I've been thinking along the same lines as you, but (to play devils advocate and help discussion):
> 1. You should not be allowed to sell a patent. The intent of a patent is to give the inventor(s) enough lead time to develop the idea, not to create a market for ideas.
Why are ideas worth nothing? If I have a breakthrough while researching, I'm screwed unless I also manage to form a company with a production line?
> 2. A patent should not be valid unless you can demonstrate it. You shouldn't be allowed to own ideas that are beyond your means to execute.
Would a prototype be acceptable, or does it have to be viable for production? Also, between coming up with an idea and being able to show it as viable might be a reasonable amount of time
> 3. There should be very low limits on patent infringement damage claims thereby removing the financial incentive of using patents and anti-competitive and extortion weapons.
This one feels the most concerning. If the limits are low, then why would anyone bother following them? You could easily make it cheaper for big companies to copy ideas and run with them, then pay people off when they get caught.
Anything that encourages people who have a new, innovative idea to hide it to try and make money is against the original aim.
I do, in general, like the approach though. Maybe there are points along the way:
1. You have a new, awesome, innovative idea. You can get a patent for it.
2. You have to show some working, or viable prototype within a certain time period (short, a few years maybe)
3. With a real example, you can then get a longer patent.
4. You can sell it, but the original time limit still applies.
This way you're compensated for coming up with great ideas (but you or someone you sell it to has to be able to make it a reality soon) and have to release how things work into the public domain (the original intent of patents).