Dealing with Unkown Unknowns

  • #philosophy
  • #software
  • Unknown Unknowns are things that you don’t know you don’t know, made infamous by Donald Rumsfeld’s comments around the Iraq War. Any sane / rational person can deal with known unknowns, it makes sense to get a grasp on things that you don’t have much context on or are uncertain about. Unknown Unknowns are more insidious and trickier to deal with. The question then is how do you take Unknown Unknowns and make them known and therefore easier to deal with?

    How can you reduce the space of unknown unknowns?

    First, before we jump into this question, lets define some properties of unknown unknowns.

    1. There are an infinite number of unknowns, you can think about this as inhabiting an infinite space. Where that space is defined as the inverse of known problems.

    2. Therefore the more you know the less room Unknown Unknowns have to surface.

    As they’re nearly infinite, it’s clear that some Unknown Unknowns are more important than others. Where should we spend time looking to uncover the ones that actually make a difference?

    Generally problems are much closer to the area you’re working in, which just makes intuitive sense. If you're working on a way to bring safe drinking water to places that don't have it consistently you're not going to be searching for possible issues in a paper about new computer chip technologies. (If someone points you that direction it's another story.)

    You should focus on building out context and retaining it, therefore decreasing the space unknown unknowns can inhabit. It's important to choose a few directions in which you want to focus, if you spread yourself thin across may subjects you'll see less benefit. Many other people have the same skills you've just acquired, making those skills less valuable. Ideally you're broad enough to see the connections, but specialized enough to understand the details.

    One person can only know so much, so it’s important to leverage your peers in understanding things that you don’t. There are certainly many things that you do not know. Seek out people who increase the footprint of your knowledge, making it more likely that unknown unknowns will be found.

    How can we make good choices when we acknowledge that there are unknown unknowns?

    Once we acknowledge that Unknown Unknowns exist, it can be hard to make a choice without spending exhaustive amounts of time trying to shake them out. Fundamentally, how can we make sound choices with limited or partial information?

    First, it’s important to realize that the choices you make are only good or bad given the information you had at the time. If you run into a bad unknown unknown it wasn’t because you made a bad choice. We can’t judge ourselves as if we had complete information about how things would have turned out.

    Second, there’s no formula or process for this! You have to uncover unknown unknowns until you’re confident you have a good understanding of the paths in front of you. You should probably be slightly uncomfortable if you acknowledge that you may hit a unknown unknown, that’s good and means you aren’t wasting time building up unneeded context.

    Knowing that unknown unknowns exist is the first step of the process, and perhaps the most important. So we’ve already checked the first and largest box.