News for hackers. Stuff that matters.
Re:On meta-design and algorithmic design systems by TheOtherHobbes on Monday February 23, 2015 @10:50AM (#9094909)
Exploring a meta-rule space isn't hard.

Exploring a meta-rule space and finding rule sets that create results that humans find interesting is much harder.

If you skip that you tend to create systems that seem too random, or too repetitive and obvious.

This is why most algorithmic art is so uninteresting. It falls on one side or the other of the randomness/repetition valley, without hitting that sweet spot that the best human creativity finds.

That's not even thinking about emotional associations, intellectual abstractions, social awareness, and models-of-mind. You might need all of them to make a genius-level creative AI, but I'd guess that the more you have, the more likely you are to get something worth keeping.

Re:On meta-design and algorithmic design systems by TuringTest on Monday February 23, 2015 @11:58AM (#9095323)
That's why I find "augmented intellect" so much more interesting as a concept than "artificial intelligence".

Truly autonomous behavior is out of reach of our reasoning systems, using most known approaches to AI. But if you put a human in charge of the creative part, and complement it with the automation capabilities that simplify the repetitive parts, the result is a win-win match combining the strengths of both.

Re:On meta-design and algorithmic design systems by lsjroberts on Monday February 23, 2015 @12:18PM (#9095464)
I guess that's the continuation of automating tasks and building up. Push more and more work to the computer / AI. We've been going through writing higher level languages, and the article is describing going a step higher again.

Fascinating stuff!

Re:On meta-design and algorithmic design systems by cheatsheet on Monday February 23, 2015 @12:34PM (#9095590)
> This is why most algorithmic art is so uninteresting. It falls on one side or the other of the randomness/repetition valley, without hitting that sweet spot that the best human creativity finds.

> That's not even thinking about emotional associations, intellectual abstractions, social awareness, and models-of-mind. You might need all of them to make a genius-level creative AI, but I'd guess that the more you have, the more likely you are to get something worth keeping.

I don't agree. I find all of those things boring in art. Mathematics and computer science are much more interesting when you think about how it relates to the mind, and why people may choose to create abstract art instead of metaphorically painting themselves naked.

Re:On meta-design and algorithmic design systems by themodelplumber on Monday February 23, 2015 @12:37PM (#9095613)
This is so true. I make fractal art and desktop wallpapers as a hobby and it can actually be pretty exhausting (in a good way). Finding results for other humans is only part of the problem--finding results that keep my own continually evolving aesthetic happy is very difficult.