Tuesday, September 23, 2008

23andMe a Good Thing?

I was reading Sergey's blog entry on Parkinsons. I am very sorry to hear that it is afflicting his family and may afflict him. I am hoping that he will be able to help turn the tide against problems like Parkinsons. He found out that he has a predaliction towards Parkinsons by leveraging 23andMe. I took a moment to check out 23andMe and it looks very interesting. I like the idea of a quick swab and being able to tell all kinds of things about your genes. Are you predisposed towards particular things? A few questions came to mind as I was thinking about this though. If you were predisposed to some issues, would you want to know ahead of time. I think that I would, but how much would it come to dominate my life and the life of my family? People are not very good at odds. If you end up with a 5% chance of having a problem, are you going to suffer your whole life waiting for the other shoe to drop?

The possibly more interesting question though, is how long before people start asking about the results of tests like these on insurance forms? Are you predisposed to Parkinsons? What do you answer? If you have not had the test, I think you might be in better shape than having had the test. Would having the test be enough to make you uninsurable? I can easily see premiums being hiked due to genetic predispositions.

We are entering an interesting age where we may know more about our futures than might be good for us. At some point, this will affect the coarse we choose for our lives and the costs of various goods and services. Now might be a great time to try 23andMe and lock in low rates for long term care insurance before they start to ask annoying questions. That being said, I went with AIG life insurance because they were the most stable thing since sliced bread. It is hard to tell who will be doing well in many years...



Monday, September 22, 2008

Contemplating Artificial Life...

I have long had a fascination with Artificial Life. My master's thesis at UCLA was evolving pack hunting and pack defense in artificial life forms. I thought it was an interesting project and it went well. I was quite surprised by the variety and completeness of the forms of hunting and defense that very simple organisms were capable of displaying. I was able to get my artificial life forms to display all of the kinds of hunting that were in a book on predatory behavior. My thesis was a parallel program written in C++. I am currently starting to port it to Python and I am going to present a talk on it at Py|Works "Exploring Artificial Intelligence with Python" . I hope that some of you can make it.

My ideal outcome would be to leverage my research and that of others to produce better, deeper, more realistic, and ultimately more fun creatures for games and education. Knowing how easy it is to evolve and/or run simple creatures that behave in amazing ways, I am constantly disappointed in the creatures that are out there in games today. I am also looking for people that would be interested in joining me in this pursuit. Do you love AI or online games? Would you love to make them better?

Let me know,
Jacob Taylor