Well, looks like the secret is out. I’ve achieved my childhood dream of being a Berkman fellow! Gosh. I’ll be starting in September.
They asked me for a blurb explaining what I’ll be doing:
Sahar Massachi’s work straddles social movements and the tech industry. He recently left the Facebook civic integrity team after almost 4 years at the company. He is researching the political economy of tech giants, the structure of the modern advertising industry, design principles for better social media, and generally how to nurture the good parts of the internet while reforming the bad.
That’s a set of big topics. And I’m still getting my head wrapped around them. But, maybe for my own understanding, I’ll take a stab at sketching out a little bit about how I’m thinking about it.
(Disclaimer: I’m just brainstorming! This is a form of thinking out loud. I have a lot of other things bouncing around in the old brain that I haven’t written down, and I expect I’ll learn things to change my mind on at least a few of these topics. Think of this as a little first-draft sneak peak teaser-trailer of a thing, a year before the movie airs, when the script hasn’t even finished being written, not the thing in itself)
If we think of large internet corporations as new para-governmental actors, can we study and interact with them as if they were indeed states?
And I do think they are quasi-states. Maybe somewhat like an executive branch fused with a court. There’s a need for “scalable” decisionmaking. Precedent matters. People inside argue using evidence but also inter-departmental bureaucratic warfare.
As the task of governing has grown, we as a society have evolved appendages to interact with agencies, legislatures, courts, etc. That looks like think tanks, lobbying, briefs amici.
Okay, so how do, say, think tanks wield power? What’s their secret? In part, they do free labor. A modern congresswoman has only a few staff, most of them dedicated to processing the increasingly overwhelming stream of feedback from her constituents. Think tanks (and lobbyists) do work she doesn’t have staff capacity to do: think deeply about legislation, get in the weeds of things, give intelligence about how different organized political groups feel, and turn her broad principles into actionable legislation.
So, what would a think tank aimed at a social media giant look like? Well, the comparison isn’t straightforward — Twitter, for example, certainly isn’t dealing with a staff of 10. But if you look at any particular team, they tend to be pretty small groups working on logical chunks of big projects. And with all the pressures and politics of working in a company, lots of work that people *wish* they could pursue isn’t being done.
Enter a think tank. Perhaps it could take on those projects that staff wish they could do.
Design principles for better social media (and democracy)
Right now, the dominant response by social media companies seems to be to rely solely on an army of content moderation “cops” to enforce “Facebook law” or “Youtube law”, perhaps armed with more sophisticated detection systems to help them find “bad guys” to arrest.
Imagine, however, changing the design of these apps to make bad behaviour less easy to do. Maybe by incorporating limits on actions that could be harmful (say, limiting the number of Facebook pages that an account could start per week, or limiting the number of subreddits a person could post a link to in the same time window). Maybe by adding friction to actions that might be abusive, in proportion to how certain the system is that it is indeed abusive. Maybe something else.
Imagine if, armed with these ideas (and quantitative and qualitative research to back them), a think tank could interact with both the high level decisionmakers, and the frontline engineers, designers, and software engineers of a company. Often, it’s those frontline workers who have a lot of autonomy to try things. Why not give them ideas of things to try?
The structure of the modern advertising industry
It’s pretty clear to me that advertising has evolved so much that the distinction between “online”, “mobile”, and “terrestrial” television doesn’t exist as much as it used to. For example, when your television sends home data about the ads you watch, do we use the conceptual bucket of “online advertising” or “old fashioned tv ads”?
For reasons, advertising-as-surveillance seems to have grown with the internet and entwined with it, but also be busting free into “meatspace”. And this advertising means surveillance. I’m particularly sensitive to surveillance because of my jewish anti-fascist, anti-police-state commitments.
So, online (and increasingly offline) advertising means surveillance. But online advertising also has market power problems. Political advertising, in particular, has democracy problems.
I want to look at the union of all those things. Because it’s all important, and I think only seeing through of those lenses tends to give people an incorrect view of the situation.
Political Economy of Tech Giants
Here’s something I’m struggling with: am I doing a project on social media giants and democracy? Or about tech platform giants in particular? (In other words, do I care about Amazon and Uber and so on?)
I’m not sure.
I do feel strongly that we should bring a political economy frame to understanding the actions of big tech. Let’s bring concepts like power mapping from politics to understand the actions of, say, Facebook. There’s already lots of journalism pointing out how it, for example, is strategically giving ground / giving gifts to the US political right.
There are also frames from history that we can use to understand what is happening. For example the “studio system” for Hollywood, and the DOJ consent decrees that ended them, could be used as a model for thinking about the distinction between production and consumption in *social* media. Let’s think about agrarian populism and The Grange. Standard Oil. So on!
And if we were to break up, say, Alphabet, how would we do it? What parts become public utilities? Which parts are broken up? Where do we promote competition? Is it possible to legislate such fast-moving things as open standards?
My past at Facebook
I used to work at Facebook. I spent almost four years there. I feel a responsibility to explain it better to the world. The bad stuff, sure, but also just the plain facts. I have a lot of respect for Alex Stamos, and I feel a certain kinship with him. Misinformed, or bad critiques of Facebook bother me. The topic is too important to get wrong.
We want contradictory things from Facebook — censor more to protect democracy / censor less to protect free speech and democracy. Protect people’s data / don’t be a walled garden. Protect us from governments via encryption / protect us from foreign election meddling by snooping on messages. Build real communities of friends / puncture people’s filter bubbles.
These all come from different, valid concerns. Each of those demands comes from an analysis of a real harm. But the proposed solutions often clash.
If we can catalog all these harms, and understand them at the same time, can we come up with proposed solutions that don’t solve one problem at the expense of another?
Wrapping up
Now that I’ve written these sketches out, I’m feeling excited! But I’m feeling that perhaps I have too many ideas at once.
Is starting a think tank too ambitious? Is the idea sound? How does one get funding to start such a thing?
Lastly, a note about heroes. I’m a fan of Louis Brandeis. He was so interesting! He was a hero on the left, but eschewed the standard methods (organizing the poor, agitation, mass politics) that come with those politics. Instead, he was creative. He organized his people — the upper middle class. A public intellectual, he wrote the book about problems with banks and then turned that into a government agency. He enlisted his friends and fellow lawyers to fight bigness in the cloak of monopoly.
Brandeis attributed much of his success to understanding, to a minute level, the workings of business or system, so that he could figure out how to fix it (business), or best regulate it (system). We could all learn from his example.
In the end, in tackling these big questions, I want to keep asking: “What would Louis Brandeis do?”