Analogy Machine Example

Start with Categories, end with Nuanced Vision

General Description:

In a vacuum of knowledge about the underlying causes of phenomena one sees, the first thing one can do to understand the phenomenon is to categorize the phenomena themselves. When a scientist groups phenomenon into categories, he may be leading himself astray – the categories may or may not have anything to do with the underlying causes. Nonetheless, the sharper and more precise the categories, the closer they may become to reflecting causes.

When these categories are in error, they are discarded outright when a sharper view of the science emerges. When these categories have merit, they still tend to take the sideline when they lose their importance. In either case, they mysteriously still get taught in elementary school.

Example:

Taxonomy (categorization of the phenomenon of biodiversity) vs. phylogeny (the evolutionary origins of biodiversity). Biologists in near pre-evolutionary times were already quite good at categorizing species based on their physiology. By the time Darwin’s On the Origin of the Species was published, they have long since discarded classification systems such as flying, land, sea or useful, dangerous, harmless or other such nonsense in favor of Linnaean Taxonomy, which reflected the best of their knowledge of the day.

Some of Linnaeus’ categories did represent true evolutionary relationships. This is because he was so careful to categorize based on morphological similarities. Some things were, of course, wrong, too. Cases of convergent evolution naturally created false positives for category matches. Paleontology did much to correct the taxonomy since Linnaeus’ time, and molecular data, more still. Now, the system itself still suffers from not correctly reflecting the underlying causes of biodiversity – there are still many paraphyletic clades (unless you want to deem birds reptiles, for example). And, the classifications highlight the sections of biodiversity we were most familiar with before the invention of the microscope, and that is but a tiny representation of one of the major clades!

So, even though Linnaean taxonomy is at the verge of being discarded outright for strict cladistics, the taxonomy itself proved quite useful in telling us where to look. Why do mammals all have such similar limbs? Such similar embryology? These questions lead to the biology we have today, and a frustrated purist who would reject early attempts at classification as simply imposing a librarian’s order on a chaotic universe would have done nothing to help a fledgling science. Oh, and the “kingdoms” are quite easy for school children to grasp.

Some More Examples:

  • Schizophrenias are still numerous (you may recognize some – paranoid, disorganized, delusional, hey, that sounds like half of my friends! just kidding, friends!), and there’s no agreement what the types are, if there really are types, or if the different types even represent the same disease. Surely the most difficult thing for the human mind to grasp is the human mind.
  • Speaking of phylogeny and genes, a good start for understanding physical anthropology was races. When the categorization was based on skull shape, not other things like skin color, it was closest to representing human history, since skull characteristics are among the least affected by natural selection. Some of the categorizations back in the 1800s were close to right, but molecular data has relegated “race” to a very minor role, if any, in describing populations.
  • Quantum theory describes a good number of quantum particles. We only know how these particles act. It very well could be that none of these exist as distinct types of particles, as some attempts at quantum field theory might suggest.

Application:

An ambitious scientist would be half-right be be suspicious of a young science’s obsession with categorization. But he should be cautiously optimistic about more and more detailed classification (read: observation) while striving for something that points to a fundamental cause. Oh, but scientists already know about all this. What can you, the non-scientist glean from this? One day, you are telling your friend over a drink “there are X types of people in this world…” and proceeding to bitch about your X, and the next, you develop a Mark Twain-esque understanding of human nature.

Be satisfied with solid observation at first, and even indulge yourself with your atavistic desire to categorize if you must. But from there, learn the underlying causes, the essential nature of things, the ways in which the categories are an illusion, or at least but a small puzzle piece.

My Idea – The Analogy Machine

Generic ideas. Idea templates. Not out of the box complete ideas, but rather generic concepts, taken from a wide variety of fields such as philosophy, the sciences, business, etc, stripped of specifics (though a book on this would give them as a history of the idea), so that it can easily be applied to different fields. This is similar to templates in C++ and other object oriented languages. Much of what makes many ideas great are in the logic structure. If an “idea” only applies to just one field, it may well be an observation, not an idea.

My machine will automate what polymaths of old (yes, they don’t exist anymore, though technology has made an unprecedented number of people think they are) did with their artful analogies. Much “innovation” is doing nothing more than this and I, myself, am good at tricking people into thinking I’m smart by, e.g., turning a joke from South Park into an insight into computer science. So take this as a warning, you people who think you’re smart – use your noggin, or I will replace you with a perl script.

Example forthcoming. Watch this space.

People like things…

Oftentimes, I’ll have ideas about aesthetics that are generic and can apply to any art form. Some principles apply to any art, especially with regards to the appeal aspect. I’ve done posts on this before, but for now on, they’ll all be posted in “art principles”. It’s my nature as a programmer to find patterns. Write generic templating code and implement specifics later…
People like things that are like people, like an individual:

  • Is complex. A good [movie] can be watched repeatedly and you find something new on each watch. Some things will always be mysterious, just like a person always dies with secrets. It’s a mistake to think that little unnoticeable things themselves make something more appealing. Rather, knowing that such things exists is enough to make it more appealing. Continue reading