InkCanon

The main difference is SS bonds are bought at market rates. CPF bonds are not.

I think whether houses should be sold on a 99 year lease vs property is a large and separate question. The vast misunderstanding of it still remains - the idea that a house is owned, not rented. Selective properties might be bought back. But the fundamental invariant of this whole situation is that all 99 year leases will worth 0 eventually. Any rise in prices now only increases the eventual depreciation. And add on that massive loans are taken out on depreciating assets, so it's not an investment, it's a liability. And the significant majority of Singaporeans bears this liability on their books.

Yes. Bluntly put, the government maximizes residence of high net worth individuals, and a 37% forced purchase of low interest bonds would be outrageous to them.

I think they all output that bold lettering, point by point style output. I strongly suspect it's part of a synthetic data pipeline all these AI companies have, and it improves performance. Claude seems to be the least of them, but it will start writing code at the drop of a hat. What annoys me in Gemini is that it has a really strange tendency to come up with weird analogies, especially in Pro mode. You'll be asking it about something like red black trees and it'll say "Red Black Trees (The F1 of Tree Data Structures)".

I'll be blunt and say most Singaporeans have a very poor idea of how these policies work. Another major one - virtually all Singaporeans believe they own their houses, and it is a point of pride and financial security. Most houses are on 99 year leases, but the idea that is deeply lodged is that this is longer than you can live so this is inconsequential. While this is true if they only cared about living in it, houses have huge financial/investment value to Singaporeans. Despite the iron mathematical law that these houses must depreciate their lease value, most Singaporeans believe house prices will continue to rise based on historical trends. The math just doesn't work out.

Singapore's economic policies are complicated and often misdirecting. I'll break down the misconceptions.

The primary purpose of CPF is not a pension scheme. It is structured as a massive forced bond purchase scheme by citizens. Financially what happens is the 37% of citizen income buys a long term bond (till retirement age, on average decades) at rock bottom interest rates (it's pegged to the overnight rate or a minimum of 2.6%). The returns are specifically decoupled from the real long term returns. This has historical roots in the government needing vast capital financing. They make enormous amounts of the delta between the short term interest rate and long term capital gains. Singapore has no oil or natural resources, but it's sovereign wealth fund has AUM in the regions of countries like Norway which do for this reason. It is not a shock absorber like the article suggests. The withdrawal terms are strict - housing, a significant medical expense and retirement are the only real ways to get money out of it.

"Trying to keep people employed" is a goal, not a policy. In fact the Singapore government maintains a large worker supply through immigration. The foreign worker population, ~30%. The main goal of the government is to maximize the absolute number of people working.

The reason it raising the retirement age is effective in workforce participation is because most people have no choice. Retirement only pays out after the age. The working life of an average Singaporean has seen 37% gone to CPF, maybe another 10% to income taxes, another 5% to GST, road tax, property tax, etc. After all this there's the astronomical cost of living. This is also intentional, to raise the number of employees.

This will be an unfair question given the limitations of a browser sandboxed compiled to wasm, but do you have benchmarks comparing this to node hosted on the same machine?

A disclaimer should be added about the heavy startup accounting used. Standard accounting revenue uses the past year revenue. Recurring revenue, used aggressively by startups, is projected future year revenue based on loosely defined contracts like subscriptions. Critically it is often extrapolated on a very short timeframe, like last month, because it gives a better growth figure. Run rate revenue is even more aggressive, it includes one time fees, contracts and other non recurring fees. They have not made that much, nor are they conservatively projected to make that much. It's a very vague measure.

It can mean many things, but clearly cannot be mean what revenue is going to be in the future. If Claude doubles revenue every six weeks, by the end of this year they would have a higher revenue than every FAANG company combined (about one trillion).

I think it borders on parody that this hyper capitalist, hype driven mindset (originally found in tech) has not only infected a lot of "art", but they are boasting about it. A more accurate title is how to make a living selling a very specific kind of popular mural/Art Basel/showroom/"elite" kind of art.

Would be interesting to see how performant it is (or how easily you can write performant code).

There are three closely related reasons why spaced repetition is much less useful in practice then in theory.

1. It doesn't train real task performance. There is a spectrum of problems that people solve. On one end it is the recall of randomized facts in a flashcard prompt->answer way. On the other end is task performance, which can be more formally thought of as finding a path through a state space to reach some goal. The prompt->answer end is what SR systems relentlessly drill you at.

2. SR is pretty costly, prompt->answer problems are also low value. If you think about real world scenarios, its unlikely that you will come across a specific prompt->answer question. And if you do, the cost of looking it up is usually low.

3. The structure of knowledge stored is very different (and worse). If you think about high performance on a real world task like programming or theorem proving, you don't recall lists of facts to solve it. There's a lot about state space exploration, utilising principles of the game, leveraging known theorems, and so on.

This is a more descriptive version of the "rote memorization" argument. There's two common counters to this:

1. Learning is memorization. This is strictly true, but the prompt->answer way of learning is a specific kind of memorization. There's a correlation-causation fallacy here - high performers trained in other ways can answer prompts really well, it doesn't mean answering prompts really well means you will becoming high performing.

2. Memorization is a part of high performance, and SR is the optimal way to learn it. This is generally true, but in many cases the memorization part is often very small.

These ideas more accurately predict how SR is only significantly better in specific cases where the value of prompt-answer recall is really high. This is a function of both the cost to failing to remember and the structure of knowledge. So medical exams, where you can't look things up and is tested a lot as prompt->recalls, SR finds a lot of use.

My own guess for the what the next generation of learning systems that will be an order of magnitude more powerful will look like this:

1. Domain specific. You won't have a general system you chuck everything in. Instead you will have systems which are built differently to each task, but on the similar principles (which are explained below).

2. Computation instead of recall - the fundamental unit of "work" will shift from recalling the answer to a prompt to making a move in some state space. This can be taking a step in a proof, making a move in chess, writing a function, etc.

3. Optimise for first principles understanding of the state space. A state space is a massive, often exponential tree. Human minds cannot realistically solve anything in it, if not for our ability to find principles and generalise them to huge swaths of the state space. This is closely related to meta-cognition, you want to be thinking about solving as much has solving specific instances of a task.

4. Engineered for state space exploration - a huge and underdeveloped ability of machines is to help humans track the massive state space explorations, and evaluate and feedback to the user. The most common used form of this is currently git + testing suites. A future learning system could have a git like system to keep track of multiple branches that are possible solutions, and the UX features to evaluate various results of each branch.

It's not even stealing. They paid OpenAI for the tokens. It violates the OpenAI TOS, which specifically forbids using it's outputs for training competing models (which is very ironic)

The other big missing part here is the enormous incentives (and punishments if you don't) to publish in the big three AI conferences. And because quantity is being rewarded far more than quantity, the meta is to do really shoddy and uninspired work really quickly. The people I talk to have a 3 month time horizon on their projects.

Could someone explain to me the difference? They both get turned to tensors of floats.

Could you explain more on what bindgen was lacking and what you did different?

They should have formally verified their website.

My own experience with Dual N-Back 1) I did it for a month. Probably some improvement, although I did not have any specific test done. Probably the most tangible use is really feeling how working memory (and losing it) is like and having better control. In the same way you might suddenly realize how breathing is and controlling it after learning meditation.

2) The study specified dual N back imposes a pretty demanding regiment. It's 12 sessions daily, consecutively for ~month. It takes about 20 minutes each day of intense focus.

3) There's naturally going to be a lot of survivorship bias in reviews. But you could argue it doesn't make your intelligence worse, so there's only net upside.

Is this the Soham?

This assumption is a pretty Western view, in that you assume that SV/USA still has technological primacy. I believe this to be generally true, but leadership has no incentive or reason it should. And if I'm being frank, there may be some bias here because it seems like companies with indian CEOs are offshoring at a much faster rate than non indian CEOs (e.g. Google vs Netflix). And non core roles could effectively be perceived to make up the majority of the company. So some cultural fondness for India plus a somewhat low view of the bulk of employees could result in most jobs being offshored.

I predict many areas will eventually become like old mining/factory towns. Middle class jobs are being offshored at an astounding rate. Just doing some math - if a tech company lays off 7% of its employees for six years and sends the jobs overseas, nearly half of its American employees will be fired

I really want to destroy this blatant lying by tech CEOs that these are layoffs because of AI, tax laws or interest rates. Most tech companies have headcounts at their peak or slightly below. Whenever these amounts are fired, often near equal amounts are hired in India (and not counting L-1 transfers into the US, H1Bs etc). AI is a fiction invented to cover up the true reason: tech companies have adopted a long term strategy of profit making by wage arbitrage. They have identified their largest cost to be people, and are working to fix this.

That probably has more to do with the absolute tsunami of CS graduates India produces.

Not according to the head of Microsoft India.

Also: Microsoft India untouched by layoffs, actually hiring

https://timesofindia.indiatimes.com/technology/tech-news/mic...

As a way to shift labour into less productive, lower paying industries? Yes.

We're watching the real time collapse and offshoring of American software base. Fun experiment you can do: go to a tech company and see openings by location.

IBM

India openings: 2816

USA openings: 312

https://www.ibm.com/careers/search

Also IBM: CEO Arvind Krishna lays off 8000 workers then rehires

https://dailygalaxy.com/2025/06/ibm-lays-off-8000-employees-...

Wouldn't your returns be destroyed by commission fees? Or do you use a payment of order flow one.

Yes, and Google has laid off ~12000+ (including some churn). Googles current head count is only ~7k below peak. The new office building alone will hold 5k, not to mention other hiring in other indian offices.

India has a (very) low base effect so it's absolute GDP growth is high. But it is growing much slower than China at the same point. Currently India is growing at about 6-7%, at the same GDP per capita China was growing at ~12%. If growth doesn't slow down, it'll take ~30 years to reach where China is now. There are lots of structural problems about the economy, and growth will likely be uneven and slower in the future.

But it is not a great place for these firms because disposable income is so low. For example, the US generates the most revenue per user for Google because it has a really high income. India is unlikely to make any significant part of tech firms revenue for a long time.

But specifically to India? There are many other countries with low income and robust education for CS and reasonable English skill. Eastern Europe for example.