I think annealed fused silica should be used for almost everything including windows and dishes. It should also be etched nanosmooth by an etching liquid that has a contact angle of more than 90° with it. I once read on the internet that corundum is a covalent network. From https://chemistry.stackexchange.com/questions/67693/can-perfect-crystal-diamond-be-grown-from-the-molten-state-under-high-pressure, it can be inferred that since corundum has a localized electron pair at each bond, it can be considered a covalent network so the molten quartz should also be stored in a crucible made of a nanosmooth single crystal substance that's made of a material that's like corundum except that a small fraction of its aluminum atoms are replaced with silicon atoms making it very non-stick, it a vacuum chamber than slowly annealed then etched into the desired shape so that all bubbles will either expand and escape or dissolve and disappear from their own surface tension if they're microscopic preventing stress formed by the desire for the air in the bubbles to contract more than the glass creating stress where the bubbles are. Before that, solid particles can also be removed from the quartz my taking a ball liquid of quartz in the international space station, then repeatedly completely freezing it into a single crystal from the inside and completely melting it from the outside. The first freezing might not get rid of all solid particles and instead the crystal will grow around them but it will push some of the solution of those particles out making the crystal purer so that after it's melted again, what ever is left of those particles will dissolve into the molten quartz.
That way, it will be extremely resistant to being shattered by any object with a significantly lower hardness like a hammer because in can be inferred from the Wikipedia article Fracture mechanics that annealed fused silica has a much higher theoretical strength than is generally observed and it's because of microscopic surface cracks that any applied tension gets magnified at the tip of the crack greatly reducing the tension needed to fracture it. It can be inferred from my question at https://physics.stackexchange.com/questions/305330/how-does-an-infinitely-hard-tip-scratch-an-amorphous-brittle-material-when-it-sl that if you slide a tetrahedral diamond tip which was made by brittle fracture along nanosmooth annealed fused silica, you will reduce its strength by an ever so large factor because atoms are really small and scratching does not follow the square cube law because it's explained at the atomic level. Because of it's really high theoretical strength, even after scratching by a diamond tip, it won't feel like you take zero force to pull apart a sheet of fused silica by hand after you scratch it with a diamond tip, and will not break from the tension caused by its own weight. Although a Prince Rupert's drop can't be shattered with a hammer because a hammer can't introduce a scratch because its surface is harder because it's under extreme compression, it only has to be scratched very lightly by a diamond tip in order to lower its compression strength below the compression stress of its surface triggering a self sustaining crack propagation that makes it instantly explode into tiny pieces, which is another reason to use annealed fused silica. I also once read on the internet that fused quartz has a very low linear expansion coefficient of 5 * 10^-7 K^-1, and will not crack if it's poured into water from the molten state. That's about 20 times lower than that of soda lime glass.
According to my answer at What are some plausible super materials?, non-stick lonsdalite with some of its carbon atoms replaced by nitrogen atoms might be a much more desirable material to use for window because the strength of a surface of it parallel to its cleavage plane probably can't be lowered by more than a factor of 2 by scratching because it doesn't support crack propagation from a surface parallel to the cleavage plane at all except for when the entire sheet is under extreme tension, and a nanosmooth surface parallel to the cleavage plane can be made by brittle fracture, and it has a high thermal conductivity and melting point preventing a sharp tip of most other materials from melting some of its away with the heat of its sliding friction.