"Everything that can be invented, has been invented"? ;-)

I agree with your point about SaaS, that people are tied to you. It stops piracy; and repeat sales are the norm. Theoretically, you could achieve the same by renting software - but it doesn't seem to work in practice (and doesn't stop piracy).

But one solution is to make a product into SaaS in an odd sense: that you keep releasing new versions - so the "service" you are selling is your on-going development. The ideal need for your software to meet therefore becomes one that can never be fully met, so that each improvement you make is needed. This differs from my perfectionistic urge to pick a problem that I can solve perfectly, and instead pick one that can't be solved.

In my software business, I experience the odd effect that when a new problem needs to be solved, people come out of the woodwork to buy it - as if they'd been using it without buying it. Before that they didn't have any problem, and so hadn't needed to contact me. Or is there a more probable and less cynical explanation?

This solution is not as good as SaaS, because you have to keep working. :-( But in SaaS, you also have to keep working, or the competition catches up. And that's a much more serious problem if the competition is free...

The last company I worked for before starting a startup did pro-audio stuff. Their stuff, generally, was cracked the same day it was released and estimates were that at least 80% of users weren't paying for the software. If you get popular enough, it tends to come back to haunt you.

I think one solution, as mentioned a couple times in the thread is to work on software for rich niches.

* The ideal need for your software to meet therefore becomes one that can never be fully met * That is brilliant