Linking directly to individual pages on a Web site instead of the home page, also known as “Deep Linking”, is a staple of blogging and the Internet in general. It is used as a means to reference sources, forward interesting articles and, generally, get information out there on the Web.
Without deep linking, social news would likely not exist, many Web 2.0 services (such as Delicious) would have to close and even Google would have to drastically change the way it operates. The Web would, almost overnight, become a much more difficult to use and less efficient place.
However, a recent lawsuit filed by GateHouse media has asked new questions about deep linking and its possible legal implications.
Though the lawsuit is clearly misguided in some ways, including the claim that the site loses advertising over deep linking, it is worth taking a quick moment to look at some of the potential legal hazards that come with deep linking and how to avoid them.
Most of the copyright issues raised by deep linking come not from linking to sites, but from linking to material that is known to be infringing. Though the DMCA has protections for “information location tools” that unwittingly link to infringing material, the protections for knowingly doing so are much less clear and, as the Project Playlist case reveals, there is much to be sorted out.
That being said, there have been cases where linking directly to content has been ruled an infringement. For example a court in Texas ruled that a Supercross site had infringed another company’s rights by linking directly to their streams. This was a very controversial ruling, in particular with William Patry, the senior copyright council for Google, who called it a “deeply disturbing” case on his personal blog.
The good news is that, generally speaking, it is almost unheard for for one to be successfully sued for simply linking to a Web page and fair use, in the United States at least, even protects your right to sample the article that you are linking to.
However, deep linking copyright disputes have been increasingly rare on the Web. The reason is that most people are simply happy to have others link to their site, which specific page being an unimportant question. When one looks at the deep linking controversies of the past, it is clear that most are well over 5 years old.
For the most part, linking to a site, whether the home page or the article, is not likely to raise any copyright issues. Linking to media files are trickier, especially if you know that it is likely infringing, but also not immediate grounds for a copyright claim.
However, there are other issues that may create problems.
Hotlinking images and media without permission is well known to be bad form on the Web. Most people think that it is disrespectful to use other people’s bandwidth to fill their own pages. That being said, whether it raises any legal issues is tough to say. Since no copy of the image or file is made nor is it made any more “public”, there is not likely a copyright infringement and other legal analogies, such as trespass to chattels, have flaws (PDF).
The real problem with hotlinking images and media files, from a practical standpoint, is that the owner of the site can remove, move or alter the files as they wish. This can cause dead links or, as the McCain campaign found, a great deal of embarrassment.
It is best to avoid hotlinking images, not only is it bad form and upsets many Webmasters, it is unreliable and causes tension. For the most part, you’ll be better off putting the content on a server that you control.
One of the more difficult and confusing issues with linking is trademark law. Trademark law is designed to prevent confusion in the marketplace and give people control over their business and product identities. This includes names, logos and other things that can make up a trademark.
The problem with linking is that it is a violation of trademark law to claim that you have a relationship with a person or company when you do not. If you link in such a way that it appears the party involves endorses your site or is a part of it, there could be problems.
Linking to an article in a blog entry is not likely to raise any eyebrows but, as with the GateHouse case above, if you keep an ongoing stream based on the site’s RSS feed and make it appear that their content it part of your site, there is a chance that others might become confused and that could upset trademark holders.
This problem is compounded by a desire by most sites to be clearly attributed for their contributions and other sites that even create widgets designed to encourage just such linking.
To make matters worse, what would and would not constitute copyright infringement is defined very much on a case-by-case basis. Though few trademark holders seem to object to their name being associated with their links, it is a rare situation where providing attribution and doing what is right can cause legal trouble, especially when dealing with an overzealous rightsholder.
The best thing you can do is attribute correctly but make it clear what content is and is not directly affiliated with your site. It may not provide perfect protection, but it is important to keep in mind, especially if you want to make links to someone’s sites a regular part of your presence.
For the most part, deep linking is safe. Linking directly to stories and even most media, including images, is perfectly fine. However, embedding such images into your site directly without permission can lead to other issues, especially if the site you’re hotlinking from does not appreciate your use.
Sites such as Google, Digg and Reddit provide excellent guides for linking policies and for using content in a fair way when providing such links.
For the most part, these sites have held up very well both in the court of public opinion and in the court of law. Though they certainly shouldn’t be treated as the final authority in this area, they are good starting points for anyone seeking out a role model.
In the end, the good news is that deep linking itself is not in any serious trouble. The legal questions from the beginning of the decade have not been resurrected. Instead, most of the issues we’re looking at now deal with specific cases that deal with relatively new issues.
Hopefully they will be resolved soon, making way for even greater clarity in this area.