work

Today, before lunch, met with M in the electronics workshop to test a photodiode. We didn't get around to making any measurements, but instead I explained Opamp circuits. After lunch: nothing useful.

Tomorrow: Mirko's thesis defense. I need to email KT about the magazine article, test the photodiode with M, finish the Matlab model of the TTFSS. But of course there will be beer-drinking in the afternoon, after the defense.

Some other things that need work:
* coil driver
* electrostatic drive
* Wenzel box

The de-friending of Mitt Romney

One of my facebook friends posted a status update observing that, if you go to Mitt Romney's facebook page, note the number of "likes", and then re-load, you can see the number of "likes" drop in real time!

This was a complete shock to me, that disappointed voters might express themselves by "unliking" their erstwhile favorite candidate after the election, and at a remarkably steady rate.

I wrote a stupid little shell script to periodically poll Facebook for the number of "likes" on Romney's page. The data from the last several days is plotted above. As you can see, Facebook users—presumably disappointed Republicans— are de-friending Mitt Romney at a consistent rate of around 10,000 defriendings per day.

There is also a clear daily modulation to the dislikification rate. I conclude that these fairweather friends do not defriend Mitt while they are sleeping, but once they wake up again in the morning, they get right back to pressing "unlike".