it_does_follow

> I haven't seen anything uses of math in any of my jobs

Math didn't "click" with me until well after college, when I eventually caught up on a lot of calculus, discrete math, probability and stats, to the point where I ended up doing math most of the day as a data scientist.

Before I learned math I used to feel like you did, and would argue that math wasn't necessary to be a good programmer.

Funny thing is that only once I learned a lot of math did I realize how insanely applicable it was to a wide, wide range of problems I was working on. Because I had to learn it late in life I learned math more enthusiastically than most of my data science peers and so find that even in that domain people don't realize how often they can use math to solve problems better and faster.

I wish I had had better teachers in HS that were able to make me realize just how important math is to so many interesting and fun problems.

It's sort of shocking to me, looking back at HS, how many math teachers didn't have a good answer to "when are we going to use stuff?". I wish I could take some of my friends making high six figure salaries with a penchant for late night partying to explain to HS students exactly how math is useful because it lets you get a job where nobody cares how much weed you smoke, how late you sleep, and pay your more than many doctors all because you can do some basic calculus tricks. Plus you get to work on really fun problems.

That distribution is still a point estimate for a multinomial, not truly the distribution of your certainty in that estimate itself. This is essentially a generalization of logistic regression, which will of course give the probability of a binary outcome, but in order to understand the variance of your prediction itself you need to take into account the uncertainty around your parameters themselves.

This can be done for neural networks, through either bootsrap resampling of the training data or more formal bayesian neural networks, both of these are fairly computationally intensive and not typically done in practice.

> a machine learning model

Not to mention that generally ML models are not useful for assessing risk. ML nearly always focuses almost exclusively on some point estimate rather than a distribution of what you believe about a value. The former case is all about expectation and the latter about variance. Correctly modeling variance is far more essential to risk modeling than expectation alone.

I recall talking to a startup that was attempting to model credit risk by building a binary classier for defaulting, and trying to figure out a way to use this to score people for credit (obviously they chose to ignore the fact that there is a huge industry with decades of experience in assessing consumer credit risk).

They focused exclusively on finding more advanced models to get better AUC without even realizing that that's not important. I mentioned that the most simplistic credit score model should at least model P(default|info) and then set the interest rate to - P(default|X)/(P(default|X)-1) to break even and they couldn't comprehend this basic reasoning. It was doubly hilarious since their population's base default rate was such that the solution to this equation was higher than the legal limit they could charge for interest.

In the early part of the current startup/tech boom there was a focus on "disruption", the idea that new ideas could easily dominate old ways of doing things. But for many industries, such as credit/lending and real estate, you should at least understand the basic principles of how these "old ways" work before trying to disrupt them.

> that’s given you the view that drug and alcohol use has increased a lot post-pandemic.

It's fairly well studied in the medical literature [0, 1] and reported on pretty regularly in popular media as well [2]

0. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7763183/

1. https://jamanetwork.com/journals/jamanetworkopen/fullarticle...

2. https://www.usatoday.com/story/news/health/2021/09/22/covid-...

I'm not sure you're aware of the future we're facing right now.

Continued unchecked growth coupled with increasing consumption is heading us towards an ecological catastrophe that quite seriously threatens the possibility of extinction, and at the very least looks like complete collapse of industrial civilization.

There is nothing worse you can do for the environment than bring another life into the world, especially if you live in the developed world.

> what today will sound absurd in the future?

I quite seriously suspect that answer will be "24 hour electricity and grocery stores where you can buy food from anywhere in the world"

> Our widespread and normalised alcohol abuse

Until pandemic we've lived in a fairly teetotaling era. I suspect for the next decade we'll see an increase in casual drinking, becoming more similar to the 1950s. If you watch any films from that era very strong cocktails are basically a standard for just about any social occasion, any time of the day. The pandemic has instantly changed peoples views on alcohol and other drugs, what's coming the immediate future will likely continue this trend.

The immediate future generations will likely look back on ours as a bizarre blend of incredibly wasteful prudes. A generation of people who gluttonously destroyed the planet while at the same time being too timid to let themselves enjoy it. I think the reaction will be a large one of revulsion "you destroyed the planet and you didn't even let yourself have fun doing it? you lived at work for what?"

> there must be a vast gulf in skill level between me and the average hn poster

I suspect it's much more a time gulf than a skill gulf.

I love AoC as well and don't so much burn out but find that after a week of coding for fun in the evening a mountain of real world responsibilities start piling up, especially around the holiday season.

When I was in undergrad I had virtually no family responsibilities during December and a large amount of free time when classes ended, so it would have been easy to spend hours a day on AoC. Many younger, single professionals are likely in the same situation.

I think for most adults with family responsibilities and full time jobs (and likely a range of side projects that also need attention) that first week of December is a rare lull, and a great time to solve a bunch of fun code problems. After that more and more other things take focus as the holidays approach.

> YAML is considered by many to be a human friendly alternative to JSON

I'm not disagreeing with the author here, but as someone old enough to remember the rise of XML as a data transmission format (and Erik Naggum's masterful rant against it[0]), it's strange because historically speaking both XML and JSON were also popularized as more "human readable".

I would be curious how many HNers (and even more so newer developers outside the HN-o-sphere) have worked extensively with or even written parsers for binary (or otherwise non-human readable) file formats. Writing an MP3 metadata parser used to be a standard exercise for devs looking to level up their programming skills a bit.

It personally feels weird to me that we would keep pushing for more "human readable" data formats when the world is increasingly removed from one where non-programmer humans need to read data. Keep your data in whatever format make sense and let software handle transforming it to a more readable or more efficient format depending on the needs, even if humans can't read it (they shouldn't need to!).

On top of all that my experience has been that JSON leads to more atrocities than XML (while fully agreeing with all of Erik Naggum's points about that) and YAML creates even worse horrors than JSON. It seems we'll soon be approaching eldritch horrors if we continue to pursue human readable data exchange formats.

0. https://www.schnada.de/grapt/eriknaggum-xmlrant.html