This is oft repeated nonsense you can disprove yourself. Make a reasonable basket of goods, write down current prices. Look up ads for those goods over time, write down prices. Compute inflation. Check BLS inflation. If there were even a tiny error “artificially down” it shows up as exponentially (in the literal mathematical sense) divergent values.

Your claim is false.

Also there is/was many other groups computing inflation such as MITs Billion Prices Project, lots of hedge funds, etc., and none have found much if any discrepancy between official BLS numbers and other methodologies.

If they were taught what was representable and why they’d learn it quickly. And those that forget details later know to chase it down again if they need it. Making it voodoo hides that it’s learnable, deterministic, and useful to understand.

Yes. A computer scientist should know how numbers are represented and not expect non-representable numbers in that format to be representable.

0.1 is just as non-representable in floating point as is pi as is 100^100 in a 32 bit integer.

Terminating dyadic rationals (up to limits based on float size) are the representable values.

I’m not sure if you got my joke but I referred to the mathematical philosophy, finitism.

There’s plenty of cases where ‘==‘ is correct. If you understand how floating point numbers work at the same depth you understand integers, then you may know the result of each side and know there’s zero error.

Anything to do “approximately close” is much slower, prone to even more subtle bugs (often trading less immediate bugs for much harder to find and fix bugs).

For example, I routinely make unit tests with inputs designed so answers are perfectly representable, so tests do bit exact compares, to ensure algorithms work as designed.

I’d rather teach students there’s subtlety here with some tradeoffs.

Of all the valuable metrics on that site, all of which grok does badly at except one, you managed to pick that single one.

https://artificialanalysis.ai/models

Removing 30% of the supply of a very important, and completely irreplaceable for most of its uses, resource isn’t hysteria.

The point is it doesn’t take giant tankers going through the Strait of Hormuz to move this volume. It could be handled by tanker trucks going to Suez….

This has nothing to do with transport. Iranian drone strikes disabled a Qatari helium _production_ facility.

You should be able to make a killing placing commodity bets right now, because you have such crystal clear vision for the causal chain currently underway

What are your top positions? You will never need to work again!

The surplus had dried up before Bush’s election, as the dot com bust tanked federal tax revenues.

Bush tax cuts did add to the deficit though.

Tons of goods companies paid tariffs for were inputs for those industries.

Instead of being mad at companies that were forced to pay illegal tariffs, who now want to recoup some of that, be mad at the cause of illegal tariffs. Letting the govt keep the money by fighting over who is a victim, hold the govt feet to the fire so they learn not to do this to begin with.

You really think Trump is capable of learning such a lesson? This isn't a 'they' situation, it's a 'him' situation. The cronies enabling him may well be cap[able of learning such a lesson, but their noses are too deep in the trough of self interest to care.

Hiring costs time, money, and other engineers time and effort. Wasting money and time reviewing a pool with less good candidates will simply lose you business over time, as you waste more resources to obtain the same result.

So you are saying the Ivy League pool is the one with more better candidates?

The same pool where connections, rather than merit count? Where the candidates have (probably) been coddled by their loaded parents all their lives? Where they fake disabilities in order to get affordances such as better housing, "support" animals and so on?

Are American universities really turning out engineers with high GPAs and relevant coursework that are unqualified? Or even create a disadvantage for their employer?

I mean realistically you don’t know how good an engineer is until they get on projects and you see their work. But that’s true for Harvard too,

Yes. Ever do hiring, or run a significant interview system? I have. Us engineers thought the same way you did, but after a significant number of interviews, we went back over resumes versus interview performance to find ways to save costs/interview more per resource used, and surprisingly to us at that point quality of college was the biggest predictor. Lower quality schools simply had a less talented or trained pool, and it was significant.

Great! Let's just continue perpetuating institutionalized bias rooted in a non-metric driven hypothesis on candidate potential!

And your line is thinking is how we end up with what is going on at the Pentagon, and less innovation in our industry :)

At significant loss to the consumer. Sure a tariff can benefit a subset of people, by costing others even more.

We could also do this without tariffs by simply taking money from some group and handing it to another.

Someone mentioned a week or two ago on HN that the point of the auto tariffs was national security (maintaining the industry/expertise/etc. in the US, I assume), not economic.

They’re wrong. Lyndon Johnson imposed the 25% Chicken Tax during trade skirmishes with Germany to make Volkswagens more expensive, and later tapes surfaced showing this was actually done for political favors. It had zero to do with national security. As a result Americans pay about 25% more for light trucks of inferior quality ever sense, as a giant handout to the car industry.

https://en.wikipedia.org/wiki/History_of_tariffs_in_the_Unit...

So as usual, tariffs cost the country imposing them, returning less net goods, and moves money into the protected class at the expense of the wider public.

Receiving money for free is different than money earned for work (even if subsidised).

It creates different incentives for the receiver.