A huge thanks to Ellen Fishbein for collaborating for months on this essay with me. It would’ve been twenty thousand words long, instead of just a thousand, if not for her.
The overarching problem of the information sphere is a signal-and-noise problem. Sorting through the near-infinite supply of available content to find relevant, high-quality information is a Sisyphean task.
This problem exists because technology has eliminated friction from the processes of creating, publishing, sharing, and curating content. Anyone can hit “publish” anytime. This led to an exponential explosion of content: the supply of content far outstrips demand. Crucially, it also outstrips our ability to process even the very best of the available content; we’re finite creatures, after all.
In the past, physical constraints—the cost of printing, the finite shelf space in bookstores and libraries—imposed the need for all sorts of gatekeepers. In the world of writing, we define gatekeepers loosely as publishers, editors, bookstore owners, and others responsible for guarding those finite resources. They had one quality in common: they had to be picky about the content on their shelves or in their publications.
While there were obvious downsides to this gatekeeping, it had big upsides: chiefly, the quality control that gatekeepers provided. Incidentally, the friction in the publishing process was conducive to intention, introspection and reflection.
Then, publishing became frictionless. The gatekeepers shrugged, and the signal-and-noise problem exploded.
Not that the old-school gatekeepers were always right. They weren’t. Some of the content on Substack today is better than almost everything the Wall Street Journal ever printed. Some blogs are more valuable than Penguin’s recent books. So, while some of the traditional gatekeeping roles remain staffed by editors and publishing executives, today’s savvy readers don’t rely on them for quality content.
What do we have instead of traditional gatekeeping?
- Search engines: They’re great for finding factual information, terrible for evaluating quality. Search engines can’t correct you or guide you when you’re asking the wrong question; they amplify existing confirmation biases and do nothing to challenge your mind.
- Recommendation engines: Recommendation algorithms used by platforms like YouTube are designed to keep you watching. They aren’t designed to expand your horizons, introduce you to different ideas, or help you learn. And they certainly don’t care whether you’ve made efficient use of your finite time.
- Social media numbers: Publicly visible metrics create a bias toward popular or shocking content. Those who “win” on social media are those able to play the game best, not the ones producing the deepest, most profound stuff. Downstream effects on self-image, depression, etc are a cherry on top.
- Aggregators: Hacker news, Reddit, etc. do a decent job prioritizing content over persona, but there’s a bit of a mob rule problem.
- A cobbled-together, low-tech“filter” that depends on people we know: In one way or another, many of us rely on friends, coworkers, and people we know to curate our content via Slack channels, texting, talking, etc. Though this filter is crude, some of us use it to the exclusion of the others. But in the end, it’s still too much.
Crucially, none of the algorithms or technical solutions were built with the incentive to educate, curate the best content, or generally control the grade of what they show you. There’s just no reason to believe they’re doing a better job than the legacy gatekeepers at ensuring the quality of published content. On top that, there’s little incentive to produce timeless content that’s worth saving or re-reading years, decades, or centuries from now. Instead, a high volume of largely disposable content is being released at top speed. “Disposable content” is a can of worms and will be a central topic in our next discussion. For now, let’s get back to gatekeepers.
If the algorithms aren’t acting as gatekeepers, how do people manage their information diets? Many people try to consume everything under the sun almost indiscriminately, pressured by FOMO, amplified by the “path of least resistance” and ease of access. Obvious drawbacks here.
In sum, we’re lacking a system that:
- Will consistently feed us great content (which might not be the absolute best, but that’s okay).
- Will have an absurdly high signal-to-noise ratio (>80%), not only on quality, but on relevance as well.
- Will calibrate the volume and medium of consumption to personal preferences.
- Will avoid creating or amplifying filter bubbles and echo chambers, and in fact present some challenges or counterpoints to our views.
- Will be finite — pushing a bit of new content at a reasonable cadence, and no more.
To bring such a system into existence, we’ll be exploring ways to re-introduce friction in a smart way, to restore balance and make the world of content more consumer-friendly while embracing the best of what new technology has to offer.