I mean, consider Apple's bouncing lists patent. When you use it, it's very intuitive, and any moderately skilled software developer could implement it in an afternoon. So it seems obvious, but the question is then - why did nobody do it before? Touchscreen devices existed previously, so that seems to suggest that it is non-obvious. Synthesizing a drug is not hard or novel, but arriving at the conclusion that this is indeed the correct drug that is the reason that pharma patents exist. So the same could be said of bouncy lists.
So how do you determine the obviousness of this, objectively? The most straightforward ways are prior art and to to consider claims from previous patents. It's mostly objective and is hard to argue against, as opposed to one patent examiner's opinion on obviousness.
The problem with software patents is that the benefits of a 20 year monopoly far outweighs the benefit of having it released to the public - which is not true for all patents. If Apple couldn't patent the bouncing list, they'd still have bouncing lists, and it would still be an advantage to their devices, even if everyone else copied them. If you couldn't have a patent on a drug that was going to take $1 billion in costs to bring to market, yet would only cost pennies to synthesize, you wouldn't even bother trying. So the question is instead, how do you objectively determine which industries require patent protection, and which don't. For the rest of the industries, copyright and trademark laws are more than enough protection.