After reading this article, I suddenly remembered an elective I took in college called “Software Archaeology.” The professor asked us to reimplement compiler exercises from the 1970s. At the time it felt useless, but later I realized that course taught me more about system design than any modern framework ever did.
Software archaeology in its most literal form would be a fantastic course addition for anyone going into a company with a medium-large codebase > 5 years old. Especially if you end up at a FAANG or something akin to it.
Being able to navigate not just a codebase but bugs/tickets attached to it, discussions in documents, old wiki pages that half work, extracting context clues from versioning history, tracing people by the team they worked on at the time...digital detective work is a serious part of the job sometimes.
If this is indeed man made space debris, who is responsible under current international rules? If it were to collide with a civilian aircraft and cause an air crash in the future, would there be an embarrassing situation where the culprit could not be found?
Sometimes when I watch my border collie chase a ball, it really makes me reflect on humans too. We’re constantly scrolling through our phones, watching short videos, playing mobile games. In a way, it feels like we’re just throwing a ball for ourselves to chase again and again.
FOSS used to win by being able to run on anything. Now hardware chooses you. If you’re not running the sanctioned OS, even the browser might be crippled. I’m not sure if that’s progress, but it’s definitely not freedom.
I've always been interested in running LLM locally to automate browser tasks, but every time I've tried, I've found the browser API to be too complex. In contrast, writing scripts directly with Playwright or Puppeteer tends to be much more stable.
What impressed me most about Claude Sonnet 4.5 is that its output structure is more stable than many other models and less prone to crashes. I ran some real world scripts from my own projects, and it exhibited fewer hallucinations than GPT-4 and performed more faithfully on code interpretation tasks. However, it can be a bit slow to warm up, and sometimes I needed more prompts in the first few rounds.
My daughter has recently become obsessed with Go and now beats me half the time. I think that's good because Go helps her slow down and think before she acts.
Medical history has seen many "miracles." Hopefully, this time, it will become something more people can replicate and learn from, rather than just a flashy headline.
I think future app stores should offer users an ad control panel. Let us choose whether to see ads, what kinds we prefer like tools, games, or education, and even which recommendation algorithm we want to use.
