Guy Reams (00:00.622)
Today is day 216, the rule of thumb.
Yesterday when I was fixing a door on a cabinet in my disorganized garage, I couldn't find a measuring tape. Instead, I used the width of my thumb to estimate where to drill the holes for the new one. This case is about three thumbs. This makeshift solution made me wonder about the phrase rule of thumb. Though it wasn't precise, it was sufficient for hanging the door. So I thought maybe this is where the phrase rule of thumb comes from.
It's a measurement that's not guaranteed to be perfect, but it's good enough for the task at hand. So I thought, that's interesting. So then I started diving into this and I made the connection that rule of thumb, that's not exactly what people mean by when they say rule of thumb. When you hear somebody say the phrase rule of thumb, they're not saying that you're actually measuring with your thumb. What they're saying is that,
you're employing a heuristic. In my field, people often use this phrase heuristic to sound intelligent, which makes it hard to understand what the word means because people use it as a catchword for fuzzy logic or things that people don't understand. So they'll say, this software uses a series of heuristics to derive a solution.
Which is somewhat ironic because in that case, they're using the word heuristic as a heuristic for sounding intelligent. Basically, you can think of the heuristic as a shortcut. An imperfect thing, an imperfect way to solve a problem, make a decision, to make a quick approximation or to make a choice without being exactly perfect. So the phrase rule of thumb actually
Guy Reams (02:01.133)
is synonymous with heuristic, or that's the way people use the phrase. It's interesting how in the English language, specific concepts that have an actual word to describe it, because people don't know that word very well, English speakers will use catchy phrases instead of those, which is really fascinating, right? So actually,
This is a fundamental thought because humans are actually a living, breathing, heuristic machine. If there's one thing that humans are really good at, it's approximating decisions for faster processing time. Because we have to survive in the world. So we have to make quick decisions. We can't just sit around. We might get attacked by a predator or we might suffer from some ailment. So...
Humans have learned to make very quick processing decisions based on limited information with an approximation of accuracy.
So humans are very good at creating heuristics. So.
So when presented, so humans are always looking for this. So this is what, in my mind, when AI came out, when tools like ChatGPT and others came out, humans adopted that so very rapidly because we can recognize and understand the value of a shortcut quickly, because that's what we do in life. That's what we do all the time. However,
Guy Reams (03:37.357)
Heuristic is a much more complicated concept than just rule of thumb. There are definitely great heuristics for solving things faster and making easier decisions, but it's important to distinguish good heuristics from bad heuristics. For example, one of the bad heuristics that humans do all the time is we will automatically accept something because it just looks familiar to us. We might have seen it.
So there's an inherent cognitive bias towards things that we recognize. For example, a ballot, we're taking, we're voting and we're looking at a ballot and we see a list of eight names that we have to choose from. And we see a name that looks familiar to, we may not even know this name, but it looks familiar to us. We become instantly biased towards that name.
And there's lots of situations like that where we choose something because it's the right color, for example. So we judge quality completely based on something we recognize.
English speakers have also come up with phrases to help us avoid this oversimplified heuristic. You'll hear phrases like, don't judge a book by its cover, or don't just eyeball it. These are phrases designed to remind humans not to use this heuristic inappropriately.
So we can have good or bad results from this, which is kind of an important thing to think about. Using cognitive shortcuts or using heuristics can serve as a double -edged sword in both our human as well as our technical implementations. So I think it's entirely contextual. In some cases, like my hanging my cabinet door yesterday, the rule of thumb, the thumb heuristic,
Guy Reams (05:36.621)
was appropriate because I didn't need it to be exact. But if I was building a vehicle or an airplane, I wouldn't want to use the rule of thumb heuristic for measurement, right? At least I hope not. So I think they can be appropriate for given circumstances, and in some cases, they can.
I think you have to also judge the amount of error you're willing to have. So in some cases, the error rate produced from using a heuristic may be acceptable. And in other cases, the heuristic that you're using may not be acceptable because the error rate would be too much.
But understand and evaluating heuristics not only aids in our informed decision making, but also enriches our insight into the way we think, the way we go about our lives. So from now on, when I hear somebody say heuristic in a sentence, I have an immediate cognitive awareness to go to this rule of thumb concept to help me understand what it is that they really mean. Thank you.