[ExI] good bexarotene article

[spike]

>... On Behalf Of BillK
Subject: Re: [ExI] good bexarotene article

On Sun, Feb 12, 2012 at 5:08 PM, spike wrote:
>>... BillK, if we manage to help Snowmed or something analogous to it go 
> mainstream and it turns into a really useful life extension tool, it 
> could be the very most important achievement the transhumanist 
> movement has ever accomplished... have the software 
> reduce all that to a digital signature and drop it into Snowmed... spike


>...Snowmed is already in widespread use around the world. ...The other
problem is that these medical records are scattered around in ... insurance
systems... BillK

_______________________________________________

Ja, BillK, another aspect you almost touched on here is that for medical
insurance companies, the understanding of the medical cost and risk
structure of a client as a function of apparently unrelated variables is
their value.  They are motivated specifically against their clientele
understanding their own risk structure, for the insurance company's bread
and butter is in having people bet against their own health.  What I have
proposed would largely obviate medical insurance companies, or would reduce
their profitability.  For that reason, we could imagine a class of people
who are motivated to intentionally degrade a database of this nature, even
if we are successful in creating it.

So we need to design in some tolerance or protection against those who would
intentionally degrade a database.  This would mean loss of privacy: all
records would need to be traceable to a particular patient, as opposed to a
privacy-robust system which would allow patients to enter anonymously and
enter their data in private.  There is a name for this where two variables
generally work against each other (anyone know?)  In my world of flight
controls, the variables are stability vs maneuverability.  In the medical
database world it would be patient privacy vs corruptibility of the
database.  

To achieve robustness against corruptibility, we would be back to making the
database volunteers willing to reveal everything about themselves, which
selects a particular cross section.  This limits usefulness, to some degree.
We could have two databases, one with traceable volunteers, the other with
untraceable corruptible database.  Then we compare the two.

spike