Are these objects actually monopoles? Well, yes and no. They fall into an interesting gray area:
No, they are not the fundamental monopoles that Dirac proposed. They are not fundamental particles, but only quasiparticles arising from the dynamics of some substrate. In this case, the substrate is quite exotic: a spin ice, which is a kind of material (dysprosium titanate) with polar atoms arranged into tetrahedra.
OK, so they're not fundamental, but they're still quasiparticle magnetic monopoles, right? Sort of. These quasiparticles still have to obey the standard laws of electromagnetism, and those laws still forbid the existence of magnetic monopoles. Every magnetic monopole is actually a member of a monopole-antimonopole pair connected by a Dirac string. To quote the paper:
In general, it is of course well known that a string of dipoles arranged head to tail realizes a monopole–antimonopole pair at its ends. However, to obtain deconfined monopoles, it is essential that the cost of creating such a string of dipoles remain bounded as its length grows.
So this is the key innovation here. A normal magnetic dipole like a bar magnet can be thought of as being like a stick: it has two ends; if you break it, both pieces have two ends; when you wave the stick around, both ends wave around. But this system is like a rope: it still has two ends; if you break it, the pieces still have two ends; but when you wave one end of the rope around, the other end can remain fixed. So the end of a rope can act like an object independent of the other end.
This makes it a great model system for playing with monopoles, as long as you close your eyes and pretend the rope doesn't exist. But it does exist, Maxwell's equations are obeyed and all is well in the universe.
