I think everyone can relate to the heavily promoted idea that removing things makes you lighter, such as “shedding a few pounds”. And maybe a lot of people can relate to reducing exposure, such as “keeping your head down”.

That’s why I find it curious to read in Behavioral Scientist a claim that people “neglect” subtraction.

The problem is that we neglect subtraction. Compared to changes that add, those that subtract are harder to think of. Even when we do manage to think of it, subtracting can be harder to implement.

The basis of this article is a cute story about parenting.

An epiphany in my thinking about less came when my son Ezra and I were building a bridge out of Legos. Because the support towers were different heights, we couldn’t span them, so I reached behind me to grab a block to add to the shorter tower. As I turned back toward the soon-to-be bridge, three-year-old Ezra was removing a block from the taller tower. My impulse had been to add to the short support, and in that moment, I realized it was wrong: taking away from the tall support was a faster and more efficient way to create a level bridge.

This says to me right away that people are naturally inclined to subtract. It’s perhaps second nature. Don’t want to get your hand wet? Subtract it from exposure to rain.

However, instead of this line of thinking the author set about using a contrived Lego set to confirm that her mistake was some sort of grand mental conspiracy instead of just her being wrong — a confirmation bias experiment, if you will.

Since I had become a professor, I had been trying to convert my interest in less into something I could study instead of just ponder. […] I began carrying around a replica of Ezra’s bridge. I tried it out on unsuspecting students who came to meet with me, checking whether they would subtract, like Ezra, or add, like I had. All the students added. I also brought the Lego bridge to meetings with professors…

All her students and close professors thought much like she did, instead of like a three-year old? Color me shocked.

At the heart of this experiment is a fundamental point that a LOT of blocks had to be added in order to construct a bridge out of Legos. Then at a crucial point a decision to add or remove is measured, conveniently ignoring the rather significant fact that blocks have been added the entire time before then.

If I add 15 blocks and then remove one, how subtractive have I been really?

And if I have 100 blocks in my collection in order to create a 10 block structure instead of a 15 block one from a much smaller collection, how subtractive have I been really?

I’m not calling the study nonsense, as it does highlight what we all know already about the need to subtract things (e.g. surface area is also targeted area, less features means fewer potential vulnerability), but in all honestly… it reads to me that the author is so insecure about intelligence they had to start a huge campaign to explain why they didn’t think of something before their toddler did.

The analysis gets really wonky and I find it underwhelming. Let’s look again at that paragraph.

The problem is that we neglect subtraction. Compared to changes that add, those that subtract are harder to think of. Even when we do manage to think of it, subtracting can be harder to implement.

Perhaps they should have subtracted a lot of words? The New York Times had a better way to describe this that probably looks familiar to everyone.

Overwriting is a bigger problem than underwriting.

But seriously, subtraction is very easy when there’s incentive and it’s even very common, in some cases overused. Overwriting is a problem because it’s easier to write long form, harder to be short form. However, when you’re in a rush because of imminent threats or you have a much smaller vocabulary then short form gets a LOT easier. Any guesses how long a PhD thesis by a toddler would be?

Speaking of incentive to subtract, a fair number of sites on the Internet track layoffs by companies overeager to regularly reduce staff, for example.

I’m disappointed these incentive and risk angles weren’t explored more by behavioral scientists, but I suspect that’s because once the author was satisfied that they weren’t the only one who made a mistake, they settled into comfort of having published articles and collaborations to prove what they believed when they started (protecting their sense of being intelligent, as opposed to understanding their own bias).

A (dis)gruntled staff write-up gives some hints of the weak tech culture at Amazon.

The most surprising thing I encountered when joining was how manual most processes are. It blew my mind how many business critical processes were managed with excel spreadsheets being shared via email chains.

These processes track a rapid turnover of staff, while not really tracking the obvious risks such as problems leading to the turnover.

When I onboarded, there was a huge backlog of work that needed to be done with PMs and engineers asking for commitments on when I could deliver things by – urgent to not block progress, but also ironic that they are asking a person who has no clue what they are doing to deliver critical work. I realized over time that this was the norm. People come and go which has the net impact of making you feel like you are just a resource.

Given how the word “dystopia” was crafted during industrialization to be an inverse of utopia, in the past I have used the following phrase to help people understand what they might be getting themselves into:

The best people can ever achieve leaves them treated like owned and controlled assets in a dystopia…

This new article is a fascinating read that lays out just how horribly Amazon treats people, especially engineers and how weak their safety culture is as a result.

To be blunt: it sucks. To make matters worse, since each team owns its infrastructure, there are varying degrees to which safety mechanisms (automated testing, one-click rollbacks, logging, etc.) are implemented. This creates a chicken and the egg problem of infrastructure vs. product development. One usually happens at the expense of the other. So when the morale on the team dips (usually related to lack of clarity or setbacks making things take a long time), people leave the team, which further damages morale and can easily result in a mass exodus.

This complete lack of safety (for both product and staff) becomes even more alarming when the author announces that building is being done before design.

…most teams don’t want to “wait around” for design to assist in the process and are anxious to start building, which usually results in half solutions. It is easier to partially solve the problem than think through things from first principles. …employees become “resources” (cogs) plugged into gaps to stop the bleeding rather than contributing out of their core competencies. This was further evidenced when one of the projects I was assigned to had an engineer who had never done web development (didn’t know HTTP, HTML, CSS, or JavaScript) was tasked with doing the front end system design for a React application as well as the API that would power the UI.

This is a great read and emphasizes clearly how a very talented and technical individual could not even last a year at Amazon because the culture is so unmistakably unsafe.

The great irony is how the author emphasizes the documentation requirements of Amazon, driven by traditional manual processes using simple spreadsheets and file sharing through email (all very non-cloud), and yet none of the number crunching seems to be directed at proper long-term goals like staff, product and ultimately consumer safety.

I’ve written about this before in terms of American love of American football, and of course it’s not the first time I’ve read an account like this of a company that cares little or not at all about some of the most important metrics for success.

I keep admiring the car companies pushing “retro” versions of their vehicles, although they all seem far less friendly to the right-to-repair movement than the original versions. Does the latest “Fix Or Repair Daily” (FORD) user interface look like it is “mod” friendly to you?

Who is going to wrench apart all those sealed plastic bits in a 2022 Ford Bronco, versus how many carbon fiber, fiberglass and welded aftermarket upgrades and modifications would you naturally find on a 1960s or even 1970s model?

…it’s hard to imagine a situation where the new Bronco would need to be modified…

Oh pleeeeaazzz. Modify it with a classic American electric engine to start (pun intended). If it was good enough for Ford in 1914, it should be good enough for her husband’s legacy in 2022.

Dell has seemingly taken the opposite approach to car manufacturers by giving people the “retro” option of modifying and repairing their newest laptop model.

Or maybe it’s the same approach, such that we’ll see a big aftermarket hot-rod movement for both the newest Dell and Ford?

In any case (pun intended) the new Concept Luna laptop seems very 1990s thinking to me, like back when Dell engineering was cool.

Last week, I was able to check out Concept Luna. Announced today, it’s a laptop aimed at helping Dell achieve its environmental goals by offering parts that are easier to repair, reuse, and recycle. The machine is built to open up easily, so you can harvest its motherboard or other parts to use in future systems. Through this process, Dell hopes that a motherboard will be able to be used in up to three machines.

It actually looks like a laptop I bought from Apple in 1989, just a few years before I became an official Apple repair technician. Yes, I used to have official Apple corporate CDs delivered by the post office, describing in detailed step-by-step videos how to take everything apart. And I used to tear machines down and upgrade or modify them for people, at one point even managing a team of engineers who would de-solder components on boards all day… up hill, both ways.

Kind of weird also to think how Apple allowing people to work on their new machines is considered something new.

For years, Apple has monopolized product repair by withholding the parts and tools that customers and independent repair shops need to fix broken products. That’s finally changing. On Nov. 17, Apple announced that it will begin sharing with the public more than 200 parts and tools for its products, starting with the iPhone 12 and 13 lineups and its new Macs. This program will presumably grow to include newer smartphone models and Mac computers.

That’s a huge breakthrough for the growing Right to Repair movement, which has been working to pass legislation to make sure that all Americans have the right to repair not just their phones but anything they buy and own.

A breakthrough in the sense that it returns us to where we were. Maybe next Apple will announce that we can even choose our own desktop backgrounds and fonts.