In response to my essay History of the Capital AI & Market Failures in the Attention Economy, a few people recommended I read WTF?: What’s the Future and Why It’s Up to Us, by Tim O’Reilly.
O’Reilly discusses many of the same themes I write about — failures of proxy metrics (mistaking the map for the territory), networks and platforms, failure of government, capital and media algorithms (cf capital as AI), technological unemployment — with lots of interesting stories and anecdotes from his time ‘on the ground’ in Silicon Valley, witnessing and taking part in the evolution of the open source movement and internet publishing, juxtaposed against the rise of corporate Silicon Valley.
As a publisher of books for self educating about open source software, O’Reilly is (I think) a bit optimistic about open source patterns fixing government and retraining as a solution to technological unemployment. While I agree about the importance of initiatives like these, I’m not sure they are sufficient as solutions to all of our problems, as I don’t think open access to educational materials and training will yield more fair outcomes when time to explore and learn (cf Alexis de Tocqueville on necessity of leisure time for governing class) and access to mentorship (eg good parents and teachers) are so unevenly distributed.
The book is worth reading, in any case, as it is a good survey of the state of things in the US right now, and the optimism is refreshing.
Notes and quotes:
On the asymmetry of privacy / openness for government and for individual citizens:
Living in a world of pervasive connected sensors questions our assumptions of privacy and other basic freedoms, but we are well on our way toward that world purely through commercial efforts. We are already being tracked by every site we visit on the Internet, through every credit card charge we make, through every set of maps and directions we follow, and by an increasing number of public or private surveillance cameras. Ultimately, science fiction writer David Brin got it right in his prescient 1998 nonfiction book, The Transparent Society. In an age of ubiquitous commercial surveillance that is intrinsic to the ability of companies to deliver on the services we ask for, the kind of privacy we enjoyed in the past is dead. Brin argues that the only way to respond is to make the surveillance two-way through transparency. To the Roman poet Juvenal’s question “Who will watch the watchers?” (“Quis custodiet ipsos custodes?”), Brin answers, “All of us.”
Security and privacy expert Bruce Schneier offers an important caveat to the transparent society, though, especially with regard to the collection of data by government. When there is a vast imbalance of power, transparency alone is not enough. “This is the principle that should guide decision-makers when they consider installing surveillance cameras or launching data-mining programs,” he writes. “It’s not enough to open the efforts to public scrutiny. All aspects of government work best when the relative power between the governors and the governed remains as small as possible — when liberty is high and control is low. Forced openness in government reduces the relative power differential between the two, and is generally good. Forced openness in laypeople increases the relative power [of government], and is generally bad.”
O’Reilly proposes some methods for algorithmic fact checking as a way to tamp down on fake news. I don’t think they are a panacea, because ensuring multiple independent accounts might be game-able or at best lead to a consistent picture of the world (w/o any guarantee on accuracy/truth) and the requirement for quantitative data is also game-able (just look at the food and pharmaceutical stats and studies that get funded and cited for the sake of marketing) — but these do seem like a good place to start. At least a consistent public discourse with some sort of refutable numbers is a starting place that can lead to some sort of progress vs emotional churn / shouting matches:
Citing and linking to sources makes it much easier to validate whether an assertion is an opinion or interpretation, and who is making it. This should be the gold standard for all reporting. If media reliably linked to sources, any story without sources would automatically become suspect.
….
Do the sources actually say what the article claims they say? It would have been entirely possible for Business Insider to claim that the data used in their article was from the FBI, but for there to be no such data, or for the data there to be different. Few people trace the chain of sources to their origin, as I did. Many propaganda and fake news sites rely on that failure to spread falsity. Checking sources all the way back to their origin is something that computers are much better at doing than humans.
Are the sources authoritative? In evaluating search quality over the years, Google has used many techniques. How long has the site been around? How often is it referenced by other sites that have repeatedly been determined to be reputable? Most people would find the FBI to be an authoritative source for US national crime data.
If the story references quantitative data, does it do so in a way that is mathematically sound? For example, anyone who has even a little knowledge of statistics will recognize that showing absolute numbers of crimes without reference to population density is fundamentally meaningless. Yes, there are more crimes committed by millions of people in New York City or Chicago than by hundreds in an area of rural Montana. That is why the FBI data referenced by the Business Insider article, which normalized the data to show crimes per 100,000 people, was inherently more plausible to me than the fake electoral maps that set me off on this particular quest for truth. Again, math is something computers do quite well.
Do the sources, if any, substantiate the account? If there is a mismatch between the story and its sources, that may be a signal of falsity. Even before the election, Facebook had rolled out an update to combat what they call “clickbait” headlines. Facebook studied thousands of posts to determine the kind of language typically used in headlines that tease the user with a promise that is not met by the content of the actual article, then developed an algorithm to identify and downgrade stories that showed that mismatch. Matching articles with their sources is a very similar problem.
Are there multiple independent accounts of the same story? This is a technique that was long used by human reporters in the days when the search for truth was properly central to the news. A story, however juicy, would never be reported on the evidence of a single source. Searching for multiple confirming sources is something that computers can do very well. Not only can they find multiple accounts, but they can also determine which ones appeared first, which ones represent duplicate content, how long the site or username from which the account has been posted has existed, how often it makes similar posts, and even which location the content was posted from.
On asymmetric power in network / platform operators and the members of the network (tho I’m not sure drivers have as much leverage as O’Reilly argues they do):
Uber has many advantages over its drivers in deciding on what price to set. They can see, as drivers cannot, just how much consumer demand there is, and where the price needs to be to meet the company’s needs. Drivers must show up to work with much less perfect knowledge of that demand and the potential income they can derive from it. Michael Spence, George Akerlof, and Joseph Stiglitz received the Nobel Memorial Prize in Economics in 2001 precisely for their analysis in the 1970s of the ways that the efficient market hypothesis, so central to much economic thinking, breaks down in the face of asymmetric information.
Algorithmically derived knowledge is a new source of asymmetric market power. Hal Varian noted this problem in 1995, writing in a paper called “Economic Mechanism Design for Computerized Agents” that “to function effectively, a computerized agent has to know a lot about its owner’s preferences: e.g., his maximum willingness-to-pay for a good. But if the seller of a good can learn the buyer’s willingness-to-pay, he can make the buyer a take-it-or-leave it offer that will extract all of his surplus.” If the growing complaints of Uber drivers about lower fares, too many competing drivers, and longer wait times between pickups are any indication, Uber is optimizing for passengers and for its own profitability by extracting surplus from drivers.
Despite the information asymmetry in favor of the platforms, I suspect that, over time, driver wages will need to increase at some rate that is independent of the simple supply and demand curves that characterize Uber and Lyft’s algorithms today. Even if there are enough drivers, the quality of drivers deeply influences the customer experience.
It is not only capitalists pursuing profit that put downward pressure on labor — it is also consumers seeking lower prices:
I like to use Walmart as an example of the complexity of the game play and the trade-offs that the various competing players ask us to make as a society. Walmart has built an enormously productive business that has vastly reduced the cost of the goods that it supplies. A large part of the value goes to consumers in the form of lower prices. Another large part goes to corporate profits, which benefits both company management and outside shareholders. But meanwhile, Walmart workers are paid so little that most need government assistance to live. By coincidence, the difference between Walmart wages and a $15 minimum wage for their US workers (approximately $5 billion a year) is not that far off from the $6 billion a year that Walmart workers are subsidized via federal Supplemental Nutrition Assistance Program (SNAP, commonly known as food stamps). Those low wages are subsidized by the taxpayer. Walmart actually pays its workers better than many retailers and fast-food outlets, so you can multiply this problem manyfold. It has been estimated that the total public subsidy to low-wage employers amounts to $153 billion per year.
You can see here that there is a five-player game in which gains (or losses) can be allocated in different proportion to consumers, the company itself, financial markets, workers, or taxpayers. The current rules of our economy have encouraged the allocation of gains to consumers and financial shareholders (now including top company management), and the losses to workers and taxpayers. But it doesn’t have to be that way.
_Excerpts from: _WTF?: What’s the Future and Why It’s Up to Us*, by Tim O’Reilly._
Read next: __History of the Capital AI & Market Failures in the Attention Economy_._
_Subscribe to essays, et cetera at _kortina.nyc