The Limitations on Engineered Biological Systems
Specifically, the "path dependency" of the development of the human bio-organism--the complexity of interacting gene systems--makes it almost impossible to make significant changes in the genome that will be more beneficial than detrimental. All of the "legacy code" of the human genome is extremely resistant to intentional tampering, and makes it far more likely that any given intervention will be harmful rather than helpful in achieving a biological singularity.
I encourage everyone with an interest in the biosingularity, and the genetic augmentation of human intelligence and lifespan, to read John Smart's well-reasoned essay.
In the coming weeks and months, I will be dealing with the issues in more detail. I am in full agreement with the basic outline and most of the main points of Smart's argument. It is likely that essentially immortal machines will acquire many of the strengths of human cognition, and human mobility, before mobile and mortal humans gain long life (200+ year lifespans) and superintelligence.
It is likely that cyborg augments with neural and vascular interfaces will extend human functionality and lifespan before humans gain the mastery of their own highly complex genomes.
But I feel that there are some loopholes in the legacy code. Recent discoveries of epigenetic mechanisms including "non-coding RNA's" and other means of influencing "families" of gene expression from "the outside", suggests that it is not necessary to change the actual genome to any great extent in order to achieve significant improvements in targeted human functions.
All futurists who write about the genetic improvement of human existence, including myself, need to take into account John Smart's clear and cogent cautions on the dangers of trying to change the human genetic design.