Slashdot Mirror
9th Circuit Court Finds 'Thumbnailing' Fair Use
mark_wilkins writes "A photographer named Leslie Kelly had sued Arriba Soft Corporation for infringing his copyrights to photos when they made thumbnails of his pictures and stored them in a public image search engine. Today the federal 9th Circuit Court of Appeals affirmed the district court's ruling that making these thumbnail copies of images for the search engine was 'fair use.' Since the applicability of fair-use defenses to copyright infringement touches on all kinds of common uses of the Internet as well as rulemaking related to the scope of the DMCA, this decision will probably have an effect on the discussion. (Note that this case was decided by a 3-judge panel and thus isn't binding precedent.)" Note that the court also reversed in part the lower court's ruling, specifically saying that the lower court should not have ruled on "whether the display of the larger image is a violation of Kelly's exclusive right to publically display his works."
Hopefully the RIAA wont object to me distributing thumbnails of music as MP3's
One small step for law, one giant leap for freedom!
Good question. All the copyright images that I use on my site are shrunk 99% of their original size before uploading...because I lawfully only publish thumbnails.
I believe this is already the case. I believe that artists/radio stations can use samples (shorter than 30 seconds i think) without having to pay for them.
"I can not bring myself to believe that if knowledge presents danger, the solution is ignorance" - Isaac Asimov
So what actually constitutes as a thumbnail? A lower quality version of an original source? This could apply to music as well, since a low bitrate mp3 is a poorer quality version of the original raw cda. In terms of electronic use, all data comes down to 1s and 0s. Does this ruling apply to other forms of electronic duplication like music, or is imagery a whole different can?
Without reading the article / ruling / whatever (I'm lazy today)... what makes a thumbnail OK?
Is it because it presents significantly less information than the original?
What if it remained the original's size, but was B&W instead of color?
What if it were saved as JPEG quality 2 instead of the original?
What if the court stated some metric? Like "must be at least 50% less than the original"... how about cutting the image in halves. Then posting both halves on your site such that they appear as one? Neither half violates individually?
Google will go a step further and remove the location of your image as well as the thumbnail of your image from their image search database if you request it. The email is dmca-images'AT'google'DOT'com
Remove an image from Google's Image Search
I wonder if they will consider changing that policy now?
Whenever the offence inspires less horror than the punishment, the rigour of penal law is obliged to give way...
I said exactly this last time /. posted this story, and now I've actually automated similar systems, so I'll repeat it more loudly this time. OK Mr. Kelly, are you listening?
:(
watermark your images
You can create yourself an action in Photoshop, or there's at least two very good free software packages that can do the same thing (Gimp and ImageMagick). If you don't want the whole world grabbing your images,
don't publish them on the web
at least, not without putting some kind of protection in front of them.
There are well known solutions to this guys problem, and he choses the courts?! I guess that's really the American way
std::disclaimer<std::legalese> sig=new std::disclaimer; sig->dump(); delete sig;
That wasn't an issue even when fair use was an unquestioned law. If you were to reproduce an entire short story in the form of 2 line quotes, no one in their right mind would believe you're not violating the copyright.
I think you're right in the sense that a "thumbnail" is arbitrary, but I think that if the court did state some metrics, the problem would be pretty much solved. At least when it comes to pictures.
Warning: Opinions known to be heavily biased.
Oh great. Another ruling from the most overturned appellate court ever.
</fat_comic_book_guy_from_the_simpsons>
A programmer is a machine for converting coffee into code.
There's some confusion here -- let's separate "binding" from "precent". Anything can be a precedent. Binding is another matter.
Circuit courts are a strange system. Each of the circuits covers several states. What a particular circuit rules is always binding in that circuit. By "binding", this means that lower courts (namely, federal district courts) should consider that ruling the "law of the land" when they make their decisions.
What this means is that the law of the land in one circuit (and therefore in one state) may different from that in another. A law found constitutional in one state may be unconstitutional in another. Only the Supreme Court can resolve these differences, and although the S.C. turns down far more cases than it hears, it almost never turns down a case that will resolve a conflict among circuits. And of course, rulings by the S.C. are always binding on all federal courts.
The "only a three judge panel" part is confusing. Usually, when a case is heard in Circuit Court, three judges (from the 10 or 15 or so in that circuit) will hear the case, and the "best out of three" wins. In some rare cases, the entire circuit will sit "in panel" to hear a case -- often if they want to review the finding of a three judge panel that seems out of whack. This is rare. But when the Circuit speaks as a whole in this way, that precent takes precedence over any previous three-panel decision in the circuit in the same case.
As I said, three judge panels are the norm. Their findings are perfect "binding" within the circuit.
Anybody who actually is a lawyer (or just knows better) can feel free to correct me -- I'm in a hurry and didn't have a chance to double-check the finer points, but in gross, this is how it works.
False. I can't believe this is in the story header, it should be changed immediately. Appellate courts like the 9th Circuit generate binding precedent every time they publish an opinion (some other appellate courts also generate precedent through unpublished opinions).
Sure, it's on the lowest rung of binding precedent. It can be overruled by an en banc panel, or it can be overruled by the US Supreme Court. But it's still certainly can be cited in other cases.
Not to give any credibility to this site, but
We have to distinguish between published and unpublished opinions in some districts, but the point basically stands.
The submission incorrectly states that because this case was decided by a three-judge panel, it is not binding precedent. The U.S. Courts of Appeal typically hear cases in three judge panels. Very rarely, all of the judges in a particuar circuit court will hear a case "en banc." While the court sitting en banc can overturn a panel decision, the panel decision is binding precedent (whatever that means, especially in the murky area of fair use) unless and until it is overturned by 9th circiut sitting en banc, or by the U.S. Supreme Court. Of course, other Circuit Courts (e.g. the 10th circuit, the 6th Circuit, etc.) are not bound to follow precedents from their sister circuits, though they often do.
IAAL, but this post is not intended to constitute legal advice. If you need advice, see your lawyer, not Slashdot.
"The dinosaurs died because they didn't have a space program." - Niven
In short, how many of our laws affect out other laws in unforseen ways?
"It's the height of ridiculousness to say for those 9 lines you get hundreds of millions."
The poster's absolutely correct. With very few exceptions (like, say, Louisiana), every court decision creates binding precedent throughout its jurisdiction. A panel decision possesses almost as much precedential value as an en banc decision; the only difference is an en banc hearing can overturn a panel decision, while a panel decision cannot overturn an en banc decision.
As a matter of practice, en bancs are almost superfluous. Looking at the statistics for the 8th Circuit Court of Appeals, they hand down unanimous panel decisions 97% of the time. This is for a court which is stocked almost straight 50/50 with appointees from hardcore Republicans and hardcore Democrats.
Only 3% of the time is a panel decision split; and less than 10% of the split decisions are sufficiently "interesting" (in a legal sense of the word) to warrant an en banc hearing.
If anyone's interested I can get statistics for the other circuits as well--I only know the 8th Circuit stats off the top of my head, though.
IANAL.
Chalk it up to the pervasive influence of the Moonie Times, Scaife funded foundations and Father Limbaugh ... proving that if something gets repeated enough, it is accepted as truth, despite facts to the contrary.
From a NY Times response by Judge Noonan Jr. of the aforementioned 9th circuit - you'll probably have to pay to get the article but here is a blurb:
Of course, you're free to adhere to the Moonie Times myth that the 9th circuit court is some aberration of justice totally out of alignment with the rest of the world and other judicial bodies...
AZspot
if you get down to page 12, you see that the court's opinion was based not only on the degradation in quality but on two other factors: the 'transformation in use' in that Kelly's work was for aesthetic value and Arriba's infringement was for informational value; and that Arriba's infringement was not competitive with the market for Kelly's product. it would seem that neither of these factors is present in mp3s of lower quality, and yet both are present in 30s clips of music, thus meshing well with law in those areas.
U.S. War Crimes blog. Email for free Mandriva support.
First, the July 7, 2003 decision IS precedent and can be cited. What is no longer precedent is their Feb. 6, 2002 decision. A lawyer in the know would have to tell you why, but my guess is that some technical glitch allowed Kelly, the plaintiff in the case, to ask for a "rehearing en banc." By withdrawing and refiling what seems to be the same decision, the appeals court created their own technicality that allowed them to rule Kelly's petition moot. Law is full of those sorts of games.
The decision mattered in my case because I'm in Seattle, which in 9th Circuit and a 9th Circuit Appeals decision like this one has the force of law at the district court level where I was fighting. If I could show a great deal of relevance between "fair use" in that case and my appeal to fair use, my defense would be on very solid ground.
That mattered because the Tolkien estate's case was built on a series of 1998 decisions in the 2nd Circuit (New York), the most important of those cases being Castle Rock, a decision that found a book of Seinfeld trivia called the Seinfeld Aptitude Text as an infringing derivative.
The court's rationale was that a fictional author creates an entire world and any use of that, whether in a trivia book or some sort of viewer's guide was infringement. The decision was much criticized in legal journals, no other circuit has followed it, and, to my knowledge, tthe 2nd Circuit hasn't repeated it. You can go to:
http://chillingeffects.org/
for a discussion of the effect that has had on Internet fan fiction. It has also made if VERY RISKY to do guide books to popular movies, TV shows or, by extension, popular works of fiction like Lord of the Rings. Books that help readers understand fictional works, rather than make academic literary comments on them, are at risk until the Castle Rock decision is buried.
My defense was that online art (as in Kelly v. Arriba) is a form of fiction and my bullet-list summaries of what happened each day the equivalent of thumbnails. Settling out of court, we will never know if the judge would have bought that argument. But I did have a lawyer tell me that if I'd won on it at the district and appeals level, the case would have headed for the Supreme Court, since the 2nd and 9th circuits would have been in conflict. Since that would have taken years, I was better off settling out of court.
This decision applies to music in a roughly similar fashion. The Castle Rock decision virtually eliminated a whole spectrum of what would otherwise be fair use categories simply because the work was art/fiction rather than fact/biography/history. It would be very easy to slip music into art, making even brief excerpts, perhaps in an Internet radio show, illegal.
The Kelly v. Arriba decision has two key factors. First, when the reduction is great enough (i.e. a picture is reduced to a thumbnail) the original purpose of the art is no longer being served, so the thumbnail is not infringing.
The second factor is that the thumbnail is part of something (i.e. an image database) that is serving a different purpose, in this case, indexing internet images. If that purpose has public value, then it is fair use and protected from charges of infringement. Think, for example, of a book that gives the basic plot of movies or an audio database that has short (stressing short) samples of music, indexed perhaps by artist, gendre, theme, etc.
That's w