Very significant that they've actually made one that works. That's a big leap. But as far as I know, we are still years behind being able to mass produce and implement them. There's a few reasons:
-
They are far more energy-intensive
-
They generate a lot more heat. If we solve 1. we might be able to solve this to an extent.
-
Theoretical designs are much bigger in size to electronic chips.
-
Due to size constraints they can process less information than electronic chips currently. While light is faster than electrons (and wavelength modulation could be used in the future to convey more complex information than 1s and 0s), photonic chips currently have less pathways than electronic chips, so less information can be processed simultaneously.
-
They require constant callibration to ensure the light source and receiver are perfectly alligned. This makes them very impractical to use for the average user. Even the vibrations inside a desktop computer, due to its fans, can be a problem. They are nigh impossible to use in portable devices, industrial machinery or vehicles.
-
They are currently really expensive to make, but I'm guessing that's because we are still prototyping. The materials and process can actually turn out to be cheaper than electronic chips.