Don Jones

Tech | Career | Musings

Grab a copy of your company’s employee handbook. Chances are, it’s more than a couple of pages long. If you’re at a large company, it’s probably huge, and filled with rules – including complex time-off provisions, dress codes, policies, and more. You might, in fact, find some of it to be pretty ridiculous.

But every rule in that handbook – and indeed, most rules in most parts of life – are the direct result of some failure in the past.

Companies didn’t even have handbooks until someone, at some point, sued some company for unevenly applying its policies. That failure led that company to adopt written policies that could be seen as evenly applied to all employees. That’s not a bad thing, mind you – but it’s important to recognize where rules like these originate.

Human nature is to learn from failure. We rarely anticipate failure and proactively guard against it; our brains simply aren’t wired very well for that activity. Ancient civilizations probably didn’t pass laws against, and punishments for, things like murder until said murders actually started happening in a big way. When we started paying attention to sexual harassment, we started documenting rules against it. When we started paying attention to poor clothing choices, we started documenting dress codes. Rules are an attempt to prevent repeated failure by outlining what’s gone wrong in the past in a sort of prescriptive sense.

There’s nothing wrong with failing, and then documenting something to prevent that failure in the future. I’m not suggesting that rules as bad, in any way. But we’re epically bad at documenting why we passed the rules. You see this in nearly every major body of rules our society has – laws, ordinances, and even the rules used to check-in code in your source control system. “All user passwords must be at least 8 characters long.” OK – why? Well, the “why” of the rule’s existence is probably to help slow down an attacker who’s trying to brute-force passwords. Fine. But why 8 characters? We rarely document the full rationale for the rule. If someone had written it as, “…because 8 characters is longer than anyone here in 1992 can possibly crack in a lifetime,” then it helps us understand the context for the rule. We could read that now, in 2017, and realize that the rule’s details are no longer true. We could know to revise the rule, get rid of it entirely in favor of some more effective rule, or something else.

You see, if rules are the end result of some past failure, and if their purpose is to prevent future failure, we have to give them more context. We have to affix them to a point in time, and we have to justify them by explaining the failure they’re meant to address. That provides future generations with the knowledge to evaluate those rules.

Take a large-scale example like the 2nd amendment to the US Constitution. Many Europeans cannot comprehend why the US is so enamored of gun ownership, because the 2nd Amendment provides almost no context for its existence. It surely came about as the result of some failure – all rules do – but what was it? In this case, it was the vexing tendency of the British government to confiscate guns from the American colonials. This had the effect of depriving them from food sources (e.g., hunting), and keeping them from rebelling effectively against British rule. That context is fairly well-baked into our culture, so perhaps it doesn’t need to be written down, but it should be written down regardless. Any modern application of the 2nd Amendment should be made with that context in mind. For example, our context suggests that one solid reason for American gun ownership is to preserve some ability for us to rebel against our government. And while that may be unthinkable to many, it’s part of our context. And new contexts since the 1700s have grown up around the Amendment, and we’ve never  really documented those, either – and so a lot of the laws we have which descend from the 2nd Amendment can seem nonsensical.

The morale: Pass rules. Prevent repeated failure. But also write down a lot of context, so your rule can be fairly evaluated as the situation on the ground changes.

%d bloggers like this: