How do IP addresses work?

Question

I kinda get how it makes sense on a local network, because it’s kinda like a list of the devices that are connected to it and your trying to communicate with other machines connected together in that same list. But this feels like a really surface level understanding and I don’t know what I’m missing.

I understand that they’re a string of 4 digits from 1-255, such as 192.168.1.1 being really common for home networks. But I don’t know what the numbers each mean. I think 192 in this case is a reserved value for home use? Same with 10? And the last number is basically the number of the device on the network I think. But I don’t understand the numbers for x.168.1.x

What I really don’t understand is how public IPs work. You hear online about not leaking or sharing your public IP or it can be used to find (pretty close to) where you live. How? How are they assigned to the billions of connected devices in the world?

I’ve been watching videos about trying to set up a FOSS router because it’s really interesting, but IPs seem like dark magic to me

In: Technology

Answer 1

“IP” addresses are meant to be a hierarchical system, where you could divide the world into 256 “regions” (much like countries… and country codes in telephone numbers…) then divide each region into 256 smaller regions (in a country those would in telephone numbers typically be “regional number”, in Sweden 08 is for Stockholm… ) And you then divide each of those region into 256 even smaller regions, and then one more division into 256 individual machines within that smallest region.

Since what matters is not always physical proximity, these regions may relate more by how close they are “electronically” (a good cable could make two regions seem more connected, even if they are further apart than some others, etc… )

The thing with “internet protocol” addresses is that there has not been a very organized way to decide who or what should get what address. Sure, it isn’t entirely organized, but, it is still a bit chaotic. So the “topological” information that you’d need to route efficiently is not always there. It can be a bit random, such as some company owning a number of addresses and using them in a way that isn’t geographically or connection-wise related.