jschveibinz

Nice start. The examples do not seem to work on Safari IOS. I will try later on my Mac.

Turabian might disagree with your use of quotation marks.

That was smart.

You have some web page material in English, and some material in either Spanish or Portuguese. You might want to check everything twice before posting on HN.

"...on the subway wall." - Paul Simon

Probably the most important thing I've read this month. The key concepts are valuable.

In Safari the refresh page button does this--found by accident.

In Safari, refreshing the page will reorder the letters. I discovered this by accident.

Ambulance chasers. If it scares people, it sells ads on CNN.

Some web research:

Car insurance premiums grew 3.8% to 5.4% annually over 30 years, compared to 2.4% average inflation rate. Why?

Complex technology and labor required to fix cars.

Costly claims for rising medical bills, increased litigation, severe climate weather damage, and distracted driving accidents from smartphones

The last 3 years, profit is way up from excess cash after post-pandemic rate increases. Before that, profit wasn't great.

Philosophical perspective:

I use AI a lot--I love its capabilities--and LLM's in themselves are the result of millions/billions of "averages" or clusters of things that have already occurred.

But I think there's an argument to be made for working from the tails of the distribution from the ensemble rather than the mean.

Humanity (and problem solving through software and computation) has moved forward based on order AND novelty. The mean provides stability. The tails drive change.

So it may be that 1000 iterations resulting in an average plan may still not be the most effective plan.

Thank you for the rabbit hole

Side note: The federal government costs about $19 billion per day to operate based on an annual budget of roughly $7 trillion.

So $1 billion is about equal to 4 hour hours of government spending.

The PhD degree process is archaic--it's basically indentured servitude. Change that, and PhD applications will skyrocket.

I get it. But information is the backbone of this forum, so I personally don't stress about it. Maybe just a note that says "AI generated"? Perhaps soon people will just post "prompts" so we can look it up or generate it ourselves. AI is probably going to be around a long time. What solutions would you offer--specifically for HN?

Nice. Keep learning and having fun!

Nicely written longform article!

I went back to school in my 50's to learn how to teach. Two of the most important takeaways from that experience were:

1) learning how to learn; and

2) using projects to learn

Writing about your experience in learning is also a powerful tool. If you can describe it to your journal (or someone else), you really know it.

Thank you

That is despicable behavior from a professional. How common is this in academia?

I think this is a great concept. I'd like to see the potential value ($) for a small to mid-size company.

imaging spectroscopy is a potential solution--it's been studied for over 30 years.

According to the article, automation (including software) and other technology "advancements" are important factors.

I think it's becoming clear that we are reaching a point where UBI must be debated in Congress subsidized by something that doesn't wreck economic growth and probably doesn't target capital investment.

One fund mentioned manages $1.4 billion, the other $300 million.

The top u.s. hedge funds by gross AUM are:

- Millennium $500+ billion - Citadel $500 billion - Ares $300 billion

There is no coherent statement's from these top folks on the "AI bubble"

I can't read the article, but if I were to venture a guess I would focus on these things for contributing to the loss of prestige:

- falsification of research results from academic researchers

- tuition increases at double the cost of living (inflation) increases...and a majority of people not able to pay these tuitions

- fewer jobs available that require a university degree and credentials

- loss of "intellectual authority" due to changes in public opinion (conservative leaning) and AI technology gains in authority

- evidence of large scale financial mismanagement and poor decision-making: my college has hired and fired 3 presidents in the past 10 years and now pays a salary all of them

- inability to keep curricula up to date relative to the rapidly changing world

- fewer academic researchers willing to teach

- higher quality programs at community colleges at much lower tuitions with better teachers

- public scandals about celebrities "buying" entrance to elite schools by cheating on exams

- public scandals about students using AI to do their work for them

- COVID changes and realizations of how colleges really work when stressed

- ...and much more

In my opinion, no coherent argument is made by the author for a "crisis" in classical liberalism.

Liberalists still believe that things work better when society and the economy works things out with minimal state interference.

Neither AI leadership nor political organizations will change this mindset. Technology marches on.

I just listened to a radio program on my local NPR station about the topic of kids and social media. From what was presented, the research shows (longitudinal study) that there is very little evidence of social media impacting mental health--which is shocking because a majority of adults think there is a connection and the politicians are pushing that narrative. I have not personally vetted the research. Has anyone else?

You can predict by using one of several techniques:

1. Naive prediction: it will be similar to now

2. Linear regression: look at several parameters associated with LLM (speed, quality, accuracy, etc.) and create linear regressions from these parameters

2a. Polynomial regression: same as above, but fit a polynomial

3. Group prediction: select 20 friends, ask them to make predictions, then find the average(s)

...and many more with increasing levels of complexity

Use peers and friends for alpha testing. Beta testing should be reserved for your champions--those likely customers who love what you are offering. If you don't have customer discovery done yet, then that is an important next step. Good luck .

I went for a 30 minute jog at lunch around 3 days a week for most of my career. As an engineer I was privileged to have that option. But today, even a 7-10 minute HIIT workout would be great for health and stress reduction.