Craigslist had to sue or otherwise anyone could scrap data from Craigslist, whether they were making "Craigslist more valuable" or not. Simple as that.
1. Why is that? I think the most relevant field of US law with that sort of provision is trademark law, which shouldn't have been at issue here at all (3taps and Padmapper were not pretending to be Craigslist or authorized by Craigslist). Copyright law, which is sorta relevant, does not have such a provision. If your song gets pirated, you can sue only the pirates named Fred without losing your right to sue. The CFAA, which is definitely relevant here, does not have such a provision: JSTOR does not have their hands full with CFAA cases against everyone who ever sent an article to a colleague or friend.
2. What prevented Craigslist from giving a straightforward license and API to Padmapper? That would have sidestepped any such concern, without setting a precedent that everyone else must get API access. I think it's pretty clear from this case that they did not want to see Padmapper exist, and regardless of whether they were obligated to sue and try to shut them down, they clearly wanted to sue and shut this particular application down.
I don't know why this was downvoted, but this was obviously the case. CL had no interest in these 3d party sites. It's their data and their terms of use, so there really aren't any debatable points. Nobody has a right to CL data unless they explicitly provide that right. If I were CL, I wouldn't have give Padmapper an API either; their arrogance was astounding. They had no respect for the CL terms of use. They built an entire company around a knowing violation of their data source's terms of use. The Terms said, "you can do X," Padmapper did "X" then complained when CL said no. It doesn't matter what X was, Padmapper just knowingly did it anyway. They had the intent to violate the Terms of Use.
It's amazing how wholesale scraping of data was being celebrated, because the scrapers "were making the internet better".
This is the first topic I've seen were the HN's usual Libertarian bias disappears and no-one challenges the underlying notion that CL is seen as part of the Commons. Channelling Dagny Taggart, I'd say 3Tap & PadMapper were a bunch of moochers.
You're making it sound like the scraping was being done to replace the functionality of CL, which, yeah, would be pretty transparently shitty to do. But they weren't doing that, especially PadMapper: They were indexing the content to make it more accessible, an action that's been taken probably trillions of times and is pretty much the main reason most of the internet is even usuable today. It's like accusing Google of plagiarizing your website because they linked to it.
That's a good point - does craigslist have a robots.txt to prevent Google from crawling it? If not, isn't Google guilty of the very same thing, by aggregating the information via search results?
Craigslist doesn't prevent Google from crawling them. Not only that, Craigslist also sued at least one company for scraping Google results in order to index Craigslist postings.
Yes, I'd agree with that. That's why I think it's incorrect to claim that Craigslist had to sue, even if there was some legal obligation (which I don't think exists). They clearly wanted to shut Padmapper down. If they didn't want to shut Padmapper down, they would have given them a license and API, and then any legal obligation would have been irrelevant, because Padmapper would be an authorized user.
I don't think people are really arguing that part. Well Techdirt is a bit with the whole whining about "b-b-but 3taps was making CL more valuable" as if that mattered (apart from being open to interpretation).
The issue is that CL and this judge have set a precedent (I think?) that private companies can make up felonies by deciding what "authorized access" is.
You wanna sue a business for scraping your site? Hey, fine, that's a civil business issue. (And really, there shouldn't be legal limits on that kind of behaviour to a publicly accessible system anyways. Or at least there should be some level of malicious intent required.) Trying to use a law with criminal/felony penalties is grossly out-of-line.
> there shouldn't be legal limits on that kind of behaviour to a publicly accessible system anyways.
Are you sure about that? So you don't want any legal recourse for someone that overloads your API/web servers with thousands of requests per second? You might say a technical solution (like rate limiting) would do - but that will cost you time & money with no upside, and can be easily circumvented (use many clients in parallel with the same effect - hammering your server and increasing your costs)
I think a law for that is fine, as a last resort. It will keep people at their best behaviour.
We likely are in agreement - isn't a rule against hammering a server 'a legal limit on the kind of behaviour to a publicly accessible system'? I was disagreeing with parent who said:
> there shouldn't be legal limits on that kind of behaviour to a publicly accessible system anyways.
The difference between "limit on that kind of behavior" and "limit on the kind of behavior" is critical.
A limit on red cars is a limit on "the color" of cars, but it's not a limit on "that color" when someone is pointing at a blue car.
So when "that kind" refers to sanely-implemented scraping, then there should not be a limit. There is a limit on "the kind", to restrict it to activities like scraping that aren't inherently harmful.
Unfortunately english isn't very precise: it's difficult to parse what MichaelGG was referring to by 'that'. I assumed he was referring to scraping (his previous sentence), but it's also possible he was referring to the CL case.
For the record, I maintain my belief that there should be a law that applies to scraping of publicly available sites.
> You wanna sue a business for scraping your site? Hey, fine, that's a civil business issue. (And really, there shouldn't be legal limits on that kind of behaviour to a publicly accessible system anyways
> You wanna sue a business for scraping your site? Hey, fine, that's a civil business issue. ... Trying to use a law with criminal/felony penalties is grossly out-of-line.
And the CFAA also happens to be the law that makes it a civil business issue, thanks to the 1030(g) section that was amended in 1996. Wikipedia has elected to make no mention of this section, but other sources do.
