The Slowest Hiker
How to improve any process.
One of the two most often requested features of Go is the support for generics. Arguably, generics are super important and they are taken for granted in most of the modern programming languages. In fact, Go developers get a lot of flak from their Haskell-minded colleagues being forced to use a language with such a poor feature set. And yet, Go is one of the fastest growing languages on Github. It seems like it became the lingua franca of the cloud, with more and more distributed systems being written (or rewritten) in Go.
How did that happen? Let’s take a look at the intro to go2 generics proposal:
The Go team (..) has been investigating and discussing possible designs for “generics” (…) since before Go’s first open source release. We understood (…) the topic was rich and complex and would take a long time to understand well enough to design a good solution.
Instead of attempting that at the start, we spent our time on features more directly applicable to Go’s initial target of networked system software (now “cloud software”), such as concurrency, scalable builds, and low-latency garbage collection.
That’s an application of the Theory of constraints.
How to optimize any process?
- Find out what the goal is.
- Find out your constraints (aka blockers, or bottlenecks, or slowest hikers).
- Improve them, one blocker at a time.
Improving constraints will improve the whole process.
How do you know if you’ve unidentified the constraint? If you speed it up, the whole process will speed up; if you slow it down, the process will slow down. Easy.
It is super important to eliminate the obstacles first, before attempting to aggregate all the marginal gains. Sure, marginal gains aggregation works for Team Sky, but then competitive cycling is already a heavily optimized process; marginal gains maybe all that is left there.
The slowest hiker
Theory of constraints was introduced in The Goal. The book offers a story of a plant manager, who was asked to supervise a boy scout team hike. The whole team is supposed to be back at the campsite together before the sun sets. However not all of the members hike with the same speed; as the result the group splits, some folks need to wait and it looks like they are not going to make it.
The solution is to put the slowest hiker, Herbie, at the front of the group. This illustrates that the group can go only as fast as their slowest hiker. Then we can focus on optimizing Herbie. He has a heavy backpack, we can distribute his load among the group. This speeds him up and the group makes it on time. The book is from 1984 in the modern, cut-throat world of Jeff Bezoses, Herbie would likely be fired ;)
What’s the most important thing to do right now?
Try to think and question the assumptions to understand what the constraint is, and how it can be eliminated. Ask why? until you find the root cause.
Focus on the most important thing right now — take a few minutes every now and then to think about the big picture and what is the biggest blocker to the realization of your goals. Fix it, and move to the next one. Say “no” to things which are not important right now (don’t aggregate the marginal gains before the obstacles are eliminated), in order to say “yes” to the most important problem at a given time.
Theory of constraints may in part explain Go success. The team focused on addressing common software engineering challenges first. The result was language perfect for codebases maintained by large teams over time. Its success in the opensource projects speaks for that. When the slowest hikers are addressed the team has enough of a breathing room to address next issues.
They’ve kept an open mind and said “no” to commonly requested features in order to focus on constraints preventing them from reaching their goal.
Theory of constraints provides a useful mental model for tackling various optimizations or delivery problems. It originated in the manufacturing space, but this model is applicable in many other spaces. Especially in the iterative world of software engineering — figure out the goal, ask yourself whats the biggest constraint right now is, improve and repeat.