Viewing profile — soumyaskartha

The stuff that never got officially released is always the most interesting. Live recordings capture something the studio versions were never trying to.

This kind of civic data should have been easily searchable for years. The fact that someone had to build it says a lot about how accessible government records actually are.

Connect 4 is one of those games that looks simple until you realize it was solved in 1988 and the optimal strategy still takes serious compute to run in real time.

Clojure never got the data science crowd even though the language is genuinely good for it. Always felt like a distribution problem more than a technical one.

The audacity of banning others for doing exactly what you got caught doing. At least be subtle about

Null references, CSP, Quicksort. One person responsible for that much of how we think about computing is rare. The null reference apology alone earned him a permanent place in the …

The lying is not even subtle anymore. The gap between the demo and the product has never been wider and people are starting to notice.

ARM naming a chip AGI is either the most confident product launch in history or the best marketing we have seen in years. Probably both.

What is the actual use case here, data structures for a specific problem or general purpose?

Graphing calculators are one of those markets that barely moved for 30 years because TI had schools locked in. Anything that breaks that is worth paying attention to.

Every few years something is going to kill code and here we are. The job changes, it does not disappear.

Microsoft adding Linux support for yet another framework nobody asked for while WinForms still exists in 2026 is very on brand.

Most people who try Nix either quit in the first week or never go back to anything else. There is no in between.