Week 21, 2024 - Frustration and Happiness

An important milestone in our young podcast series with Jeremy this week: we had our first guest on the show! We talked with the CPTO of Lunii about their introduction of a 4-day workweek and the use of AI in educational toys. Watch it here if you missed it!

In other news, I spent more than half a day trying to set up an e-ink display with a microcontroller. This was not the usual joyful tinkering, with small setbacks and motivating excitements when overcoming them. No, this was pure, relentless, demotivating frustration. I couldn’t get it to display a simple Hello World - I messed up wiring the devices or missed something in the ESPHome configuration. I know when to count losses and move on, everything's all packed up in a box for a more ambitious day.

📋 What I learned this week

📟 Failures are learning opportunities, but it’s hard to find the lessons without much feedback, and an empty screen is anything but. Pointers I left for my more energized future self, for a time when I’ll feel ready to play with the e-ink display again:

• The 7.5” e-paper has different versions, some require config changes for better display quality and longevity;
• This controller includes a display driver board unlike the one I bought, giving hobbyists less chance of mixing up the wiring.
• The Rev2.3 version of my Driver HAT has an extra power pin compared to previous versions, which will probably need 3.3 volts.

Good luck, future Péter, I believe in you.

📝 Our collaboration with Péter Batiz, my recruiter consultant friend resulted in a huge advice article about Perfecting your CV and LinkedIn profile. While our series is aimed at Engineering Managers, the post is full of useful tips for anyone looking for a job in tech now. This time I got inspired by The 20% Rule and shared my draft earlier with him. This gave him a better chance to provide input, which had a big impact on the end result.

🎯 What I want to try next week

Most of the week will be occupied by the events in and around Craft Conf: besides the talks during the conference, I’m planning to attend a meetup with Gergely Orosz, where he’ll talk about how and why the software development career is changing now. I’ll write about the conference for HWSW, one of the most prestigious Hungarian tech publications — and I’m sure I’ll have something to share here too.

🤔 Articles that made me think

Soft Delete is an Anti-pattern

I love reading about technical details like this. The post is not as opinionated as its title, it lists legitimate reasons when just pretending to delete data is an acceptable choice. But the point is that often soft delete is a sign of a lack of commitment or trust, and in both of these cases, it’s better to look at the root causes and use one of the alternatives instead. The way I see it, soft delete has the danger of building tech debt, by postponing today’s questions for tomorrow to solve. Also, I’m not sure how GDPR-compliant it is to only set a “deleted” flag on data when a user is explicitly requesting deletion.

Further Struggles with Measuring Productivity

This is a candid view into the thinking of a VP of Engineering figuring out how to track productivity. The solution he came up with is measuring where engineering time is spent and equalling productivity with value-adding work. (Defined as “The percentage of engineering time spent on adding value for users and customers.”)

I’m a bit skeptical about this metric, not because it’s easy to game (as the author is pointing out), but because it says nothing about how the time was spent on value-adding work. To show my point with an absurd example: if the entire engineering organization spends a month developing a single red button on the homepage and nothing else, then they achieve 100% productivity if that button adds value. Still, I like how this metric focuses on user value, I just wish it wouldn’t use relative percentages but absolute values too, about the number of items shipped.

Also, it’s important to recognize that 100% productivity (in this definition, meaning time spent adding value for users and customers) should not be the goal. Not balancing engineering investments will inevitably lead to a lot of unplanned work, interruptions, and having to manage the complexity of the resulting system, as John Cutler pointed out a few days ago.

All these aside, it’s always a fun opportunity to peek into the mind of an engineering leader, so I’m grateful for James to share his experience.

🌍 Something cool: What Makes Us Happy?

I saw this post on Reddit the other day, listing the differences in time spent eating and drinking in a day between OECD Countries:

This is super interesting, but I wondered: are people who spend more time eating happier? I would assume, but what does the data say?

The amazing Our World in Data site has data on self-reported happiness around the world:

I downloaded both data sets and massaged them in Google Sheets to calculate the correlation between Happiness and various daily activities.

Here are the results – big positive numbers signal a strong positive correlation, and similarly, negative ones signal a strong negative correlation.

A few things jumped out to me:

• Traveling to work or study is the activity that has the biggest negative correlation with self-reported happiness. In layman’s terms: we hate commuting, twice as much as working. No wonder people vehemently oppose return-to-office initiatives.
• The activities that are the best indicator of happiness in a country are shopping, volunteering, and sports.
• Spending time with family and friends, even if caring for household members, has a moderate positive correlation with happiness.
• Somewhat surprisingly there’s only minimal correlation between self-reported happiness and time spent doing activities like job search, routine housework, sleeping, religious activities, and, to answer my original question, eating and drinking.

That’s it for today, have a happy weekend, for example, by figuring out how to volunteer for a local sports club,

Péter