I've just been informed that the web streaming address has changed. I guess that's the price of being at the experimental edge. The web streaming address is actually
I saw the laser boot up in MSWindows. Yikes! Blue screen of death with a laser pointed in my eye? Pretty scary!
But I don't regret it.
Best advice I got while researching concerned (a) doctors and (b) machinery.
(a) Doctors. 1. Ask how many procedures they've done. The number of operator errors and procedural problems goes down dramatically after the first 1000 procedures on a particular machine. If the surgery has just got a brand new machine, it's time to wait for a few others to do the beta testing.
2. Ask what percentage of cases the doctor advises against the procedure. This will tell you about how conservative they are about your risk. The more people they turn away, the more they care about the long term safety of your cornea. Corneal bulge has been known to occur when the cornea is ablated too thin. There are varying opinions in the literature about what is too thin. I'd go with conservative!
3. Ask who they refer their problem cases to. Generally there is one doc in the area that does the clean-up on the problem cases. I drove 50 miles to go to the guy that others referred their problems to. He was also the most conservative. Surprise!
(b) Machines.
Google is your friend. New machines are coming out all the time. There is a difference in (a) the smoothness of the resultant cornea and (b) the diameter of area that can be ablated. Do your homework.
With all that said, I'm in my 50s. My vision was stable and was extremely near sighted and astigmatic. I can sometimes see halos around point sources at night, but basically I'm now 20/20. I have no trouble driving at night. I still need reading glasses like most people over 50. My enjoyment of the 3-D world around me is much greater now. Even 2 years later I continously marvel at simple things like trees and buildings and pretty girls.
I've also found that Bausch & Lomb ICaps (one cap a day) do a great job of alleviating the dryness problem caused by the reduced blink reflex post Lasik.
With an arbitrary number of parameters we can fit a curve to anything and it doesn't help at all. True.
However, that does not mean that useful knowledge does not arise from mathematical modeling. To give a different spin on things, consider a bird. A perfectly self-consistent explanation for why a bird can fly and we can't is, "God made it that way." But, given some appropriate experiments and a mathematical model for their resulting aerodynamics, one can project that it is possible to build an airplane. So often there are many explanations. But it is easier to tell what will work when you try to implement a mathematical model than when you implement an explanation.
Now, we could rest with the explanation that, "Men want sex and women want security." But it is unlikely that things are nearly so simple. And just like the "God wants it that way" explanation, these types of explanations are self consistent, but self defeating. The true use for mathematical models of human behavior is to help us understand how the behavior of a complex system such as interpersonal interactions might be usefully changed. In other words, the hope for such models is to allow better learning and adaptation by the individuals involved. And, in cases where couples seek help, how a therapist might usefully, rather than destructively, intervene.
Visualization of data is far more common than sonification. Both techniques have their uses. The visual system is good at spatial relations. Good visualization techniques take into account the strengths of human visual perception. Good visualization is a lot less common than it ought to be. We are awash in USA Today style visualization on the web. For instance we are very good at detecting differences in texture. Visualizations as texture are rarely used, but they can provide information about high dimensional relationships that are just impossible to convey in most of the commonly used methods.
Sonification is best used when high dimensional temporal relations are what is of interest. The human auditory system is adapted to extract high dimensional covariances in sonic signals. This is very difficult to convey visually.
While I was impressed with the vision and scope of the B5 series, I'm revolted by the notion of sending $20 to a right wingnut group whose anti-open source work couldn't be more blatant. I refuse to view anything presented by "real" networks.
"That's Big-Boot-Ay" -- John
The Psychology Department at Notre Dame has been attempting to hire such a person for over a year now. It isn't that we can't pay a competitive salary, it isn't that the work would be boring, it isn't that there aren't a lot of people for whom this job would be perfect.
Our problem is we just can't reach the people who would be interested in this job. The traditional methods of advertisment and head hunters are all geared towards finding someone who is going into the business world. I'm guessing that people who would want the job just don't look there, because we haven't had so much as a nibble from anyone who is qualified. Further complicating the problem, it is considered to be poor practice to hire from within the University student population.
The people who would be most interested are members of communities such as slashdot where "trolling for candidates" is generally looked on as something disgusting. We haven't broken these taboos and so having this opportunity to post is somewhat liberating.
I know the job is interesting, because while we are lacking help, I have some of the best and brightest ND engineering students coming over to my lab to work for fun. We do, for instance, 3-D motion capture of human subjects interacting with each other and with animated avatars. It's good toys and interesting projects.
The problem as I see it is how to get those individuals for whom these research jobs would be a "dream job" together with those people who have these dream jobs available. After reading through the comments, I still don't have an answer to that question.
However, [SHAMELESS PLUG] for those who are interested in an all-around programming position primarily working in the interface between cognitive science, robotics, perception and memory, and who are also interested in working as we used to say, close to the iron, my email is sboker@NOSPAMnd.edu [/SHAMELESS PLUG].
Slashdot Mirror
I've just been informed that the web streaming address has changed. I guess that's the price of being at the experimental edge. The web streaming address is actually
rtsp://diatom.uchicago.edu/vibrio-live.sdp
I saw the laser boot up in MSWindows. Yikes! Blue screen of death with a laser pointed in my eye? Pretty scary!
But I don't regret it.
Best advice I got while researching concerned (a) doctors and (b) machinery.
(a) Doctors.
1. Ask how many procedures they've done. The number of operator errors and procedural problems goes down dramatically after the first 1000 procedures on a particular machine. If the surgery has just got a brand new machine, it's time to wait for a few others to do the beta testing.
2. Ask what percentage of cases the doctor advises against the procedure. This will tell you about how conservative they are about your risk. The more people they turn away, the more they care about the long term safety of your cornea. Corneal bulge has been known to occur when the cornea is ablated too thin. There are varying opinions in the literature about what is too thin. I'd go with conservative!
3. Ask who they refer their problem cases to. Generally there is one doc in the area that does the clean-up on the problem cases. I drove 50 miles to go to the guy that others referred their problems to. He was also the most conservative. Surprise!
(b) Machines.
Google is your friend. New machines are coming out all the time. There is a difference in (a) the smoothness of the resultant cornea and (b) the diameter of area that can be ablated. Do your homework.
With all that said, I'm in my 50s. My vision was stable and was extremely near sighted and astigmatic. I can sometimes see halos around point sources at night, but basically I'm now 20/20. I have no trouble driving at night. I still need reading glasses like most people over 50. My enjoyment of the 3-D world around me is much greater now. Even 2 years later I continously marvel at simple things like trees and buildings and pretty girls.
I've also found that Bausch & Lomb ICaps (one cap a day) do a great job of alleviating the dryness problem caused by the reduced blink reflex post Lasik.
With an arbitrary number of parameters we can fit a curve to anything and it doesn't help at all. True.
However, that does not mean that useful knowledge does not arise from mathematical modeling. To give a different spin on things, consider a bird. A perfectly self-consistent explanation for why a bird can fly and we can't is, "God made it that way." But, given some appropriate experiments and a mathematical model for their resulting aerodynamics, one can project that it is possible to build an airplane. So often there are many explanations. But it is easier to tell what will work when you try to implement a mathematical model than when you implement an explanation.
Now, we could rest with the explanation that, "Men want sex and women want security." But it is unlikely that things are nearly so simple. And just like the "God wants it that way" explanation, these types of explanations are self consistent, but self defeating. The true use for mathematical models of human behavior is to help us understand how the behavior of a complex system such as interpersonal interactions might be usefully changed. In other words, the hope for such models is to allow better learning and adaptation by the individuals involved. And, in cases where couples seek help, how a therapist might usefully, rather than destructively, intervene.
"Laugh while you can, monkey-boy" -- John Wharfin
Sonification is best used when high dimensional temporal relations are what is of interest. The human auditory system is adapted to extract high dimensional covariances in sonic signals. This is very difficult to convey visually.
For more information on sonification, try: (1) Scientific Sonification (2) ICAD -> International Community for Auditory Display
I teach a course on exploratory data analysis at Notre Dame. Go to the bottom of this page for the link to the course.
The MS mascot is a BUG!
While I was impressed with the vision and scope of the B5 series, I'm revolted by the notion of sending $20 to a right wingnut group whose anti-open source work couldn't be more blatant. I refuse to view anything presented by "real" networks. "That's Big-Boot-Ay" -- John
Our problem is we just can't reach the people who would be interested in this job. The traditional methods of advertisment and head hunters are all geared towards finding someone who is going into the business world. I'm guessing that people who would want the job just don't look there, because we haven't had so much as a nibble from anyone who is qualified. Further complicating the problem, it is considered to be poor practice to hire from within the University student population.
The people who would be most interested are members of communities such as slashdot where "trolling for candidates" is generally looked on as something disgusting. We haven't broken these taboos and so having this opportunity to post is somewhat liberating.
I know the job is interesting, because while we are lacking help, I have some of the best and brightest ND engineering students coming over to my lab to work for fun. We do, for instance, 3-D motion capture of human subjects interacting with each other and with animated avatars. It's good toys and interesting projects.
The problem as I see it is how to get those individuals for whom these research jobs would be a "dream job" together with those people who have these dream jobs available. After reading through the comments, I still don't have an answer to that question.
However, [SHAMELESS PLUG] for those who are interested in an all-around programming position primarily working in the interface between cognitive science, robotics, perception and memory, and who are also interested in working as we used to say, close to the iron, my email is sboker@NOSPAMnd.edu [/SHAMELESS PLUG].