Nadeesha Cabral


Opinions are a liability

July 19, 2020

    Opinions that we hold, that do affect the world external to us, are a liability. The very nature of opinions dictate that we can’t really test them and come to a conclusion. If we do, it crosses a boundary and becomes a fact.

    In large engineering projects, we will have to make a number of trade-offs. Some of these will find answers in a shared technical vision. For example, “working software at 99th percentile vs. software that survives a catastrophic event that spans multiple continents”. But in more murkier waters where observations are limited, and we don’t have as much data, lie opinion.

    Opinions are necessary

    Opinions fill the void that hypothesis and fact cannot. Not everything can be analysed quantitatively and tested. Opinions are the blinders on the horse of me that guards against decision fatigue and analysis paralysis. This is especially true when the road ahead is not clear.

    For example, I hold the opinion that early stage software projects without crazy throughput requirements should store data in relational databases. This is informed by my observations that early stage software projects pivot regularly, and do not hit the scaling requirements needed until significant value is delivered. I might wrong in some cases, but I can avoid the decision fatigue for most of the cases.

    It takes work to build an opinion

    Opinions are not free. It takes work to build an opinion. It takes analyzing the observations, and carefully arriving at a personal belief. Holding an opinion without doing the work would be akin to claiming that the house that my land sits in is “sovereign land”. You might amuse a few, but you’ll still be paying tax.

    Work you put into forming the opinion either results in you concluding the opinion is not correct, or making it as hard as possible for another person to conclude that it’s not correct.

    It takes work to maintain an opinion

    The (inconclusive) opinions that we hold has a set of assumptions. The underlying data of these are constantly in flux. New data comes to you all the time. Either the opinion must be re-evaluated to encompass the altered finding, or abandoned.

    An opinion that doesn’t get enough investment for it’s defense will ultimately lose it’s value and become obsolete.

    It takes work to get the buy-in

    No matter how right you are, if you don’t want to do the work to establish consensus, people just don’t want to give you the satisfaction. On the other hand, even if you’re not sure whether you’re right about something, giving others a stake in the opinion will create a shared responsibility for evolving the opinion and maintaining it.

    Good fortified opinions are easy to defend. Opinions with consensus and shared understanding preempt attacks.

    Having vs. not having opinions

    Much like how a great empire wages great wars for amassing land and extending borders, having opinions of significance will increase the overall effort you must expend to defend them. On the other hand, not having them in circumstances that matter will make it harder for you to set your agenda and limit your influence.

    Energy is limited and the cognitive burden of having an opinion and defending it is a cost that must be budgeted for.

    Opinions are a liability. As other things I write here, this is my opinion. And just like other essays, this is a liability.