I understand Google screwed up in this area before, when they captured network traffic. (Even then, it seems they captured that data due to oversight, not some nefarious scheme.)
There are lots of huge privacy issues in the world; I don't think a database of approximate wifi locations ranks particularly highly. I don't care if Google map my hotspot, and I'm happy to have my phone locate quickly, when I'm out and about.
And they are now providing a way to opt-out, for the people that feel strongly about it; its not trivial to implement, but I can't think of a way of making it easier.
It seems a lot less complicated for an end-user than robots.txt is in the related scenario of web scraping.
That blog post is full of vitriol; I don't think its adding much constructive discussion, or insight; I think its just blowing things out of proportion.