jacobs, the "live players" are the robustly self-sustaining cities; what happens elsewhere is solely dictated by the forces exerted by them
Thus the centralising threat of AI to the rest of the world is in reducing the number of those cities to 1
hopefulmons.com/p/1980s-jane…
People point out that GenAI mainly means "AI that I don't like" but even if the term has no technical definition to it as commonly used it ended up correctly delineating between mere tools and a deal with an alien-minded devil
I like kanji aesthetically but reading a Japanese tweet, spotting a kanji conversion mistake (i.e. the machine's first guess at disambiguating homophones guessed incorrectly), yet still being able to understand without much of an issue, makes one wonder what's the point