Hacker News Reader

Comment by eclipxe | Hacker News Reader

You're likely breaking Zillow TOS, so expect a C&D before you monetize

I think scraping websites that don't require authentication is legal in the US nowadays.

Monetizing data acquired through scraping could run afoul of copyrights, so the love letters would probably only arrive once OP attempts monetization.

>I think scraping websites that don't require authentication is legal in the US nowadays.

The LinkedIn lawsuit resolved to that conclusion, though I think LI is appealing again.

It basically says the site can't take legal action against scrapers of public (non-auth protected) info. Still, there's nothing that says they have to make it easy for you, to include deploying anti-scraping measures, rate limits, redesigns and moving data behind auth.

I thought that USA doesn't allow copyrighting facts which to me numeric values certainly are.

OP Here! Yes I don't know the law well but common sense tells me that its one thing to copyright a brand, logo, movie, or anything creative, and it's another to copyright numbers on a database.

The only argument Zillow can against me (imo) is that I'm being a burden on their servers at scale.

I would be very happy if my project gains attention from a product manager from Zillow and they say “Why don’t we just let him pay for the data”. I’m unable to get their attention and they’re very cryptic about their GraphQL API spec.

Hell yeah brother, greyhat on.

Alas, you'll likely hear from a lawyer before a PM, but good luck nonetheless.

It's usually the PM who prompts the lawyer, "hey that person is stealing my limelight, please fix it".

Yeah. I feel like Zillow will pretend like all this data somehow belongs to them when it belongs to the public.

IANAL but just because it is public data doesn't mean that Zillows aggregation and capture of the data can't be subject to copyright.

Good point. IANAL too, But I can’t imagine copyrights on work that didn’t even belong to Zillow. It’s not creative content like a movie. They are just collecting data off existing sources.

I could also argue that I’m redistributing the computed data done by my web app like the average, median, and the graph. Then I’d have to hide the Zillow results source table.

Facts by themselves are not copyrightable, but certain arrangements of facts may be.

https://sco.library.emory.edu/research-data-management/publi...

If this is the case, why not go to the source of the data? (I think it’s because the source of the data is not guaranteed to be free but hitting Zillow’s public infra is)

The problem is the source isn’t one place. The source is a collection of 500 MLSes across North America. They have a vested interest in keeping the data accessible only for licensed realtors. Striking a conversation with 500 MLSes would be a big undertaking, I would need contacts and connections in the real estate world.

Hmm the source Zillow and other realtor company’s get their data from are local governments. They all offer tax, ownership info, last sale price etc. they all also charge differing amounts from free (if you walk into their records office) to cost per dataset to as costly as cost per data request. Then there’s the sheer number of locales you must query for this info, typically on a per county basis. So you head to a data broker who has collected all of this information but they charge for their time and efforts which is $&$$. So that’s why I mentioned Zillow’s public servers are cheaper

Contacts and connections are easy. Money is harder :)

Might be worth noting: There is a DOJ antitrust lawsuit against NAR over MLS practices, though I don't know if your use-case is part of that.

https://www.justice.gov/opa/pr/justice-department-files-anti...

Reply on news.ycombinator.com