Category: Thinking

What should kids study when AI is going to eventually know it all?

~ Veselin ~ 18 Comments

I have a 7th grader and in Bulgaria, this means we are preparing for exams and for applying to high schools. High schools have a specialization. You study more math, or a foreign language, or some sciences. Unfortunately, most if not all schools are focused on skills that have quite significantly changed over the last decades of computerization and will further change under the rule of AI. Schools around here teach kids languages they can naturally learn from Youtube and probably don’t need. They teach them math that no longer works that way because of online solvers. Also literature with words that no longer mean anything.

So, I’ve been asking myself that, what do they really need, given where things are heading? My current belief is that kids should learn how things in everyday life work. Like everything possible. From microwaves to airplanes.

How does the water come to the faucet? Where does it come from, where does it go, why does it matter, what’s inside it. How do I fix a small problem in the area? A big problem? Dams, floods, safety.

How does the TV work? What makes these leds produce light? What is light? What’s electricity? What do you do if it breaks? How does the software on it work, how can we change it the way we want? How does the Internet behind it work? All these protocols, switches, routers. What the hell is the WiFi? How does it work? What’s the CPU, the GPU, how to change them, how to fix them, if something breaks.

How does a car work? What’s engine, fuel, fire? What’s a motor? What’s behind the doors, behind the dash, what’s OBD, what’s a multimeter and how to use it.

How does a house work? A bank, a loan, a mortgage. Where does the trash go?

We are fighting with poems at home and hand written texts about cows and rebellions. Well, if you want to teach kids about the 19th century struggles with farming, at least show them where the milk comes from and how cheese is made. City kids have no clue about any of that and don’t even have the vocabulary to talk about it.

There’s lots to learn out there for a curious future adult. Knowing how things work and understanding what the AI changes will be essential in maintaining the civilization alive. I feel like we’ve already outsourced most of the understanding to computers, and are quickly offloading to the probabilistic AI, where understanding doesn’t even mean much. Humans not knowing how things work won’t end well for us just like sooner or later, a Windows won’t be fixable with a restart. Big machines being controlled by AI won’t end well for us either. What if… it decides to do something harmful? How do we control a car that has no steering wheel? Kids will have to know how things work to deal with problems like that and to know they are problems and not let them happen. Worst case, they won’t be hopeless if something in their surroundings break.

Unfortunately, most high schools are language schools and teach math and languages. Meanwhile, translators are not in high demand, and anything math is in crisis.

The video that inspired me to write this.

The most underrated life skill

~ Veselin ~ Leave a comment
Daily writing prompt
What’s the most underrated life skill?
View all responses

Listening to understand.

When listening other people speak, our brains create a fog of thoughts that wants to get out like AI’s running on a sandbox. We grab those valuable thoughts and stop listening, or we rather start listening to interrupt.

It’s difficult to just listen to understand and let your own thoughts just fly away.

A metric for an experimentation system’s maturity

~ Veselin ~ Leave a comment

Experiments in the context of software involve showing different groups of people different UIs or features and measuring the impact. This is one way to make informed decisions based on data rather than opinions. If experiments aren’t used, decision-making is usually subjective and prone to errors.

Humans are not good at predicting the future. We may be able to guess where the bus will stop next, but try guessing the price of Bitcoin next year. Being able to properly evaluate the outcome of decisions has real value, while relying on opinions can lead to irreversible losses.

Experiments are one way to get that data, and they’re something I’ve practiced a lot over the last two decades. I can say I have some experience, good or bad. But how far did I get with my own understanding of what a mature experimentation system should be?

The book I’m reading introduces a very interesting metric for experimentation maturity, based on the number of experiments, run per year. Metrics like this can certainly be criticized as superficial, it’s comparable to lines of code written or numbers of PRs. But having a metric is better than not having one, so here’s the maturity model from Trustworthy Online Controlled Experiments, Chapter 4. Numbers apply for large online services, like Google.

  1. Crawl phase: <10 experiments/year, few metrics, no internal system, manual post-experiment analysis
  2. Walk phase: up to 50/year, establishing standard metrics and some setup and data validation procedures, some internal systems and scorecards are available
  3. Run phase: up to 250/year, adoption at a level where most decisions are verified; the focus is on scaling, ease of use, increased number of standard metrics (100s to 1000s)
  4. Fly phase: 1,000s/year, a memory system is developed, the process is built around experimentation, shipping is easy, proper learnings and knowledge are preserved, metric drill downs (mentioned later in the book), thousands of metrics

The book isn’t exactly clear about when you’re out of Run and into Fly, given that one is around 250 experiments and the other is 1,000+, but it at least gives you a way to estimate roughly where you are if you’re running experiments.

It also doesn’t talk much about metric classification. As experimentation matures, metrics should be grouped into categories, with drill-downs for important characteristics. Some of the examples of what should be available felt very strong, like things I wish I knew sooner.

This book is great. I’m about 20 pages away from the end, and I have enough notes to write a chapter rather than a blog post. The last chapters are full of formulas and take time, as some of the formulas are not something I’ve encountered before. I learned stats with pen and paper, making countless calculations of averages, standard deviations, p-values, and such by hand. When I encounter a new stats formula, my brain wants to see the tables behind it, the charts behind it, the history behind it, and it refuses to move on by just accepting it.

Bad choice of words

~ Veselin ~ 18 Comments
Daily writing prompt
What’s a word or phrase that annoys you?
View all responses

I’ve developed an allergy to using the right buzzwords in the wrong order, or overusing them with the intent to convey a false sense of expertise. I wrote an article about this and pinned it to the right menu on my blog a few years ago but I’ve been observing this phenomena since the beginning of my engineering career.

I believe experts should explain difficult problems without resorting to lingo, adjusting their language according to their audience. If they don’t, we can’t really distinguish an expert from a bullshitter.

Our hyper-contextualized multimodal AI orchestration layer leverages adaptive quantum-inspired neural synergies to optimize scalable human innovation.

Long-term plans

~ Veselin ~ 6 Comments

There’s an old saying that says “when humans plan, God laughs”. Yet, I believe in long-term planning. I know, it’s often just nonsense. So many things can go wrong. You can be struck by lightning. The path can lead to some other place. You can lose the ability to execute it. Things can go wrong. I get it. It can be a bad plan. The longer a plan takes, the less likely it should be to accomplish it.

But there’s a trick with long-term plans. Essentially, keep showing up. Like a drop falling over and over at the same spot until it carves the stone. Showing up changes things. That’s something we can normally do. We can manage ourselves, we can be the drop no matter if it will eventually carve the stone or not.

And while you keep showing up, you learn, you set the bar for persistence, you give an example, and you may add 0.01% to someone else’s inspiration. If you do that a hundred times, the 0.01% becomes a full unit of inspiration. Just like reverse-buying chocolate. Bring that home, eventually everyone will eat it. Don’t bring that home, who knows. Maybe nobody brings it home today.

If the goal is worthy, the world becomes a better place even if it’s not fulfilled.

Speaking of goals, despite the series of injuries, I’m still showing up to the pull-up challenge. I started taking selfies on day 18, missed some. This screenshot is also missing some. But it captures my general belief.