How not to give a girl your phone number
The idea behind cryptography is to take a piece of information, render it sufficiently incomprehensible without prior knowledge and/or considerable effort, and to be able to decrypt the information efficiently, accurately, and without the original message being compromised. The closer to random or "noise" one can render a signal and extract useful information out of it later determines how successful an encryption/decryption method is.
The bane to cryptography is pattern recognition and predictability. For something to be random, there needs to be no discernible pattern for the problem you're trying to solve. Humans, for limited amounts and types of data, are great at pattern recognition. Computers, for limited types of problems and data set sizes, are arguably better. A classic pattern in encrypting/decrypting English messages is that given a large enough message, there will be a greater number of E's than any other letter. Starting with this knowledge, someone can get a good idea about what a message will look like after encryption if there are more of one symbol than another, and work backwards.
The Internet, when considered on a grand enough scale, is a noise generator. It is the theoretical situation of monkeys typing on keyboards in order to generate Shakespeare, but never reaching that point. Not only this, but upon human users generating random information, even further information is affixed to their records in the form of unique identifiers, usernames, passwords, etc. Some millions of systems, replicating and generating information and metadata in order to service their human noise generators.
However, with little effort, some knowledge about how web pages and HTTP are put together, one can start extracting useful information from the noise of the Internet. Huge amounts of information can be mined from dynamically generated content, particularly when it pertains to human-assigned information, for example, email addresses, phone numbers, street addresses, latitude/longitude coordinates, zip codes, likes, dislikes... anything, really, given data availability and a known pattern.
There are methods to combat this, with most effective of all to never publish such information in the first place. The second most effective is to make it difficult for computer programs to scan text for well-known patterns (either by embedding your information in images, CAPTCHAs, keep it hidden behind passwords, etc.) However, with a world full of enterprising but bored computer hackers, "difficult" soon becomes "30-40% success rate," after which it just becomes a game of time and numbers.
So I had to choose something unconventional when I got a new phone number and decided I need to let people know what it was without communicating it over what I consider to be a "compromised" medium (Facebook, etc.). I also had to make the message only make sense in a given context, something that humans are able to perform quite well, but computers fail miserably at. I had to provide people with an "encrypted" message, appearing to be noise given the context (figuring out my phone number), and let them "decrypt" it.
I won't give away the answer just yet. The message is linked to from the front page of idkfa. Which link you ask? How about the one that doesn't make any sense, given the context of idkfa. Also, remember that you're looking for a number. That's something you probably shouldn't lose sight of. Also, I haven't strewn numbers about this post or anything like that, so don't bother counting E's or anything.
There are a lot better methods to pass messages given today's cryptographic technologies. Still unsolved, mathematically and philosophically, is how to ensure that only a human being can access information whereas a computer can never do so. In my case, I made use of our ability to contextualize, but I can't help but think that there are better ways out there.
I hope that things like this catch on eventually as people realize the need for security, privacy, and anonymity on the Internet. Sure, it pisses off the people who are used to having all of their information provided to them instantaneously without any analysis on their part, and the ones that spent hours looking at the HTML source code on idkfa (sorry), but it ensures that I don't get strange calls from solicitors, or other abuses of one's phone number. After all, I might have it for a while. I wish to remain part of the noise for as long as possible.
- Previous: Darkness, silence, lense flare
- Next: The Wedding Ring