What protection do we have for links?

Wired on Friday: What's in a link? You could argue the web - and, by extension, the net's - popularity has come from those simple…

Wired on Friday: What's in a link? You could argue the web - and, by extension, the net's - popularity has come from those simple blue highlighted words, which when you click seamlessly take you from one site to the next, writes Danny O'Brien

The hyperlink is the stitching that holds the web together, and makes a network of millions of computers feel like an approachable corpora of knowledge. But what a link really signifies - in business culture, in licensing contracts, and in law - is still greatly in flux.

Take a look, sometime, at your own corporate website, or that of a competitor. Buried at the bottom, next to the privacy policy and the slightly overlarge credit to the web design company, may well be some sort of "website user agreement". Click on that link, and you'll often find a demand that other websites get permission in writing before they link to the website.

It's a clause from the Stone Age of the web, probably copied and pasted from another clueless website's own contract. It's a largely unenforceable demand: creating a link to a site is simple, largely undetectable to a non-tech savvy site owner, and unstoppable. But worse, it's practically a suicide note for the site itself. Links toward a site are what makes a site popular, and read. Unless someone physically types a website address into their browser (a vanishingly rare occurrence), incoming links are the only way to reach most websites, outside search engines. And most post-Google search engines use incoming links as a rough guide to how important, and high-ranking a site should be. A site that encourages other web users to link to it climbs the top of the search engine listings.

READ MORE

But what about outgoing links? Clearly, if you run a website, you can have complete control over the content. For many new immigrants to the net, putting links to other people's sites has been an anathema - especially if those other sites are your competition. If you're running a site paid for by the New York Times why would you link to a competing paper, like the New York Post?

It's a subtler lesson, but sites that link elsewhere are the popular sites. Why? Because they're a stopping off point, a beginning to your visits around the web, or a guide. Think of Yahoo or Google. How many of the links it offers are to itself, and how many to other sites.

A site that links only to itself is a dead end, no matter how much hand-rubbing marketing executives demand that no competition should be brooked. There's an exception to this - social networking sites like MySpace.com and Bebo, where links stay within your circuit of friends, and on the same site. Perhaps that will be their downfall, however. Commenters have already noted how fickle social-networking site users are.

Maybe that fickleness is accelerated by those sites attempting the cheapest way of holding onto customers, by not letting them wander on their own. There's another reason for linking that has less to do with users travelling the web and more to do with the internal culture of respect online.

Because incoming links are so valuable, even the most selfless blogger can be driven to fury by others not crediting them as a source - and linking to them. That pressure feels like a transitional phenomenon; "cultures of respect" have never lasted long online, no matter how hard dedicated communities attempt to make them scale from dozens to millions of creators. And the parts of the net that profit most from incoming links often rely on the free flow and incremental, communal development of ideas to such an extent that it is often impossible to locate an "original" source in the first place.

Which brings the legal world back into the picture. In the Stone Age, a notorious British court case, Shetland Times v Wills, a judge made the judgment that external sites could not "deep" link into individual websites, without linking to the original homepage, and that a defence that the linking site was providing a benefit to the linked site by giving it greater exposure was immaterial.

In the US, a law that attempts to prohibit American use of online poker sites contains a cautious provision which allows links to poker sites removed from ISPs or websites. The law treads cautiously because of the recognition in the United States that censoring links may violate the First Amendment.

Nonetheless, laws like this in such a free-speech friendly jurisdiction don't bode well for businesses elsewhere. Last week, in Greece, a website owner was arrested when his site provided an extract and link to content that the government deemed illegal - even though the content and link were unrelated to his site, and automatically generated. Just as fewer sites demand written permission for incoming links, such extremes of law should fade over time. But if the wheels of law grind slowly, intimate knowledge of what a link is for, and why it should be protected, spreads even slower. Maybe we'll only see sense when the Chief Justice is blogging, and demanding his own hyperlinked credits on the latest judgments.

Danny O'Brien is activism co-ordinator at the Electronic Frontier Foundation