*Update: as of 8/8/2012, the youtube video has been pulled at the request of the TED organization. They've also asked me to help them figure out how to keep crackpots like this out of their conferences. I turned them down: if they want help, they've got the money to hire a legitimate professional, someone who actually knows what they're doing - i.e., not me.*

I'm not a big fan of the TED phenomenon. In my opinion, it's basically an elaborate scheme to help make a bunch of self-important rich guys show off their importance by getting people to come give them speeches that, ultimately, serve to reinforce their self-importance.

But there's another reason that I dislike it. In addition to the self-importance of its audience, as it's grown, it's also turned into a forum where other self-important twits can pretend that they're actual scientists presenting important scientific work.

A great example of this is something called "Vortex Math". An idiot by the name of "Marko Rodin" came up with this ridiculous idea, wrote up a website talking about how wonderful it was and how brilliant he is, and then worked out an invitation to talk at one of the TEDX conferences, which he's then used to further promote himself - after all, he *must* be a genuine genius in order to have been allowed to present to such an important group of people!

http://youtu.be/c1hLzQPio_8

Vortex math is an example of what I called numerical pareidolia. For those who haven't heard the term, pareidolia is seeing patterns in randomness. For example, the common event seeing the image of jesus in a piece of toast, or in a mildew stain, or... We find these images, and then believe that they're not just an illusion, but that they're a real, deliberate, meaningful reality.

Paradeidolia isn't limited to seeing images. We humans are natural pattern seekers. We can find patterns in dirt, in words, and in numbers. Numeric pareidolia is finding patterns in numbers. The most common version of that is numerology, where you assign meanings to numbers, and then find more meanings by performing arithmetic to combine numbers with arithmetic, and finding meaning in result.

In Vortex math, Mr. Rodin has done something interesting, for some definition of that word. He's found a numeric pattern, and he believes that not only is that a pattern, but it is *the* pattern, the fundamental structure of the universe. And what is this oh-so-awesome pattern, this vortex that defines the entire nature of the universe?

It's a sequence of 1-digit, base-10 numbers. To get it, you start, obviously, with 1. So take the number 1. Double it. You get 2. Double it again, and you get 4. Again, 8. Again, 16. But 16 is two digits - so add them together: 7. Double 16 again, and you get 32, 3+2=5, so 5 is next. Double 32, and you get 64. 6+4=10, 1+0=1. Etc. You get a repeating cycle.

That's it.

Of course, this only works in base-10. It's a result of the interaction of doubling with the base. So you won't get the same pattern in any other number system! According to Rodin, because of the significance of this pattern, that means that base-10 really is the only *correct* number system.

And what is the significance of this base-10 pattern? Let's let Rodin and his supporters explain, shall we?

Marko Rodin has discovered the source of the non-decaying spin of the electron. Although scientists know that all electrons in the universe spin, they have never discovered the source of this spin. Rodin has. He has discovered the underpinning geometry of the universe, the fabric of time itself. He has done this by reducing all higher mathematics – calculus, geometry, scalar math – to discrete-number mathematics.

With the introduction of Vortex-Based Mathematics you will be able to see how energy is expressing itself mathematically. This math has no anomalies and shows the dimensional shape and function of the universe as being a toroid or donut-shaped black hole. This is the template for the universe and it is all within our base ten decimal system!

The potential scope and breadth of the Rodin Solution is staggering; it is universally applicable in mathematics, science, biology, medicine, genetics, astronomy, chemistry, physics and computer science. The Rodin Solution will revolutionize computer hardware by creating a crucial gap space, or equi-potential major groove, in processors. This gap space generates underpinning nested vortices resulting in far higher efficiency with no heat build-up. The Rodin Solution replaces the binary code with a new code called the binary triplet which will revolutionize computer operating systems. It will transform physics and astrophysics by finally answering how black holes and pulsars work. Space travel will be revolutionized by reactionless drives that are unaffected by the weight they pull, making the present day combustion engine obsolete. The revolution brought on by reactionless drives will far surpass the societal changes wrought by the shift from steam engines to the present day combustion engine. The Rodin Solution can even be applied to ending pollution and drought by creating an inexhaustible, nonpolluting energy source. Because Rodin´s Vortex-Based Mathematics enables him to condense a trillion-fold calculation to only a few integer steps and because he is able to solve all the mathematical enigmas, the Rodin Solution will revolutionize computer information compression.

Pretty impressive, eh?

And what would crackpottery be without at least a bit of conspiracy? See, the government knows all about it, and they're actually *secretly* using it to protect us:

Rudimentary versions of the Rodin Coil, or Rodin Torus, have been created and tested by leading scientists and are presently being used by the U.S. Government in antennas that protect the four corners of the continental U.S.. Life-saving medical devices based on crude approximations of the Rodin Coil Torus are being manufactured and used in the treatment of cancer patients. Microsoft´s former senior researcher is using the Rodin Coil to research, develop and patent new computer information-compression schemes.

Nifty!

Alas, it's all bullshit. It's not worth spending too much time on this, but I'll grab a couple of the claims that are close to my interests, and briefly explain why he's so full of shit.

One of the claims in the passage above is how he'll revolutionize computer operating systems:

The Rodin Solution replaces the binary code with a new code called the binary triplet which will revolutionize computer operating systems

Suppose for a moment, that we replaced binary in our computers with a different underlying representation - *any* underlying representation. Ternary, quadrary, decimal, or his "binary triplets", whatever they are. How much difference would that make?

None *at all*. We've had the capacity to create ternary computers for a long time - there's just no reason to. We *have* built decimal computers. For many years, IBM had computers for financial systems that used a representation called BCD - binary coded decimal. BCD can be useful in financials, because it's easier to control rounding errors. Floating point math is a bit weird, because numbers that should be precise don't necessarily have precise binary floating point representations, so you can get some odd rounding errors if you're not careful. You don't need BCD to do this - you can use a variety of notations, so long as you're doing fixed point instead of floating point, but using a decimal fixed point representation makes it all easier.

The thing is, you can't do anything with different representations that you can't do with binary. It doesn't matter. So we don't build hardware using different representations. We don't use binary because we don't know how to build anything else; we use binary because it's easiest to build binary hardware, and there's no benefit to making the hardware more complex.

More important, one of the beautiful things about computers is that computers don't really do binary *numbers*. Computers use binary to represent things. Numbers are one example of something we can represent. But we can represent anything we want. When I was in college, one of my assignments in a CS class was to implement ternary arithmetic. It's a simple enough thing that it makes an *easy* assignment for an undergrad introductory CS class! We can build any representation that we want, and use it. We do this *routinely*. We're constantly building new representations for particular purposes. Some of them are so useful that they've been enshrined in hardware. For example, computers used to only come with integer hardware - that is, the only mathematical operations that were implemented in the hardware were operations on integers. The computers still did floating point math - you just needed to implement the representation in software. It was so useful that we added it to hardware in order to make it faster. But it's not fundamentally different. And if a new representation that worked better than simple binary worked, we could implement it using a standard binary computer.

So Rodin's magic vortex binary-triple computer? There's just nothing special about it. It's not going to revolutionize computers.

Another example is compression:

Because Rodin´s Vortex-Based Mathematics enables him to condense a trillion-fold calculation to only a few integer steps and because he is able to solve all the mathematical enigmas, the Rodin Solution will revolutionize computer information compression.

Again, it's stupid. The problem with compression isn't that it's too hard to compute. The problem is more fundamental than that. We *can't* compress everything - it's impossible. (I described more about the reason why it's generally impossible to do universal compression in this post.) The science and math of data compression are based on the fact that we *don't* actually want to compress arbitrary things; we want to compress specific types of things: text, images, video. For each of those, common representations contain a lot of redundancies, and compression tries to remove those redundancies. So, for example, by finding regions in successive frames of a video that don't change, we can reduce the size of a video file. But that technique won't do anything for a still image or a text file. We exploit the specific properties of the medium to find an effective way of compressing that specific medium.

In fact, we can do better at specific kinds of media with customized hardware. People build custom hardware for things like mp4 compression all the time. But that's for a specific medium. It's got nothing to do with general compression. General compression remains impossible, vortex math or no.