Wednesday, February 13, 2008

On Taste

Note: Before reading this entry, you should probably read Paul Graham's essay on the same topic. In fact, you should read all, they're quite good.

So, yesterday I was talking to a friend about how a mutual acquaintance of ours, while being a nice guy, had remarkable poor taste. Most of his favorite authors, television shows, video games, etcetera, were almost uniformly what one would charitably refer to as "trashy", or less charitably refer to as "bad".

Well, since you read Paul Graham's article I pointed you to at the beginning (you're so studious!), you understand the premise that no, he isn't entitled to his opinion in this case: Some things just aren't as good as others, and liking those things equates to bad taste. So, my question is, can you learn better taste? I'm sure that you could learn to like things that are better—if we took away some hypothetical person's bad music, movies and books, and had them just listen to Tchikovsky, watch Citizen Kane, and read Shakespeare (or whatever we'd agree upon is good, for some value of "good"), they'd probably get used to it eventually—but does that necessarily mean one will have better taste? Would one still seek out better materials, or would they remain unable to tell the difference in quality between Dan Brown and William Faulkner?

This is not a trivial problem.

I'm really not too sure what the answer is. In fact, as writing this, I realized how poorly defined "taste" is: When giving my examples of "good" things to force feed someone, I knew full well that people would have issue with some of the items. How can we objectively assess how good something like art/entertainment is? I do agree with Paul's thesis that there is such a thing as good taste or bad taste, but I think that one can only really assess it by what they create, since judging what they like is too hard, and will probably be reflected in their creations anyway.

Of course, all this does is move the problem from defining judging if what they like is good to judging if what they've made is good. However, I would intuitively say that this is a more tractable problem: If they've made something elegant, that works well, and is appealing to some statistically significant (for some value of "significant") audience. Essentially, I think that the popularity of the creation is positively related to one's taste, if not directly proportional. Then again, if that's a valid metric, why can't we use it to assess what people like (I think I'm starting to tread into a Zen and the Art of Motorcycle Maintenance-esque "what is Quality" question, which is an established Hard Problem...)? Is there some fundamental difference in judging someone by what they make, rather than by what they like?

Again, not a trivial problem

That being said, I think there is a difference. When you create something, you have to actively think about how this system is going to be designed: You have to consider all aspects of it, while when one is the passive recipient of something, you typically only notice the most superficial level. Therefore, when choosing what you like, it will probably be based on popularity first, since that is most likely the criteria you encounter. On the other hand, if you create something, then you see every aspect of it: At that point, it's being good or bad is more dependent on your taste, as it controls every aspect of what is being designed.

Well, not sure if that made sense to anyone else, but whatever. No-one reads this anyway (except you, of course. Thanks!).

In any case, my original question is still unanswered: Is it possible for someone to learn good taste? With my new pseudo-definition of taste, I would now say yes, it is. You can improve your taste (at least as it pertains to a particular domain) by good designing things. Determining how to design good things is left as an exercise to the reader.

Next time!

So, with my Algorithm X implementation done, I probably should design the sudoku solver that it was originally for...but, know what I think will be more fun? Writing a lexer/parser for ECMAscript 4. Hopefully, this will turn into the front-end for the compiler I keep planning on writing...

Anyway, laughs aplenty sure to follow!

No comments: