When software and hardware engineers talk about reliable systems, we talk about uptime. This is the idea that a system is reliably doing what it is intended to do, and not in an error or failure state.
Uptime is usually expressed in a number, like 99.9%. That would be known as “three nines” of uptime.
Thanks to this handy uptime calculator, we know three nines is:
level of 99.9 % uptime/availability results in the following periods of allowed downtime/unavailability:
- Daily: 1m 26.4s
- Weekly: 10m 4.8s
- Monthly: 43m 49.7s
- Yearly: 8h 45m 57.0s
Those times above are the amount of time spent in failure state. So an uptime calculation is actually saying that at 99.9% uptime, a system is down ten minutes a week.
Don’t worry, this isn’t a post about engineering uptime. It’s about people.
See, humans don’t have the concept of uptime in our personal lives or in society. And by uptime I don’t mean non-sleeping time, I mean time when you are alive and not in an error state.
What’s an error state for a human? Well, it would be up for debate, but I would suggest that times when you are consumed by depression, addiction, anger, fear, etc represent the failure states of being a human being.
What is an error state in a society? Perhaps it is when we are consumed by political minutia instead of real issues, when we vote based on fear and ignorance, when we are led by imbicles, when we allow our nation to commit atrocities unchecked.
But as any engineer knows, it’s not enough to say “we are at 100% uptime!”. You have to measure the error states and track it over time.
When we ask questions like “will technology X help or hurt society or individual people?”, It shouldn’t be simply a slew of opinions. Questions about ethics over time should be able to be classified into error or success states.
Does this approach deemphasize trying to figure out how to maximize uptime to give us the highest highs? It most certainly does. That’s the point, which I guess makes this uptime approach similar to classical Stoacism.
This approach doesn’t tell you how to live a good life, but it would measure when you are having a bad one.
So how would we measure our uptime, or downtime, as a society or individual? Well, we try to currently with clumsy metrics like the Dow Jones Industrial Average or our own bank accounts. Ugh. It’s like trying to figure out an error state by watching the movements of distant stars.
We need better measurement instead of the classical pseudosciences of Economics, Philosophy, Psychology, Sociology, polling, etc.
Each individual should be able to clearly define an error state, and then track when they are in it. And the health of a society is computed by aggregations of those metrics.
We will likely need AI to do this effectively. But not AI controlled by folks trying to shove ads for toothpaste down our throats. We will need systems that can observe us, and tell us when something is wrong, and guide us to healthier choices in life to get us our of error states. But we don’t need them quite yet – we can track our own uptime/downtine today.
Because you can’t fix something if you can’t measure it. And if you want to get to three nines, you may need more than an occasional session at a therapist.
So try to conduct an experiment on yourself in determining the amount of time you in an error state, which we will define as being in a foul mood, being in pain, anger, etc.
Don’t worry about registering how happy you are, just in registering when you are not. Count a good night’s sleep as uptime, but a bad night’s sleep as some amount of downtime.
Just record the downtime to the minute or hour. Doesn’t have to be exact to start.
I will share my uptime data once I collect a few weeks of it.
Image from reddit somewhere.