Event Response: Jonathan Zittrain’s “Minds For Sale” and Ubiquitous Human Computing
Strapped for cash (or have some time to kill)?
Here’s a deal for you: If you can figure out how to control the bubble size in carbonated beverages, or can find a novel approach to protecting corn from insect damage, the website Innocentive will broker a deal where your idea could be purchased for $20,000.
Or maybe chemical compounds aren’t your thing? Head over to Amazon.com’s Mechanical Turk program, and make $1 for identifying in 100 pictures whether the person in the photo is male or female, or earn 5 cents for every city and country you match with the correct overseas zipcodes.
Still need more work? If you successfully pass the interview process at LiveOps.com (also known as the “contact center in the cloud”), you could soon be a call-center employee taking someone’s drive thru order from the Jack-in-the-Box from across town, simply sitting at home on your couch connected to the drive-thru module via your laptop.
Each of these is an example of Ubiquitous Human Computing, a term coined by Harvard Law Professor Jonathan Zittrain used to describe the trend to network and distribute mindpower as a fungible resource on the web.
Last Week, Zittrain came to speak about this topic at Santa Clara University in his lecture entitled “Minds For Sale”, where he dynamically discussed the myriad of issues we are faced by this new wave of the internet.
Out of the many interesting topics Zittrain covered, a few ideas stood out to me:
One question that Zittrain raised was what effect ubiquitous human computing would have on society. With contracting jobs available at our fingertips, how will the lines between “work” and “home” continue to be blurred?
The company LiveOps.com, for example, allows people to work from their home as call center employees for everything from fast-food chains to political campaigns, and represents a different model for employment than the typical worker/boss hierarchy, Zittrain says. LiveOps enables people to work anywhere (presuming they have a computer), and anytime (whether for 5 minutes, or 5 hours) depending on the person’s availability; thus, if a person wants to take some Jack-in-the-Box orders for 10 minutes while in between errands, Liveops allows him or her to do so by simply signing in and signing out at his convenience. Zittrain presented an advertisement with the slogan “Become a Mompreneur“, intended to target stay-at-home mothers looking to be employed in small increments throughout the day.
But he questioned what kind of impact these types of “contracting” jobs could have in the future. Will we grow into communities where people are sitting in parks on their computers, one person ordering a pizza from a Liveops contractor sitting across the way? In his article on Ubiquitous Human Computing, Zittrain writes,
“One can visualize in the near future a subway car packed with people, each far less attuned to the local environment and to each other than even with today’s distractions of newspapers and iPods. Instead, they will stare into screens even for just a few minutes and earn as much money in that time as their respective skills and stations allow.”
And he rightfully notes that with computer-mediated convenience, also comes computer-mediated control. Liveops, for example, documents its contractors performance based on statistical metrics of timeliness and efficiency, measured even down to the second of how fast someone answers a Liveops call. This introduces a categorically different model from the typical workplace, Zittrain suggests, where employees don’t have any flexibility or slack like they would as a typical employee, but are instead measured solely by these quantifiable measurements. And if a contractor is inefficient, Liveops has no obligation to keep them employed — they can be dropped and replaced by another contracted employee at anytime. Does this model lead us down a track towards people being treated less like human beings and more like computers, Zittrain asked? And one wonders this in general: will ubiquitous human computing make us more like task-mastering computers, and less like humans?
This issue seems even more relevant to the next topic: Amazon’s Mechanical Turks.
Amazon’s Mechanical Turks is a service that allows people to complete tasks for small amounts of money– often 1 to 15 cents –that are considered exceedingly simple but require human intelligence because computers are, for whatever reason, unable to carry them out. Examples of a “mechanical turks” include identifying keywords to describe a photo on Google image search, summarizing a website for a search function, or matching names with e-mail addresses — tasks that, at this point, are too complicated for a computer, but are fairly simplistic for a human being.
Most people think of these tasks as mindless and just something to pass the time – but Zittrain pointed out some interesting ethical questions that can arise related to these tasks. He pointed out the issue of alienation, which refers to the idea that people are being asked to do a task having no idea of what it relates to. (In other words, with mechanical turking, you can complete a task but have no knowledge of what company or person you are completing the task for.) Why is this issue? Well, consider the following scenario:
You are completing the “mechanical turk” of clicking through a series of pictures, simply identifying if the person in the photo is male or female. You’re earning some small sum of money — a few dollars, maybe–while watching TV and just breezing on the internet. But what if, in helping categorize the gender of a person in a photo, you are in fact helping to identify and catalog the images of protesters at a rally who will then be arrested based on information you helped provide? Because you have no knowledge of who commissions the turk, you could be contributing to a cause — political, religious, social — that you might not condone if there were full disclosure. Zittrain raises this as an important question going forward: If we don’t have transparency on who is commissioning these tasks, how can we know if we are contributing to something that may be out of line with our moral principles?
Zittrain’s talk also raised a lot of questions about honesty on the internet in regards to marketing and consumer reviews. He cited one website called Subvertandprofit.com, which allows a person to pay money to have his or her article “Digg’d” (for non “diggers,” Digg.com is an aggregate social news site where users vote on popular stories — and the more votes, the more publicity). You can also pay money to get people to give your product a 5/5 positive review on Amazon.com or other review sites. Essentially, if you shell out the cash, you can control the type of reviews your product can get. Should we condone this type of practice online, Zittrain asked? How will we know who to trust online if reviewers are being paid to give good reviews?
Captcha codes are another area that breeds dishonesty online. Captchas are codes which were created to eliminate online spam–they require a human user to read and type out the letters presented, thereby confirming that the website user is in fact a person and not a computer program trying to spread spam:
But spammers have learned to get around this hurdle by eliciting their own “mechanical turks” to solve Captchas for them. For example, some programmers have designed free pornography that can be viewed if one just simply enters in a Captcha code; basically, spammers trick pornography viewers into solving Captchas for them, and then use those solved codes to spam whatever blog or website featured that Captcha. On a massive scale, this leads to Captcha codes being solved left and right by people who don’t even know that they are contributing to this internet spamming scheme (again raising the issue of alienation on the web). Zittrain even described what he called “Captcha sweatshops”, where people are commissioned to solve these codes all day long–a new type of digital labor for the 21st century. As ubiquitous human computing advances, one wonders, what will happen to honesty, trust and ethics on the internet?
The last point I thought was fascintating in regards to ubiquitous human computing was the idea of citizen surveillance. Zittrain pointed out a website marketed to engage citizens in border control, where people can watch surveillance footage of the borders and report to the website if they “see anything unusual.” If enough people respond with concern, patrol men will be sent out to the border to survey the problem. Essentially, people can watch the borders 24 hours a day, looking for any behavior they might deem suspicious.
But is citizen surveillance something we should be encouraging in this way? And one wonders, what might this look like in the future if surveillance intensifies, and people are constantly being commissioned to report on the behavior of others recorded on surveillance camera footage streaming online?
All in all, Zittrain’s lecture raised many fascinating questions about the future of the internet in regards to the distribution of human mindpower. Zittrain himself admits feeling ambivalent towards this trend: excited about its possibilities, but also anxious about its downsides. I share his ambivalence; there are obvious benefits (like idea sharing on Innocentive) and significant drawbacks (like turning us into cogs in a machine) to this type of networking.
But what I admired most about Zittrain was his approach to the topic — he is really the consummate example of a Technological Citizen. His goal is to get people talking about the path we are going down, and have us all evaluate whether it is the right path to be on. He encourages people to get out of the all too popular camps that arise around issues in technology, the first which claims it’s “too early to tell if this will be a problem” and the other that insists “it’s too late to do anything about it now anyway.”
If we examine these issues now, he says, we can have some sway in bringing about changes that help move us in a direction where privacy, autonomy, and human values are taken into consideration. If we wait too long or never address these questions, he warns, we might continue down a road that leads to a place we don’t want to end up in, where our minds are for sale, and we’re nothing more than human machines.
What do you think about ubiquitous human computing? Do you share Zittrain’s concerns, or have additional concerns of your own? Alternatively, what benefits do you see to ubiquitous human computing?
Watch Jonathan Zittrain explain Ubiquitous Human Computing in the interview below:
To check out Jonathan Zittrain’s Book, The Future of The Internet And How To Stop It, click here.
It is interesting to ponder the future based upon these innovations Zittrain discusses… People, in these times, are being treated less like individuals and more like tools or “cogs” to achieve an end. Long gone are the days in which job security was assured for most workers. As Zittrain explores the idea of “mechanical turks,” we are confronted with the morals of anonymity and how alienation of information can subjugate workers into supporting immoral goals. He brings up valid points, yet now at the time of recession, easy money for those struggling to find employment is appealing and many will find solace in the ignorance anonymity provides.
In regards to LiveOps, such technological advances are characteristic ventures of a growing global community. Although its implications sound alarming and potentially “mechanic,” we cannot disregard the benefits to stay at home/single parents. My mom works from home as a legal accountant and its benefits to myself and the rest of my family has been resounding. I do agree that the example of traveling on a subway train or answering a drive order on a park bench in Montana for a Jack in the Box in Colorado is chilling, it is apart of the ascent towards McLuhan’s “global village.” Its inevitability is real, and although its potential mechanization of human interaction is different, it is something to which Zittrain says we must question and alter now or forever conform to its consequences.