Kelly vs. Arriba (PDF)
February 11, 2002 10:55 AM   Subscribe

Kelly vs. Arriba (PDF) Arriba search engine has been determined to be infringing on the copyright of photographer Leslie Kelly. The reason? Arriba displays thumbnails of copyrighted images in their search results, and displays the original page in a frameset. What kind of precedent will this set for Google and the rest of the web?
posted by johnjreeve (17 comments total)
tha't strange. according to newsbytes, the court ruled that thumbnails ARE ok. The issue at hand, and where arriba really lost, was that they were presenting the full images also.

"Writing for the three-judge panel, Judge Thomas G. Nelson said that the indexing and use of thumbnail images was a fair use, in part because it was "transformative" or added value to the work, and also because it did nothing to diminish the market for Kelly's works.

But the judge also ruled that "Arriba's display of Kelly's full-sized images is not a fair use and thus violates Kelly's exclusive right to publicly display his copyrighted works."
posted by bliss322 at 11:03 AM on February 11, 2002

"In conclusion, all of the fair use factors weigh in favor of Kelly. Therefore, the doctrine of fair use does not sanction Arriba's display of Kelly's images through the inline linking or framing processes that puts Kelly's original images within the context of Arriba's web site."

sorry. thumbnail are ok. Framesets are bad.
posted by johnjreeve at 11:04 AM on February 11, 2002

Ok. I'll be the first to bite.

In my sloped-headed, uneducated point of view, I'd tell Leslie Kelly to remove the photos if he's so concerned about being infringed upon. This is a risk any artist takes when choosing to display their work in a public forum such as the internet. Just ask any indie artist. It's not like Arriba was running off copies, matting and selling them.
posted by KevinSkomsvold at 11:07 AM on February 11, 2002

I had also read that the court found in favor of thumbnails as ok and not aninfringement. Now: can I take a copyright picture and turn it into a thumbnail which can be opened up to full picture and not be in violation?
posted by Postroad at 11:09 AM on February 11, 2002

It's unfortunate that the judicial system cannot not know enough about the technology to distinguish the common from the necessary, but it's also inevitable. And so once again we cripple anything that shows a chance of damaging existing systems of control, for fear of (horrors) reducing someone's profits.

posted by Mars Saxman at 11:23 AM on February 11, 2002

I would just watermark or put a url on the picture so anybody who creates thumbnails or uses framesets ends up providing free advertising.
posted by srboisvert at 11:47 AM on February 11, 2002

I stumbled upon this site that has resources about copyright, but takes a pretty hard line view of thumbnails, linking and fair use.
posted by ajayb at 11:50 AM on February 11, 2002

The court's decision seems pretty sound to me. Thumbnails are okay, because they have a legitimate use (in this case, guiding users of the search engine to the artist's site.) Framesets are not, because they use the artist's work while depriving the artist of the benefit (people coming to his site & generating traffic for advertisers.) I'm not sure what the implications are for a company like Google that stores & offers up cached versions of a web site (presumably incluing images), but I notice that I just did a search on Google for Leslie Kelly & found plenty of references to this court case, but no links to his site or cached versions thereof.

KevinSkomsvold - "This is a risk any artist takes when choosing to display their work in a public forum such as the internet.
You might just as well say that anybody who publishes a book of their artwork deserves to be ripped off, since photocopiers abound. Nevertheless, we do have copyright law. I admit that there is no practical way to prevent individuals from grabbing your work off the web once you put it there, but going after a company that is making commercial use of your work & possibly depriving you of the profits thereof seems within the bounds of reason.
posted by tdismukes at 12:12 PM on February 11, 2002

I think one thing that would protect a search engine like Google in a case like this is that Google presents the full-sized image in its original context (using its own frame to also display the image independently). This means that any text-based copyright message or whatnot that accompanies the picture in its original context will still be displayed with the result of the search. Which Arriba did/does not do.

Mars Saxman -- did you READ the result of the case? This finding doesn't "cripple" (or even chill) the internet or search engines. The ruling is very sound and quite knowledgable about the technologies at hand.
posted by bcwinters at 12:39 PM on February 11, 2002

tdismukes: the point your argument misses is that no copying occurs. Technologically speaking, there's no difference between embedding an image and providing a hyperlink to one. The browser really doesn't care whether a page provides the address of a picture on the same server or the address of a picture on a different server. Many sites take advantage of this fact, separating pictures onto a separate server (or array of servers) to reduce load on the page server.

but going after a company that is making commercial use of your work & possibly depriving you of the profits thereof seems within the bounds of reason.

The company is making commercial use of your work by directing other people's web browsers to your server. If you're not taking advantage of the referral, it's your own fault for throwing away potential business.

But it doesn't appear that way, given an uninformed glance at the web browser, so I doubt the law will care very much.

posted by Mars Saxman at 12:44 PM on February 11, 2002

bcwinters: drat. I read a chunk of it, but apparently it was too early in the morning for my brain to deal with pages of oversized PDF legalese. It appears that I completely misunderstood.

sorry, folks, please ignore everything I've said in this thread!

-Mars (off to make some coffee)
posted by Mars Saxman at 12:54 PM on February 11, 2002

The way I understood it is that displaying the site from which the image came, in the context of an Arriba frameset, was an infringement of copyright. Or was Arriba linking to the image and not the site? (I read the PDF over and over and i'm still confused).
posted by johnjreeve at 1:45 PM on February 11, 2002

In the 'real world' copyright symbols on the item itself denote whether a given work can be duplicated. Why can't the search engines search for copyright symbols?

As a more tech-savvy answer, why cant the artist simply use a robots.txt file to specify that they don't want their pages walked?

If this puts too much onus on the content creator, then flip it, and only index those sites that specifically have a robots.txt or meta tags that give permission for such third-prty indexing and display.

All the better for me. I'd put the tags in and get a bunch more hits since users would be searching a smaller pool.
posted by kfury at 1:51 PM on February 11, 2002


Copyright symbols aren't necessary for works after March 1st, 1989. They are highly recommended, but optional. A good page on copyright is at: Circular 1 - Copyright Basics, which explains the reason for that date, and has a lot of other information on copyright. So searching for a copyright symbol alone isn't sufficient.

The problem with web robots is that there really isn't an official standard for robots and robot.txt files, and not every spider follows them, nor are they required to. The closest we have is the Robot Exclusion Protocol from 1994. this "standard" was developed on a robots mailing list. This is what they have to say about the status of their protocol:

"It is not an official standard backed by a standards body, or owned by any commercial organisation. It is not enforced by anybody, and there no guarantee that all current and future robots will use it. Consider it a common facility the majority of robot authors offer the WWW community to protect WWW server against unwanted accesses by their robots."

The big case about frames and copyright violations and trademark infringement was Washington Post Co. v. Total News, Inc. That case ended in a settlement, but it was the inspiration for many people to consider carefully the use of frames to access another persons material on a different web site, without permission or payment. This case actually provides some guidance from a court on the subjects of the use of frames, and thumbnails. In many ways, it appears to be a workable result.
posted by bragadocchio at 2:41 PM on February 11, 2002

I own the rights to my photos. I get to determine how they are used. I get to say whether someone can use them commercially. The image search engines aren't all honoring those rights. Court decisions like this are good. They remind business that they can't steal intellectual property and use it commercially.

On my site, I've got a robots.txt file, no robots metatags, text copyright notice and ugly copyright notices on each full-size photo. Nevertheless, the site gets indexed. It would be nice not to have to ruin the presentation of my art because of the search engines.
posted by neuroshred at 2:48 PM on February 11, 2002

OK, help. I do copyright law for a living, and I can't parse out what the court here means by "framing." Here is the Court's verbatim description of the "framing" it had a problem with:

Alternatively, by clicking on the "Source" link or the thumbnail from the results page, the site produced two new windows on top of the Arriba page. The window in the forefront contained the full-sized image, imported directly from the originating website. Underneath that was another window displaying the originiating web page. This technique is known as framing. The image from a second website is viewed within a frame that is pulled into the primary site's web page.

Can someone fill me in here? Does this description sound like framesets in the way that Google does framesets? They are talking about new windows -- are the new windows purely the stuff described, or do they include Arriba-specific stuff and just inline link the full image? Later on in the decision, the Court says that part of Arriba's problem is that it "actively participated in displaying Kelly's images by trolling the web, finding Kelly's images, and then having its program inline linke and frame those images within its own website."

posted by IPLawyer at 5:24 PM on February 13, 2002


Take a look at the totalnews website, which has a similar frame structure to what it had before. At one point, when you clicked on a link on their site, the new site would open within the frame to show the news source with the totalnews frame still around it, including totalnew's advertising. Thats the type of framing that had them at the wrong end of a legal settlement. These days, a link from totalnews opens in a new window.

If the Google frame which opens upon clicking upon the thumbnail contained advertising, and didn't have a link to "remove frame," I'd think that they might be skating upon a similar thin patch of ice.
posted by bragadocchio at 10:22 PM on February 14, 2002

« Older   |   Who wants to be a saintly millionaire? Newer »

This thread has been archived and is closed to new comments