Honey Monkey!

Do you know who Steve Gibson is? He is the author of Shields up! He is also the author, or at least co-author of “Security Now” a new podcast that showcases security content.

--
I was surfing around the net and found an interesting web site about Microsoft’s Honey Monkey research program.

ftp://ftp.research.microsoft.com/pub/tr/TR-2005-72.pdf

The Honey Monkey Research program is Microsoft’s way at taking a proactive approach to fighting spyware. Basically, the project involves many computers setup to mimic basic users surfing the web. The computers log and record any URL’s that attempt to infect a pc with any malware/spyware etc.


Ironically, Steve Gibson’s podcast was about this very topic. Give it a listen.

Digg This... Mice Regenerate! Forget the cloning of humans just regenerate!

Scientists have long known that less complex creatures have an impressive ability to regenerate. Many fish and amphibians can regrow internal organs or even whole limbs. "

Pretty freaky! I found this article while surfing Digg.com.    

Sorry, the link above is no longer valid. It was pretty interesting though!

Discreet Mathematics - Random Numbers

Back in College, I wrote the paper below about Random Number Generation. Personally, I really enjoyed creating this paper and thought I'd like to share it with you. You should see a link to a spreadsheet that accompanied this document. Enjoy!

Discreet Mathematics
"Random Number Generation"
Jeffrey R. Byrns


Many people take for granted how computers generate random numbers for any particular program that they may be using. What people generally donÂ’t understand, or even realize is that computers are incapable of creating random numbers. They do however generate numbers that appear to be random, or arbitrary. Many forms of random number generators are in practice today, all of which are nothing more then algorithms with math equations behind them.

One of the most basic, and most popular, forms of random number generation is the algorithm developed by D. Lehmer in 1951 (Sedgewick, p. 511). It is called the linear congurential method. We can generate a set of random numbers with the below algorithm.

A [0] = Seed
b = Constant
M = Constant

For i := 1 to N do
a[ i ] := (a[ i -1] * b + 1 ) Mod M

In order to get a random number we take the previous random number (a[ i -1] ), and then we multiply it by ( b ) and then add 1. Take the result of this and get the remainder when we divide it by M (Mod M). It can help if you think of a[] as being an array.

One of the first questions to be raised is what should the constants for b and M be? D.E. Knuth (born in Milwaukee, Wi) came up with a few rules governing the choices for b and M. According to Knuth, M should be a very large number. It is convenient for computer processing to make M a power of 2 or 10. The constant b should be smaller than M. It shouldnÂ’t be too small however. A good rule of thumb is to use a b with one less digit than M.

Below is an example of the above algorithm applied in an Excel spreadsheet. A scatter plot has been provided to show the “Randomness” of the output. In the example a “Seed” of 15 was used, b is set at 1277, and finally M is set to 131072.



A spreadsheet has been included with this report, Random.xls. This can be used to demonstrate the above example

As you can see above, once you know the “Seed” value you can reproduce this seemingly random number scheme. Seed values can be created by any number of ways. Visual Basic uses the randomize statement to generate the Seed value. In turn it uses the internal clock to create a seed value.

You may have noticed that the resulting numbers generated can be extremely large. Usually, these types of random numbers aren’t what are required. Often what is required is a random number between one and whatever. This can easily be achieved by adding one extra step, Mod it. Probably the best-known application of using the mathematical Mod function is to “Wrap Around”. For instance, lets say you wanted a number between 1 and 10. You could take any of the results from the linear congurential method and mod it by 10, and then add 1. You can be certain that your results will be between 1 and 10.

Exploring other methods of random generators can be done by looking at Microsoft’s spreadsheet application, Excel. Looking under the Data Analysis menu, one can find “Random Number Generation” in a list of analysis tools. Once opened you will find that Excel provides seven different methods of generating random numbers. The web site http://www.vuse.vanderbilt.edu:8888/es130/lectures/lecture7/random.html provides a lot of insight on how to use this functionality. The seven different methods are; Uniform, Normal, Bernoulli, Binomial, Poisson, Patterned, and Discreet. Bernoulli and Binomial will be the remaining focus of this document.

Excel Bernoulli method:

Using the Bernoulli method will result in either a 0 or a 1, pass or fail if you will. It is based on the principle of the Bernoulli Trials. Giving the probability of success at 50% or .5 and 10 trials should reveal a result of 5 1Â’s and 5 0Â’s. The table below demonstrates the output of Excel for this demonstration. An excellent application for this method would be to reproduce a coin-flipping scenario. Given only a pass or fail, any application that requires only two outcomes would be a candidate for this method.

Excel Binomial method:

The Binomial is an expansion of the Bernoulli method. Using a variable number of Bernoulli trials, you can sum them up. In the table, an example of using the Binomial method is displayed. In this case, the probability of success is .5 and the number of trials is 100. Notice the output for 10 separate generations. The numbers hover around 50, or half the amount of the number of trials.

As you can see, computers donÂ’t generate random numbers; they generate numbers that appear to be random. In a sense, one could say that a computer generates arbitrary numbers not random numbers. Arbitrary meaning that what we require is any number, and it doesnÂ’t really matter what is returned. The most common algorithm used to date is the Linear Algorithm, and it is incredibly easy to implement. The math involved is very rudimentary and the steps involved are few. Although the focus of this document has focused on the Linear Congurential method, other methods of creating random numbers exist. Using ExcelÂ’s data analysis tools, one has a plethora of choices when choosing a randomization method.


References

Dewdney, A. (1989). The Turing Omnibus. Rockville. Computer Science Press.

Sedgewick, R. (1988). Algorithms. USA. Addison-Wesley Publishing Company, Inc.

Vanderbilt University. (Unknown). Random Number Generation. Available: http://www.vuse.vanderbilt.edu:8888/es130/lectures/lecture7/random.html

Skype International

Well, I had my first international Skype call yesterday! One of the projects I've been working on at work is to roll out a centralized support framework for a certain system we use. I have chosen to use Skype for overseas communication between them and myself. The conversation I had with one of my associates in Brazil went flawlessly. Ironically, the international call was the best quality Skype call I've had so far.    

4 Meters!

One of my good friends, Brad, kindly let me know today that Germany is under water... "4 Meters" under. You can tell that Brad is one of those people that just adores Germany. I mean really... what tall, blonde hair, blue eyes, German - American male says 4 meters? Real Americans say 12 feet! (Actually 13.123 feet)


Hmmm, for some reason I feel like playing Castle Wolfenstein!

All in good fun Brad!

Insta Blog! I knew that cell-phone camera would have a use someday!

At .25 cents a pop I can send stupid pictures like this directly to my Blog!

Woohoo!

Call For Help!

Hey, G4 is going to do “Call For Help” again. According to the TWIT website, http://thisweekintech.com/node/3920, G4 has purchased CFH for viewing in the US. This is great news for me… but not for my wife! Get your TIVO setup, it is on at some really sucky time slot.

Google is speaking to me!

Out of the blue today one of my friends, Jason, sent me an invite to Google's new voice service. Being one who loves Google, and especially loves VOIP (voice Over IP), I was excited to give it a try. The interface is like G's main page, simple and easy to use.

My first experience with the new toy was miserable! My 1 minute conversation with Jason had decent sound but a 10 second delay. Jason later replied, in chat, that I was choppy and couldn't be understood.

Later in the day, I was sent another invite to use the GV service. This time it was an old co-worker that I hadn't heard from in quite a while. I accepted Josh's invite and within seconds the software began to ring. This time, WOW... How do they say that in Mortal Combat?? "Flawless Victory!". The conversation had zero lag and perfect audio.

I guess only time will tell how well Google's new service pans out.

Picture of the kids helping me wash the Harley 

Remember ZDTV? How about TechTV? Well, for those that do, the guys from those shows have gotten together once again. They are now doing TWIT... "This Week In Tech". At this time it is a very popular podcast. http://www.thisweekintech.com/

If you like listening to a bunch of guys talk about Tech stuff.... Give it a listen.