> Europe's highest court is considering whether every hyperlink in a Web page should be checked for potentially linking to material that infringes copyright, before it can be used.

A hyperlink is just an address of some kind that client software interprets in a certain way allowing users to easily send network packets to it. Whatever response comes back to a request directed at that address is entirely determined by whoever controls the address, and outside the control of the publisher of the hyperlink.

So if one published, say, a phone book, with all of the addresses and phone numbers of consenting residents in a city, when a reader dials one of those numbers, or sends a letter to one of those addresses, it's entirely possible that a reply could come back containing copyrighted material. Should the publisher be required to check every address and phone number for such copyright-violating replies before publishing the phone book?

The response that the controller of an address sends (if any) to requests sent to it may change from request to request, for any reason at all. Would it be illegal to publish a hyperlink to an HTTP URL that has a 50% probability of responding with copyrighted material? 10% probability? 1%?