Anton Chekhov was a lot of things. A doctor, a playwright, and a short story writer are a few of the commonly given descriptions, but among those he is one of my favorite writers. My initial exposure came to Chekhov from a collection of short stories titled 'Lady with Lapdog and Other Stories', which I purchased at Dog Eared Books in San Francisco. Immediately after reading, one understands that Chekhov's style was different than many of his counterparts. The writing is brief, striking, and can often hit you in the head with surprise if you aren't careful. But what does Chekhov know about technology?

Chekhov's Gun is a principle which was popularized by none other than himself in a series of letters he wrote in his life. The verbiage I particularly like is the following:

"Remove everything that has no relevance to the story. If you say in the first act that there is a rifle hanging on the wall, in the second or third act it absolutely must go off. If it's not going to be fired, it shouldn't be hanging there." — Sergius Shchukin (Memoirs, 1911)

The idea is that every single thing in a story must have meaning. A gun is one example, but a character's words can be another. If a character reflects on the rain outside, then the rain MUST either flood the home or lead to amazing crop growth or anything in between. However, the rain must never have no purpose. That, to Chekhov, is a cardinal sin.

I think that this principle is very apt when it comes to technology. There is more complexity today than there ever was. Databases, networks, and storage still exist, and on top of that we now have infrastructure and artificial intelligence behind almost every single system in the world. In other words, whatever was complicated in the past is still complicated and is being layered even more with today's complexity. Think about any application you have ever used for something big. For instance, the first thing I think of is Zoom. I used it for the first time in the back half of 2018, and I use it now at least 3 days a week for whatever purpose. Zoom, at its core, is a collaboration software built to serve a similar purpose to FaceTime (with a few added office features). However, as time has gone on, Zoom has added a multitude of features - reactions, virtual backgrounds, chats, transcripts, breakout rooms, whiteboards, and many more. These are all amazing features that I have used, but the theme here is increasing complexity. And regardless of all of these features, Zoom is known as a collaboration software. There are no features in Zoom that do not enable communication and that are not frequently used.

Zoom is an example of a product adhering to Chekhov's gun. In a time when there is so much complexity, every single feature has a role that it must serve. Every single feature must have a customer that champions it, otherwise there is likely no purpose in including it.

I think a follow-on effect of this principle is that A/B tests are generally not as important as one might think. To qualify that statement, if we liken a feature to something hanging on a wall, it is hard to imagine an A/B test in that same sense. I think that Chekhov would say that one might be better off understanding what characters lament about, and then introducing a wall piece to address that. For instance, if Chekhov's character complains that clothes are always wet, perhaps he would add a clothesline outside. In much the same way, I am sure that Chekhov would advocate for tight customer iteration loops instead of A/B testing proposed flows and features.

I think that he would advocate for a different type of A/B test, however. He would likely be in favor of running an A/B feature test, not an A/B flow test. Perhaps he might hang a hunting rifle and a bow and arrow on the wall, and let the characters decide between those two instead of deciding between a hunting rifle and a regular rifle. Features are not the same as flows.

There are a variety of ways to look at these things. The bottom line is the following:

less is not necessarily more, but less is almost always more than less

I also wonder what Chekhov would think about AI!