In this article we are going to talk about how engineers can make huge mistakes when they do not exhibit the attributes of good engineering. And, we will also see how people without engineering training can make better engineers (better than you even with your training) when they do exhibit these attributes.
Thus, this article is both a cautionary tale for engineers and a lesson in how you can always be the best engineer.
So let’s talk about robots – specifically, humanoid robots.
To start this conversation, take a look at the video below to see how this journalist with little science training takes apart supposed engineering and robot experts like Elon Musk. It goes for about 20 minutes, but it is very interesting, it will help inform what I will say next, and does provide a good engineering case study in its own right.

What’s your take on it?
For me, the following were the significant takeaways:

• People like Elon Musk and Bernt Børnich do seem to be victims of attachment – they want the humanoid robot.
• It is not unusual for people to want a human robot servant – the series of times people seemed to want to believe such technology was possible or close to being possible is congruent with such a desire.
• There was at least an attempt by Elon Musk to make it sound like he was being logical – the world is made for humans so robots should be too.
• But I think that is a case of the core psychological principle that all in sales and marketing understand and use – people make decisions based on emotions and then justify those decisions with logic later on.
• The journalist actually sounded more like the expert engineer:
  • Showing no attachment to the idea of humanoid robots.
  • Using systemic thinking by expanding the scope to having multiple robots dedicated to specific tasks.
  • Using first principles to understand the difficulty of bipedalism.
  • Using first principles to understand how much data would be needed for training and why the established datasets are not sufficient.
  • Noting the framing of the designers who conceived and developed robot vacuums.
  • Bringing all of the above together through logic to reach the conclusion, which would also help establish a frame, that robots should be designed for the job that they do – and thus avoid the unnecessary complexity that comes with the humanoid robot.
• While I would not be investing in 1X, I just can’t see their Neo robot going anywhere, I do think that the challenge of producing such a robot could produce some remarkable and valuable spin-off technologies.

​What do we take away from this?

First thing, it shows how easily we can ignore what we know about good engineering practice when there is something that appeals to our emotions.

In a global context, this can be even more significant; different cultures would, because they have different aesthetics and values, likely be emotionally aroused by different things.

If you are in a mixed team, then you might experience your colleagues getting excited by an idea, and then losing all engineering judgement and expertise, that leaves you feeling very uninspired. That means you can bring them back to good engineering practice.

The second thing to note is how powerful these principles of expert engineering – framing, systemic thinking, and first principles – can be. If they allow a journalist with no engineering training outperform Elon Musk, then think what they can do for you as a trained engineer.

So on the other hand, if you are in a mixed team, you might find your colleagues can bring you back when you are excited by something that appeals to you, and only you, due to your background. And that can help you be a better engineer.

Ideally, we can all do this ourselves, but there is no harm in having colleagues who help. And this is more likely in a mixed team.

If you want to know more about mastering these attributes, and you have not acquired a copy yet, then check out my book on the topic.

Think back now about times your attachment to an idea made you forget good engineering principles. Can you stop this happening again; do you wish there had been someone there to point it out to you?

Or When fear and shame override logicWelcome to the next “What would an Engineer Do?” article.
As a reminder, these articles take current issues that sit outside engineering and look at them through an engineering lens.
The goal is twofold:

1. To help you better understand the core attributes of engineering expertise by seeing how they apply elsewhere. Sometimes the different context makes things easier to understand.
2. To show how those same attributes can be applied outside of engineering. This allows you to leverage your engineering skills even more globally.

In this article I am going to apply good global engineering principles to shaken baby syndrome.

Why shaken baby syndrome?
Depending on the country you are in, you might have seen debate about the validity of evidence used in shaken baby syndrome convictions.
You might also be in a country where courts now require an independent witness. The physical evidence alone is no longer considered sufficient.
At the very least, you may remember a time when that physical evidence was accepted as proof.
It is in a state of flux so it is a timely topic, which makes for greater interest. It is also well outside of what many would assume is the domain of engineering.

Some background
You can read more about the science and controversy around shaken baby syndrome here, but the key points to note are:

• It isn’t ethically possible to run proper double-blind experiments in this context.
• It hasn’t been proved that no other mechanisms can produce the same symptoms: subdural haemorrhage, retinal haemorrhage, and encephalopathy.
• Other medical conditions are known to sometimes cause similar effects.
• Reviews that claimed to support the shaken baby hypothesis often relied on circular reasoning: the cases examined had already been classified as abuse victims based on the assumed evidence.

If you think like an engineer, and especially like a global engineer, you know that logic and first principles must guide your thinking. You also know that first principles are found through the application of the scientific method. And in this case, there are no first principles that justify concluding that “shaken baby syndrome” has occurred.
And that means something more concerning.  Because the evidence that was used as first principles cannot be treated as first principles, around the world people have been convicted of a crime they did not commit. And a terrible crime at that – so terrible they would never have committed it.
And yet still, when courts are confronted with reports challenging the status quo based on the above, some judges have responded by saying words to the effects of:

• The argument that the scientific evidence is not actually scientific is radical.
• It seeks to set aside decades of study.
• It stands against other respectable scientific opinion.

For any engineer, that kind of reasoning is known to be flawed. It reveals a misunderstanding of how science works. And when it comes to the scientific method, an engineer should think like any other scientist.
You likely recall Albert Einstein’s response to the book titled 100 Authors Against Einstein. He said “Why one hundred? If I were wrong, one would have been enough.” This shows that science works on facts and logic – not popularity – and a single piece of evidence that contradicts a theory disproves that theory.
Science is not based on consensus or longevity of an idea. It rests on evidence and logic. And in this instance, it seems the logic has been lost.

How are we in this situation? More use of engineering expertise principles
Did the judges not understand science? Or, was there something else going on, something that would be familiar to the global engineer?
Imagine if you were a judge who just had it suggested to them that a key piece of evidence that the legal profession relies has come under question. For context, the legal profession relied on this so much that some defendants said that their own lawyers did not believe them.  You would start thinking that maybe many innocent people have been wrongly convicted. That is not a pleasant thought, you would be attached to the original idea that the evidence is strong and your profession has done nothing wrong. You would be fixated on it – this would make it hard to accept contradictory evidence.
Ideally, this attachment would not result in a fixation that would override the proper application of first principles.
As an engineer, you know that once contradictory evidence emerges, previous conclusions must be revisited.
So, we would hope and expect, that an engineer would, when in such a situation, understand the weakness of the theory, and acknowledge that all prior decisions made (under the assumption the theory was a strong one) are not justified.
First principles should override fixation and attachment. But this is not what seemed to happen with these judges.

The takeaway for the global engineer
This case highlights a deeper professional lesson.
Are you willing to hold yourself to the same standard; detaching from your own preferred theories, your past assumptions, and maybe even your professional pride when the evidence shifts?
That can be the challenge of genuine first-principles thinking.
Think back to a time when you were attached to an idea that clouded your judgement. Or when you saw a colleague resist evidence that contradicted their preferred model. Anyone can do it. The key is to notice it and then to do your best to let go of it – no matter how serious the issue at hand.
 
References used
https://en.wikipedia.org/wiki/Norman_Guthkelch
https://www.theage.com.au/national/australian-court-ruling-in-shaken-baby-case-was-ignorant-and-embarrassing-20251013-p5n25z.html
https://www.theage.com.au/interactive/2025/diagnosing-murder
https://www.brisbanetimes.com.au/national/this-man-spent-six-years-in-jail-but-experts-say-his-case-has-question-marks-all-over-it-20251029-p5n6c9.html