The story, outlined in a New Republic piece published in the mid-60s, was that a group of comedians hanging around a New York Deli came up with an informal model that could predict how long Broadway shows would last. To figure out the lifespan of a run, you simply needed to take the amount of time that show had already been open, and multiply it by two. A show that had been running for 2 years? You can probably expect it to run two more years. A show whose run was only sitting at two weeks? You can probably only expect two more weeks out of it. If it happens to make it through that two week period and the show is still running? Well, that’s because its expected expiration date is actually 8 weeks, because it’s been open for 4 weeks.
Benoit Mandelbrot is probably best known as the guy who came up with the slightly psychedelic fractal patterns which allowed Generation X computer geeks to demonstrate that they not only had a color graphics display but also a math coprocessor. In some of his off time, he picked up this ‘Lindy Law’ and ran with it. Pivoting to the comedians themselves, he theorized that you could predict the number of tv appearances a comedian would make in the future based on the number they’d made up to the present point. Of course, being a mathematician determined to formalize a bar joke, he escalated and connected the whole joke to power law distributions.
Nassim Taleb took it even further in ‘Antifragile’, extending it far beyond Broadway and comedians, to just about anything humans have created that we consider to be non-perishable. Books, technologies, ideas, cultural practices, institutions; these all became targets to be understood through the Lindy effect. Humans have an expiration date, but the things that we design don’t necessarily have to fall under the same constraints. Taleb argued that every year a given design has survived, the probability of its continued survival goes up.
We see this in literature. Don Quixote is 420 years old, and considered to be one of the greatest novels ever written. The Lindy effect tells us that if it’s survived this long, it’s likely to stick around for another 420 years. The internet tells me that ‘Woman Down: A Novel’ by Colleen Hoover is currently at the top of the fiction charts as I write this. Will people still be reading ‘Woman Down: A Novel’ in 420 years? To the extent that we can use the Lindy effect as a magic eight-ball, the answer is probably leaning towards ‘outlook not so good.’
There are a lot of reasons behind this, though it can probably best be boiled down to the notion that time is the only honest critic. The list of things that converged into optimal solutions or cultural touchstones with universal appeal that span multiple music, literature, technological, and other artistic movements to pass the test of time is large. The list of things that did NOT is far larger. Bicycles yes, hoverboards no. Mozart yes, Debbie Gibson no. Fork yes, spork…well, it’s too soon to call it on the spork. The ones that remain are the sort of things that have converged on optimal solutions to problems that don't change, or that have provided emotional impact that resonates across generations, and they just... stayed.
Of course, when it comes to tangible goods there is a slight hitch regarding their potential longevity: modern corporations seem to be determined to turn modern manufacturing in the direction of defeating the Lindy effect artificially. Objects that should be more durable than their predecessors, modern flat panels versus CRTs, smartphones versus old Nokias, automobiles, etc, are instead engineered to die faster through planned obsolescence, software abandonment, and deliberate incompatibility. If it seems counterintuitive that the longevity of an object got worse as the materials and manufacturing processes got better, that’s because it is. This is almost entirely an artificial construct.
I’ve recently found myself enamored with makers who focus on restoration and rejuvenation with the goal of taking something old and making it usable in the modern age. Indeed, my own efforts seem to be leaning more and more in that direction. I call this Lindy curation. This is the practice of identifying objects with some sort of durable core trapped inside their corporate-mandated artificial obsolescence, and intervening to unlock their latent survival fitness.
Sometimes this is about extending the lifespan of an object another 5 years. You can take an old computer, upgrade the processor and the memory to the max it supports, add a modern storage solution, and then stick some new software on it. This results in a system that can still do useful things, because the core of the object still effectively does what it was designed to do, and as much as we’d like to think otherwise, for the most part, our relationship with technology hasn’t really changed. That is to say, what we actually use computers for hasn’t really changed all that much. Despite what software vendors would have you believe, word processing has not changed that much in the last 30 years. The fundamental task of putting words in order, checking their spelling, and saving them is virtually unchanged. With a bit of expectation management, these older machines are still perfectly acceptable compute platforms, if not as ‘general purpose’ as they once were.
Consider taking something like a bicycle from the early 90s, cleaning up the frame, and replacing the legacy components with modern versions that will basically give the bike another 35 years of life. Indeed, with these sorts of projects, it’s often possible to leave things in a state that’s as good, if not better, than the day they rolled off of the factory line. Hundreds of thousands (millions?) of 26” mountain bikes were produced in the 90s. It follows that there’s still money to be made producing parts that are compatible with this, and that is indeed the case. There is a rich ecosystem of replacement and upgraded parts available for almost all bikes made within the last 50 years.
Or take something like an iPod, where you can leverage new components built to add modern conveniences like increased storage, bluetooth connectivity, and better battery life to help extend the object far beyond its original capabilities. The ‘modern’ capabilities that this doesn’t have, like having the capacity to store every song ever recorded, or to automatically jump to the next algorithmically selected album? Not having these things is a feature, not a bug. Hobbyists and enthusiasts viewed the iPod as worthy of an investment, and took the time and effort to modernize it, creating this path forward. Any competent Lindy curator should first examine not only the inherent usefulness and quality of the core of an object, but follow that up with a careful examination of what efforts will be required to maximize the potential of that core.
Again, in all of these cases, this isn’t done with the sole goal of preservation. These are not (and will likely never be) museum pieces, but rather pieces that can continue to serve their original purpose, or at least something adjacent to it. Of course, the core of these objects is not enough to keep moving them forward without an ecosystem to support that core. Sometimes that ecosystem emerges intentionally, as with the iPod modding community; sometimes it's incidental, as with Linux distributions that happen to breathe new life into old hardware. The Lindy effect suggests that the practice of repair and reuse is itself ancient and durable, and that the disposability culture we find ourselves living in is actually the historical anomaly. Lindy curation isn't nostalgia, but rather an alignment with a deeper pattern of how durable things actually work.
I won’t go so far as to say that there’s nothing inherently better about a new computer, or a new bicycle, or a new music player. Modern processors are inarguably faster. Disc brakes on a bicycle are clearly more performant than older center-pull brakes. Modern music devices (read: phones) have much better support for wireless bluetooth connections. Lindy curation asks, “Do I need a faster processor to write a blog article?” It asks, “Do I need disc brakes on a bicycle I ride down to the grocery store?” It wonders, “Why not just add bluetooth support to this perfectly serviceable music player from 15 years ago, instead of buying a new one, or allowing my phone to constantly interrupt me while listening to music.” At the heart of it, I think, it also asks, “If this was good enough a decade ago, why isn’t it good enough now?” Often, the answer simply boils down to: the manufacturer would like you to buy a new one.