Anthony, thank you again. These are well articulated, well thought out issues and for whatever it's worth, I sincerely appreciate you taking the time to write them. I think we're on different sides of the river, but I can at least see why you prefer your side to mine. I won't convince you mine is better, but here are my thoughts on it.
To respond to your limitations issue, I hear what you're saying - most of the claim is scaffolding and there's really only one interesting nugget of invention. This is often the case because there needs to be context and antecedent basis. It sounds like what you'd prefer is if everyone wrote a Jepson claim. A Jepson claim is where you say:
A system|method for blah blah blah ... wherein the improvement comprises:X.
In those claims you are acknowledging all the prior art and laying out the single improvement, which is X. Then everyone automatically knows what's novel and what's not. The problem with those from a patentee standpoint is: everyone automatically knows what's novel and what's not. What if I thought steps A and B of my process were obvious when in fact they weren't and I said the improvement was just step C. Well I just gave up all rights to A and B. And to turn your own issue on its head, there is no way for me know - no seer I can consult - to know if what I am giving up is actually in the prior art. Let's consider the yard and fence analogy. If a surveyor didn't exist to say where exactly your property line was, when you set up a fence around your house, do you set it up as far out as you think you're reasonably entitled, or do you put it as close to your house as you can so you don't offend anyone? It's the same thing. Why would I intentionally give up claim scope? I can understand why that's a greedy approach, but I hope you can see - it is to me at least - that it's a reasonable one.
Re: the uncertainty, you can get no more certainty that you don't infringe than you can that someone hasn't copied unlicensed code into your codebase. You can take steps to address the latter, e.g., code monitoring and review, but you never really know unless you have a development team of one - yourself. That's a risk of doing business. You can mitigate the risk - for code hire a code scrubber (which admittedly has a finite data set to compare against), and for patents get a freedom to operate opinion - but you'll never be free of the uncertainty. I don't know what to tell you.
As a final question: where are all the stories of big companies completely crushing the little guys in legal fees? Everyone throws it around as a problem, but it seems that more often than not the big companies are fighting each other or the trolls are going after the big guys. I know David and Goliath battles are academically possible, but where are they in reality? I'm not even looking for a case cite; I'm looking for even just a news article that says Company X was a big meanie and used its patents to crush li'l ol' Y. It's bandied about as a scary possibility; I just want to see a real-world example of it.