Can Scraping Non-Infringing Content Become Copyright Infringement… Because Of How Scrapers Work?

from the this-seems-troubling dept

Earlier this year, we couldn’t figure out how Facebook’s lawsuit against made any sense. tried to aggregate various social networking accounts in a single place, so you could manage them all at once through a single interface. Yet Facebook charged the company with all sorts of complaints, including copyright and trademark infringement, unlawful competition and violation of the computer fraud and abuse act. asked for the case to be dismissed, but last month the judge sided with Facebook, but did so in a troubling way, by basically suggesting that since Facebook’s terms of service prohibited these uses, it made it copyright infringement. Michael Scott points us to lawyer Jeff Neuberger’s take on the ruling, and separately Tom O’Toole has a good analysis of the ruling. Neuberger states the following:

Judge Fogel concluded that the allegations of the complaint made out a sufficient claim of copyright infringement because Power Ventures “need only access and copy one page to commit copyright infringement.” The court also found that the ToU prohibited downloading, scraping or distributing content from the Facebook Web site content except that belonging to the user, and that in any event, using automated methods, i.e., “data mining, robots, scraping, or similar data gathering or extraction methods” to access any content were also prohibited by the ToU. Thus, the court found that the allegation that Power Ventures accessed Facebook via automated means constituted made out a claim of direct copyright infringement, while the allegation that Facebook users utilized the interface to access their own profile pages made out claim of secondary copyright infringement.

Thus, because the terms of service said you can’t do any automated scraping of the site, it’s suddenly infringing? Even worse, the court found that even though the data being used by isn’t owned by Facebook (it’s the users’) the scraping was still copyright infringement, because in order to scrape the non-infringing content, had to first “scrape” the whole page. O’Toole explains:

OK, so far the court has found that made unauthorized copies of the Facebook Web site. What about the fact that Facebook does not own the copyright in its users’ profile data? Facebook surmounted this hurdle by arguing that the content of the Facebook page that surrounded the user’s data is copyrightable and is owned by Facebook. According to Facebook, the scraper operated in a manner that required it to copy the entire Web page in order to extract the user’s profile data….

Note that the court is conditioning its ruling on the assertion that the Power Ventures scraper necessarily copied the entire Web page before it processed the page and extracted the profile data. That comports with my (limited) understanding of how a Web scraper works. But is it true? If it were true, couldn’t an argument be made that this is a fair use of the page? I’ll leave that for better lawyers.

All of this seems a bit troubling, as it would effectively rule out scraping even non-infringing content, just because the scraper had to first read through copyrighted content to get to the non-infringing stuff. But, that seems to go against the entire purpose of copyright law. The fact that the scraper reads copyrighted content shouldn’t mean that it’s infringement. It’s not doing anything with that content other than using it to find the content it can make use of. Anyway, this ruling probably doesn’t mean all that much, since it was just to reject the dismissal request, but it does seem odd that the judge gave so much weight to Facebook’s terms of service, and seems to indicate the mere act of scraping can be copyright infringement.

Filed Under: , , , ,
Companies: facebook, power

Rate this comment as insightful
Rate this comment as funny
You have rated this comment as insightful
You have rated this comment as funny
Flag this comment as abusive/trolling/spam
You have flagged this comment
The first word has already been claimed
The last word has already been claimed
Insightful Lightbulb icon Funny Laughing icon Abusive/trolling/spam Flag icon Insightful badge Lightbulb icon Funny badge Laughing icon Comments icon

Comments on “Can Scraping Non-Infringing Content Become Copyright Infringement… Because Of How Scrapers Work?”

Subscribe: RSS Leave a comment
Anonymous Coward says:

I guess what the judge is implying is that if the terms of service forbid access by automated tools, then any access by automated tools is unauthorized and is therefore “infringing.”

As far as I know, copyright infringement requires a medium – in other words, it is not infringement for to view copyrighted Facebook material without authorization, but it may be infringement for to republish copyrighted Facebook material without authorization. Admittedly I don’t know exactly how’s application works, but it would seem that if they are only republishing user information (info that Facebook does not own the rights to) then their scraper tool would fall into the first category.

Anonymous Coward says:

Perhaps facebook should have an option asking whether or not a user wants his information to be displayed on third party sites, kinda like how google and search engines have an option allowing webmasters to prevent sites from showing up on search engines (ie: using metatags). It should ultimately be up to the users, perhaps users don’t want their information being easily searchable by others and they should have that choice.

Will Sinclair says:

This is an interesting topic.

What does the law actually say regarding automated retrieval of content?

I write scrapers, and the very way that they work requires you to use either an HTML POST or GET command to retrieve a target url, the same way your browser works, then you parse the results for the content you are after. Mostly I do all this client side, so it isn’t done by my company’s servers but rather in javascript in the client’s own browser.

So really it’s not my company’s server that’s accessing the content, but a client’s computer, and in the case of a facebook profile, presumably the client has agreed authorized access with facebook. Surely therefore it’s not breaching the ToS to access your facebook profile with your own browser through a secondary interface (scraper)??

No webmaster likes their precious site being scraped, but in all honesty there’s not a great deal they can do about it unless they put a CAPTCHA image on every page, which would seriously detract from the user experience.

Anonymous Coward says:

Re: Re:

the same way your browser works

This is what popped into my head when I read the post. Why would a Browser not be a problem but a Scraper would be?

Also, as far as I know, it’s nmot copyright infringment until you produce a copy — if the product is just information FB don’t control, there should be no case for infringement.

Andrew says:

Dirty scrapers

What about Firefox and IE? To my knowledge both of them make a copy the raw code from the facebook website, enabling the user to extract information from it.

Ok that’s a forced argument – but the point is, if you don’t want information to be available, don’t publish it on the web. And this is as nothing to Facebook asking for my email password. What a stupid question.

Allen Harkleroad (user link) says:


Generally speaking in order to display content from another website using a page scraper, the entire page is copied and most time Regular Expressions (RegEx) is uses to extract specific information and then display it.

We often scrape our own websites content and then display it on another of our websites to reduce the number of times we publish content.

I am sure one could write a scraper that only copied specific content directly from a full web page and display it.

Doctor Strange says:

The rationale seems to be indirectly based on the Ticketmaster decision.

The argument, it seems to me, goes thusly:

When a program or a browser or some other software accesses a website, it is making a copy of the content of that site. That content may be copyrighted. By default, this would be a violation of copyright law. However, the Ticketmaster website included some terms of use that permitted this – effectively granting a limited license for some uses, but not others. Thus, an ordinary user with a browser was not in violation, because they were accessing the content under a license.

RMG’s product, on the other hand, explicitly went outside the bounds of this license, and thus fell back under the aegis of copyright law, where such copying is prohibited. I suppose RMG could have mounted a fair-use defense, but it appears they explicitly did not.

Now, you could argue, what about the millions of websites that do not have terms of use granting a license? Is just browsing these sites a commission of copyright infringement?

The answer is likely no, based on the Perfect 10 ruling. This ruling indicated that some caching can be fair use; in that case, the caching was “noncommercial, transformative, and has a minimal impact on the potential market for the original work.” The difference in Ticketmaster was that the fair use defense was not adequately asserted, and moreover the RMG software was in pretty clear violation of Ticketmaster’s terms of use.

Killer_Tofu (profile) says:

Re: Re:

Your post once again proves to me how stupid copyright laws are in the technological world we live in today.
None of those tiny details should matter at all.
Copyright is a horribly outdated pre-internet system that just isn’t needed anymore.
Scraping like this should be less than no problem. Simple.

Granting tiny limited licenses? C’mon. Its dumb.

Anonymous Coward says:

Re: What about Google?

Google’s crawler (and other search engines) must be infringing then as well as violating the TOS. Should google stop crawling facebook?

I thought the same thing, but it goes beyond Facebook. Is this yet another ruling that would effectively criminalize Google if it were applied to everyone (of course it won’t be)?

inc says:

So Facebook puts a page available to be viewed by users, or scraped by their browser and eyes. So using a different access method also allows users to see the same available page. Also how can a ToS create law? Last I checked in the U.S. only Congress had the power to create law not the courts or a ToS. Just because it’s in a term of service doesn’t extend copyright law or give the courts power to interpret any law in a different way. Unless is actually breaking Trademark laws by claiming they are Facebook then this is just monopoly tactic.

Add Your Comment

Your email address will not be published. Required fields are marked *

Have a Techdirt Account? Sign in now. Want one? Register here

Comment Options:

Make this the or (get credits or sign in to see balance) what's this?

What's this?

Techdirt community members with Techdirt Credits can spotlight a comment as either the "First Word" or "Last Word" on a particular comment thread. Credits can be purchased at the Techdirt Insider Shop »

Follow Techdirt

Techdirt Daily Newsletter

Techdirt Deals
Techdirt Insider Discord
The latest chatter on the Techdirt Insider Discord channel...