I recently saw a tweet about Tobi Lutke, CEO of Shopify, operating on Crocker’s Law.
What is Crocker’s Law?
Well, Crocker’s a Wikipedia editor who asked people to not apologise for editing his pages. He just cared about making his pages better and wanted them to do the same. If he was offended, it was his fault.
What a guy! (Takes a lot of self-esteem I bet). If we follow Crocker’s law, we don’t let ourselves be affected by feedback—instead, we use it to fuel our improvement. Here’s what Tobi had to say about it:
“Feedback is a gift. It clearly is. It’s not meant to hurt. It’s meant to move things forward, to demystify something for you. … Just give me the raw feedback without all the shit sandwich around it.”
Once opinions are formed, they persist. In 1975, Stanford researchers conducted an experiment where participants were presented with a pair of suicide notes—a real note by someone who committed suicide and a fake note by a random person. Participants were asked to identify the real notes.
Some participants had a talent for this—they identified the real note more than 95% of the time. Others had a success rate of 40%, worse than a coin flip.
But this was a ruse. The scores were made up (though half of the notes were indeed real). Participants who were deceived that they were 95% right were not more perceptive than those who were 40% right.
After revealing the truth, participants were asked to estimate how many they actually got right. Here’s the punchline: Participants who mistakenly thought they did well (95% correct)—due to the deception—continued to estimate a high score; participants who thought they did badly (40% correct) continued to estimate a low score.
Even when presented with disconfirming evidence, once those participants formed an (incorrect) opinion (of themselves), those opinions persisted. That’s belief perseverance–it explains why it’s so difficult to change our views.
But what if we agree with the feedback, and choose not to accept it? Well, it gets uncomfortable on the inside–here’s why.
Clashing opinions can lead to cognitive dissonance. Cognition means thinking and reasoning; dissonance means tension or clash from contradictory elements.
Cognitive dissonance happens when you simultaneously have two (or more) contradictory views—such when you agree with the feedback but don’t want to accept it. Thus, your existing views clash with new evidence (i.e., feedback) and causes mental discomfort.
You can resolve the inconsistency and alleviate the discomfort in two ways: Change your views, or ignore or deny the new evidence. Guess which is easier? (It’s the latter.) Cognitive dissonance helps explain why people vigorously defend, justify, and excuse their incorrect beliefs in the face of new evidence, or in this case, feedback. (We can be pretty stubborn, huh?)
Some other factors affecting how we accept feedback:
Hey, I get it. Feedback is uncomfortable. Ironically, the most useful feedback is often the hardest to accept. There’s a Chinese saying: “良药苦口 (Good medicine is bitter)”. To coax myself to take this bitter medicine, I’ll briefly list the medicinal properties.
Feedback helps you learn. When you receive feedback, especially that which contradicts your existing views, you gain a perspective you might not have considered before. Learn from it and perhaps revise your outdated or incorrect knowledge.
“The illiterate of the 21st century will not be those who cannot read and write, but those who cannot learn, unlearn, and relearn. ” — Alvin Toffler
Feedback helps you make better decisions. When you get feedback, you might change your initial opinion or decision. Were you wrong previously? Does this show you didn’t really know what you were talking about? Well, not necessarily. Here’s the advice Jeff Bezos shared with 37signals (now Basecamp):
People who were right a lot of the time were people who often changed their minds. … The smartest people are constantly revising their understanding, reconsidering a problem they thought they’d already solved. They’re open to new points of view, new information, new ideas, contradictions, and challenges to their way of thinking.
Thus, changing your mind could help you arrive at better decisions. It’s kinda like SGD–take a quick decision, get feedback, update decision, repeat. With enough iterations, you arrive at global optima.
Feedback helps you improve. An effective way to improve your skills is through deliberate practice. For this to work, repetition and tight feedback loops are essential. Repetition is pointless if you don’t know whether you’re getting better, or how to get better—thus, feedback (read: supervised learning) is key.
In The Making of An Expert, Anders Ericsson (author of Peak) and his team found that elite performers knew what they were doing right and focused on what they were doing wrong (i.e., they used AdaBoost). “Real experts seek out constructive, even painful feedback”, Ericsson wrote.
Receiving feedback—especially negative feedback—can be difficult. (I’ll admit, I’ve a secret fear of being criticised; I can’t shrug it off as well as most people can.) Here’s advice that helped.
Reframe your thinking. Who stands to gain most from feedback? The receiver! You! Feedback really is a gift. (Though granted, some criticism, disguised as feedback, is meant to injure). If someone takes the effort to go through your work and provide feedback, be grateful.
Here’s what a mentor of mine, a Principal Engineer at Amazon, had to say about this:
“I love getting feedback on my docs, especially if they lead to substantial changes in my thinking, or make me scrap the idea completely. This means I’ve saved myself a lot of work in the wrong direction. It’s good to learn this upfront from feedback, rather than after you’ve done the work.”
Weigh feedback based on believability. I got this from Ray Dalio’s book, Principles. (Ray Dalio is the founder of Bridgewater Associates). He was discussing it in the context of decision-making but I think it applies to feedback as well.
Believability weight the feedback you receive; weigh the opinions of believable people more. How do you determine believability? Dalio shared that believable people have done two things:
If the feedback source has high believability, try to understand their perspective and learn from them. If the person is similar in believability, debate to arrive at the best outcome or conclusion. If the person is less believable, consider their objections and feedback, but don’t spend too much time on it. Be like L2 regularization.
Don’t be too affected by negative feedback and outright criticism. As you put your work out there, you’ll get feedback. Some of it will hurt. It could be in the form of straw-manning your work, baseless criticism, or sidestepping it to attack you. Recognise when this happens and don’t be too affected by it. Or better yet, ignore it like L1 regularization.
“It never ceases to amaze me: we all love ourselves more than other people, but care more about their opinion than our own” – Marcus Aurelius
Don’t stop creating, shipping, writing just because you got some negative feedback. Please.
Learn from Crocker. Learn from Tobi. Learn from the other giants mentioned here. Feedback is meant to help–it really is. Believability weight the feedback, and don’t take it personally.
“We all need people who will give us feedback. That’s how we improve.” – Bill Gates
But remember, Crocker’s Law doesn’t mean we can insult others; it means others don’t have to worry that they are insulting us.
P.S., Jeff Bezo’s description of “people who are right a lot” sounds a lot like intellectual humility, covered in this recent video here.
Thanks to Yang Xinyi, Marianne Tan, Gabriel Chuan, and Chng Yee Siang for reading drafts of this.
If you found this useful, please cite this write-up as:
Yan, Ziyou. (May 2020). How to Apply Crocker's Law for Feedback and Growth. eugeneyan.com. https://eugeneyan.com/writing/how-to-apply-crockers-law-for-feedback-and-growth/.
or
@article{yan2020feedback,
title = {How to Apply Crocker's Law for Feedback and Growth},
author = {Yan, Ziyou},
journal = {eugeneyan.com},
year = {2020},
month = {May},
url = {https://eugeneyan.com/writing/how-to-apply-crockers-law-for-feedback-and-growth/}
}
Join 9,300+ readers getting updates on machine learning, RecSys, LLMs, and engineering.