article

Ten years after its untimely death, the team that built the much-beloved feed reader reflects on what went wrong and what could have been.

Rendering text, how hard could it be? As it turns out, incredibly hard! To my knowledge, literally no system renders text “perfectly”. It’s all best-effort, although some efforts are more important than others.

Obliterate toil: automate it.

Automate ruthlessly. This is where I have seen the most surprising pushback. We’re programmers. Automating processes is what we do! People will flinch about this, afraid of time spent automating things that won’t pay off. Yes, we’ve all been there. So don’t do that. Don’t automate things that are really one-offs. If there’s any chance you have to do the same thing more than five times, automate it. If it’s complex and difficult for a human to do, automate it. If the blast radius of the explosion caused by a human doing it wrong is large, automate it. If the end results need to be the same every time, automate it.

Infrastructure should be automated as far as you can push it.

The upside of automation is that the software that does the work for you can be instrumented.

The discovery raises intriguing new questions about the role wildfires may play in spurring the growth of microscopic marine algae known as phytoplankton, which absorb large quantities of climate-warming carbon dioxide from Earth’s atmosphere through photosynthesis and are the foundation of the oceanic food web.

“Our results provide strong evidence that pyrogenic iron from wildfires can fertilize the oceans, potentially leading to a significant increase in carbon uptake by phytoplankton,” said Nicolas Cassar, professor of biogeochemistry at Duke’s Nicholas School of the Environment.

The algal blooms triggered by the Australian wildfires were so intense and extensive that the subsequent increase in photosynthesis may have temporarily offset a substantial fraction of the fires’ CO2 emissions, he said. But it’s still unclear how much of the carbon absorbed by that event, or by algal blooms triggered by other wildfires, remains safely stored away in the ocean and how much is released back into the atmosphere. Determining that is the next challenge, Cassar said.

It’s been over a decade since we’ve heard from the elusive White Stripes drummer. Could renewed attention over a Rock & Roll Hall of Fame nomination coax her back into the spotlight?

The issue with hCaptcha’s strange AI generated prompts highlights two issues with machine learning systems. The first is that the AI systems require an enormous amount of human input to not be terrible. Typically image labeling is outsourced to foreign workers who do it for pennies on the dollar. The other is the issue of data drift. The longer these machine learning systems run, the more input they require. Inevitably, they begin to use data they’ve generated to train themselves. Systems that train on themselves long enough become AI Hapsburgs, churning out requests to identify incomprehensible objects like “Yokos.”

Notes let us forget and remember, simultaneously. No more loss aversion; we can have our ideas and forget them, too. We can cut and trim and still keep our darlings.

We need to feel safe that our memories were not in vain, that they’ll be there if we want them again. Only then can we let go.

Then the cracks appear. You read something new, think new thoughts. Then you go to save it and feel a tinge of déjà vu, think you’ve seen this thing before, yet you couldn’t find the memory. And, come to think of it, you never did use all those murdered darlings, either. Your faith in the second brain falters.

Today, the world belongs to TCP/IP. Those two protocols (together with UDP) govern most of the remote communication that happens between computers. But I think it’s wonderful that you can still find, hidden in the plumbing of the internet, traces of this other, long-extinct, evocatively named system. What was Chaosnet? And why did it go the way of the dinosaurs?

These days, we provide thermal comfort in winter by heating the entire volume of air in a room or building. In earlier times, our forebear’s concept of heating was more localized: heating people, not places.

They used radiant heat sources that warmed only certain parts of a room, creating micro-climates of comfort. These people countered the large temperature differences with insulating furniture, such as hooded chairs and folding screens, and they made use of additional, personal heating sources that warmed specific body parts.

I quickly learned that writing one giant query with a bunch of joins or even bunch of Python helper functions could get me stuck. My transformation functions weren’t flexible enough, or my joins were too complicated to answer the endless variety of questions thrown my way while keeping the numbers correct.

Instead, the easiest way to be fast, nimble, and answer all the unexpected questions was to prepare a giant table or dataframe and limit myself to it. As long as I understood the table’s contents, it was harder to make mistakes. I could group by and aggregate on the fly with confidence.

The main reason we don’t create more affordances, however, is pure egocentrism. When we just say whatever pops into our heads, we may think we’re making craggy, climbable conversational rock walls, when in fact we’re creating completely frictionless surfaces. For example, I’m thrilled to tell you about the 126 escape rooms I’ve done, but my love for paying people $35 to lock me in a room blinds me to the fact that you probably do not give a hoot. I may even think I’m being generous by asking about your experiences with escape rooms, when my supposed giving is really just selfishness with a question mark at the end (“Enough of me talking about stuff I like. Time for you to talk about stuff I like!”).

Calorie intake, protein intake, and training volume are by far the most important factors determining your body composition and degree of swole-ness. Add in training intensity and specificity, and you’ve got the major factors determining strength as well.

Everything else is just details. And it’s not that details don’t matter; it’s just that unless you’re an elite-level athlete trying to eek out an extra percentage of performance, they don’t matter very much. In a complex, redundant system, details generally get lost in the noise, and end up having at most a trivial effect.

I wanted to explore the malicious, confusing, and deceitful things that occur after signing up for digital services, as well as how design can nudge us to forget about a free trial or accidentally sign up for things that we didn’t intend to.

Dark patterns are often most egregious with subscriptions and free trials, especially when attempting to cancel, so I focused on those.

Though it makes me feel like a grandmother on her deathbed to admit it, I remember the days when the internet was vast, when there seemed to be more places to go than anyone could ever visit and infinite things to read. What you saw was not determined by some highly protected coded algorithm that lives somewhere in the cloud. You could just go out and find it.

The tech industry is abuzz about a new standard for social networking that is more open, more user-centric, and potentially more powerful than Twitter and Facebook. But we’ve been here before.