TonyAlicea10

All those .BAS files were an education.

I did the same in QBasic (and Tandy Color Basic)!

I feel like the old Choose Your Own Adventure books were a sort of precursor to programming for a lot of us. Basic if/then/goto.

For me the spark is reinvigorated. I can try ideas before I didn’t have the time for, prototype and test and only pursue what proves to be worth the time.

Same. This was true for so many of us. An open source education.

There’s a marked difference in quality depending on context, usage of sub-agents, initial specs, and model-wise a big difference from what Opus would produce and Fable.

I’m not saying I’m hand coding, but saying getting good outputs doesn’t take any technique means your definition of good outputs probably isn’t very refined. And saying it hasn’t taken technique for 4 years is just flat wrong.

Funny I made some very similar points awhile back in a blog post, thinking of it in terms of mode collapse: https://tonyalicea.dev/blog/single-mode-burnout/

Thank you! I tried to add just a little snazz without changing the core.

I've made two updates: the wind and round text should be more legible with a slightly transparent background under it. The camera is less whiplash-y and pauses for a moment after the banana hit if it's zoomed out and the zoom in is a bit slower.

I understand, I'm not going for accuracy, just suggestion. For someone like you it won't feel right, I totally get it.

Nope, haven't changed it! I think it's an aesthetic preference thing.

I think the first one for me was The Oregon Trail followed shortly by QBasic Gorillas and Nibbles. It was my first step in coding beyond Turtle Graphics on an Apple from even earlier in childhood.

Opening up the QBasic source code was eye opening for me: that computers weren't the realm of hackers typing incomprehensible text on TV and swapping floppy disks, but something that could start to make sense. My TRS-80 came next, and, for me, the rest was history.

Good memories. I wonder what the memories will be for kids right now in the age of LLMs.

Code-wise, mostly LLM. I gave it the original QBasic source code so that's not mine either.

This experiment was a guidance experiment for me with Fable. Heavy spec and planning architecturally and UX, and strictly no JS frameworks or dependencies.

I do these kind of experiments with new models, and Fable has been the most successful so far for capturing the intended UX and feel per spec.

Yes! Fun things that either you’d never have taken the time to build otherwise, or recreations of things that you know well.

Things where you know how they’re supposed to make you feel (like games from childhood), and getting that experience match is an exercise.

I did let the LLM handle the mouse and touch aiming for v1. That’s an experiment as I prefer typing in the boxes or using arrow keys.

I think it needs tweaking to let you hone in a value better. What are you thinking?

Fair. This is a direct port in some areas of the 1991 game, which strategically was more try something then dial it in. I’ll ponder. It’s almost arcade vs simulation in some ways.

That is, if you can calculate it too easily then the first player always has a huge advantage.

I hadn’t seen the other one! I grew up with this as well.

I like the camera grace window idea. I’ve been pondering how to handle the post-hit zoom in.

Good call on the wind text. I like the filter (?). I’ll look at your link.

Thanks for the feedback!

I don't think this is equivalent. The calculator won't occasionally hallucinate a wrong answer. LLMs are a far leakier abstraction which means skill atrophy impacts evaluation and verification ability.

It would be inconvenient if maintainability still mattered because their products default to making it worse.

I think this is the most important line from this piece.

Incentives matter. The AI companies are incentivized to have us believe that LLMs are the new compiler. That’s ridiculous (a coding LLM is a very very leaky abstraction) but I hear coders, especially coders with poor fundamentals, say it all the time.

This entire AI period has been a study in marketing disguised as futurism.

I say this as one who uses and teaches AI. What fantastic, amazing, unreliable tools. Extraordinary in the right hands, but engines of cognitive and technical debt.

We’re going to disagree about the quality of LLM learning experiences.

As an educator I’ve used them a lot for myself and in teaching others. Good for one-off topics if you check facts and a lot of human educators have built good quality content that it rips off.

LLMs are rather terrible at an extended progressive curriculum.

That said, the core thesis was simple: it’s dangerous to teach children to trust LLMs.

Not what they are. That they’re right. Which they aren’t. A lot.

You said you largely agreed with that. So, what’s the approach within this service to mitigate that?

I appreciate you responding but you’re throwing a lot of “but X” when the question is “what about Y”.

We actually work hard to make sure a child knows they aren't talking to something real - that they're talking to a computer program.

The problem is not talking to a computer program. Kids have been using computers in school long before LLMs.

The problem is teaching a child to trust what the non-deterministic system is guessing at them. That will carry forward in their life.

You’re teaching them that an LLM is right, not just what it is.

As an aside, I’m also not sure about the certainty of what “making sure” a 4-year old knows is real or not is.

If your kid gets everything they need from their school, their home, their private tutor and so on, then great.

That’s a powerful emotional point made to shut down criticism. Except it has nothing to do with putting a kid in front of LLMs.

I expressly mentioned this in my comment. You could create a mountain of educational material, apps, and tools with LLM assistance and verify it. Same result.

You were skeptical that LLMs would hallucinate on topics for children and you said it was common sense.

But now you’re saying the common sense is that they will hallucinate.

I'm not sure how common sense says that LLMs will hallucinate less on topics for children. That's not how they work.

And no, LLMs are not perfect. They're non-deterministic.

And no, even if they were perfect my statement was that it is irresponsible to teach a child to trust an LLM. Even if it was perfect on these topics, as they move forward in education that trust would later fail them.

That is a problem both ideologically and pragmatically. You have to think long-term.

On a side note: ask an LLM to write a story at a 4-9 year old level. Work with it for awhile doing iterations and it will hallucinate a previous detail or change a character name as context rots. Lots of other examples of "basic" topics that still produce hallucinations.

a hardcoded, "Let's get back on track" button would probably be needed

Very true! An adult might hit that button. 4-9 year olds I think are less likely to.

You’re applying human cognition expectations to an LLM. Just because the topic is simple to a human doesn’t mean it improves the likelihood the LLM will behave better. It’s all inference to the machine.

Anyway no learning is needed for 4-9 year old education.

This is an extraordinarily naive statement. Learning how to teach well takes a lifelong commitment. A topic being basic for an adult does not mean it is easier to teach to a child. In many ways it makes it harder.

That's just such a weird point to make.

No, it's not. The claim that "teacher's make mistakes too" suggests that AI is equivalent to a human teacher in value. I posit other ways in which a human teacher is potentially more valuable.

Also, teachers who love their students tend to want to educate them well.

lol no. Have you ever tried correcting a teacher?

Do you mean as a student? If so, that's a different dynamic. Because teachers certainly have accountability processes in many institutions.

Reading books does not have the psychological effects of passing cognition to an LLM.

EDIT: A "luddite" is opposed to new technology. I'm a coder and technical educator. I teach people to use LLMs.

Responsible use of a new technology is not being a luddite. And there are equivalent concerns of the effects of social media on children as well.

My point is that the hallucination rate of modern LLMs, while not zero, is so low that this is no longer an issue in practice.

That's just not true. Talk to an LLM about a subject you know deeply. They make up things all the time.

2024 called and wants its talking points back.

It's a classic example of the kinds of questions LLMs get wrong. There's plenty of others. Not sure your point here. We can easily find things a 4-9 year old will talk about that an LLM will get wrong, hallucinate, etc.

a child tutored by a good quality state-of-the-art LLM with a good teaching-focused harness could have better learning outcomes than a child without it.

There's a lot of work being done in 'could'. And it's entirely ignoring the dangers.

I'm not saying "no LLMs in education". I'm a technical educator and I give students LLM prompts and agent skills that I've built to help them learn.

This isn't that. We're talking about giving an LLM to a 4-9 year old and saying "this is your teacher".

They're not going to struggle at all with things a 4-9 year old is learning.

Ask them how many r's are in strawberry.

Also it's not like teachers never make confident mistakes.

That's an extraordinary false equivalency that is popular around here. Teachers don't make confident mistakes because the student asked the wrong way. Teachers can be held accountable. Teachers can learn. Teachers can love their students.