Urthor

[Urthor @Urthor]

@Jukanlosreve F4Q is a pretty traumatizing acronym ngl.

Thx tho.

[Michael Pettis @michaelxpettis]

7/16
As long as the associated debt-servicing costs could be rolled over, the economy could continue capitalizing these losses and adding fictitiously to GDP, but because no one has infinite debt capacity, eventually those fictitious losses have to be recognized.

[Michael Pettis @michaelxpettis]

6/16
In China's case, however, the growing tendency of the economy to shift to areas that did not operate under hard-budget constraints meant that these losses could be capitalized rather than expensed, in which case they boosted GDP growth and created fictitious wealth.

[Callie Cox @callieabost]

Many of the stock market's best years over the past three decades started with low expectations.

Earnings expectations grew 9% last year, and analysts now expect the biggest increase in S&P 500 profits this year since 2018.

Expectations matter.

A low bar is easier to hurdle.

[Dan Luu @danluu]

What's the top ad for an gambling addiction management app?

Of course it's a gambling app, because the CLV (customer lifetime value) for a gambling addict on a gambling app is very high and the CLV an addict on an addiction management app is zero.

https://t.co/F5oRwLoj8m

[Corey Quinn @QuinnyPig]

My eldest asked if Santa was real, so I taught her about Occam’s razor.

“Which is likelier: that Santa is real, or that millions of adults are in a global conspiracy to fool children into thinking he is? I remind you that you and your sister can’t keep secrets.”

[Yann LeCun @ylecun]

Current LLMs are trained on text data that would take 20,000 years for a human to read.
And still, they haven't learned that if A is the same as B, then B is the same as A.
Humans get a lot smarter than that with comparatively little training data.
Even corvids, parrots, dogs, and octopuses get smarter than that very, very quickly, with only 2 billion neurons and a few trillion "parameters."

[Matthew Anderson @MattAndersonNYT]

Things native English speakers know, but don't know we know:

[Mike Cvet @mikecvet]

Technology companies which measure impact by easily-observable (and game-able) metrics like commit-count or lines-of-code mistakenly treat their software as an asset, rather than the liability it actually is. Their products and capabilities are the assets, not the code

[Alexander Granin @graninas]

Empowering the pragmatism is, of course, a mandatory yet not the only thing that needs to be done for a language to become popular and sustainable.

[Alexander Granin @graninas]

The most respectful people gain a lot of attention in the community, their voices are strong, their beliefs are being explicitly or implicitly supported.

This is what the outside world sees and makes its conclusions about if the language is worth using or not.

[Alexander Granin @graninas]

There is an obvious correlation between what people are respected in the community and what system of values is dominant in it.

[Alexander Granin @graninas]

Making a language popular means doing a lot of things.

But everything boils down to people because it's people who do those things. And people are the carriers of their value systems.

[Alexander Granin @graninas]

The borderline lies between what people flourish in the community:

It's self-serving persons for whom the language is nothing more than a local pile of mud they can demonstrate their power,

Or it's pragmatic people who want the language to be big and popular.

[Alexander Granin @graninas]

It's very clear to the outside world when a language first declares and second seeks popularity (== wide industry adoption), and it's very clear when it doesn't.

All the tiny occurrences of the language contribute to its image and brand, and usually, the image is correct.

[Alexander Granin @graninas]

People in small languages struggle to realize it.

They've built an echo chamber that is about them, not about the language, and feel okay with that.

Making the language more popular would mean destroying their narratives and empowering other people.

[Urthor @Urthor]

@zuhayeer Absolutely beautiful writing and formatting.

[Zuhayeer Musa @zuhayeer]

A memo written by Steve Jobs proposing a ban on Thursday meetings at NeXT to preserve uninterrupted time for individual work.

1986. Lot of timeless nuggets in here.

[Steph Mui @stephmui]

there's a popular expression in dating -- 'if he wanted to, he would' -- and i can't emphasize enough how relevant i think it is to hiring

if a prospective candidate isn't literally hiring a PI to find your address, flying across the country, and begging you to work at your company, it's a no

[Riza Respecter ✝️ @SpiritualGenin]

Reminder that this was "science" in textbooks for something like 30 years after it was debunked in 1974