Strong opinion, weakly held


I enjoy having very interesting, but at the same time, very heated technical discussions with people. These discussions made me reflect on two feedback that I was given throughout my career:

  • Carl, your opinion is too strong, it's not possible to convince you.
  • Carl, you change your mind too quickly, what you are suggesting now contradicts what you were suggesting before.

These two feedback are not wrong, they both reflect real scenarios where I was either holding a very strong opinion or I was changing my opinion very rapidly. But they might sound contradicting to each other, since how can a strong-minded person change their ideas all the time.

You might have heard about the very catchy phrase: "Strong opinion, weakly held", this framework of thinking will lead to exactly the phenomenon that people describe me in the above feedback. The idea of the Strong opinion, weakly held framework is very simple: when we are making decisions for new things, we are effectively doing forecasts, and in order to extrapolate our past knowledge into the future, we will need data. Instead of waiting for unguided data gathering before making a judgment, we form a series of hypotheses based on expert intuition, common sense and the available data. Then we take a step toward testing the hypothesis that was just made, based on the outcome of the test, refine the hypothesis and repeat again.

During these 2-step iterations, you will see exactly the types of behavior that I was given feedback out above:


Steel Man the hypothesis

In order to start the process, we need to first Steel Man my hypothesis, we can do so by gathering all the data / facts that are available, then forming them into a logical, deductive thought process, the goal is NOT to prove that my hypothesis is right, the goal is to prove that my hypothesis is not wrong given the knowledge that is available. While doing this interactively during a discussion, we need to be careful to not let "Absence of Evidence" disprove our hypothesis.

On the other hand, if someone else is having an opinion and tries to convince you, we should do the same, try to help them to steel man their argument and see whether they can rationalize it. However, if the other person can not provide the reason that holds the hypotheses, and all you've got is the overconfident statement like "I'm telling you X because I'm the expert in this domain.", then I personally won't take the argument very seriously.

Note that the most important thing to make Strong opinion, weakly held work is that the person that come up with the hypothesis has to come up with the steel man reasoning, not the other way around, demanding others to provide strong evidence to prove the hypothesis wrong.


Take the action to test the hypothesis

An hypothesis in our work context usually takes the form of "solution X will work". Treat this as a null hypothesis and the easiest way to invalidate this hypothesis is usually: to prove that under the most simplification of the problem, "solution X" still doesn't work. If we can prove this through a quick experiment, then the original hypothesis is nullified, and there is no need to invest further towards proving the original hypothesis.

However if the simplest example using solution X works, it's NOT a proof that the original hypothesis is true. This information of a simple success should lead to further investigation of what part of the solution made it work, therefore enabling us to form more hypotheses around each part of the solution.


Rethink, given the new information

Any new information that is added after the initial hypothesis was formed should change the validity of the hypothesis. This information could either come from the test that we just did, it could also come from the change in the environment as a new global constraint. For example, when making a project decision in the beginning of 2022, the lack of engineering capacity to fund it might not be a reason to disprove a project proposal. But given the current much limited capacity after the layoff, this critical constraint will invalid many previous sound hypothesis, therefore it's not only reasonable, but also critical to evaluate all the previous decisions.

There is a technique that I learnt from the book Essentialism: The Disciplined Pursuit of Less to conduct the re-think process. It's as simple as "starting from zero again, given the new information". The book used the example of "how to clean up your wardrobe as an essentialist" and the method that the book suggested was to assume you don't have any clothes in the wardrobe, and now you want to rebuild your collection of clothes, which ones that you currently have would you still buy. Applying to conducting R&D projects, the analogy becomes: given the current knowledge and context, what would you do to the project if you were to get all your investment back and restart the project all over again.

Comments

Post a Comment

Popular posts from this blog

Risk vs. Uncertainty

Image
It's the planning season, "Risk" is one of the most discussed topics during this time. But when we say "Risk" we might mean different things. Therefore I took some time to reflect on the two most common confusions that we might have when talking about "Risk", that is Risk vs Uncertainty. I took a lot of inspiration from this note and I'm sharing some of my own thoughts and realization that might be useful. What is risk? Risks are bad things that can negatively impact the outcome, so we don't want them to happen. Risks are certain, therefore usually measurable / quantifiable. Given the nature of its certainty, we can usually mitigate them using tools like insurance, contingency and parallel paths etc. In the context of R&D, dealing with risks ultimately leads to solving a trade-off problem: you put in more effort, spend more money, extend the timeline or compromise the feature set, etc. An intuitive example : You are on the roof of a 20 flo...
Read more »

Feedback Loop

Image
Over the past few weeks, I've been reading "Thinking in Systems" by Donella Meadows . One of the mental models that I've reflected on is about feedback loops and how it is relevant to our work.  The concept of a feedback loop in systems is fairly simple, as in the attached photo: you do something, measure the outcome, analyze what leads to the outcome, correct the mistakes, then repeat the action. The biggest problem to deal with is system delay: when there's a significant delay between the first action, the first measurement, and the updated action, oscillation happens. We either commit too little action or too much, which both lead to undesirable results. An analogy and a concrete example can help to understand this. Analogy : Imagine you are taking a shower and you want the perfect temperature for the water. But your hot-cold water mixer is quite far away from the showerhead. What's going to happen is that there will be a few seconds' delay after you tu...
Read more »

Three tools to help think clearly

Image
Part of our job at a research lab is to clarify an ambiguous area and create a clear path to navigate it. To do this effectively, we need to dedicate significant time to think. I've found Mental Models , or tools for thinking, to be extremely useful during this process. I'd like to share three that I use quite often, in the hopes that others might also find them beneficial. Thought Exercise While there are numerous well-known exercises, the one I've found most useful involves hypothetical scenarios, or "what ifs". The basic idea is that you construct a hypothetical scenario which differs from the current reality. Then, you explore the problem within this alternative reality. Here are a few examples: What if I could discard the current system and rebuild it from scratch? What would I do differently? What if I could, hypothetically speaking, "fire" the entire current team, regain the headcount, and rehire one by one? Who would I want on the new team? What...
Read more »