There is a fundamental problem with IT that is not new to the digital calculating device, nor even to the human experience.  Look at the brain itself, and discover that there are challenges which IT is ignoring, and which will continue to lead to great peril.

Evolutionarily, the brain plan advances in starting new systems, but setting aside old systems, retaining their structure, but diminishing their usefulness.  The Thalamus, for instance, contains an extraordinary amount of structure which achieved very high functional activity in lesser vertebrates.  We have retained it, but delegate the mainline functions to a newer, higher, more accurate system.

We are not particularly troubled by the multi-tracking nature of pain.  The neocortex assembles the sensation of pain from widely scattered mechanisms, and coordinates it into a single concept of pain in our perception.  Unmyelinated Class C fibers have a non-localized, crushing pattern of sensation; the myelinated fast fibers have a more specific and “bright” type of pain, with pathways through more advanced systems.

The problem is in IT systems – unlike in evolution, where old designs might come in useful in the future – parallel tracks are NOT torn out, but added to.  The old systems are not unplugged – they are still used.

This is the disaster unfolding in ICD-10 before our eyes.  I find that ICD-10 sucks FAR beyond my expectations.  After using it, I am astounded to see that it is “informationally orthogonal” meaning that it adds no precision to the database, but may even, in circumstances, be more vague than ICD-9.

But ICD-9 has not gone away, and I expect it never will.  I expect this dual-coding track to remain perpetual – the requirements of BOTH coding systems.  Killing ICD-9 would require an intelligent, wide-ranging policy decision at a high level, with no immediate benefit to the organziation deciding to eliminate it.  So it’s going to be kicked down the road, and we will dual-code everything.  Maybe tomorrow I’ll get to the kvetch about ICD-10.