It is much easier these days to invoke informational models of DNA as a kind of encoding. Just consider that real working DNA computers have been successfully designed and built. But biological DNA evolves, the genes are a relatively small number of bases compared to large sections which don't appear to encode genes, but do have some use. This partly-useful DNA has to be evolutionary, an advantage, or it wouldn't be (even partly) useful. Moreover, the DNA replication process has viruses to contend with, and here again evolution has produced cells that survive by integrating the viral gene advantageously. Evolution also steps in when errors aren't repaired, these ones might be a bad mutation and die out, a good mutation and survive, or have no effect. But we still don't know much about this so-called "junk DNA" except it isn't junk, and there's a lot of it. I speculate that it's connected to the way viral resistance evolves which is basically cells learning to slow single-gene replication and preventing the formation of viral particles. So the virus "family" is actually a major evolutionary driver, it "infects" all other forms of life from bacteria to us mammals. I think that understanding junk DNA is going to show that it's really about how what we call life, processes genetic information, and that viruses that can invade the system and hijack the code will be shown to be absolutely necessary for life to exist as we know it, Jim. Which is to say, viral forms of "encoding", are advantageous for evolution in general. Otherwise they would have died out and we wouldn't find much evidence in the DNA record.