Derivative vs. original ideas
The other day, I was listening to old music on the radio, and I realized that I could guess the decade a particular song was from without having heard that specific song before. This made me realize how similar music from a certain period must be to be recognizable as a group in this way.
I think the insight here is that most "original" songs are actually derivative, based - unconsciously or otherwise - on other music that came before it. The same could be said of most other cultural phenomena such as fashions, colloquialisms, and even memes. Nearly all of these are destined to be washed away by time, or at best relegated to some historic anthology as being particularly representative of its era.
Of course, some timeless things that never go out of fashion come to mind. Plain t-shirts. Fairy tales. Bicycles. Wine. Rice.
The common denominator of the timeless is their simplicity. This is because the simpler a solution is, the more difficult it is to improve on it.
What about original ideas? From what I've seen, few original ideas stick around because most of them are not improvements over what already exists. The constant stream of startup obituaries is evidence of this.
And yet, the world needs original ideas - and the visionary designers, inventors, scientists and entrepreneurs to develop them - because without them, mankind experiences no progress.
“The reasonable man adapts himself to the world: the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man.” - George Bernard Shaw
Substantial benefits accrue to those who develop original ideas that take hold, or to those who get involved with them at an early stage. The challenge is to recognize the ideas that will take hold, because only these are valuable. Not everyone is good at this. If we're self-aware enough to realize we're in this group, then a good strategy might be to follow those who are the likeliest to have or support valuable new ideas. I can think of two groups of such people.
Experienced early adopters
The first version of a new idea is almost always worse than existing solutions, so it takes a special kind of person to recognize its potential. The first iPhone was a worse productivity tool than the Blackberry (for those who remember it). Airbnb in the beginning was such a weird idea that no one funded it (Paul Graham Airbnb because he liked the founders).
Of course, early adopters are not always right - just ask the 1,000 people who purchased a Better Place electric car.
However, they just need to be right enough of the time to come out net ahead. This is easier to do with experience, which they gain by sheer volume of early adoptions and by being wrong a lot in the beginning. Experienced early adopters generally aren't too difficult to locate - significant subcategories of them are angel investors and early-stage venture capitalists.
Domain experts with "crazy" ideas
As Paul Graham noted, domain experts are normally rational people - so they should know when an idea would sound "crazy" to others. When they propose such an idea anyway, the rest of the world should pay attention (though most people don't).
The reason few heed such experts is because the grip of the current paradigm is so strong. We're mostly followers of the prevailing fashion and unable to see or appreciate things that are truly revolutionary. Thankfully, the process of testing most innovations has accelerated to the point where the truth is visible to all within a matter of years rather than decades, as it mostly had been in the past.
These days, it's still lonely to be in the extreme minority with correct beliefs, but often quite lucrative as well. If you're not a domain expert with an idea yourself, lend your support to someone you know who is.