I doubt a heat source can achieve nm scale density. On the other hand patterned media requires special lithography tools, like imprint or ebeam for disks not wafers. If they have something like self-assembling or self-patterning patterned media, that might work.
To some extent the lead is buried here. Spinning-media disks are getting to the point where they are too big. I am finding myself nervous about trusting 2 GB or more of data to a single mechanical system. Right now I am much more interested in SSD interfaces which don't treat these devices as spinning media. These standards will open up a major bottleneck in system performance. The IDEMA work towards that mentioned in the last couple of paragraphs is critical towards that end.
I am very interested to see how this research group develops. With the stated scale of costs associated with the next technological leap, this kind of cooperation may be the only way to afford reaching the next level of storage density. This group may be able to fund research that significantly shortens development time. Then the individual manufacturers can continue to differentiate themselves by how they implement the new technology (managing costs, implementing hardware and firmware designs, etc.). I remember when I worked for Vertex Peripherals back in the mid-80s hearing about Perpendicular recording. It took over 20 years before it became a reality. Perhaps the next step can be taken in a much shorter time frame.
While I still feel confident in spinning media disk, I believe SSD is the technology that is the most interesting right now. Large OEMs are already starting to utilize this technology and it seems to be gaining popularity every quarter. I would expect to see a large growth of this particular type of storage in the coming months. I don’t think we will see the impact of this particular research group for some time.