Irish newspapers created quite a stir when they demanded a fee for incoming links to their content. Actually, this is a mere prelude to a much more crucial debate on copyrights, robotic scraping and subsequent synthetic content re-creation from scraps.
The controversy erupted on December 30th, when an attorney from the Irish law firm McGarr Solicitors exposed the case of one of its client, the Women’s Aid organization, being asked to pay a fee to Irish newspapers for each link they send to them. The main quote from McGarr’s post:
They wrote to Women’s Aid, (amongst others) who became our clients when they received letters, emails and phone calls asserting that they needed to buy a licence because they had linked to articles in newspapers carrying positive stories about their fundraising efforts.
These are the prices for linking they were supplied with:
1 – 5 €300.00
6 – 10 €500.00
11 – 15 €700.00
16 – 25 €950.00
26 – 50 €1,350.00
50 + Negotiable
They were quite clear in their demands. They told Women’s Aid “a licence is required to link directly to an online article even without uploading any of the content directly onto your own website.”
Recap: The Newspapers’ agent demanded an annual payment from a women’s domestic violence charity because they said they owned copyright in a link to the newspapers’ public website.
Needless to say, the twittersphere, the blogosphere and, by and large, every self-proclaimed cyber moral authority, reacted in anger to Irish newspapers’ demands that go against common sense as well as against the most basic business judgement.
But on closer examination, the Irish dead tree media (soon to be dead for good if they stay on that path) is just the tip of the iceberg for an industry facing issues that go well beyond its reluctance to the culture of web links.
Try googling the following French legalese: “A défaut d’autorisation, un tel lien pourra être considéré comme constitutif du délit de contrefaçon”. (It means any unauthorized incoming link to a site will be seen as a copyright infringement.) This search get dozens of responses. OK, most come from large consumers brands (carmakers, food industry, cosmetics) who don’t want a link attached to an unflattering term sending the reader to their product description… Imagine lemon linked to a car brand.
Until recently, you couldn’t find many media companies invoking such a no-link policy. Only large TV networks such as TF1 or M6 warn that any incoming link is subject to a written approval.
In reality, except for obvious libel, no-links policies are rarely enforced. M6 Television even lost a court case against a third party website that was deep-linking to its catch-up programs. As for the Irish newspapers, despite their dumb rate card for links, they claimed to be open to “arrangements” (in the ill-chosen case of a non-profit organization fighting violence against women, flexibility sounds like a good idea.)
Having said that, such posture reflects a key fact: Traditional media, newspapers or broadcast media, send contradictory messages when it comes to links that are simply not part of their original culture.
The position paper of the National Newspapers of Ireland association’s deserves a closer look (PDF here). It actually contains a set of concepts that resonate with the position defended by the European press in its current dispute with Google (see background story in the NYTimes); here are a few:
— It is the view of NNI that a link to copyright material does constitute infringement of copyright, and would be so found by the Courts.
— [NNI then refers to a decision of the UK court of Appeal in a case involving Meltwater Holding BV, a company specialized in media monitoring], that upheld the findings of the High Court which findings included:
– that headlines are capable of being independent literary works and so copying just a headline can infringe copyright
– that text extracts (headline plus opening sentence plus “hit” sentence) can be substantial enough to benefit from copyright protection
– that an end user client who receives a paid for monitoring report of search results (incorporating a headline, text extract and/or link, is very likely to infringe copyright unless they have a licence from the Newspaper Licencing Agency or directly from a publisher.
— NNI proposes that, in fact, any amendment to the existing copyright legislation with regard to deep-linking should specifically provide that deep-linking to content protected by copyright without respect for the linked website’s terms and conditions of use and without regard for the publisher’s legitimate commercial interest in protecting its own copyright is unlawful.
Let’s face it, most publishers I know would not disagree with the basis of such statements. In the many jurisdictions where a journalist’s most mundane work is protected by copyright laws, what can be seen as acceptable in terms of linking policy?
The answer seems to revolve around matters of purpose and volume.
To put it another way, if a link serves as a kind of helper or reference, publishers will likely tolerate it. (In due fairness, NNI explicitly “accepts that linking for personal use is a part of how individuals communicate online and has no issue with that” — even if the notion of “personal use” is pretty vague.) Now, if the purpose is commercial and if linking is aimed at generating traffic, NNI raises the red flag (even though legal grounds are rather brittle.) Hence the particular Google case that also carries a notion of volume as the search engine claims to harvest thousands of sources for its Google News service.
There is a catch. The case raised by NNI and its putative followers is weakened by a major contradiction: everywhere, Ireland included, news websites invest a great deal of resources in order to achieve the highest possible rank in Google News. Unless specific laws are voted (German lawmakers are working on such a bill), attorneys will have hard time invoking copyright infringements that in fact stem for the very Search Engine Optimization tactics publishers encourage.
But there might be more at stake. For news organizations, the future carries obvious threats that require urgent consideration: In coming years, we’ll see great progress — so to speak — in automated content production systems. With or without link permissions, algorithmic content generators will be able (in fact: are) to scrap sites’original articles, aggregate and reprocess those into seemingly original content, without any mention, quotation, links, or reference of any kind. What awaits the news industry is much more complex than dealing with links from an aggregator.
It boils down to this: The legal debate on linking as copyright infringement will soon be obsolete. The real question will emerge as a much more complex one: Should a news site protect itself from being “read” by a robot? The consequences for doing so are stark: except for a small cohort of loyal readers, the site would purely and simply vanish from cyberspace… Conversely, by staying open to searches, the site exposes itself to forms of automated and stealthy depletion that will be virtually impossible to combat. Is the situation binary — allowing “bots” or not — or is there middle ground? That’s a fascinating playground for lawyers and techies, for parsers of words and bits.