None of this would make the site lose its primary function, which is by their own admission to do a background check on the landlord you’re about to sign an agreement with.
[0]: https://www.google.com/search?q=Aberdeen+St%2C+Chicago%2C+IL...
It only takes one. And if it’s been automated, you’ve just made it plain text but with extra steps.
Bcrypt has key stretching, brute-forcing every address in existence would cost a lot of CPU even for one city. How will the attacker get compensated for that?
I honestly don’t get why I even have to explain this. The original question was how anonymous it can get. Any practice that reduces the amount of personal information, or the ease to access it, helps, period. Dismissing one because it doesn’t offer perfect protection is like not using condoms because they are ineffective against mono. There is no reason not to implement them - that is, if the maintainer actually cares about privacy.
What it does solve is the problem of this data being visible to every landlord who types the address of their property into a vastly more popular search engine known as Google. They need not even know of this site, it gets served to them on a silver platter.
And if you think a robots.txt alone solves that, you’re mistaken. A robots.txt is just a recommendation. Web crawlers are not obligated to honor it. The only way to solve it is to make sure the private data isn’t even there in the first place.
Regardless, if the info did get leaked or the landlord did some basic investigation, or the property was just small enough that the data pool was small, what’s to keep the tenant from litigation by the landlord?
I’m thinking about all those review-sites for hospitality where a bad review starts a whole lot of grief for the reviewer.