j7ake

Not every subject has simple explanations.

In math or theoretical fields it’s not unheard of to have young professors. Terence Tao was full professor at 24. Wolfram at 21

Why particularly in tech? Find an industry where EU salaries are better than USA and you’ll have much better traction.

Sounds like asking for the public to buy them out .

Switzerland is not a good place for working moms. It is a very conservative country, for example working moms are systematically discriminated. The society, police, insurance, and banks enforce a conformity to a conservative, male-dominated society.

https://www.businessinsider.com/female-breadwinner-moved-swi...

Didn’t say guaranteed, I said more likely.

To succeed, you have to execute many things well and avoid bad luck. Add to this an extra constraint that one also needs integrity or morals and it will drastically reduce your success rate.

My statement will be true as long as having integrity or morals could hinder growth in some way.

Also suggests 100% WFH is not as good as some weekly face to face time if part of the goal is to keep people motivated and disciplined

It’s also survivorship bias. Given that a start up has explosive growth, it way more likely that they are amoral.

The ones who play the game with integrity and morals never have the type of growth that makes headlines. But I’m sure they exist.

Please name a few such places

Why simulate? We can already do it experimentally

This summarizes what irks me about the community.

They use technical terms (eg expected value, KL divergence) in their verbal reasoning only to sound rational, but don’t ever mean to use those terms technically.

The core is that hyper rationalists mistaken their model of the world with the world itself. They then lecture with confidence their opinions in the real world, ignoring the hairy details and unknown unknowns in the real world.

Does 20k really make a difference in quality of hire?

It starts early : closed source software offers discounted or free licenses to schools to hook kids in early.

This is how matlab continues to have a hold in engineering fields.

Isn’t “write everyday” a goal?

The author seems to be distinguishing goals that are externally motivated (eg win award) vs internal (eg practice every day)

Honestly even if a prediction comes an experiment, and they know exactly how the experiment was done, it takes month to years to follow up and verify.

Generative AI is basically going to flood the field with more predictions, but with little explanation of how, and doing nothing to alleviate the downstream verification process.

Also predictions in biology take months or years to validate, so they lack the fast feedback loop of the vision and NLP world where the feedback is almost instant.

Combine this with the fact that In vivo data in biology is extremely limited, and we see copying the NLP and vision playbook into biology is challenging

There is a typo: it’s a nature communications paper not nature. Difference is vast

Yes the standards are different. Also true for restaurants and sanitation.

Must be an article for westerners because the average Asian household would have 4+ different soy sauces in their kitchen eg light, dark, steamed fish, raw fish, meat marinades.

Use Lee kum Lee for cooked stuff, Japanese ones for sashimi.

For restaurants they sometimes cook their own soy sauce with different oil and herbs for extra flavour

Guido van Rossum is 69, he would be an excellent colleague at any tech company

It’s funny because there are people in USA who actually believe health and quality of life are irrelevant because they’re not measured by GDP.

It’s a faster Google search

On paper? Oral exams are mush better in my opinion

1000 scientists is a lot. The entire Microsoft research department is about 1000 scientists, and they are a productive bunch

Two people affiliated with Google won Nobel prize last year.

Great article.

I am surprised at how close cooper became with the stampf family. There were others that found Bills body but only cooper stayed close.

Interesting that they don’t consider any eastern cultures to have produced any books worth studying.

It’s essentially the pvalue hacking we see in social and biological sciences applied to machine learning field.

Once you set an evaluation metric it ceases to become a useful metric.

The only paper they cite was the bat virus paper in cell, which isn’t evidence for lab leak.

The review paper I mentioned shows dozens of papers showing evidence of market as initial spread.