Written: 2024-04-02

Embrace the boring technologies

The main idea is developed in a presentation by Dan McKinley (and the short blog/essay version)

I love the simple concept.
It resonate with what I've seen in the aerospace business.
It resonate with my demoscene activities where smart tricks are often simpler than what it looks like.

If your core business is not about selling software, but is about selling a service or any complex product, then industrial maturity is key.
Fancy technology is probably not in line with you domain expectations.

To me the main outcomes are:

• the known unknowns of a boring tech (e.g, well-understood failure modes) are a lot less risky than unknown unknowns of anynew fancy tech (where does it end?)
• being polyglot tech needs a lots of knowledge and resources and increase the cognitive overhead
• happiness comes from delivering meaningful work
• production / integration / testing / deployment are hard. It takes a lot of time to achieve the right level of production-ready maturity.
• Keep It Simple Stupid: as told so many times making things simple is hardwork, but making complex things perfectly working is even harder...

Common biases

In no-specific order:

• optimist assumptions and over-simplification of the reality
• attraction for novelty and playing the "bad cop" is not popular
• hero-plan validated by people which are not part of the plan (ownership)
• technology seen as the only way to be innovative
• "old-school" is not "shiny" enough to spent R&T budgets
• decision making based on believes/concepts rather than real-world numbers

I have no silver bullet to overcome those biases. Beyond setting clear and measurable expectations, if one, it would be on the engagement level and trustability of those pushing the tech before starting.

• How much do they personaly engage in the operational deployment?
• In the contrary, how much do they push back as soon as a critical milestone reaches?
• Who is accountable in case of failure?

Good reasons to switch

You achieved high level of mastery and the bright new tech will help you jump in another stellar productivity level:

• be highly suspicious about the demonstration (excess of confidence)
• discuss and check the numbers several times (confirmation bias)
• think of the human factor and non-elite learning curve (does it scale on human skills?)

The gold skill: from fancy technologist to old-school boring tech

Any company needs to try-and-learn new stuffs and fancy technologies are one of them.

When you try a new technology, it is to learn how to use it, if it matches or disrupt expected performances, or if it opens new doors in your product features or in the ways of working and operate.

One day I had to disrupt a system engineering process based on an over-hyped architecting tool. I focused on what really matters which was the global project velocity rather than a shiny interface. Few people were mocking like: "you're not gonna use this Grandpa'tech!". The grandpa'tech was a spreadsheet tool.

But the grandpa'tech was known by everyone, so people scalability was a no-brainer and mass-editing and integrating high volumes of data became smoother. Then I was able to focus on the system to software transition without loosing time on a flawed-by-design interface originally meant to model boxes and wires (I over-simplify for the story...)

Later, other people were able to modernize the editing process to make it a state-of-the-art collaboration tool and replaced the grandpa'tech, with low impact on the model, process and operational data.

With boring techs, comes boring engineering management: for continuous industrial operations, plan changes step-by-step and avoid loosing maturity on all technology stacks at the same time.