Context is intuitively important, but people rarely put themselves in the LLM's shoes.
What would be eye-opening would be to create an LLM test system that periodically sends a turn to a human instead of the model. Would you do better than the LLM? What tools would you call at that moment, given only that context and no other knowledge? The way many of these systems are constructed, I'd wager it would be difficult for a human.
The agent can't decide what is safe to delete from memory because it's a sort of bystander at that moment. Someone else made the list it received, and someone else will get the list it writes. The logic that went into why the notes exist is lost. LLMs are living the Christopher Nolan film Memento.
I looked at Screenpipe but it doesn't fill me with confidence. The ideal solution would be a device-agnostic self-hosted service for archiving and searching screenshots. Would make data migration super simple. And then the app snapping the screenshots could be almost trivial.
It's a bundle of 24 high quality games, by talented, diverse and motivated game developers for ~$8 each. And it comes with a free console. Art is worth paying for.
I feel like this is missing the point of Vim a little bit. It's not about using hjkl instead of arrow keys. It's about progressively learning a vast and useful language specifically made for editing text efficiently. And then realizing you can use every single part of that language in a macro. And then realizing you can use that same language to orchestrate those macros.
The music industry has been fighting against the natural economic theories of capitalism ever since the mp3.
Digital files have infinite supply. Draw a classic supply and demand diagram using that information and look at where the price should be. The supply curve is a vertical line at infinity on the x-axis. The demand curve crosses it at $0. Everyone knows and feels this intrinsically.
The artist is the thing with extremely limited supply. Something like Patreon will end up being the correct model.
Without those other "negligible" aspects there is no incentive for anyone to run nodes to validate the chain. There is no blockchain. It's those very things that allow these systems to function.
Oh boy. If people think this is an example of being unacceptably direct and gruff, I need to reexamine the way I talk during technical discussions.
I've always thought it best to just cut straight to the core of issues like he's doing. I suppose it's a jerk move to corner people into admitting they're wrong even when it's obvious. Maybe there's a nicer jedi-mind way to do it.
The analogies damage this article because they feel wrong. For example the "never synchronous" example is more like this:
You order a burger at the cashier window, then go to the pickup window. If the burger is already made, it's already at the pickup window when you get there.
The author wants a special case where if the burger is already made, they hand it to you immediately at the cashier window. This might seem more efficient, but both in the restaurant and in code it makes logic way more complex.
Absolutely, click "No extra automation" on the last version and play it normally. It makes you question everything you previously thought was a random guess. Makes the game vastly better, because you're richly rewarded for thinking through every possibility. And you're incentivized to spend time thinking because you're not punished when it ends up being a truly random choice. Great, great work. Too bad minesweeper is not suited to mobile devices because I'd play this version all the time.
That is exactly what the cryptocrazies are stipulating. It's cryptocurrency vs fiat. Seen from the fiat perspective, it's an obvious bubble. But if you're on the cryptocurrency side, it's barely the start of a massive and inevitable shift.
Context is intuitively important, but people rarely put themselves in the LLM's shoes.
What would be eye-opening would be to create an LLM test system that periodically sends a turn to a human instead of the model. Would you do better than the LLM? What tools would you call at that moment, given only that context and no other knowledge? The way many of these systems are constructed, I'd wager it would be difficult for a human.
The agent can't decide what is safe to delete from memory because it's a sort of bystander at that moment. Someone else made the list it received, and someone else will get the list it writes. The logic that went into why the notes exist is lost. LLMs are living the Christopher Nolan film Memento.