I'm curious about what happens to the art of programming when Moore's Law goes away.
Pretty clearly, programmers (as a society) have let Moore's Law carry the ball. We've quit doing assembler, we've quit doing C, we've started to do interpreted stuff. If you're an Arch Linux user, you'll have noticed the 10x or 100x size difference between a GUI and a text interface system. So, even in terms of user interface, we've let Moore's Law carry the water.
After speed and multi-threading and parallel programming stop giving rewards, do we go back down the abstraction chain? Do we start programming in C or assembler again to gain additional speed? Do we all give up (at least to some extent) on GUI programs?
Herb Sutter wrote a good article on this a few years ago. It was written in 2005 and centered mostly on multi-threading and parallel programming, but definitely worth a read.
http://www.gotw.ca/publications/concurrency-ddj.htm
Pretty clearly, programmers (as a society) have let Moore's Law carry the ball. We've quit doing assembler, we've quit doing C, we've started to do interpreted stuff. If you're an Arch Linux user, you'll have noticed the 10x or 100x size difference between a GUI and a text interface system. So, even in terms of user interface, we've let Moore's Law carry the water.
After speed and multi-threading and parallel programming stop giving rewards, do we go back down the abstraction chain? Do we start programming in C or assembler again to gain additional speed? Do we all give up (at least to some extent) on GUI programs?