The internet and World Wide Web have become invaluable communication tools to our society. Though it was relatively unheard of 15 years ago, it is now nearly inconceivable that any school project, business transaction, or teenage masturbation could take place without it. Every day, over 75 million people worldwide use the internet in some form or another, resulting in the transfer of almost 150 terabytes of data.
This information exchange is far from free, however. Last year alone, over $12 billion dollars were spent on the servers, switches, routers, and fibre optic channels necessary to support the data flow. With internet traffic increasing at an exponential rate, these numbers will only increase. Any attempt to curb the world's insatiable demand for information or pornography will surely meet with failure. Therefore, the only effective way to reduce internet traffic is to employ new methods of data compression.
There are several existing methods of compressing binary data. Run Length Encoding searches for continuous streams of 1's or 0's, and compacts them together. More sophisticated algorithms, such as LZW, search for frequently repeated patterns (such as the word 'the' in a novel) and compress them together. As good as these algorithms are, they can only achieve about 2:1 compression ratios, and only on certain types of information. Most importantly, they do not attack the biggest source of internet traffic:
Recent studies show that 73% of internet bandwidth is spent on pornography. Of this data, nearly half is dedicated to the female nipple. Clearly, any compression scheme that can reduce this high traffic area could save billions of dollars every year. JPEG (jpg) image compression is the best solution to date. It offers lossy image compression ratios of nearly 10:1. Though this was revolutionary when it was created, it is no longer an effective solution: 1/10th of an astronomical number is still too big. Clearly, a better compression method is needed.
At 72 dpi, a standard 1" nipple requires 72 * 72 = 5184 pixels. (While the proverbial 'silver dollar' nipple can require significantly more.) One byte each is needed for the red, green, and blue (RGB) channels, resulting in a total size of over 15,000 bytes. JPEG compression can reduce this to approximately 1500 bytes. A significant savings, but still not good enough.
We need to step back and see what we are trying to do: There are only about 2 billion women in the world, totaling approximately 4 billion nipples. We can, therefore, uniquely describe any nipple on the planet simply using a number from 0-4,000,000,000. Basic algebra tells us that this number can be encoded using only 32 bits, since 2^32 = 4,294,967,296. This revolutionary theory tells us that we can now encode a nipple using only 4 bytes of information, representing a compression ratio of almost 4000:1!!
The astute reader will notice that this does not take into account the nipple's surroundings, state, or spatial position. These can easily be addressed by adding one byte each for scale, orientation, skew, and percentage erection. Even still, the total size of the compressed data is only 8 bytes, providing a 200:1 improvement over the next best compression scheme.
Figure 1:
One limitation of this scheme is that it requires a comprehensive nipple database to be stored locally on every PC. (Many would argue, however, that a similar database exists on most teenager's computers.) Though the 100 terabytes required for this list may seem inconceivable today, by the time this worldwide system were put into place, such hard drives would be commonplace. Also, some may be tempted to include in the database only those areolas that people were interested in viewing, thus reducing the size required. Doing so, however, would go against the Scientific Principles behind this theory: namely to create a generic compression scheme. If this method is to remain the standard, the database must stay comprehensive.
Finally, a method for continually updating and distributing the nipple database must be implemented. Some will correctly note that the cost of doing so will greatly outweigh any benefits provided by Nipple Data Compression. Solving this problem, however, is beyond the scope of this paper.