It is a truism that slow-moving law cannot keep up with fleet-of-foot digital technology, so that makes the rare court decision dealing with the details of how people use the Web of particular importance. Here's an interesting case that has just been handed down.
Reading the entire judgment (I did so you don't have to), I don't think the judge did a bad job. The big problem is simply that copyright law just cannot cope with what is going on routinely a billion times a day.
The case was about a company indexing content from a group of newspaper publishers. The former had a licence from the latter, so that was not in question. But the issue was whether the customers of the indexing company needed a licence from the newspaper publishers to view the following that was sent to them:
A hyperlink to each relevant article ("the Link"). The Link is a citation of the headline ("the headline") from the article. A click on the Link takes the customer through to the article as it appears on the Publisher's website.
The opening words of the article after the headline ("the opening text").
An extract from the article ("the hit extract") showing the context in which the agent appears. This will reproduce the agent and some words immediately preceding and following it.
The total extent of all the copyright material per article did not exceed 256 characters "including spaces".
In the end, the judge decided that customers of the indexing service did indeed need their own licence to view even those limited, 256-character items.
To arrive at this decision, the judge seemed much exercised by the fact that to view those items, a copy of them had to be made. This is, of course, not a bug but a feature: it's how the Web works. Then the judge considered whether fair dealing covered that copy. You might have thought that would be a no-brainer – after all, 256 characters is only about 40 words, which is tiny by blogging standards. But amazingly, the judge decided that it was not covered by fair dealing – in part because the indexed items didn't explicitly include the author's name. As a result, she found for the newspaper publishers, and that end-users of the indexing service needed to have a licence from the former to view the snippets.
What is particularly depressing about this decision – however correct it may be from a legal viewpoint (and IANAL) - is that one consequence of the indexing service is to encourage visits to the newspapers' sites to view the relevant articles in their entirety. That is, it would drive traffic to the publishers. I doubt whether all of the end users are going to take out paid-for licences from the newspaper group to do that, so the net effect of "winning" in this way is likely to be less traffic, and less visibility for their articles.
The problem is quite simple: copyright is an 18th-century intellectual monopoly that was designed to promote the creation of books in a world where making copies was hard, and likely to be carried out intentionally with the aim of defrauding the copyright holders. Trying to apply it to the digital world where copies are not just easy to make but the inevitable consequence of content travelling around the Net and being viewed on end-user devices is just folly. Until copyright is brought into the 21st century – and tinkering at the margins does not count – we can expect more of these court cases with counterproductive outcomes.