“People don’t want to buy a quarter-inch drill. They want a quarter-inch hole!” Theodore Levitt, Harvard Business School
It used to be that if you wanted to run a website, you had to buy a physical server, networking equipment, storage. Then you needed to install an operating system, webserver, database, and other software and tools required to run the website.
You don’t have to do that anymore. Owning physical hardware didn’t add differentiating value. Managing an operating system didn’t add differentiating value.
What do you do when something doesn’t add differentiating value? You outsource it.
Those activities don’t disappear, they’re just done by someone else. Someone (attribution required) cleverly represented this with the analogy of “Pizza as a Service”. Every time you eat pizza, those activities are going to be involved (except possibly soda, because I’m not a big soda drinker). There will always be cheese, toppings, sauce, dough, fire, an oven, fuel, a place to consume it.
“Serverless” computing is a misleading term. In essence, whereas you used to need a server to run for 24 hours a day listening for certain things to happen in order to execute code, companies like Google Cloud (my employer) now have services where you don’t have to leave that server running anymore. They’ll run the server that’s listening. When a request comes, they’ll run your code. Instead of paying for the server to run 24/7, you’re paying for the 10 minutes per day that your code runs. You’ve eliminated waste (muda) in the system.
Now on to the point of this post. What if we applied the principles of serverless computing to our jobs? We don’t need to be up and running for X hours per day, we need to deliver the output that is expected of us. Our differentiated value.
The concept isn’t new, and I’ve thought about it in some variation for years, but it hit me in a different way today. There is still a contractual agreement, written or unwritten, that employees will work XX hours per week. We don’t know how to get beyond that. There were experiments with Results Only Work Environments (ROWE). A lot of people have explored the space. Nobody’s quite nailed it yet.
Idling some machines in a system generally leads to better system outcomes than overproducing just to stay busy. Read “The Goal” by Goldratt. There is so much of Lean in agile, devops, and hardware and software architecture, as well as business models.
In the spirit of challenging traditions, I’ve decided to start challenging one myself. I started this website and blog to share what I was learning and thinking with others. That’s no longer the intent. If you look at the frequency of my posting, you’ll notice that it’s down to yearly. At the same time, my list of things to blog about has over 200 items on it. Where is the disconnect? Me wanting blog posts to be perfect for other people to read them.
My solution? This blog is now going to be my online notebook. I’m going to publish my first draft of my thoughts. If my thoughts evolve or I want to add more detail, I’ll make changes. If some of the writing is rough or half-formed, it’s because the biggest theories in life take years to work out. Read about The Slow Hunch in Stephen Johnson’s book “Where Good Ideas Come From”.