jmoiron.net

A question has really plagued me for quite some time now. What makes something "good"?

I started thinking about this a long time ago in college. Back then, good design to me was kinda like pornography: in the immortal words of justice Potter Stewart, I couldn't define it but "I know it when I see it." For me, it isn't enough to just recognize something as good, I have to know why I recognize it as good.

I don't mince words here. I firmly state here that some things are good and some are bad and that the difference between these things is not merely subjective. It's not that I think something is better than something else, in some sense I know it's better. It's easiest for me to assert this on things that I am an expert on; I know that some programs are better than others, and I know when code is bad and I know when code is good. In this world, there are some easy measurements for goodness (space & time complexity) and some difficult measures for goodness (cleanliness, elegance). It's the difficult measures of goodness where people run into trouble.

I originally started to think more deeply on this after reading Paul Graham's Taste for Makers. In it, he describes how designers (whether they be designing paintings or programs) get "better" as time goes on, which strongly implies some objective and universal measure of 'goodness'.

Some years later, I read Gödel, Escher, Bach. Within, Douglas Hofstadter explains Gödel's incompleteness theorem, the artistic works of M.C Escher, and the intricate puzzle canon's of Johann Sebastian Bach. He uses these concepts along with a host of other muses to talk about how systems themselves can hold and process information not implied by its constituent parts (like the human brain or a colony of ants) to talk about AI. The meat of this book, for me, though, was the inherent beauty and elegance of Bach's work.

Music is largely considered a matter of "taste" in the United States, and at least people of my generation were taught (virtually brainwashed) that matters of taste were purely subjective. In the "easy case" described above, where there are simple numerical constructs that can define goodness and badness, people are able to yield that one thing is better than another: for instance, Lionel Messi is a better soccer player than my Grandfather, because his statistics are superior. But Hofstadter introduces the genius of Bach and his music (and contrapuntal baroque music in general) via mathematics, an utterly stringent framework in which elegance is immediately obvious to trained professionals and the relative complexity and difficulty of problems is very well explored.

Recently, I watched a program on PBS' Nova called "Secrets of the Parthenon. I was amazed to learn from this program that the Ancient Greeks designed the Parthenon to have "perfect imperfections"; the parthenon is completely devoid of straight lines and right angles. A popular theory for this and the use of entasis (slight curvature in columns & foundations found in the Parthenon and other Greek structures) in ancient Greek architecture is that it subdues or corrects for the effects of optical illusions that you would get in perfectly straight architecture. Over time, the Greeks would have understood that if you line up a bunch of straight lines and view them on a horizon, they no longer appear straight. The ratios between height and width of their temples and the width and spacing are also carefully constructed and come from common ratios in nature (the "golden" ratio, and others).

The Greeks not only came to understand the nature of beauty (which to them, was defined by nature itself and the ratios therein) but were able to duplicate in their works, from beautiful statues to magnificent architectural feats. 200 years ago, Goethe, one of the most intelligent polymaths in modern history, began to unravel basic laws of human perception and revolutionized the understanding of optics and color. And yet today, we've thrown away their hard work and generations of experience; it's all just a matter of personal preference.

This isn't to say that there is a final universal answer to whether or not Pele is indeed better than Maradona, or if The Beatles are better than The Rolling Stones, but merely that there is "good taste" and "bad taste", and the difference is not merely subjective.

I've found out that this idea is sometimes met with hostility, because if you accept that design can be good or bad and that your opinion of something is not necessarily right, it brings up the possibility that things you like might be crap and things you hate might be great. It also means that, although you might be fine with your particular tastes, they can be improved.

To not accept it is worse yet, I think, for anyone who aspires to design anything. It allows you to dismiss any critics you don't agree with without cause, and also means that you can't improve upon your skills. If you hold someone to be a better designer than yourself, it's difficult to explain this and to reconcile the differences in design skill within this framework of thought.