"Number of characters gives little information about the complexity of a program. Functions is a better metric."
When the human genome was sequences (it was a ten year project), lots of pseudo-philosophers predicted the number of genes that would be found, with numbers ranging from thousands to millions. The genome in hand, we found about 25000 genes, which make up a few percent (1.5-3%) of the genome. We then looked more closely, and now we can't even agree on the defintion of "gene" anymore, which is one reason why I can't give an exact percentage.
It turns out that one gene can produce more than one protein, or sometimes less. There are lots of genes whose only function (assuming "function" is even well defined) is to regulate other genes, a bit like a program could have functions of higher order. We're slowly grasping the fact that we have no idea how complex the genome and the machine interpreting it really are. Number of genes really is no metric at all.
When the human genome was sequences (it was a ten year project), lots of pseudo-philosophers predicted the number of genes that would be found, with numbers ranging from thousands to millions. The genome in hand, we found about 25000 genes, which make up a few percent (1.5-3%) of the genome. We then looked more closely, and now we can't even agree on the defintion of "gene" anymore, which is one reason why I can't give an exact percentage.
It turns out that one gene can produce more than one protein, or sometimes less. There are lots of genes whose only function (assuming "function" is even well defined) is to regulate other genes, a bit like a program could have functions of higher order. We're slowly grasping the fact that we have no idea how complex the genome and the machine interpreting it really are. Number of genes really is no metric at all.