So I have a really high rubric for my projects: “does the world need new software?” one.

The more this “ #AI” craze drags, the more convinced I am that the world actually doesn’t need more.

True, a person with spare $1K can buy a #Claude subscription and slop up a “product” scratching their own itch. The necessity of scratching either itch aside, I doubt that this product will be innovative and useful, with all its single-purpose undocumented APIs and flat faceless #Tailwind design.

There are projects solving all kinds of issues already. The problem is not creating them. The problem is learning and extending them. That’s often a hard, but nonetheless important part. Because if we invest in good #FOSS projects that are flexible and useful enough, we’d get better foundations to solve problems on. Without the need for slopifying a new project every time there’s some difficulty we encounter.

The world doesn’t need more software. It needs better software.