News for hackers. Stuff that matters.
Re:Imaginary Problems Are the Root of Bad Software by hinkley on Sunday June 18, 2023 @02:43PM (#36383060)
There’s a couple of woodworking hand tool companies who among other things make replicas of old school Stanley tools, the way Stanley used to make them (materials and tolerances). They also fuse the best elements of several eras or manufacturers to make slightly better versions. Surfaces from this one, handles from that one, adjustment mechanism from a third.

I hope that I live to see a time when software applies modern algorithms to classic designs and produce “hand tools” in software.

Re:Imaginary Problems Are the Root of Bad Software by Aerbil313 on Sunday June 18, 2023 @04:03PM (#36383879)
The field of software is maturing as we’re reaching the end of Moore’s Law and time passes. Times of constant innovation is very slowly coming to an end, the curve is slowly flattening. You can already see it with general trends like type-safety, DX features universal in all languages (linting etc.), browsers finally becoming the universal OS (Wasm, WebUSB, GPU), more and more things being standardized every day.
Re:Imaginary Problems Are the Root of Bad Software by hinkley on Sunday June 18, 2023 @07:39PM (#36385668)
Proebsting's Law says compilers double code efficiency every 18 years. I wonder what the doubling interval is for algorithmic performance. I expect it would be tough to calculate, like the cost of living, because algorithmic improvements rarely anymore affect all aspects of code performance. Incremental improvements in sorting efficiency likely have one of the broadest reaches, followed by concurrency improvements and object lifetime analysis. Then there's a long tail of niche applications that only apply to certain domains.