The article that this article cites doesn't go into what exists and what is new and what the implcations are most likely because the author doesn't have the background. So this article reflects that. The original article in Nature makes no false claims AFAIK.
Multiple wavelengths of light on a fiber (or core in this type of fiber) is wave division multiplexing, in use since mid 1990s. Optical amplifiers is similar timeframe. Multiple cores per fiber is only applicable to very short distances (relative to size of the earth) and the original paper in Nature points this out. The longest production DWDM (dense WDM) as of about 2010 is US west coast to Australia without stopping for signal regeneration in Hawaii. That is 6,000+ km with fiber and amplifiers at about 100 km intervals (I think). I don't know the configuration but I think it is 160 waves at 100 Gb/s each but may be higher. Infinera gear on either end. Details probably still burried on the web site blog. One core because the power to drive 100 km and keep signal integrity for that many amplifications would melt the fiber if multicore.
They also didn't invent photonic integrated circuits nor are photonic integrated circuits new. Infinera was the first to use them commercially, shipping first product in 2005. Current product 800 Gp/s per chip pair (transmitter chip and receiver chip), soon 1.6 Tb/s, planned 4 Tb/s. But this is for ultra long haul with claimed 10,000 km reach. Heat is a huge problem in the transmitter chip as well as for the fiber itself.
What is new and the paper in Nature is clear about this from the title, abstract, etc, is the source laser. They don't use a single laser per wavelength or laser array (small number of lasers) but rather use a single polychromatic laser on a chip and a microcomb ring resonator. The paper in Nature points out this works for very short distances (relative to earth, therefore the Internet) and their demo is 7.9 km. This limitation is due to be very low power produced this way. That is not to say it has no use. It might be great for data center use where multiple building nearby is all the distance needed.
It is unfortunate when a one mainstream publication botches the details and imagines something revolutionary and that every bit of technology mentioned is new and unique (because they never heard of it) when in reality the contribution is useful, very impressive, but much more limited than the mainstream publication imagines it to be. It is worse when the mainstream publication that has no idea what they are talking about becomes a cited source all over the Internet. None of this flawed reporting is intentional. (And maybe I got some facts wrong too). It does remind me a bit when TV news anchors tried to explain what the Internet was in 1994 and 1995 or worse yet tried to give the layman's version of how the Internet works and completely botched it (and now famously). Lower consequences and lower exposure here. Now someone please explain to Redit and elsewhere.
Facebook
Twitter
Instagram