Hacker News new | past | comments | ask | show | jobs | submit login
Internet hyperlinks do not infringe copyright, EU court advised (reuters.com)
523 points by jonbaer on Apr 7, 2016 | hide | past | web | favorite | 107 comments

The 90s are back! Clinton is running for president, and the legality hyperlinks are up for debate again.

In 1999, a similar lawsuit was brought up against 2600 Magazine over the issue of linking to source code that could have the potential to violate copyright:



And don't forget the renewed crypto wars ...

..and bubble burst in a few years?

HN's Law: The deeper the comment chain, the higher the probability someone talks about bubble burst

Might as well call it Weirdy's Law.

I'll take it!

I mean, the deeper the comment chain, the higher the probability of someone talking about X, where X can be literally any topic.

That sounds rather optimistic. You really think it'll be a few years?

Maybe today, the average judge has a basic understanding of what a hyperlink is.

Everything old is new again.

Everything new is old already.

"Plus ça change..."

The older you get, the more you realize just how true this is.

It's the 90s on the European internet in other ways, or almost.

For instance broadband penetration in Belgium is about where it was in 2003. There are lots of internet cafes.

Edit: This is incorrect. European broadband is fine.

I think you are reading the wrong statistics. What has stagnated in Belgium since 2003 is the percentage of broadband connections that are ADSL. Between 2003 and 2008 alone broadband penetration more than doubled.

ADSL is not broadband it is fraudband.

Not when its faster than broadband :)

/dev/null 21%[======> ] 301,70M 15,8MB/s eta 1m 40s


This is NOT a court ruling and NOT law. Yet.

Note the word 'advised'. This is a legal opinion from an important individual who looks at the case and tells the court what he thinks. But (i) the court can ignore it or look at the case themselves and make their own opinion; and (ii) any ruling they do make does not bind any EU government to implement it (though it should strongly decisions in EU countries). It's like asking your mechanic friend what he thinks of the car you've got lined up to buy.

As the CJEU press release states:

> NOTE: The Advocate General’s Opinion is not binding on the Court of Justice. It is the role of the Advocates General to propose to the Court, in complete independence, a legal solution to the cases for which they are responsible. The Judges of the Court are now beginning their deliberations in this case. Judgment will be given at a later date.


FWIW my wife puts the odds of victory on this at something like 70% at this point. She's a director at Stanford's Center for Internet and Society, working on freedom of expression.

That's based on this study:


You're correct, but the practice is that the advice of the Advocate General is followed by the Court of Justice. So normally it's just a matter of time.

The same page says that it would be detrimental to the Internet if hyperlinks were fallen under such restrictive laws

And since it's not ruled, there's still time for bribes to judges.

Off-topic: anyone else find the article picture really odd? (who uses a laptop that way?)


The caption is kind of amusing with it's literal-ness and complete irrelevance to the story:

> People are silhouetted as they pose with laptops in front of a screen projected with binary code and a Central Inteligence Agency (CIA) emblem, in this picture illustration taken in Zenica October 29, 2014.

It is; I think it may have been intended as alt-text rather than a caption. It's precisely the sort of description that you'd give to a blind person to describe "here's what this image is, in case you can't see it."

> who uses a laptop that way?

I didn't notice before, but now you mention it, it _is_ really strange that none of them are pointing or fake-laughing at the screen.

I don't understand why they're not wearing balaclavas. Don't they know we can see who they are now?!

>>who uses a laptop that way?

Steve Jobs. http://i.imgur.com/cMjKysz.jpg


I did that once and my brand new laptop allmost fell on the floor. Cannot recommend.

I've used mine in a similar fashion on many occasions. However, I have more of a death-grip on it, where I rest it over my forearm horizontally, and grab the right side of the base with my left hand.

Don't try that with your Surface Pro!

People with no tables or chairs?

If I was in that situation, I'd sit on the floor.

And if the floor was really dirty?

I wouldn't agree to pose for pictures.


L33t hackers at the CIA, duh!

I wonder how this would have affected the UK TVShack.net case [0]. Richard O'Dwyer was subjected to the absolutely awful UK-US extradition agreements over linking to copyrighted content.

[0] https://en.wikipedia.org/wiki/Richard_O%27Dwyer

This comes at the right time, because the EU Commission was just preparing to put "copyright infringement by hyperlinking" into law:


France was also planning to ban "hyperlinking without permission":


Of course, an opinion that hyperlinks aren't currently copyright infringement by themselves could be seen by some as evidence that a new law is actually necessary.

Or a sign that European nations need to stop propping up an obsolete business model with internet breaking laws.

Seems absurd to legislate this when there's a simple technical solution. If you don't want to allow "deep links" on a public web site, simply block any request where the HTTP referrer isn't from your domain (or from domains whose owners have paid you for linking rights, if you can find anyone inclined to do so.) You could go even further by encoding a base64-encoded expiration date + HMAC into the URL, so that your deep URLs can't be saved at all.

I think doing either of these things is silly - why even run a public web site if you're going to do this? But it's their site, and their right to make bad decisions with it.

If a site owner chooses not to do this, it's safe to assume they intended to allow deep links, since links are a fundamental, well-known feature of the medium they chose to use.

I don't think that's what this is about. It's about linking to content that violates copyright laws, e.g. running a torrent tracker.

> The case arose in the Netherlands where the GeenStijl website had provided a link to an Australian site showing pictures of a Dutch celebrity taken by Playboy magazine. The Australian site did not have Playboy's consent to do so.

If a site owner chooses not to do this, it's safe to assume they intended to allow deep links, [..]

Or it could mean that the site owner doesn't have the technical expertise to block deep links in the manner you describe. I agree that it's silly to block deep linking but there are people that want to do so and a lot of those people don't know how (other than not publishing those links).

The correct answer is to hire the expertise to get it done. Can't blame the neighbors for looking at your lawn if setting up a high fence is easy and you chose not to.

I think the better analogy is you can't be upset at people viewing you nude when you're walking around in public without clothes.

My own basic point is this:

If the behavior that somebody finds to be unwanted is trivial to protect yourself from, then only in very special circumstances it should possible to legally ban it. The easier the defense is, the higher the bar for getting a ban through. (Besides the usual need to prove harm.)

Because the law should only get involved when you need a strong third party to get involved for protection, preserving order and to arbitrate.

Legality aside, it just seems counter intuitive in the first place.

If you're putting content on a publicly facing server with web routes open to all (i.e. walking nude in public), you shouldn't be surprised when people look.

Lots of people block referral headers. Browsers also drop them across HTTPS->HTTP boundaries.

The expiration date+HMAC is still an option. Plus, the number of users blocking referrer headers is probably small enough that a site owner doing this wouldn't care. And browsers dropping the header from HTTPS -> HTTP isn't an issue if you only want to allow access from your own domain. (If you want to authorize third-party sites that use HTTPS, you'd just have to use HTTPS yourself.)

> Lots of people block referral headers

Source? I would think most people run stock browsers that don't manipulate referrer headers.

Just take a look at the amount of times https://chrome.google.com/webstore/detail/referer-control/hn... has been installed. And that's just one extension for one browser.

100K users out of 3B+ internet users [1]? I don't think that qualifies as 'lots of people'.

[1]: http://www.internetlivestats.com/internet-users/

100K users of a single extension for a single browser. I'll leave it to you to get the numbers for all other extensions for all other browsers. And don't forget that uBlock can do this as well.

They're not as popular as they used to be, but many popular security packages (ie Norton Internet Security I believe) also blank all outgoing referrers.

> If a site owner chooses not to do this, it's safe to assume they intended to allow deep links

That's not how the law works, though; you don't have to make it impossible for someone to infringe on your intellectual property for it to have protection.

That's not to argue that there is any IP law protection here regardless.

I saw a video from EU parliament where Gunther Oettinger urged that paper pritning news industry will die thanks to hyperlinking and so we, the EU, must do something about it (ratify this law). I think it's pretty clear where his second wage is coming from. Or he's absolutely stupid and doesn't have a clue about economy and prosperity.

> I think it's pretty clear where his second wage is coming from. Or he's absolutely stupid and doesn't have a clue about economy and prosperity.

Look into his relationship with lobbies and consumer rights' organisations some time. Also, his speech about the Taliban-like evils of net neutrality (yes, really).

EDIT: And if you understand German, there's even a great remix: https://www.youtube.com/watch?v=fkORC6FcSH4

Ah yes, Oettigner, EU commissioner for digital economy and society, no less. From the country that brought you greatest hits like "Leistungsschutzrecht".

    A link to a website which publishes photos without authorisation of the author does not in itself constitute a copyright infringement
Sorry for my naivety here but the way this is worded is creating confusion for me. Does the article mean:

1) Linking to photos that you don't have authorisation to use on your website isn't copyright infringement.

2) Linking to a website that doesn't seek authorisation to use photos on their website.

To me it reads like the latter, in which case how is one meant to know if a website has sought and gotten authorisation for each photo they use?

It's the latter. The specific case that prompted this general inquiry was of an Australian website posting a copyright infringing photo, and a Dutch website linking to that photo. Assuming it's given that the Australian site is infringing copyright, the question being considered is: does the Dutch site also then infringe the copyright?

If the court rules that yes, the Dutch site also infringes copyright, it basically breaks how the internet works. You are liable for copyright infringement on any content linked to from your site. Imagine a forum, for instance..

>"Hyperlinks which lead, even directly, to protected works are not 'making them available' to the public when they are already freely accessible on another website, and only serve to facilitate their discovery," the opinion said.

First case seems to be fine as well. Unless of course you're hosting the images yourself, then you need to get permission.

It sounded to me like the latter as well. I think the Advocate General is saying you can't ban hyperlinks unless they are the only way to access a copyrighted work.

As an example, I would think a link that allowed someone to get around the requirements to login to view something (maybe username/password as query parameters or something?) would be considered illegal.

There was also the comment about being freely available already. So a page behind username / password would not be freely available and you would be breaking the law.

Now, if a user on a private page, marked something as public, but did not publish the link, then it would be legal as you could argue that its freely available.*

* Not a lawyer, so just my own understanding.

I read it more that if the GreenStijl website which was in the EU jurisdiction was posting the links to the pages hosted on a computer within another jurisdiction, but that the pages were not generally available (ie they weren't being hosted on an actual website where they were part of that that website) then GreenStijl would be liable for copyright infringement because the only way to access the published content would be through them.

so GreenStijl can't just host the pages anonymously on mega.com and link to them.. in that case they would deemed to be part of their website.

Oh thank god I'll be able to link to Irish news websites again without fear of prosecution!


Reminds of when Spain tried something similar: http://arstechnica.com/tech-policy/2015/07/new-study-shows-s...

However in this case I believe it centered on snippets of the news content itself, not only a link.

In Spain they illegalised links to infringing content (mostly movie/music downloads) by creating a "Commission of experts" which would rule on takedowns. The reason for this, of course, is that judges and courts were deeming hyperlinks were legal.

This is El Pais singing praises: http://elpais.com/elpais/2012/05/01/inenglish/1335894807_391...

Techdirt is second-hand reporting, but fair: https://www.techdirt.com/articles/20120104/04252517273/spani...

If you can read Spanish, google "Sinde-Wert jueces". This one is good: http://www.eldiario.es/zonacritica/ComisionSinde-jueces_6_16... (its author is David Bravo, a copyright lawyer who's recently been elected to Spanish Parliament, and one of the most vocal personalities on copyright in Spain).

Thanks for the good info. I can read Spanish, but too rusty at this point to be reading about law.

What if I put base64 encoded content into the hyperlink? Will it be legal still? :)

While I am not a lawyer, I suspect that a reasonable person would say that encoding the content in a string turns the link into the content rather than leading one to the content.

Perhaps you should read the first paragraph of the article to receive clarification on this :)

Try magnet links instead. The opinion only really covers pointers.

It was a joke, magnet links are base64 encoded hashes of the content iirc.

I wonder if this will have an impact on DMCA takedowns against search engines?

So I guess card catalogs in libraries are safe as well.

I honestly can't believe this is even being debated.

So the pirate bay is legal?

This was my first thought, however there was this at the bottom of the article:

  The advocate general added that this also depended on whether the link was indispensable in making the photos available, a matter which he referred back to the local Dutch court.
I believe the argument is that while The Pirate Bay only links to copyrighted content, they are integral to making that content available (since very few sites have these link listings).

If widely available search engines indexed magnet links then I think TPB would be legal under this opinion.

There are widely available search engines indexing magnet links; it's just that they're all other torrent sites. Each individual site isn't integral to making that content available, but the class "torrent site", taken as a whole, may be. I'm not sure how the interpretation would work here - "would be integral if it were the only extant illegal source"?

So, if Google is the only site that has spidered my content and my content is infringing then Google is also infringing?

Obviously Google are too rich for the argument to work against them, but say a popular blogger finds a website with little traffic and some great content and links to it; they then would seemingly become responsible for the original site's infringement?

One can see why the media lobby would want to leave in the option to declare TPB illegal but it seems an unworkable reservation if we're to hold to the rule of law. Kinda like (stealing someone else's analogy up-thread) if you charge someone with public indecency for pointing out there was a person being indecent - inappropriate nudity, say - if not many people had noticed the indecency prior to the declaration. Seems pretty mad to me.

I'm not sure TPB is what they're thinking about. I think it's closing the loophole where EU sites could host all their infringing content under another jurisdiction which doesn't care about copyright, and then just linking to it.

> they are integral to making that content available (since very few sites have these link listings).

What is "very few" in your opinion? There are dozens, even hundreds of torrent sites out there. They often also meta-index data from each other, serving as a sort of ad hoc replication.

It was morally always "legal". The paid inquisition may have made it illegal (and looks imoral), but things should get back to the rightful state of legality sometime in the future.

In its original jurisdiction of Sweden, yes.

Yes it is. Always was.

For freedom loving countries at least.

Given the legal uncertainty introduced by the new Spanish law, I think that is not worth the risk to put links to spanish webs anymore in the 90% of the cases. Definitely, not more free linking to news on online journals related with AEDE by now.

Unless absolutely necessary, I now give just a couple of clues and let people find the page for themselves in google if they want. Stupid measures for stupid times...

Just my opinion.

Things like this are why I often wonder why people hail court verdicts so much. We've clearly established that courts are rarely even close to logically consistent, yet a court ruling in favor of or against an issue being discussed is often hailed as absolute truth.

I don't understand how a link could ever be judged to infringe copyright. No copy is made, any more than if I tell my friends, in person, to go and check out the cool, copyright-infringing photo that one of us have.

The text of the link itself could contain copyright material. Google says the practical limit 10 years ago was ~100Kb, not sure now.

This doesn't really go into hotlinking, does it?

What's the legality of hotlinking? = embedding an image in your page by using an URL directly to an image that is hosted & controlled by someone else.

No, not hotlinking. I don't think hotlinking inherently could break a law, I think that would require some understanding that the host provides the content for one particular use and any other use would be a violation. There probably are sites with license agreements/terms of service that state as much in which case hotlinking would be breaking the contract. If ad blocking were illegal, it would be for the same reason.

Hotlinking could be involved in criminal or civil actions, e.g. used to perpetrate fraud or used maliciously (intended to use up host's resources or even DDOS).

I'm no lawyer, but from an end user's perspective if you hotlinked an unauthorized image instead of linking to a page containing the image, then you aren't just "pointing" to the image anymore, you're practically displaying the image on your website.

The content you have merely contains a link to the provided resource. The user can fetch your page without violating the law, the user can also legally fetch the image from the owner. So neither you nor your user have committed a crime.

You could try to argue that the user requires a license to combine the 2 in ways not considered by you however this has several problems.

- Its laughably unenforceable

- In the us it would probably be fair use

- It would make literally the entire internet illegal

- It would make adblocking on your end illegal

- It would make you a world wide source of negativity and scorn.

Basically there is a technical solution to most hotlinking issues and every imaginable legal solution would be a cure worse than the disease.

The Ninth Circuit says hotlinking is not infringing: https://en.wikipedia.org/wiki/Perfect_10,_Inc._v._Amazon.com....

Thank you for a very relevant link, yet (as legal things go) this case has its special setting. They were using thumbnails, and they were not pretending or that the image belonged to them or their own content.

If I write a news article, and I hotlink in an illustration with no credit or comment, that still seems problematic.

I'm not sure the distintion is relevant. In the decision[1] which affirmed the lower court's, the judges applied the "server test"; essentially, they've said it's never copyright infringement if you don't actually distribute it yourself, but merely tell the browser where it can get the image.

This excerpt is particularly relevant:

  Perfect  10  argues  that  Google  displays  a  copy  of  the  full-
  size  images  by  framing  the  full-size  images,  which  gives  the
  impression  that  Google  is  showing  the  image  within  a  single
  Google  webpage.  While  in-line  linking  and  framing  may
  cause some computer users to believe they are viewing a sin-
  gle  Google  webpage,  the  Copyright  Act,  unlike  the  Trade-
  mark Act, does not protect a copyright holder against acts that
  cause  consumer  confusion.

So a URL of JPEG file put in 'img' tag is infringing and the same URL put in 'a' tag is not? Sounds rather silly to me.

hotlinking gives the impression the image is part of your page.

It is part of your page. Even if you link to it. It's also part of your page if it's just a comment in the source.

link to the CJEU press release, with details (PDF) http://curia.europa.eu/jcms/upload/docs/application/pdf/2016...

This is at least reassuring. I don't think it's a big leap from saying that a link to a site with infringing content is itself an infringing act to saying a link to a site that links to a site [... etc] is itself infringing.

So does such an advise also include things like magnet links, which do not include the content, or its location, but instead a description of the content by means of a hash or checksum?

"If I have seen further it is by standing on the shoulders of giants" --Isaac Newton

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact