Hacker News Reader

Comment by ericd | Hacker News Reader

You're right, I was hoping I could go on, but I wasn't stealing anyone's content, I was posting links pasted geographically located with stat summaries. Look at a listing on PadMapper - there's no listing text, and when you click through, it goes to the original listing.

View on HN

My apologies. I may have used some improper wording there.

But if I may... you must admit that you have taken "something" from Craigslist.

Whatever you want to call it... be it geographical location points, links-to-listings, content, data, images, etc... it is inevitably some form of information that was originally curated by Craigslist through their 10+ years of existence online and is thus available to be used (and not used) at their discretion.

Canada Post for example charges $5,500/year (http://www.canadapost.ca/cpo/mc/business/productsservices/ma...) just to have access to a postal code verification system. It's all publicly available information... if you walked to every city, suburb, nook and cranny in Canada and asked the local people what their postal code is, they would tell you without hesitating.

But if Canada Post gets wind that you may have copied any of that publicly available information from their databases instead of doing the hard work yourself, you can expect a lawsuit at your door fairly quickly: http://www.cbc.ca/news/business/story/2012/04/13/technology-...

I don't think Canada Post had a case in the above example because the company in question didn't actually copy anything from their database, but you did when scraping Craigslist (for however insignificant you found the information to be), see what I'm saying?

Just because you are sending them traffic, doesn't give you the right to take (insert whatever might have been taken here) from them without their permission.

That being said, I love padmapper and have used it numerous times already for myself and friends. I wish you all the best going into the future!

Thanks for the explanation! I've talked with lawyers about this in the past week, and facts, totally rewritten summaries of fact, and collections of facts aren't copyrightable in the US, so it's not really an issue of their content, as far as I understand. They do have a right not to be crawled by me, though, especially if it causes a burden on them. I wrote that crawler specifically to be very light on their servers including making tradeoffs to usability (it never recrawls listings), but that's still their decision to make. If there are ways to get the data that don't involve touching their servers, then the TOU no longer applies, if I understand the law, but it would obviously make them angry, and so I'm most likely not going to do that.

What about copyright traps?

Say Craigslist created a fake listing or two for say a 2 bedroom apartment in the middle of an industrial complex? (ie, an apartment that doesn't exist, but was placed there by Craiglist to serve as evidence against your copying).

If the Craigslist lawyers find that fake listing on PadMapper, could they do anything to charge you with regards to a copyright violation?

I'm not trying to drag this on. I'm only trying to help.

Here is a smoking gun article on Rand McNally's copyright traps (or lack thereof): http://www.straightdope.com/columns/read/1058/do-maps-have-c...

Right, but there they were wholly copying what is essentially a drawn representation of geography, which is an original work. If you made a rough sketch of that map, it wouldn't be copyright infringement.

I agree that this isn't a copyright issue. If you're operating in California, the 9th Circuit is probably your best bet for holding that your scraping of their data doesn't damage or impair their computer system. But given your other comments, it is unlikely you want to raise the funds or spend the time to battle it out in court.

But there is an another way - let us (all of us) get the Craigslist data for you. You could create a browser plugin that would save craigslist posts a user views, and then upload only those posts to a central database that feeds padmapper. With just 1,000 users you would get most but not all the data you currently are scraping.

This is the same approach used by Recap (https://www.recapthelaw.org/) as a response to the federal government's paywall to access public court filings.

Thanks for the advice! But yeah, court sounds like a pretty big clusterfuck with a potentially bigger clusterfuck at the end of the rainbow. I have a feeling that the plugin would get me sued as well, though I suppose that would be a more winnable battle.

Reply on news.ycombinator.com