How is the fact that there is old code in modern Linux distributions suddenly becoming empirical evidence on the viability of a business model? A business model does not become viable because you as the customer are somehow able to work on the project after it is no longer commercially available. By the way: there are a lot of commercial projects having escrow agreements to solve that problem without having to be open source.
I simply don't buy that we all should be living off donations to support our software development. It's utterly clear that only the biggest projects can pull that off, and even there — only to a certain degree, with some notable exceptions (RedHat, Linux).
> I simply don't buy that we all should be living off donations to support our software development.
Most "software development" is internal software that exists simply to solve some organization's bespoke issues, and is not open to the public - FLOSS vs. proprietary is a non-issue there. As far as off-the-shelf, non-internal stuff goes, software maintenance is actually a bigger issue than software development per se, and it's entirely appropriate to say "I will not be putting any effort into maintaining this unless you pay me some real $$$ for the trouble." So, I'm not even sure that we're disagreeing about anything of relevance - except inasmuch as "being able to work on the project - or to sponsor work on the project by third-parties - no matter what" is often very important!
I simply don't buy that we all should be living off donations to support our software development. It's utterly clear that only the biggest projects can pull that off, and even there — only to a certain degree, with some notable exceptions (RedHat, Linux).