I know this is hacker news, and this'll get down-voted to oblivion, but alas its worth saying, so I'll say it.
The number of possible customers , who care whether your product is ooen-sourced or not, iz a rounding error away from zero. [1]
The pie chart ideas follows:
A) the big slice goes to companies who've never heard of open source.
B) the next slice are ones who've heard of it, but they lack the tech to make it a useful feature.
C) then there are those who think it's a great idea, but ultimately want a good program now, not some source code in the future.
To illustrate the point, drive down your high street, or ho yo your mall, or visit your local industrial park. Go past each business in turn. Wonder if he cares about open source. Wonder if he runs Word/Excel or Open Office.
In my career I've had a small number of customers who cared about the source code. I simply burned it onto a CD and gave it to them. Quite frankly it's useless to them. He the time they've hired a developer, and he's spent time just getting the code so it'll compile, he's spent 10x more than the software cost.
I agree selling is hard when you don't have a "history". For every product we ever made, the first 10 sites are the hardest to sell. You're asking people to take a risk. You need to convince them that you are worth the risk. You need to reduce that risk as much as possible, but IMO providing source code does nothing. [2]
[1] this can be product dependant. If you are selling to Linux OpsDev then sure. But assuming some not-sold-to-techies thing, it's reasonable safe to assume that your customer base has never heard of open source.
[2] in cases where we've built custom software for folk which is how we started, we've negotiated ownership of the code. Typically we own it, but we give it to them when we no longer want to be involved in the project. This allows dome other programmer to maintain it going forward.