And any push to use 2 to build infra to make 1 easier is hard to sell when a lot of engineers think AI will be able to perfectly do 1 in some nebulous time in the near future.
It reminds me a lot of adderall's effect on people without ADHD. A pretty universal feeling that it's making you smarter, paired with no measurable increase in test scores.
I'd believe this more if companies weren't continuing to use words like reason, understand, learn, and genius when talking about these systems.
I buy that there's disagreement on what intelligence means in the enthusiast space, but "thinks like people" is pretty clearly the general understanding of the word, and the one that tech companies are hoping to leverage.
The problem is that intelligence isn't the result, or at the very least the ideas that word evokes in people don't match the actual capabilities of the machine.
Washing is a useful word to describe what that machine does. Our current setup is like if washing machines were called "badness removers," and there was a widespread belief that we were only a few years out from a new model of washing machine being able to cure diseases.
If I came up with something novel while watching a sunrise, which I wouldn't have come up with had I not been looking at it, where did the novelty really come from?
The backend has plenty of complexities, but frontend developers have to deal with something just as complex - the user.
Given ramp up time, most backend engineers could build a bad frontend, or build a good one if they have a really good UX team that thought through everything and are just implementing their work.
In the real world though where UX is understaffed and often focused on the wrong problems - I've had to rescue too many frontends built by backend focused teams to share your confidence.
The problem is less that those high level engineers are only good at deterministic work and more that they're only rewarded for deterministic work.
There is no system to pitch an idea as opening new frontiers - all ideas must be able to optimize some number that leadership has already been tricked into believing is important.
Every day I am more convinced that LLM hype is the equivalent of someone seeing a stage magician levitate a table across the stage and assuming this means hovercars must only be a few years away.
I was unclear, I meant chunks of the world becoming unlivable from climate change could easily be the catalyst for a nuclear war. So the worst case for both is the same in my mind.
Could the recipe limitations be because of the danger an incorrect recipe could put the user in? It's probably unlikely you'd make something toxic, but a made up recipe could easily be a fire hazard.
My guess is that while it may not be too much effort to get a mostly accurate emulator that works well enough for hobbyist use, it'd be a lot of effort to get something up to the compatibility and usability standards of an official product.
Many older apps may use undocumented functionality or non obvious quirks of the system that an emulator may miss, which means you'd need to have a QA team testing individual apps for compatibility.
Part of Apple's brand is usability and a lack of rough edges. The downside of that is that building a tool like this up to their standards would be prohibitively costly.