I support the collision risk management system for a few of the Earth observation satellites at NASA. Pollution of low Earth orbit is a big deal. There are thousands of individual objects being tracked up there. Some of them are satellites that have exceeded their lifespan, some of them are parts of later stage boosters, and some are debris from previous collisions.
When a satellite nears the end of its life expectancy, US law requires that the remainder of the propellant be used to decelerate it so it disintegrates in the atmosphere over the ocean. Presumably European and Asian space agencies have similar rules. These rules didn't used to exist, and there's stuff that's been up for decades.
Yes, these objects are very far apart, but they are moving very fast, and there are surprisingly many of them. Operational satellites are extremely valuable, and if we anticipate anywhere close to a 1 in 1000 chance that one of these objects will collide with one of our satellites, we will burn some of the irreplaceable propellant to avoid it.
The problem is we want a lot of things to be in geostationary orbit, which is only possible at a specific altitude. That effectively transforms the "space" from three dimensions to two because the third relative coordinate for all of those things has to be the same.
Which is when relative motion becomes a problem. The Empire State Building and the Trump Building are at approximately the same altitude (i.e. sea level) but you don't have to worry about them crashing into each other because their relative position doesn't change. In space everything is in relative motion which effectively costs you another dimension.
Then you no longer have "space" you just have "a line" and if two things are at the same point on the line, they eventually crash into each other.
Actually, the useful part of the geostationary orbit is 1 dimensional, because the satellites have to be at the right height and over the equator. It's called the "Clarke Belt": https://en.wikipedia.org/wiki/Geostationary_orbit
But anyway space is very big and the probability of collisions is still low. Two geostationary satellites out of control would probably not collide, they will pass a few hundred of meters away in any direction. The big distance between them is mostly to avoid interference between the radio signals.
The things in geostationary orbit don't crash into each other, they just constitute a relatively high density of "stuff" all in one place that non-geostationary objects at the same altitude could hit every time they cross the equator.
The worst case would presumably be something large and heavy put into orbit around the equator at the same altitude and speed as geostationary orbit but going in the opposite direction.
We've already had our first accidental crash between intact satellites. If SpaceX and other low-cost launch providers succeed in making stuff like low-orbit satellite internet a reality, we'll need to figure this out.
The problem is overall pollution of exit trajectories... Imagine how much more difficult it will be in 100 years to launch anything without hitting something on the way out. It should be a requirement of any satellite manufacturer to have a decomm plan. A plan which will eject the thing from orbit, or let it burn up on re-entry.
That's not a very responsible way of looking at it.
It's not an infinite amount of space, so without taking some precautions it would fill up over time, and it's undoubtedly easier and cheaper to avoid the problem altogether than it is to fix it once it becomes a problem.
"________ is huge, we don't have to worry about it!" is how we ended up with so much pollution on earth, so why make the same mistake in space?
I mean low earth orbit is at least as much space as we have on the surface, which is already pretty damn big.
Yeah, speeds and orbits and stuff change the game a bit, but i can't imagine even a few thousands of satellites cramping a space that large.