#97: The Coroner at Dreamgate Frontier

The El Farol bar problem is a game theory problem based on a bar that actually exists in Santa Fe. A finite population of hypothetical people want to go to the El Farol bar on a Thursday night. If more than 60 percent of them go to the bar, it will be too crowded and everyone will have a worse time than if they’d stayed home, but if fewer than 60 percent go, all who show up will have a great time. Everyone must decide whether to go at the same time, and then they learn whether they made the right decision or not. The point of the El Farol bar problem is to illustrate an aspect of collective decision-making—not to offer a practical solution to the issue of crowded restaurants—but the thought experiment’s contrast with how shortages and surpluses are resolved in real life is the reason I think about it a lot. The El Farol bar problem exemplifies a mismatch between supply and demand (for space, in this case) and much of life is organized around trying to prevent those situations from happening. The El Farol bar is a place, and physical places are where people’s ability to properly sort themselves are perhaps the least sophisticated—a point that the El Farol bar problem unintentionally makes.

In other contexts, people are sorted more efficiently. Crowdedness isn’t just a meatspace phenomenon. On the internet, you never jostle for elbow room or push past someone blocking an entryway in the literal sense, but sometimes you figuratively do. While the constrained nature of physical spaces, from restaurants to houses to cities and countries, is tangible and well-established, digital crowdedness isn’t directly experienced as much as it’s transmuted into anxiety, anger, or mental fog. Social networks are a perfect example. Eugene Wei describes how social media requires “interest rate hikes” to balance their seemingly infinite flow of content with the limited real estate that is our attention: "If a social network achieves enough success, it grows to a size that requires the imposition of an algorithmic feed in order to maintain high signal-to-noise for most of its users.” As an example, a chronological Twitter feed becomes chaotic once it comprises thousands of followed accounts, but Twitter restores order algorithmically, making that feed less crowded by regulating which messages actually appear in it.

What this example should indicate is that the El Farol bar problem doesn’t occur in environments where those running the system exercise enough control over the users. Physical space, which humans have inhabited as long as we’ve existed, is frequently free space, and our collective behavior in crowded areas is probably as disorderly as it was a thousand years ago. With more control and less individual agency, however, the El Farol bar problem fades—an algorithm can decide which 60 percent of people get to go to the bar based on a variety of criteria (the idea of an airplane being overcrowded is absurd for this same reason). Today, due to the growing interpenetration of digital and physical environments, there are more levers available to similarly gate and rationalize meatspace, ensuring that certain people show up and everyone else stays home. If we understand that process of decongestion as an interest rate hike for the physical world, similar to the algorithmic Twitter feed, we should also note that most American cities and public spaces are too empty, not too crowded, and consider ways to lower their rates instead of raising them.