this applies to every single skill should one desire to be maximally competent, even in a single bounded domain; there is but one Skill, of which all others are but projections
learning just enough of everything to identify good work makes your BS meter antifragile
---
RT @normonics
Everyone should learn a bit of modeling and simulation so that they can become appropriately skeptical of it
https://twitter.com/normonics/status/1471110980054704132
if you start a company, for ex., you'll inevitably have to delegate critical work functions to collaborators; would you rather operate entirely on trust, or offset your risks with the ability to validate their outputs?
some would call this soulless, but it's merely due diligence
in the real world, not only is it quite rare to find business collaborators you trust on the level of close friends/family, but those too are capable of making mistakes; is it not then an act of kindness to release the from the burden of perfection thru mutual accountability?
I've found that I can build much deeper trust in someone's competence thru iterated monte carlo output validation than by simply trusting them based on character or recommendation; this validated trust allows me to release cognitive load which would otherwise be spent worrying
there'd a russian saying: доверяй но проверяй, translated as "trust, but verify"; this has very much been a motto for me, and has been quite instrumental thruout the torturous process of learning to delegate work to others, which I've always struggled with, having high standards
never shy away from learning something far from your core competence, nor from underpinning your trust with evidence; both practices will make you a more effective collaborator in the long run
thru thus process I've learned the value of having skilled collaborators in domains you're only minimal familiar with, but still enough to check their work; the trust they earn professionally filters back out to the interpersonal realm as well, building stronger relationships