We live in a vast and expanding universe with an observable limit of about 46 billion light years. This edge continues to expand away from us such that, rather than exposing more of the universe to us, we are actually seeing less of it. You can think of the edge of the observable universe as a kind of funnel. The bottom is in the past, about 13.77 billion years ago, only a few hundred thousand years after the Big Bang itself. This light rode the wave of expansion of the universe outward, wavelengths stretching redder and redder. Because of the stretching of space, the light was able to reach us in only 13.77 billion years despite being from 46 billion light years away.
As the observable universe expands away, however, the light from the most distant objects becomes redder and redder and those objects appear slower and slower, until, like an object falling into a black hole, those objects freeze and their light disappears.
Indeed, if the universe existed for an infinite amount of time, we would only be able to see a finite amount of it. This forever limits our ability to know exactly what size of universe we live in and whether it is actually infinite.
It also means that we cannot know what the “edge” of the universe is like. A better and more mathematical word for this is boundary. Some conjecture that space has no boundary because it is like the surface of a hypersphere, like a sphere but with one additional dimension. This universe is called a de Sitter space and was first conjectured by Willem de Sitter, a contemporary of Einstein, in the 1920s. As on the 2D surface of the Earth, it has no boundary and if you could circumnavigate it, you would end up back where you started. The problem is that the universe cannot be circumnavigated without exceeding the speed of light. Nor, indeed, can parts that are outside the observable universe necessarily ever even influence us here on Earth.
Others conjecture that the universe is really anti-de Sitter. This kind of space has two time dimensions instead of one. Thus, you can travel in circles in time unless you make certain assumptions that make it impossible to do. Ordinary space exists in a linear rather than hyperspherical region. Combined with the circle of time and the linear space, you get the surface of a hypercylinder. Instead of moving up or down the one longitudinal dimension of an ordinary cylinder, you have three dimensions.
Anti-de Sitter spaces are interesting because of something called the AdS/CFT correspondence which shows that you can create quantum theories (Conformal Field Theories) on the boundaries of anti-de Sitter spaces. This has become important to some theories of quantum gravity. As far as we can tell, our universe is not anti-de Sitter and all evidence points to a nearly de Sitter configuration. If our universe is anti-de Sitter at the quantum level, something strange would have to happen at larger scales to flip it to de Sitter. Besides that most of the interesting results rely on the universe having more dimensions than the usual four.
A de Sitter universe has the unique property that it can originate from a Big Bang, which is just a hot dense early universe, that happens across a region of space. That is, the Big Bang would not have been a point. It only appears to be a point. Each point of that extended Big Bang forms essentially its own universe separated by a horizon from the others. That horizon is the edge of the observable universe. If this is true, then the edge of the universe is the observable universe and there are other, causally separate universes lying beyond it that grew out of the same Big Bang.
Thus, far from being just one event, the Big Bang may be our small part of a vast surface of a boiling universe generator, like water in a kettle letting off bubbles.
“Bubble” universes aren’t just a convenience science fiction. In inflationary theory, universes can exist in their own bubbles of space and time, yet the shining surface of our bubble is something we can see just by looking up on a dark night (particularly if we are using a radio telescope tuned to the Cosmic Microwave Background).
Yet, because our bubble is expanding faster than light can keep up, our bubble is actually, from a causal-connection standpoint, growing smaller rather than larger. The universe is expanding into the unknown, but also breaking apart into smaller and smaller regions, disconnected from one another. Eventually, the universe will have “shrunk” to the point where individual particles will have a universe to themselves, and nothing else will be reachable.
This is true also of worlds that may be close to the edge of our observable universe. For them, we are the ones receding away into oblivion.
The observable universe is everything that lies within our past lightcone, where a lightcone is simply all the points from which light can reach us. The further back in the past you go, the larger your light cone is, and light is reaching us from that past all the time. And yet, objects that generate that light, because of the expansion of the universe, move outside of that light cone over time, meaning we are forever separate from them.
The region of space inside a past light cone is bounded by an invisible 2D boundary called a horizon. I have been talking about two different kinds of horizons so far. The first is the light cone boundary which is called the particle horizon. This is the horizon inside which particles can have reached us, including particles of light. Another horizon is a visual horizon. Unlike for most of the universe’s history, the early universe was opaque to light. Electromagnetic radiation did not propagate but was continually absorbed. Therefore, the oldest light we can see was actually emitted about 300,000 year after the Big Bang.
The particle horizon reaches further out than the visual horizon, back to the Big Bang itself. But the only way we might be able to see further than the visual horizon is by detecting ancient gravitational waves, older than the CMB.
Another kind of horizon is the event horizon. Unlike the particle horizon, which changes from moment to moment as our past light cone changes, the event horizon includes all things that can ever reach us or be reached by us by light. Sometimes this is called the future event horizon. Our future event horizon is finite in size because of the expansion of the universe. There is only so much matter that we will ever come into contact with.
In the bubble model for the universe, you can theoretically also have bubble collisions. These would mean other universe could have a measureable imprint upon our universe. In this case, for example, the bubble wall of another universe can influence our early universe within our particle horizon but outside our visual horizon, so we couldn’t see it except indirectly. Whether this could happen depends on how many universe are created versus how quickly they expand. Unless they impact one of our horizons, they are inaccessible.
Bubble universes are largely conjecture and will probably always remain so. Whether they can pop is another story.
Matter and energy may have nucleated out of a false vacuum. A false vacuum is a vacuum that has energy which can then produce matter. A true vacuum, on the other hand, cannot generate anything and moreover if it comes into contact with matter or false vacuum will cause it to disintegrate into nothingness. If a true vacuum appeared inside our bubble it would eventually consume the entire universe in nothingness.
Pop.
Ellis, George FR, and Jean-Philippe Uzan. “Causal structures in inflation.” Comptes Rendus Physique 16.10 (2015): 928–947.