One day some scientists worked out how to join her glass palace up with the glass palaces of other princesses to make silicon circuitry on computer chips.
"Did you see any diatoms?" asked R's Mummy when I enthused about the uni-cells (and the awesomely destructive rotifer) that I'd seen under the microscope in my 1st year Cell Bio lab. "What's a diatom?" I asked, so she showed me some pictures and described diatoms in such loving detail that they immediately became part of my personal mythology (I learnt a lot about passion for research from R's parents).
R's D, who also worked on diatoms, once described them as "siliceous micro-fossils", which rather lacks romance and (pick the geologist) also ignores the fact that there are thousands of diatom species still in existence, but does prove that diatoms rock!
I suspect the computer chip thing is pretty speculative (and maybe a bit of manoeuvring for grants), as what they have actually reported is identifying 75 genes involved in silicon processing in one species of diatom (Thalassiosira pseudonana) from a screen for genes expressed during silicic acid starvation. They've chosen 30 genes for further study, of which 25 have no similarity to known genes, so I'd guess they have a lot of knock-outs to make and/or proteins to express and a lot of characterisation to do before they fully understand the function of even one of these genes, let alone describe the process and learn to manipulate it in any meaningful way... So unless they have an enormous and unusually well-financed team, diatom-fabricated computer chips are probably a lifetime away. Still, it's a magical idea, and to me, it's ideas like these that justify pure research in fields as apparently esoteric as diatom genetics.