Giddings and Mangano, "Comments on claimed risk from metastable black holes,"
http://arxiv.org/abs/0808.4087
The basic answer is that according to what we know about physics (a) they can't be produced and (b) even if they could, they would evaporate immediately. (Radiation shielding doesn't prevent an object from emitting radiation.) If we turn out to be wrong about what we think we know about physics in two different and very major ways, then both a and b could be wrong. The Giddings paper has calculations of how long it would take the black hole to eat the Earth in that case.
Zero people have been killed by radiation at Fukushima. 20,000 people have been killed by the earthquake and tsunami. Personally I would find a black hole far scarier than either of those.
rogerl said:
Oh. I'm mistaken. I think Lee Smolin said our Big Bang might be the result of White Hole from a Black Hole on the other side. So White Hole doesn't have Event Horizon.
Smolin's theory of cosmological natural selection has been disproved by the recent discovery of a high-mass neutron star. It's still possible that black holes form baby universes, but that is speculation about a possible result from a possible theory of quantum gravity, which we don't have yet.
rogerl said:
String Theory and Loop Quantum Gravity both blur the Planck scale by making the smallest object Planck size. But we heard the Big Bang came from inside the Planck scale. You mean it's just conjecture that all came from inside the plack scale in the Big Bang and it's possible the entire universe with billions and billions of galaxies where maybe the size of an egg instead?
No, not the size of an egg.
FAQ: How big was the Big Bang?
There are two kinds of cosmological models. One is spatially infinite and always has been spatially infinite. The other is spatially finite and always has been. Current measurements of curvature are statistically consistent with flatness, which puts us on the fence between these two cases.
In either case, it's not valid to imagine the Big Bang as an explosion with a certain size that happened against a backdrop of empty space. In all cosmological models, the Big Bang is a process in which space itself stretches out. Every region of space on a cosmological scale increases its own volume over time. In the early universe, all of space was always uniformly filled with hot matter and radiation.
In the spatially infinite case, the universe has been spatially infinite at all times, so there is no way to measure the diameter or volume of the Big Bang with a real number.
In the spatially finite case, the universe wraps around on itself spatially, like a sphere. There is no edge. It does have a well-defined volume at any given time. According to general relativity, this volume approaches zero in the limit as time approaches zero. GR does not describe t=0 itself as a moment in time.
The above picture of the finite case would presumably be modified by quantum effects at early times. We don't have a theory of quantum gravity, but there is only one length scale that you can make by combining Planck's constant with c and G, and that's the Planck scale.
So in summary, there are three possible answers to the question: infinity, approaching zero, or the Planck scale.
A somewhat different question is the initial size of the *observable* universe. The current radius of the observable universe (i.e., the region of space from which light has had time to reach us since the Big Bang) is about 46 billion light years. (This is greater than the result you get by naively multiplying the age of the universe by c, because space has been expanding.) Yesterday the observable universe was smaller, and tomorrow it will be bigger. In all cosmological models (in GR), the radius of the observable universe for an observer at time t approaches zero as t approaches zero (although of course there were no actual observers present in the very early universe). A hypothetical theory of quantum gravity might change this answer to the Planck length.