3 ms·
This article seems to be using abstract academic hand waving to say "it's impossible to simplify anything". But it seems so obviously false. A "hello world" pr
by QuadrupleA 6y ago
This article seems to be using abstract academic hand waving to say "it's impossible to simplify anything".
But it seems so obviously false. A "hello world" program could fetch every character from a separate database; same output but enormously more complex process.
A small car could have 16 wheels, 2 engines, and be controlled by typing commands on a keypad ("steer left 3 degrees").
Don't tell me these examples are just exposing some complexity that was already there, or that complexity is a conserved constant, etc.
It's amazing what can be made to sound plausible with an erudite wiring style.
- asdfgeoff 6y agoAgree that treating complexity as a constant is a bit extreme. The author lost me at the bit about Tesler's law. That said, it feels like there is likely some sort of Pareto efficient frontier with regards to complexity vs. usefulness. You can certainly have a poorly-designed system which is arbitrarily more complex without being anymore useful. But if you make things "as simple as possible" for any given amount of complexity, there will always be some underlying trade-off. Brings to mind → https://www.joelonsoftware.com/2002/11/11/the-law-of-leaky-abstractions/ https://www.joelonsoftware.com/2002/11/11/the-law-of-leaky-a...