This anecdote from history feels timely given the recent shift of Apple’s iWork suite (Pages, Numbers, Keynote) from being bundled with Macs to being a freemium subscription.
Instead of trying to think of just any animal, I found it easier to add a constraint…
1. Animal that starts with A
2. Animal that starts with B
3. Animal that starts with C
…
(I also appreciated the easter eggs: “Are you Australian?” and “You listed both dingos and dogs, so I gave you the benefit of the doubt, but there's disagreement on whether the dingo is its own species of canid, a subspecies of grey wolf, or simply a breed of dog.”)
Hinges on whether this new high level -> low level transformation becomes reliable enough to build watertight abstractions on top of it. If AI code becomes good enough that you don't have to worry about the low-level representation 99.9...% of the time, absolutely. But we're pretty far from that at the moment and it's impossible to say where things will be in another 50 years.
What we could do is increase the number of IP addresses available. Just imagine if we enlarged the IP address space from 32 bits to 128 bits: Every device on the Internet could have a unique IP address!
That sounds apocalyptic. What if street addresses were unambiguous? Think of the security implications. Anyone could just walk into your house. Much better to just have "local street 10 b" etc.
The thing is, this upgrade you two are praising is designed to satisfy the original article's needs and no one else's.
Why do all those devices need to talk to each other btw? It's never specified. Is it a user need or a data collection/spyware need?
In a world where security articles make the news saying that you could obtain access to something IF the attacker already has local root and IF the moon is in a quarter phase and IF the attacker is physically present in the same room as the machine and this means the sky is falling...
... we should be questioning why disparate devices on unrelated home networks need to talk to each other.
Peer-to-peer requires that devices from different home networks talk to each other. Gaming, audio/video chat, screen sharing, file sharing (torrents), etc.
The whole idea of the internet from the beginning is that devices can talk with each other.
The need is real. You are a service provider. You need to manage equipment at customer sites. You need to access them simultaneously. But all the customers are using the same subnet...
If Bell gave out cellphones with the same phone number, how can you call anybody? But they still do.
Many devices have cloud access, but every manufacturer is different. It is a nightmare at scale.
The issue is that we DO NOT want every device to have a publicly routable IP address. It does make sense for some machines, but you probably don't want your your Internet-of-Shit devices to have public IPs. Of course you can firewall the devices, but you are always one misconfiguration or bug away from exposing devices that should not be exposed, when a local network is a more natural solution for what is supposed to remain local in the first place.
We did. It's called IPv6. It's 20 years old and still not usable universally. At the high end, like enterprise or telcos, it's fantastic. But at the grass roots level of residential and small businesses, it's still a nightmare.
The extant Tower of Hercules is based on the original plans for Pharos. The Tower of Hercules stands 55 meters tall—half the height of Pharos—and has remained intact for about 2,000 years.
> Using the stars as a reference, Maskelyne’s team found that the plumb lines on either side of the mountain pointed just 0.0152 degrees apart.
I'm really interested in knowing how they could get such a precise measurement (even accounting for errors), especially in the field (outdoor). There's no figure depicting the apparatus they used, I wonder how it looked like.
Sometimes, I just ponder at how ignorant I am. If I was tasked with the same assignment, I'd definitely fail and this was performed 250 ago!
> The first caliper with a secondary scale, which contributed extra precision, was invented in 1631 by the French mathematician Pierre Vernier (1580–1637).[1] Its use was described in detail in English in Navigatio Britannica (1750) by mathematician and historian John Barrow.[2] While calipers are the most typical use of vernier scales today, they were originally developed for angle-measuring instruments such as astronomical quadrants.
So it would have been a contemporaneous technique with that initial angle measurement, and the use of a Vernier scale for angular measurements would have itself been common.
The most avid members of the Cartographers Guilds had even proposed a Map of the Empire several times larger than the Empire itself to depict microscopic details that would otherwise be invisible. Such proposals were considered the peak of academic excess after the Study of Cartography fell out of favor.
I do sometimes wonder if we will get "detailed enough" vector embeddings in LLMs to bring the grain of resolution down below human perception - like having enough bits to fully capture what's on tape in audio world. Maybe this is never possible, and (I hope) some details are unresolvable, but it will be interesting to see.
LLMs are already used in signal processing so the idea is explored.
Simply put anything that can be encoded is a language, so you just need sensors to capture and classify the incoming data and build that into a model. The real question is post training the model to behave correctly as these places are far less explored than things at the human scale. RLHF may be a poor choice because the models may see actual behaviors that humans don't and humans will discount it as being incorrect.
I suspect the curse of dimensionality makes this an optimization dead end. You hit prohibitive latency limits on retrieval long before the resolution approaches human perception. Even with current dimensions, the trade-off between index size and query speed is already the main constraint for production systems.
>When I was a child I used to ask my mother—of course—all sorts of ridiculous questions that every child asks, and when she got bored with my questions she would say, “Darling, there are just some things we’re just not meant to know.” I said, “Will we ever know?” She said, “Yes, of course, when we die and go to heaven, God will make everything plain.” So I used to imagine on wet afternoons in heaven, we’d all sit around the throne of grace and say to God, “Well, now, why did you do this?” and “How did you do that?” and he would explain it to us. “Heavenly father, why are the leaves green?” And he would say, “Because of the chlorophyll.” And we’d say, “Oh.”
1:00:09
But in he Hindu universe, you would say to God, “How did you make the mountains?” And he would say: well, I just did it. Because what you’re (Text sourced from https://www.organism.earth/library/document/out-of-your-mind...) asking me for—when you ask me how did I make the mountains, you’re asking me to describe in words how I made the mountains, and there are no words which can do this. Words cannot tell you how I made the mountains any more than I can drink the ocean with a fork. A fork may be useful for sticking into a piece of something and eating it, but it’s of no use for imbibing the ocean. It would take millions of years. So it would take millions of years, and you would be bored with my description long before I got through it, if I put it to you in words. Because I didn’t create the mountains with words, I just did it. Like you open and close your hand. You know how to do this, but can you describe in words how you do it? But you do it. You are conscious, aren’t you? Don’t you know how you manage to be conscious? Do you know how you beat your heart? Can you say in words, explain correctly, how this is done? You do it, but you can’t put it into words! Because words are too clumsy, and yet you manage this expertly for as long as you’re able to do it.[1]
Why do I need to help? Is this an experiment to see if it can do it on its own, or just another "project" where they give AI credit for human's work for marketing purposes?
https://www.macrumors.com/2026/01/28/apple-updates-keynote-n...