There is always something new in science, this article caught my eye...

http://www.physorg.com/news201795438.html

I've thought of making a true random number generator from radioactive elements for years. There are a lot of potential sources for a hobbyist that are mostly harmless. Granite, for example, is mildly radioactive.

 

Very interesting article. It will be interesting to see how this pans out. I'm sure most folks would tend to believe an instrumentation problem (as I do), at least until enough (tens or hundreds) of others have independently confirmed the rate changes.

If it is real, then it's genuinely new science and we'll learn something interesting about the universe we live in.

 

That is an interesting article. Plus it's easy to confirm independently.

For some reson, I think of Isaac Asimov's thiotimoline, an fictional substance that dissolves exactly one minute before water was added to it. Does a drop in the decay rate of any substance mean a solar flare must follow?

You are mildly radioactive. There is a methodology of determining the fat in your body by comparing the emission of particles from the breakdown of potassium and calcium in the body.

 

random.org uses atmospheric and radio noise. This actually seems like quite a low cost way to generate random numbers. Get 6 radios, tune them to a blank spot on the dial (preferably outside commercial FM or AM if you can modify them to do so) and choose the 'most random' between the 6 to get your bits. Or, turn the white noise into a bitstream and XOR the bits together.

Also, Linux and a few other operating systems have true random number generators in them (/dev/random). The numbers are sourced from peripheral noise, but it's a slow source (1-3 blocks every second, each block about 5-10 bytes.)

 

...and choose the 'most random' between the 6 to get your bits.

I suspect each of the six should be random or none are. Remember that episode from Numbers about random numbers? The punch line: what appears to be "most random" is probably not random.

John

 

You can base your random selection on analysis of each stream of data; whichever has the closest to a true random distribution (that is, equal probability of either bit, equal average distribution over time, expected results for Monte Carlo simulations etc.)

 

The easiest to get sources of radioactivity are salt-free salt, because of the weakly radioactive potassium-40; a Coleman lantern mantle, which contains thorium nitrate; and the active surface from a smoke detector (technically illegal), containing americium-241. Handle the latter two with disposable gloves.