The problem with this is that, even though it can store a massive amount of data, the data retrieval would be relatively slow. Edit: Never mind, they seem to be making advances in that area with "random access" rather than entire sequencing... This is rather exciting.