Can an insulator have magnetic poles? Is it possible to make an insulator have two poles like a magnet?
If yes, how to make it?
 A: In principle I can see no reason why an insulator cannot be magnetic, but in practice all pure materials that are ferromagentic seem to be conductors. This includes iron, nickel, cobalt and Gadolinium. Even the known plastic magnets are conductors or semi-conductors ( see http://en.wikipedia.org/wiki/Plastic_magnet )
Magnetism can be produced by circulating electric currents round a coil. In this case the material clearly has to be a conductor. In ferrous magnets the magnetic field is produced by aligning the spins of the atoms, so the current that produces the magnetic field is really just the electrons orbiting the nucleus. It will only be an electrical conductor if the electrons are free to move between atoms. The properties of the materials that make it possible to move the atoms into alignment coincide with the properties that make electrical conductivity possible, but I don't think there is any general law of physics that says this has to be the case.
One solution would be to make a mixture of fine iron filings in a plastic insulator. The iron filings could be magnetised but the plastic would make the material an insulator.
Neutrons have a small magnetic moment so in principle you can line them up in a magnetic field and switch off the field to leave a residual magnetism. Since they are neutral they cannot conduct a current. That may not count as an insulating material in the normal sense.
A: you would have to explain what you mean by the poles. Insulators don't inherently create any inherent kind of an "insulator field" around them, so this non-existent field cannot have any poles, either. An insulator is a material that doesn't conduct - heat or electricity.
Another thing is whether insulators can behave like dipole sources of the known fields such as electric and magnetic fields. 
First, permanent magnets I know of always have to display ferromagnetism or superconductivity. Ferromagnetic materials are similar to iron; superconducting materials are ... conducting as well. So permanent magnets have to be conductors, I believe. After all, if one is able to change the magnetization, it means that the angular momentum of the electrons or their spin is changed as well. If this is possible in an atom, the electrons may effectively jump to an adjacent atom, too, I think. But even though this proof is not complete, I do think that it's true that permanent magnets have to be conductors just like iron. As far as I know, there is no known counterexample.
Another thing the insulators could have are electric dipoles. Yes, insulators may act as electric dipoles, but only in an external electric field: the relative position of electrons in each atom moves a little bit relatively to the nuclei. This effect is known as polarization.
They don't stay electric dipoles, however. They are not permanent dipoles. If a material is not a conductor (of electricity), it means that its electrons can't move much, so they're sitting in their molecules and in every molecule, there is an energetically optimal position where they want to sit. So I believe that (good) insulators can't be (good) permanent dipoles.
Best wishes
Lubos
