Month: February 2020

Recently, I read an article about the history of the long-defunct human telephone switchboard operator profession:

“Users of the telephone in the late 19th century and early 20th century couldn’t dial their calls themselves. Instead, they picked up their handset and were greeted by an operator, almost always a woman, who asked for the desired phone number and placed the call.”

“…An operator did more than simply connect a customer to his or her desired number, however. In the early decades of the industry, telephone companies regarded their business less as a utility and more as a personal service. The telephone operator was central to this idea, acting as an early version of an intelligent assistant with voice recognition capabilities. She got to know her 50 to 100 assigned customers by name and knew their needs. If a party didn’t answer, she would try to find him or her around town. If that didn’t succeed, she took a message and called the party again later to pass the message along. She made wake-up calls and gave the time, weather, and sports scores. During crimes in progress or medical emergencies, a subscriber needed only to pick up the handset and the operator would summon the police or doctors.“

https://www.richmondfed.org/publications/research/econ_focus/2019/q4/economic_history

At its height around midcentury, this field employed 342,000 people in the U.S. By the end of the century, the field was dead, with nearly all of the jobs replaced by machines. Phone users dialed numbers into their devices themselves, and the calls were automatically connected. As the article makes clear, the automation process was partly driven by cost: as real wages for low-level jobs rose and workers came to expect higher pay, it got increasingly hard to entice people to be operators with salaries that would let the phone companies remain profitable. While automatic switchboards were faster, many “free” conveniences previously offered by human operators, like wake-up calls and news announcements, disappeared.

That is, until the rise of cell phones, smartphones and digital personal assistants. Smartphones can be easily programmed with lists of other people’s phone numbers, and can understand and execute verbal instructions like “Call Dad,” which is little different from how people interacted with human operators in the past. This amenity has returned. Also, if a person you are calling doesn’t pick up his or her phone, you don’t need to task a human being with re-calling them until they pick up–you simply leave them a voicemail message or text them. The fact that most people have switched from landlines to cell phones that they always carry on themselves sharply reduces the frequency of missed connections, anyway. Smartphones can be customized in a few minutes to display continuous updates on weather, sports game scores, or anything else. Pulling a small device out of your pocket and glancing at its screen is a faster way to get information than calling a human, vocalizing a question, and waiting for them to find the answer and say it to you. Finally, voice-enabled personal assistants like Siri can understand and execute spoken orders from their users requesting medical or police help.

Even though technology destroyed hundreds of thousands of human jobs, in a sense, it also brought the jobs back from the customers’ perspectives by providing the same services, and for less money. This made me wonder which extinct jobs formerly done by humans would “come back” in the future thanks to better technologies. Here’s what I came up with in short order:

Gas station attendants

It used to be common for gas stations to have workers who inserted gas nozzles into customers’ cars and filled them up. They also did minor car-related tasks like cleaning windows, checking fluid levels, and checking tire pressure. It saved a little bit of time for drivers, and was more convenient for them since they didn’t have to exit their cars. Gas station attendants disappeared after the oil crises of the 1970s, when gas station owners were forced to cut costs.

New cars in most advanced countries are now required to have computerized tire pressure monitoring sensors that alert their drivers if the pressure in any tire is low, so technology has already resurrected that service. In the short-to-medium run, the other tasks formerly done by gas station attendants will be done or rendered moot by ride hailing services like Uber and by electric cars.

From an Uber customer’s perspective, all of the maintenance and repair needs of the vehicle they are riding in are satisfied unseen, and on someone else’s time. The Uber cab always shows up clean, in good running order, and fully-fueled. Electric cars also don’t need oil, eliminating the most important vehicle fluid, and thus sharply reducing the frequency with which the fluids in aggregate should be checked.

In the longer-run, robots will do all the tasks that human gas station attendants used to do, making the conveniences commonly available once again, this time to private car owners and Uber vehicle owners. Imagine your robot butler plugging a power cable into your electric car every night and unplugging it before you left for work the next day, or robots doing the same at charging stations. In the future, governments could mandate the installation of simple sensors in cars that continuously monitored fluid levels, or robots at your house or at charging stations could periodically lift the hood to visually check them. Wiping down car windows will also be automated once autonomous cars become common, and owners can push one button to tell them to drive to the nearest car wash to have themselves cleaned. Additionally, if the private car ownership model is overtaken by the ride hailing Uber model, the dominant companies will have large facilities where their vehicle fleets are routinely serviced and recharged by machines.

Slaves / servants

The history of slavery is well-known today, but it is often forgotten that, in the past, low-paid, often live-in servants were also common. Even lower-middle-class households in the U.S. and elsewhere usually had servants of some kind to do menial household tasks like cleaning floors and cooking meals. In the U.S. and Britain, it was widespread until the 1960s. The labor rights movement and the opening of better job opportunities led to the near-collapse of the servant industry, and for generations it has only been accessible to rich people who can afford to pay high enough wages.

As I said in a previous blog entry, I predict robot butlers and intelligent personal assistant AIs will allow average-income people to have servants once again. There will be a day when the thought of washing and folding one’s own laundry again becomes unthinkable to average people.

Radio actors

From 1920 until the popularization of television in the 1950s, radio broadcasts were the primary means through which people got entertainment and news content. And just as there are TV stars and movie stars, there were radio stars during that medium’s era of dominance. Though forgotten today, those radio actors were once household names, were masters of diction and the art of speaking, and performed plays, mystery shows, soap operas, and many other types of shows that commanded audiences of millions.

Radio will never reclaim its primacy over motion picture entertainment, but I think audiobooks will have a renaissance thanks to new technologies, and the high production standards and outstanding voice acting that listeners commonly enjoyed during the Golden Age of Radio will return. Text-to-speech computer programs can of course be used to easily convert any book into an audio file, though the subtler aspects of the text, like the pacing of the words and the emotional intonations and volumes of the character’s speech, are not carried over. In the near future, these technical problems will be solved, and there will be simple and powerful computer programs that let anyone make high-quality audiobooks in little time.

After a book’s text was uploaded, the program would be smart enough to tell which text corresponded to which characters, or to the third person storyteller. The human user would be made to assign each character a different voice picked from a wide selection of options (varied with respect to sex, age, race, nationality, and other qualities), each of which would sound like a real person. The voices of famous people could even be used. The user would then listen to the recording from start to finish, and would be able to stop the recording to fix mistakes the machine made, such as assigning lines of text to the wrong character voice, or voicing a line of dialog with the wrong tone, speed, or emotion. Making those kinds of corrections could involve a simple process where the user speaks the words into a microphone in the correct manner, and the machine reproduces that detailed speech pattern, but in the character’s voice. The program would also let the user easily add sound effects–many of which would be artificially generated and not recordings of real noises–and background music.

This kind of technology is entirely within reach, and merely builds upon those that already exist, like advanced voice mimicry and sound-effect-generating algorithms. It will empower untrained individuals to, at little or no expense, make high-quality audio productions that rival the professionally made radio shows of the last century.

Mudlarks

Mudlarks were the “dumpster divers” of the 1700s and 1800s–they were poor kids and old people who sifted through the muddy banks of city rivers looking for objects they could resell. At the time, waste management practices were primitive, and people would throw their trash into waterways. Things often fell overboard from boats plying the waters through cities. While the end of mudlarking as a job was a positive development for society, something of value was lost to society since there was no one left to find and recycle useful things that had been dropped in the water.

I think technology will bring mudlarking back in a sense. For one, if robots and AIs take most human jobs, then people will have more time to indulge in their hobbies, even if the financial returns are negligible. That means the number of people who do things like mudlarking, magnet fishing, and metal detecting will increase. In the longer run, those pursuits will be automated, and we’ll use millions of cheap, autonomous machines to gradually comb over the Earth’s surface–including riverbanks and the bottoms of bodies of water–for hyperaccurate mapping, archaeology, waste removal, and recycling (particularly in the case of metal objects since it’s much cheaper to recycle existing metal than to mine and refine metal ore). This derives from my more general prediction that robots will clean up all of the garbage created in human history by the end of the next century.

Links:

1. If you yell “Hey Siri, call 911” at your Siri-enabled smartphone, it will make the call. https://www.ooma.com/blog/how-to-dial-911-using-siri-on-your-iphone/
2. A short history of domestic servants in the U.S. https://www.theatlantic.com/business/archive/2015/09/decline-domestic-help-maid/406798/
3. Machines can clone the voices of specific humans. https://www.theverge.com/2019/6/10/18659897/ai-voice-clone-bill-gates-facebook-melnet-speech-generation
4. Google Assistant can mimic all the nuances of human speech. https://www.npr.org/sections/thetwo-way/2018/05/09/609820627/googles-new-voice-bot-sounds-um-maybe-too-real
5. Computers can generate realistic sound effects with minimal user input. http://news.mit.edu/2016/artificial-intelligence-produces-realistic-sounds-0613