Once again, a variety of things I’ve read and enjoyed - both non-tech and tech. Hiring, AI categories, decision-making, tech debt, Chrome-in-a-console, Playboy centerfolds, and more.

Assorted

I was looking through some old notes on hiring processes the other week and came across the Google re:Work piece on “Use Structured Interviewing” - in short, stop relying on Vibes, drop the “brainteasers”, and instead use a structured approach with standardised questions and rubrics - and they’ve got the data to back it up, guidance for getting started, and nuance to consider.

Drew Breunig’s “The 3 AI Use Cases: Gods, Interns, and Cogs” feels like a really neat categorisation to aid in the many, many discussions about AI that we all seem to be having nowadays…!

Don’t be misled by the title - David R. MacIver’s “How to make easy decisions” is an excellent piece about decision making, how so-called “easy” decisions are more common than you might think, and how valuable it is to get better at them.

Mandy Brown’s “Accountability sinks” discusses the eponymous(?) concept identified in Dan Davies’s “The Unaccountability Machine” - organisations forming “structures that absorb or obscure the consequences of a decision such that no one can be held directly accountable for it”. Per the article, now I’m aware of the concept, I see it everywhere - and alas I’ve yet to see any good solutions proposed. So I guess this is slightly cursed knowledge, and I’ve just inflicted it upon you - sorry!

“Destigmatize being dumb” is a great piece on how, sure, we should probably admire and reward intelligence more, but we should also make it more ok to be wrong or “look stupid” - that doesn’t mean celebrating it, but we want people to try, to fail, to admit they haven’t got it right, to ask, to learn, and more.

Finally, I got a bit grumpy about people throwing around the term “the root cause”…!

Tech

Jeremy Morrell’s “OpenTelemetry Tracing in 200 lines of code” is a really neat first-principles exploration of tracing. As is so often the case, it’s one of those things where the underlying core is relatively simple - the complexity arises from the supporting infrastructure - and diving into the fundamentals and stripping the rest away is a great way to gain a better understanding.

Should you set an upper-bound version limit on your library’s dependencies? In short: almost certainly not. In long: Henry Schreiner’s “Should You Use Upper Bound Version Constraints?” dives into the nuance and considerations and tradeoffs, providing comprehensive persuasive reasons why no, you almost certainly should not.

The phrase “Tech Debt” bugs me in an “If By Whiskey” sense - too often I see it used incredibly fuzzily to mean whatever the user happens to intend it to mean. But apenwarr’s “Tech debt metaphor maximalism” leans into the “debt” metaphor and covers a number of ways in which it’s particularly apt - certainly a level of nuance and consideration that I’d like to see in future discussions of the topic.

I was looking for a console-based web browser a little more featureful than Elinks/Links/Lynx, and stumbled upon Fathy Boundjadj’s Carbonyl - “Forking Chrome to render in a terminal” is a great explainer of how it all works under the hood.

Finally, if you’re looking for a test image, you might consider reaching for the classic Lena image - maybe don’t. Try Morten Rieger Hannemose’s “Recreated Lena Picture” instead, featuring a model who actually wants the image to be used.

So

Ending these always feels odd - it’s a list of links - I hope they were interesting to you! Feedback always appreciated, as well as shares to other folks you think might enjoy:

Share

Cheers, Kristian

Thanks for reading Burn After Reading! Subscribe for free to receive new posts and support my work.