Reading Time at HSNY: “You Must Hate Smartwatches, Huh?”

News - Horological Society of New York News - Horological Society of New York Horological Society of New York

This post is part of a series, Reading Time at HSNY, written by our librarians. Today’s post was written by St John Karp. 

This is a question we get often at the Horological Society of New York (HSNY). And the short answer is: No! We don’t! HSNY’s Jost Bürgi Research Library has many books about quartz movements and digital watches, but smartwatches are more than just electronic, they’re a whole computer/cellphone on your wrist. HSNY’s Executive Director Nicholas Manousos often wears a smartwatch and says that the best watch is the one that works for you. I appreciate that answer because it embraces all the reasons that people wear watches: as status symbols, for sentimental value, as mementos, for fashion, for a love of horology, to track fitness, or just for the fun of it. It’s a big, colorful world out there and we appreciate all the ways people wear watches. But we all know I wouldn’t be writing this blog post if all I wanted to give you was the hippie love-fest answer, so what is the real deal with smartwatches?

The long answer is that smartwatches are also very interesting from a horological perspective. Horology isn’t just about pendulums and escapements (though to be fair, a whole section of the library is dedicated to pendulums and escapements). The problem with the way people view smartwatches comes from the fact that computers present people with a flat screen that looks like a dead end, but in truth, every screen is really a window; only when you look through it can you see the rest of the story. Behind every smartwatch, behind every computer, is a vast network of infrastructure that exists solely to let your gizmo know what time it is. So what is this vast infrastructure and how does it work? Is the internet full of gnomes and gremlins toiling away in the salt mines to enable teenagers to post selfies on social media? How is any of this horological at all? Let’s find out!

Our story begins with quartz, the rebellious bad boy of the mineral world that revolutionized the watch and clock industry in the 1970s and simultaneously put many clock- and watchmakers out of business. Nicholas Foulkes in “Patek Philippe: The Authorized Biography” writes that Patek Philippe’s belief in the irreplaceability of mechanical watch movements led it “to nurture quartz technology without knowing that it would grow to be a voracious cuckoo in the nest; to bring it within the city walls only to find that it was in fact a Trojan horse.” He later describes the collapse of the Swiss watch industry at the hands of quartz as a “Götterdämmerung” (Twilight of the Gods). You can’t accuse horologists of not having a flair for drama! HSNY’s library had that very Trojan horse on display in our 2024 exhibit “The Evolution of Seiko and Grand Seiko,” which included a Seiko Quartz-Astron, the first commercially available quartz watch.

Prior to the invention of quartz movements, all timekeeping depended on the physical action of some sort of regulator such as a pendulum or a balance wheel or tuning fork. By contrast, quartz movements, as Benjamin Matz (Trustee Emeritus of HSNY) writes in “The History and Development of the Quartz Watch,” work by passing an electrical charge through a small piece of quartz and measuring the resulting vibrations. According to Alex Newson in “Fifty Watches That Changed the World,” the Quartz-Astron was precise to within 0.2 seconds per day, significantly more so than a wholly mechanical movement could be. Which isn’t to say mechanical watches aren’t still cool — I wear one every day — but computers require clocks too and it might detract from the cool factor if your sleek, sexy Macbook ran on an old-timey pendulum.

“Pish-posh!” you may say. “Flimflam and flapdoodle! My computer doesn’t have a clock.” Actually, your computer contains two clocks, each of which has its own important job. The first clock is required to synchronize the operations of the computer’s components, just like the conductor of an orchestra sets the pace and coordinates the musicians. Remember back in the old days when your desktop computer ran at 33 MHz and had a turbo button you could use to jazz it up to 120 MHz? That’s the speed of the computer’s clock signal. The faster your computer’s clock can run, the more quickly your Facebook timeline loads. If the computer’s clock is like a conductor, then the turbo button is like giving the conductor three shots of espresso. Every consumer computer in the world contains one of these conductors, including your laptop, your smartphone, your car, your TV, your fitness tracker, and your smartwatch, and, just like a quartz watch, your computer’s clock signal runs off the vibrations of a piece of quartz.

Your computer’s second clock is a real-time clock, also quartz-based, that keeps something closer to what we’d call human time. In most cases, your computer keeps track of human time by counting the number of seconds that have elapsed since some arbitrary starting point. In the case of Windows computers, the starting point is January 1, 1601, while in the case of Unix-based systems such as MacOS and Android, it’s January 1, 1970. This has resulted in an impending “year 2038 problem” because in the year 2038, the number of seconds elapsed since 1970 will have grown so large it will exceed the computer’s ability to store it, resulting in a potential crisis similar to the millennium bug in 2000. For now, however, we can safely punt that problem down the road and let our kids worry about fixing it. All you need to know is that this timestamp can be converted into a proper date and time whenever you want to read the clock on your desktop or your phone. It’s not, however, a simple calculation. It has to take into account all the leap years, leap seconds, daylight savings transitions, time zones, and locale irregularities that have happened between 1970 and now. I for one am very glad that it’s someone else’s job to deal with that mess.

For a while, your computer’s real-time clock was sufficient, but then along came the internet, causing trouble as usual. Sure, your smartwatch could keep ticking over just fine using its own internal quartz clock. But because your smartwatch connects to the internet, it has to integrate and synchronize with every other machine connected to the internet so they can all agree on what time it is. If a bunch of machines talked to each other but all assumed the conversation was happening at different times, chaos would ensue.

The way all these gizmos keep in sync is by regularly pinging a nearby time server. The time server is a computer with the sole task of knowing what time it is and distributing that information to anyone who asks. If you could spy on your smartwatch and see all the messages it sends into the ether, you’d find out that it’s bothering a nearby time server with lots of questions on a regular basis. Almost annoyingly regular, if you ask me — I’ve broken up with exes for being less needy than that.

But now all we’ve done is move the goalposts back. Your computers don’t have to know what time it is because they ask the time servers, but how do all the time servers know what time it is? Dr. Demetrios Matsakis, former chief scientist for the Time Service department at the U.S. Naval Observatory, gave one of HSNY’s monthly lectures on this topic in 2016. His lecture, Theories of Time, is available online, but visitors to our library might take the opportunity to read the book he authored with Dr. Parameswar Banerjee titled “An Introduction to Modern Timekeeping and Time Transfer.” In their book Drs. Banerjee and Matsakis go into great detail about computer network time synchronization, which is the process by which all the world’s networked computers keep the same time. The time servers mentioned above take their cue from a high-precision reference clock such as an atomic clock.

But what do atoms know about the time anyway? Who put quantum particles in charge of our smartwatches? HSNY’s public demands answers! Up until 1967 the second was pretty much an afterthought. First people had days, which were self-evident because people could see the sun rise and set. Then they divided the days into 24 hours, and the hours into 60 minutes, and the minutes into 60 seconds. The lowly second was therefore a byproduct of dividing up the larger units of time. In 1967, however, the whole thing was flipped on its head when the second was given an official definition based on the vibration rate of a cesium atom. Minutes, hours, and days were then defined in relation to the second instead of the other way around. This, Chad Orzel writes in “A Brief History of Timekeeping,” marks the point at which we divorced modern timekeeping from astronomy. Our timekeeping system no longer depends on the rising and setting of the sun but rather on the vibrations of atomic particles. This is the reason why we now have all sorts of leap-seconds and other newfangled contrivances — the vibration rate of cesium is constant but the length of astronomical years can vary slightly due to irregularities in the Earth’s orbit.

Frankly, none of this would have been an issue if we’d stuck with astronomical time. Hellfire and damnation, why not just go back to sundials like we had in the good old days? Mike Cowham’s “A Dial in Your Poke” documents a myriad of portable “pocket” sundials that, unless the Earth has started to orbit a different sun, work as well today as the day they were made. You can even buy modern pocket sundials and sundial wristwatches and, unlike a smartwatch, they never need to be recharged. But for all its other faults the smartwatch is a small miracle of modern horology. People see a gizmo and they treat it like a black box — they think the story stops there. Look through it instead at the whole picture. There is an enormous global network of timekeeping technology required just to let your little smartwatch know the time. So no, we don’t hate smartwatches — we think they’re pretty nifty.