This is actually not a great decision so far.
I skimmed the decision from PACER (so the below may be wrong or missing facts), and it looks like (and remember, this is just a preliminary decision by the court) that the court found
1. The individual posts were copyrightable
2. The compilation (whose only creativity is being organized by geographic area), was copyrightable (because "“deciding which categories to include and under what name,”" met the minimal level of creativity necessary for copyright)
3. Craigslist's eventually lost the copyright claims because the TOU was not an exclusive license (in the US, only an exclusive rights owner can sue for infringement, licensesees cannot)
4. Craigslist did not need to register the actual posts it's suing over, only the compilation
#1 is maybe right (i think it depends on the post)
#2 should be clearly wrong, though there is a bit of bad precedent here
#3 is good, but the judge basically suggested with the right set of words buried in the terms of use, craigslist could have transferred copyright. This is actually really dangerous in practice (you want people to know exactly what they are getting into), because you really want copyright transfers to be clear and knowing by users, and buried somewhere in a separate TOU document, plus a button says "by clicking you accept the TOU", shouldn't be enough.
Of course, this is just a preliminary decision based on minimal briefing, so i'm hopeful the judge would have gotten it "righter" had those claims not been dismissed.
#4 is probably wrong, in the sense that registering "craigslist.org" as a compilation, should not count as registering all the works that have ever appeared on craigslist.org.
At best, it should be registering the works that appeared as of the copyright registration date for craigslist.org.
The compilation registered in 2012 is titled "Craigslist website 2012 (Homepage and primary supporting pages)"
That should not be "all posts".
There are also two compilation registrations for post.craigslist.org, one in 2004, one in 2008.
None of these should be sufficient to claim what they did.
Keep in mind, from the judgement, section II Legal Standard:
In determining facial plausibility, whether a complaint states a plausible claim is a “context-specific task that requires the reviewing court to draw on its
judicial experience and common sense.” Id.
at 679. Allegations of material fact are taken as
true and construed in the light most favorable to the non-moving party.http://3taps.com/images/pics/430_138734862-Judge-rules-in-fa...
So essentially these findings are made while reading all facts in the best possible light for craigslist. So it's not necessarily saying, for example, that the compilation is copyrightable - just that it is conceivable that it could be.
I agree in theory, but at least my complaints are basically "even if what craigslist said is completely right, they still should have lost on the law", which is in line with the judgement standard :)
"Craiglist has been recognized for a failure to innovate and having among the internet’s worst terms of service’s."
Wow, that is quite a way to open an article! Isn't journalism supposed to be slightly more neutral than that. I understand that most of us (myself included) are quite against copyrighting, DRM and closed source etc. but let's not harshly rebuke a company with 30 employees [1]. Simply put, copyright is a contract between the society and an individual that encourages the individual to create, innovate and be artistic. Craigslist did things that they thought were right to protect their users' and their own interests. They found a market. Padmapper used their data. Courts said it is ok to do so since they didn't the copyright to begin with. Fine, let's move on.
No reason to write such a bitter sounding article peppered with "Let this be a lesson to all tech companies", "hopefully this will discourage craigslist", and "sledgehammer to scare innovators". tsk. tsk. Forbes.
Yep - I spearheaded that campaign. We got a bunch of bills introduced in Congress, White House reversed their position, FCC is still investigating. I'm trying to get the best bill through Congress while we lay the groundwork for an overall copyright reform bill.
It's not every day I say this to a Republican -- in fact, I'm not sure I ever have before -- but seriously keep up the good work. We're all counting on you. :-)
Hopefully some day Craig Newmark will be held responsible for being the internet pariahs he is. Craig is a sweet, awkward man with a lot of power which he has allowed to be abused by others, and he is responsible for that. He needs to step in and control his company in the direction of good, or be seen for what he is - an appeaser of greed and evil.
Does this mean it's okay to scrape / clone any user generated content without legal repercussions? I can make a HN clone that copies all the comments and everything and things would be just peachy legally?
No. For one thing, this ruling simply dealt with copyright claims. If you scrape a site you are subject to their terms of service, most of which prohibit scraping. So there would be contract and possibly even CFAA claims to deal with.
Also, the specific reason the copyright claims were dismissed in this case is that craigslist does not have an exclusive license on its posts. If Hacker News' terms exclusively license comments (which I doubt, but I haven't checked), that would make this particular ruling less relevant. It is also possible that the comments' authors could make claims against you, should they wish to do so.
Can terms of use prohibit scraping for reuse, while still allowing Google, et al., to index?
I genuinely wonder this. Can I say, "it's OK for XYZ to scrape, but not ABC?" (Or is it general use -- scraping OK/not OK?)
I think it's different for API usage, because you enter an agreement with an entity by signing up, etc. I'm speaking specifically to the unauthenticated scraping/indexing.
I don't know, but I suspect that they can. If I own some data, I should be able to give (or sell) it to you and not to someone else.
That said, once I've given it to you, if you choose to give it to someone else, I have no claim against that someone (although I might against you, depending on our contract). That's the heart of 3taps' position as I understand it; they get craigslist post data from third party sources like search engines, instead of scraping craigslist, and so do not have a contract with craigslist.
I'm pretty sure I agree with you, if you have a specific agreement with one of the parties.
What about "anonymous" scraping, however? (I put it in quotes because I know you can reverse DNS/UserAgent/etc. identify... but that's simple to spoof, and we're discussing within a sophisticated context.)
Can I say that "OOgle" can use my data (because I want visitors) but others can't? (To complicate, can I exclude only certain, specified, parties and include all others?)
Yes, you're right. What I meant to say was that only the copyright claims were ruled on, but even that is not strictly accurate. (And in fact, a conspiracy claim was also dismissed, so doubly inaccurate.)
To clarify, my meaning was that among the various potential legal impediments to scraping and displaying website contents, this judgement would set precedent for copyright only, if anything.
Not yet. That's just my blog. I did take into account people's comments here and adjusted accordingly. I wrote this to break the story - while I was involved with other work. I have headed many of this forum's suggestions.
Err... he's not necessarily being paid for this writing, and even if so, it's probably not equivalent to full-time wages. Keep a look out for a great place for Derek!
Is it really hard to create a Craigslist clone? Just curious. I say 'clone' because of the unethical practices followed by Craigslist, and we can have some one more open-minded replace them soon (with enough marketing). I'm not sure why no YC companies (or anyone else for that matter) want to disrupt Craigslist, though.
The internet works the way it is now because of the concept of 'free and open' information. Attempting to harm the ones that enable such open-ness isn't exactly the right recipe for a long term successful Internet based business model.
Creating the clone is fairly easy. For every type of service that Craigslist provides there have probably been 1-5 clones over the past few years.
Currently both buyers as well as sellers have been trained that the place to go is Craigslist. Give that it would be impossible for a new service to have anywhere near the number of buyers and sellers as Craigslist, no rational individual is going to choose an alternative.
The problem becomes even tricker since Craigslist is for the most part free. They currently use the revenue gained by job postings to subsidize the other listings on their site. This makes it harder for a niche service to start up as they are competing with free.
For some reason, Kijiji is dominant in most of Canada. Craigslist is almost never used here. I'm honestly glad, because I occasionally check Craigslist and it's never a rewarding experience. Without the network-effects monopoly, it's a truly awful site.
There have been many discussions on HN about CL.a quick search will be extremely helpful.
Afai recall, CL is being pushed into corners by people who desperately want to eat it alive.
Further - users like CL. As much as people wax on about how features and a UI will make a big difference, fact is that in its geography people don't really care about that.
There's also a chart out there showing that CL is being disrupted, in that several silos its present in have alternate service providers.
I must further point out that the ideological posturing at the end assumes that the next firm who comes in, and adds yet another middleman to the chain, will let the data be free and open.
Should they end up consuming/transitioning most of CLs base, who is to say they won't institute a lock in of some sort.
And a YC firm definitely can't operate on the same budgetary and revenue constraints CL imposes on itself.
I'd hate to see a significant rework of the UI for Craigslist--it's so bloody simple that it exactly solves the problems I have, and the site is so barebones that I can browse it on near anything without issue.
I probably save 95% of my apartment search time with Padmapper instead of Craigslist. What is needed is an API so that fans of Craigslist can keep their torturous interface and the rest of us can get the info we need and go on with our lives.
They can afford to be unethical (assuming they are) thanks to their popularity. So any clone would also have to clone Craigslist's ability to attract posters. Otherwise the site is useless, no matter how ethical or better designed it was.
If this is an accurate summary of the legal situation (not at all something one can assume with this publication) then it sounds as if Craigslist simply needs to change their terms so that a user agrees that Craigslist has a perpetual and exclusive copyright on the content of listings. Then you may be able to aggregate the data in some form, but would have to be careful how you used the content of the listings.
Copyright transfers must be assigned explicitly and in writing (digital signature would be okay).
"(a) A transfer of copyright ownership, other than by operation of law, is not valid unless an instrument of conveyance, or a note or memorandum of the transfer, is in writing and signed by the owner of the rights conveyed or such owner's duly authorized agent."
So the standard TOS at the bottom of a page would not be okay to transfer copyright.
According to the decision, craigslist moved to a click-through eventually.
I think they tried this for a bit (last summer?), but it didn't last long. The few times I need to use Craigslist I can't afford to give them an exclusive license.
Exactly. There was also speculation that they realized it could open them up to liability for the content of posts, which obviously they would not want.
There's a post on the EFF blog regarding some other elements of the decision (particularly relating to exclusive copyright licenses and to the CFAA): https://news.ycombinator.com/item?id=5641151
1. The individual posts were copyrightable
2. The compilation (whose only creativity is being organized by geographic area), was copyrightable (because "“deciding which categories to include and under what name,”" met the minimal level of creativity necessary for copyright)
3. Craigslist's eventually lost the copyright claims because the TOU was not an exclusive license (in the US, only an exclusive rights owner can sue for infringement, licensesees cannot)
4. Craigslist did not need to register the actual posts it's suing over, only the compilation
#1 is maybe right (i think it depends on the post)
#2 should be clearly wrong, though there is a bit of bad precedent here
#3 is good, but the judge basically suggested with the right set of words buried in the terms of use, craigslist could have transferred copyright. This is actually really dangerous in practice (you want people to know exactly what they are getting into), because you really want copyright transfers to be clear and knowing by users, and buried somewhere in a separate TOU document, plus a button says "by clicking you accept the TOU", shouldn't be enough.
Of course, this is just a preliminary decision based on minimal briefing, so i'm hopeful the judge would have gotten it "righter" had those claims not been dismissed.
#4 is probably wrong, in the sense that registering "craigslist.org" as a compilation, should not count as registering all the works that have ever appeared on craigslist.org.
At best, it should be registering the works that appeared as of the copyright registration date for craigslist.org.
The compilation registered in 2012 is titled "Craigslist website 2012 (Homepage and primary supporting pages)"
That should not be "all posts". There are also two compilation registrations for post.craigslist.org, one in 2004, one in 2008.
None of these should be sufficient to claim what they did.