This also happens to be pretty relevant to my current life status (grad school decision, hurray!), so I'm going to write about it.

Fair warning: this is amateur philosophy and psychology. If you have an allergy to that stuff, ye might best skip to the 5 section.

In the Beginning, there were Games

If one were to ask me to trace my exposure to ideas back to the earliest source I can remember, I would more often than not stop around the time I became active in the MMO community at large (beginning primarily in *Anarchy Online and Shadowbane*). Optimality is one of the most central tenets of online gaming culture.

MMO communities (and probably communities in general, but I'm trying to speak from experience) have a tendency to focus on playing their games of choice in the 'optimal' fashion according to some metric. This may manifest as maximization of XP rates during leveling, attempting to optimally pack daily quests for reputation gain rates, or setting up elaborate macro systems to ensure complete buff coverage (looking at you, Shadowbane). The pervasiveness of this focus on optimality had a significant impact on my view of the subject. Around my junior year of high school, I reached the point that I began to outright reject the pursuit of optimality.

That was about six years ago. A lot has changed since then, but I feel even more confident now that the pursuit of the 'best choice' is often not worthwhile. I'm going to attempt to outline my reasons for this, and then cover some ideas that I've found as beneficial substitutes for optimum-focus.

The State-Space Exploration Problem

The first and most significant problem with pursuing the best decisions, the best actions, the best techniques, the best X is not with convincing oneself to take the best action, but simply determining what the best action is! The mental model I hold as an explanation for this is derived from experience with optimality problems in mathematics and computer science.

The two primary reasons that choosing the best X is hard, in my view, are:

• The state space is enormous
• It is difficult to define accurate comparison heuristics

From Wikipedia: "A heuristic technique, sometimes called simply a heuristic, is any approach to problem solving, learning, or discovery that employs a practical methodology not guaranteed to be optimal or perfect, but sufficient for the immediate goals." In this case, by "comparison heuristics" I mean techniques for comparing choices and their outcomes. We use these all the time without even thinking.

Consider a small example: you are having a conversation with a friend. They make a statement. What is the best response? How does one approach determining what the best response is? Well, first we need to have some way of saying response A is better than response B. Then, we need a way to navigate the space of all possible responses in order to find the best choice.

But given an arbitrary statement, how many possible responses are there? Even if we think only about verbal responses (ignoring all non-verbal things like view intensity, line of sight, posture, etc), there are an incredible number of options to choose from!

Suppose they say "I don't feel prepared for my Calculus exam tomorrow." Do you respond humorously in order to help them relax or seriously to help keep them focused? Do you offer to help them prepare for it or maybe refer them to someone else who could? Suppose you respond humorously. Do you use sarcasm? Tell a funny (to you) story? Use a comically bad metaphor? I already feel overwhelmed with the number of responses, and we aren't even at the point of choosing words yet!

Then there's the question of comparison. Even assuming we can reasonably explore the state space, how does one compare a light-hearted response with a serious offer of help? Which is the best choice? It seems to me that it depends a lot on the situation that you're in and who you are talking with (which has the implication that maybe learning uniformly best responses is also impractical).

The choice of heuristic comparisons and possible responses is simply overwhelming, and this is only for a very simple, everyday interaction! What about life-altering decisions like what graduate school to attend or what topic to choose for your dissertation? The consequences are potentially enormous. How does one choose the best decision?

This overwhelming feeling is one that I associate strongly with feelings of indecisiveness. Being unable to quickly make decisions is something I'd like to avoid. Inability to reach decisions is something that is easy to notice in others, yet the common advice of 'just pick something' – while practical for deciding what kind of burrito to get – simply isn't something I'm comfortable with for any sort of significant decision.

Degenerate Behavior

Indecisiveness is something that I'd call a degenerate behavior. The behavior is not something that I want in myself, but is not far from behavior that I do want (thoughtfulness: thinking before acting). In some sense, it has degenerated from desirable into undesirable. Why does this happen? How can we prevent it?

One reason that indecisiveness can strike is a heuristic that just takes a long time to evaluate. If your method is trying to find the best option among hundreds of millions of options, it follows that it might take awhile. Great! We're done, right? Problem solved?

Not quite. People are really good at doing these sorts of searches. Every moment of every day, we make decisions about how to move, how to think, how to speak. Each of these involves an efficient search of a space of possible motions/thoughts/sayings. Take motion as an example. When walking down stairs, I see the stairs and make a decision on how to place my limbs. If I make a poor choice, I could twist an ankle or fall down them.

So is the problem that these searches are actually slow, or is there more to it? My experience in online-gaming communities leads me to believe that it is often not a slow search that causes indecisiveness, but rather conflicting heuristics that cause it!

Suppose you're playing World of Warcraft. (If you've never played, just pretend. My point isn't dependent on this one example.) You're leveling a new character. What is the best way to do this? There are several different schools of thought on this very straightforward-seeming question. Some insist that the best way to level a character is to blitz through dungeons from 15 - 85 and then race through quests for the remaining levels. Others take a more balanced stance by proposing to maximize experience gain without significantly dampening your enjoyment of the game. Still others consider experience gain to be a very low priority in the leveling process (I myself fall in this camp).

What we see here are three different – but still valid – comparison methods. Which one should I use? The community as a whole pushed the 'zomg XP' heuristic for a long time as the metric. It is still very much dominant, but others are seeing more discussion (in my experience). I began using that heuristic because it was well established and very common.

However, over time I discovered that I didn't really enjoy this. I entered a transitional period, in which I was constantly debating whether to go for the XP or just enjoy myself. In the end, I fell into the latter method and am happier for it. This is not to say that everyone should do the same! One of my good friends (you know who you are (Rodger)) applies the XP-maximizing method and enjoys playing WoW more than I. My point is twofold: our environment can shape our methods of decision-making, and conflicting heuristics can cause indecision.

We can observe similar situations of borrowed heuristics in everyday life. I learned to be conservative with food spending from my mother, but learned the joys of good lunches hanging out with friends at my (well-paying) internship last summer. These two desires regularly come into conflict. How can I resolve this conflict? Everything has come down to the problem of deciding between heuristics. I propose a simple solution: change the problem statement.

Rephrasing the Problem

We began this adventure by thinking about the problem of finding the best choice. What if we don't? Let's instead think about determining if a choice is merely good. Then we might also be able to decide if another choice is better and – more importantly – why.

Going back to the food example:

• Getting the House Special Pho for lunch from Vy's Kitchen is good.
• Lo Mein from Red Bang Bang is much less expensive (and thus good).
• Today, I consider getting Lo Mein for lunch better because I am more concerned about watching my spending than eating delicious Pho.

While there is a core algorithmic difference between the two (determining best requires considering all options, but deciding good doesn't), I consider the key difference between this approach (henceforth good-better) and the best-first approach to be awareness. With the best-first approach, I have the tendency to let my brain run until it gives me an answer I am satisfied enough with to take action on. I tend to be dissatisfied with this choice because it isn't the one I was looking for.

With the good-better approach, I am actively aware of what options I'm evaluating. When I reach a time that I have to act, I can simply do so with my currently optimal choice. I almost always feel better about the result because I know my choice is good, although better choices probably exist.

Further Reading

I am certainly not the first person to think about this. Although I have not read any of these papers, the citations of the Wikipedia page Satisficing appear very closely related. The whole page is a good read. Really, I could have probably just linked that page and been done with it, but where's the fun in that?

In Conclusion

This is the approach that I've been using for the better part of the past year, and have only just now formulated it concretely enough to write about. I know that it is based wholly on my own experience, and I'm okay with that. My objective was not to somehow prove that my way was better, but merely to describe it.

I have noticed a marked improvement in my general mood following decisions after starting to use this approach. I have also noticed that I act differently when alone doing things. When playing WoW I do all the same things as before, but focus on doing good things rather than the best things. I enjoy it more as a result.

This approach is also informing my decision on which graduate school to attend / whether to jump into the industry for a while. I still often get stuck thinking about what the best option would be, but when I regain control of my thoughts and focus on good-better I tend to relax significantly. It's comforting to think about deciding whether one good choice is better than another, rather than thinking about why my options aren't the best they could be.

I've broached this subject to a few different people, and the responses seemed generally positive. My decision-making process is constantly evolving, but this change made a noticeable improvement in my life. Here's hoping that it might improve others' lives too.

P.S. Leave comments.