Also don’t forget - we still have to write certain routines in Assembly even after so many years. Number of abstraction layers doesn’t mean complete automation.
Re:Will humans write code in 2040?
by zitterbewegung on Wednesday December 06, 2017 @10:09PM
(#15867319)
Going from high level to low level code will not go away. But, for the most part you would probably try to get something working and then optimize it.
http://wiki.c2.com/?PrematureOptimization
Re:Will humans write code in 2040?
by greenhouse_gas on Wednesday December 06, 2017 @10:30PM
(#15867413)
>I agree, the level of abstractions will just change. We might end up writing something more like formal logic in a declarative language and the AI will then implement and optimise it.
We've tried. The problem with layers of abstraction is that they're leaky, and don't always work.
For example, simple SQL is easy. Optimizing SQL requires intimate knowledge of your db engine.
Yeah, I imagine the formal logic being quite strict, maybe requiring formal verification. It wont stop business-leaky-cases, but it might give us something deterministic that is better than what we have - we dont need perfection, an order of magnitude (or 5) jump in productivity would do...
Re:Will humans write code in 2040?
by seanmcdirmid on Wednesday December 06, 2017 @09:52PM
(#15867239)
Doubtful, debugging would be so hard at that point. I think we’ve already reached peak abstraction where we are right now, and more indirect forms of programming will be increasingly non-viable.
Kind of the way a compiler generates optimised machine code from our higher level languages now.