Always on information flow.

The New York Times ran an article about How Depressed People Use the Internet.  In a survey of undergrads at Missouri University of Science and Technology, the researchers determined depressive symptoms.  They then compared the results to data usage from the school's IT department.

Students who were depressed showed a pattern of internet usage that was heavy on messaging and data sharing, and low on staying in one place for any length of time.  Games and difficulty concentrating sounds a lot like depression itself.

The article wrapped up with some applications for the research, which should be read in light of mental health collapses and shootings like at Virginia Tech.  Their proposal:

"We hope to use our findings to develop a software application that could be installed on home computers and mobile devices. It would monitor your Internet usage and alert you when your usage patterns might signal symptoms of depression.  [...]Such software could also be used at universities, perhaps installed on campus networks to notify counselors of students whose Internet usage patterns are indicative of depressive behavior. (This proposal, of course, raises privacy concerns that would have to be addressed.)"

The idea of a program doing this kind of assessment on me is not as much of a non-starter as one would expect.  Do it.  Get the information and start compiling.

But tell me before or at the same time as you're telling a doctor.  Let me read the results along with the health practitioner.   This goes for all the measures we're developing with always-on medical assessments.  Tell me and let me digest the information.  The challenge will be making sure that the information is usable when it gets to me.

I look at the top of this page and see one word.  "Dashboard."  It's really a descriptive word wrapping a wholly difficult idea.  What is a dashboard but an output of filtering mechanisms from multiple sources.

It a descriptive word because, for most people over the age of 15 1/2, this is a concept that comes pretty naturally.  A lifetime of half-second glances, and we know wether to change our actions or to start worrying.

But so much of the sensor-to-analysis-to-output system in our cars is simply a black box.  Rumors abound and misunderstandings are easy.  For example, my check engine light was on for a couple of days.  It addled my mind because the car was running well and had just been checked out.

That is what must be avoided in a big data synthesis system.  It's the first rule of Wikipedia: citations needed.  Tell us where the data is coming from and WHY it is pointing in one direction or another.