Sean has obtained for us the full text of the Brussels Court of First Instance February 13th ruling, the original one being appealed by Google, in French [PDF], and the official English translation [PDF]. I'll present it in plain text here as well. He also obtained permission from the BBC for Groklaw to let you hear a discussion [Ogg] from last March on the BBC including Google's CEO Eric Schmidt and Margaret Boribon of Copiepresse. But I'll let Sean tell you about it himself.

Mainly I'll let him tell it, because I find the ruling infuriatingly clueless about the tech. The "expert" advising the court obviously did not understand Google cache. Nor does the court comprehend that a robot can't do anything that a human hasn't told it to do. Nor does the court understand the difference between an information portal and a bunch of search engine results, which is all Google News is.

If I were Google, frankly I'd simply exclude the lot of them, and see how they like it. I say obviously because now Copiepresse members are simply putting the noarchive tag in their web pages which they could have done prior to litigation and instead of suing over cached pages, and it is an option for anyone who wishes to avoid Google cache, an option that the "expert" should have told the court about.

Did he? There is no mention of any such information, and Copiepresse significantly refuses to release the "expert" report. You will see his report referenced several times, and if you are like me, you will in each instance find yourself saying, "But that's not technically accurate." With such a report to rely on, how could the court reach a correct conclusion?

I try to be patient, because that is after all why I started Groklaw, to explain tech to legal folks and vice versa. But to me, to waste everyone's time and money going to court about caching, when all you have to do is put a noarchive tag in your HTML is ridiculous. If you don't wish to be included in Google at all, just use the robots.txt and Google won't include you. How hard is that? So I view the litigation here more like this: when you become successful, people sue you.

But this case challenges more than Google. It challenges the Internet itself. If the law as set forth by the judge here were to stand, it really is the end of the Internet as a functioning entity. The fundamental problem, as I see it, is that the Internet has more value than any single piece of copyrighted writing. So that means that the public has a stake in this, not just the Copiepresse authors, who all have self-help mechanisms to protect them if they don't want to be cited or referenced.

Similarly, everyone has the option to simply not allow Google to enter the premises at all by using the robots.txt option. We do that here on Groklaw and have for years, although we let them on certain pages. All authors have that choice. If you wish to be included, you can. If you don't wish to be included, you can avoid it. Of course, if what you want is to be paid, neither option will thrill you or satisfy you.

Computers copy. It's what they do. If you apply copyright laws, as written for an earlier age, strictly to the Internet, you have to shut it down, except for those willing to go by another arrangement, such as Creative Commons licenses. Now, folks can disagree about tradeoffs in such a result. But to me the Internet has value. I think of those children now receiving their laptops from OLPC. I think about all I have learned personally and can still learn from being able to use search engines like Google. And it pains me to read this ruling, because it attacks the Internet itself, not Google alone, and that means it is threatening something that is precious.

*****************************

Update on Copiepresse-Google
~by Sean Daly

After the court's ruling mostly confirmed the previous default judgment that had been handed down prior to Google's participation, Google appealed, but they also opened talks with Copiepresse, as you can see in this (joint press release [PDF]) and reportedly asked the judge at a hearing last week for more time for talks. Shortly before the ruling, Copiepresse revamped their website, in French, with litigation pages describing their procedures against Google, Yahoo and Microsoft MSN. The homepage of their site has since been updated with news of the Google litigation.

The decision, as you will see, refers to two expert reports, one commissioned by Copiepresse and the other by Google, both of which are unavailable. I wrote to the court seeking permission to inspect or publish them, but the chief clerk replied these documents can only be released by the parties to the litigation. Copiepresse declined to make the Golvers report available, unfortunately, citing the costs of having the report done. The ruling relied in part, some might say heavily, upon this expert's findings, and not being able to see what he wrote is a huge gap in our knowledge.

Google responded to the court's ruling, stating in part:

Today we heard that the Belgian court, which last year ruled against us in the Copiepresse case has reaffirmed its original decision. This judgment is clearly disappointing, and we intend to appeal it because we believe that Google.be and Google News are entirely legal and provide great value and critical information to Internet users. However, we are very pleased that the judge agreed Google should be given notice of articles and other material that content owners want removed. As we have in the past, we will honor all requests to remove such materials.

You may remember I interviewed the Secretary General of Copiepresse, Margaret Boribon, last October. Copiepresse is an association of French and some German language daily newspapers in Belgium. When Google opened its Google News service in Belgium in early 2006, Copiepresse, likely encouraged by the ongoing litigation of Agence France Presse against Google -- a case since settled under confidential terms -- asked Google to seek *prior permission* before publishing links and thumbnails to stories, in accordance with Belgian copyright law. This is what the Author's Guild also is demanding. The Copiepresse titles apparently hope to negotiate a deal with Google similar to that of the Associated Press: not pay-for-content per se, which Google is firmly opposed to on practical grounds, but some sort of larger partnership.

Google has been careful to keep the terms of its agreements with AP, the New York Times, and the AFP confidential. Microsoft has also signed its own far-ranging deal with AP.

The Copiepresse lawsuit touched a nerve, placing into question the venerable "robots.txt" tradition which is the de facto Internet standard, although not a formal one, and has been characterized by some as stacking the deck in favor of search engines by imposing an all-or-nothing choice at the page level to web publishers. In Copiepresse's view, robots.txt has no legal basis. Additionally, it claims the jurisdiction of the Belgian courts over any site visible on a browser computer located in Belgium. It also challenges Google's right to adjust search results and propose contextual ads on keyword searches, and attacks Google cache, which Copiepresse claims undermines the ability to offer paid content on the newspapers' websites. There was a US ruling [PDF] in January of 2006 which held that Google cache is fair use, but fair use is a US legal concept.

In an interesting BBC Analysis radio segment aired in March [Ogg, used by permission] featuring Rachel Whetstone and Eric Schmidt of Google, Lawrence Lessig, and others, Mme Boribon said:

[3:24] "We are investing a lot to publish newspapers. We are paying journalists, paying many people to produce that content, and we can't accept that anyone takes that content and reuses it without authorization and without remuneration."

Most would agree that Google is ahead of Microsoft in the "social networking" and "user-generated content" spheres, with its recent acquisition of YouTube. Microsoft's response seems to be to position itself as a champion of copyright, most recently in a widely reported speech. However, while one might reasonably wonder if Microsoft has had a role in the Copiepresse action, in the sense that it backed the SCO litigation, it must also be noted that Copiepresse is continuing its litigation against MSN and Yahoo on the same basis as the Google litigation.

Google seemed confrontational at first during the litigation, refusing to talk to Copiepresse and at one point last November announcing the abandonment of two of the third-party intervenors (SOFAM and the deliciously named SCAM) on the courthouse steps in a surprise move -- changes cited in the ruling below. However, since the ruling, Google seems to have softened their stance by resuming dialogue with Copiepresse and reindexing the newspapers' websites in the main search engine. They have opened a Google News blog to better communicate with publishers and readers. And in particular, they have taken steps to embrace Belgium, naming a Belgian as country director (management was previously based in the Netherlands, overseen from Ireland) and announcing a 250 million euro investment in a huge datacenter sited in the French-speaking Wallonia region.

With that introduction for context, here's the official English translation of the ruling in Copiepresse vs. Google. I have linked to cited jurisprudence and essays where possible (the Belgian documents are in French, the European documents are in English and other languages). Note that I have adapted the pagination and formatting from the French original, and corrected several minor typographical errors in the English translation.

***************************************************

1 (first page)

In. Exp. DS

Ref. no.: 7964

WE, ALBERT II, KING OF THE BELGIANS,

MAKE KNOWN TO ALL PRESENT AND FUTURE:

That the Court of First Instance sitting in Brussels,

Has pronounced the decision of which the text follows:

COPY

Issued to the Party Copiepresse

1

In the case of

The company under American law GOOGLE Inc. with registered office situated at Mountain View, 94043 California, USA, 1600 Amphitheatre Parkway;

original defendant, plaintiff in the opposition, represented by Mes. Erik Valgaeren and Audry Stevenant [address]

Versus

The civil company in the form of a scrl COPIEPRESSE, entered in the BCE under Nr. 0471.612.218, with registered office situated in 1070 Brussels, Boulevard Paepsem, 22;

original plaintiff, defendant in the opposition, represented by Me Bernard Magrez [address]

And in the case of

The scrl Société Multimédia des Auteurs des Arts Visuels (SOFAM), entered in the register of civil companies under Nr. 312 and in the BCE under Nr. 419.415.330, with registered office situated in av. Frans Courtens, 131 in 1030 Brussels;

voluntary third party intervening, represented by Me. Carneroli and Me. Alain Berenboom [address]

2

Versus

The company under American law GOOGLE Inc. with registered office at Mountain View, 94043 California, USA, 1600 Amphitheatre Parkway;

original defendant, plaintiff in the opposition, defendant in the third party intervention, represented by Mes. Erik Valgaeren and Audry Stevenart [address]

And in the case of

1.	The civil association in the form of a cooperative limited liability company SOCIETE DE DROIT D'AUTEUR DES JOURNALISTES (SAJ), entered in the BCE under Nr. 0455.162.008, with registered office situated at avenue Roger Vandendiessche, 36 in 1150 Woluwe-St- Pierre;
2.	The company under French law SOCIETE CIVILE DES AUTEURS MULTIMEDIAS (SCAM), with registered office situated at avenue Vélasquez 5 in 75008 Paris;
3.	The civil company in the form of a cooperative limited liability company ASSUCOPIE, entered in the BCE under Nr. 0466.710.748, with registered office situated at rue Charles Dubois, 4/3 at 1342 Ottignies- Louvain-la-Neuve;

voluntary third parties intervening, represented by Me Carine Doutrelepont [address]

Versus

The company under American law GOOGLE Inc. with registered office at Mountain View, 94043 California, USA, 1600 Amphitheatre Parkway;

original defendant, plaintiff in the opposition, defendant in the third party intervention, represented by Mes. Erik Valgaeren and Audry Stevenart [address]

3

And in the case of

The limited company PRESSBANKING, entered in the BCE under Nr. 471.483.841, with registered office situated at rue de Birmingham, 131 in 1070 Brussels;

voluntary intervening third party, represented by Me. Bernard Magrez [address]

Versus

The company under American law GOOGLE Inc. with registered office at Mountain View, 94043 California, USA, 1600 Amphitheatre Parkway;

original defendant, plaintiff in the opposition, defendant in the third party intervention, represented by Mes. Erik Valgaeren and Audry Stevenart [address]

***

The submissions and pleadings in this case were made in French at the public hearing of 24 November 2006;

After having deliberated the Chairman of the Court of First Instance pronounced the following judgement:

Considering:

- the summary ruling pronounced on 22 September 2006, and the antecedents of the procedure stated in it;

- the petition in voluntary third party intervention of the SOFAM submitted to the clerk of the court on 6 October 2006;

- the petition in voluntary third party intervention of the SAJ, of the SCAM and the scrl Assucopie submitted to the clerk of the court on 10 October 2006;

- the petition in voluntary third party intervention of the limited company PRESSBANKING submitted on 17 November 2006;

- the submissions of the party Google Inc submitted to the clerk of the court on 30 October 2006 and its additional and summary submissions made on 22 November 2006;

4

- the submissions of the scrl Copiepresse submitted to the clerk of the court on 9 October 2006 and its additional and summary submissions submitted there on 14 November 2006;

- the submissions of the SAJ submitted to the clerk of the court on 13 November 2006, 14 November 2006 and 24 November 2006;

- the submissions of the SOFAM submitted to the clerk of the court on 6 October 2006, its additional and summary submissions made there on 13 November 2006 and its submissions for the withdrawal of a lawsuit submission at the hearing of 24 November 2006;

- the submissions for the withdrawal of a lawsuit of the SCAM submitted at the hearing of 24 November 2006;

Having heard the pleadings of the counsel for the parties;

***

ANTECEDENTS:

The original claim transformed by the company Copiepresse and introduced by summons of 3 August 2006 according to the summary forms through the application of Articles 587, 7° of the judicial code and 87 of the Act of 30 June 1994 relating to copyright law and related rights intending to hear:

- it stated that the activities of Google News and the use of the Google "cache" notably violate the laws relating to copyright and related rights (1994) and on the database (1998),

- to order the company Google to remove all its sites (Google News and "cache" Google under any name whatsoever), all the articles, photographs and graphic representation of the Belgian daily press, French and German speaking represented by the company Copiepresse dated from the notification of the order under penalty of a fine of € 2,000,000 [euros] per day of delay,

- in addition to order the company Google visibly, clearly and without comment on its part to publish the entirety of the judgement to be pronounced for an uninterrupted duration of 20 days from the date of the notification of the ruling under penalty of a fine of € 2,000,000 [euros] per day of delay on the 'google.be' and 'news.google.be' home page;

5

By ruling of 5 September 2006, pronounced by default with regard to the company Google, the summary judge declared the claim admissible and well founded under reserve of developments with regard to the conditions of the measures ordered; That the company Google was in this way ordered:

- to withdraw from all its sites (Google News and "cache" Google under any name whatsoever), all the articles, photographs and graphic representations of Belgian daily press, French and German speaking represented by the company Copiepresse within 10 days from the notification of the ruling under penalty of a fine of € 1,000,000 [euros] per day of delay,

- clearly and without comment on its part to publish the entirety of the judgement to be pronounced for an uninterrupted duration of 5 days within ten days from the date of the notification of the ruling under penalty of a fine of € 500,000 [euros] per day of delay on the 'google.be' and 'news.google.be' home page;

The company Google were served with this ruling on 8 September 2006;

By ruling of 19 September 2006, the company Google made an opposition of the order pronounced on 5 September 2006 requesting:

- In principal, to hear:

To hear all the clauses of the judgement objected against withdrawn,

To pronounce the original claim unfounded and to dismiss the plaintiff,

- In subsidiary order, to hear:

To reform the judgement objected against in relation to the order against Google clearly and without comment on its part to publish the entirety of the judgement to be pronounced for an uninterrupted duration of 5 days within ten days from the date of the notification of the ruling under penalty of a fine of € 500,000 [euros] per day of delay on the 'google.be' and 'news.google.be' home page,

To state in law that this publication is not required to take place;

In an entirely subsidiary order, to change the conditions of the publication or limiting the publication of the entirety of the judgement on the result pages that Google can no longer display in execution of the obligation to withdraw all the articles, photographs and graphic representations of the editors of the Belgian daily press in French and German represented by

6

Copiepresse from all its sites (Google News and "cache" Google under any name whatsoever) and to state in law that this publication should only be required within 30 days from the date on which the decision becomes effective as a judgement;

- From the introductory hearing, if applicable in accordance with Articles 19, paragraph 2 of the judicial code and/or 735 of the judicial code, to suspend the execution of the injunction of the publication until a ruling is pronounced either in a defended action and until the matter chosen has become enforceable;

By ruling pronounced on 22 September 2006, the summary judge stated that there is no cause to grant the claim of the company Google intending to have the execution of the publication injunction suspended and reserved judgement for the remainder;

VOLUNTARY THIRD PARTY INTERVENTION:

By petition in voluntary third party intervention submitted on 6 October 2006, the scrl Société Multimédia des Auteurs et Arts Visuels (SOFAM) requested to be received as a third party intervening voluntarily in the existing proceedings between the scrl Copiepresse and the company Google in order:

- to establish that Google may not maintain any exception stipulated by the Act of 30 June relating to authors rights and related rights,

- to establish that the activities of Google News are in breach of the Act of 30 June 1994 relating to copyright and related rights,

- to order Google to withdraw all photography of the photographs from the daily press represented by SOFAM from the date of the notification of the ruling under penalty of a fine of two million euros per day of delay,

- moreover to order Google clearly and without comment on its part to publish the entirety of the judgement to be pronounced for an uninterrupted duration of 20 days within ten days from the date of the notification of the ruling under penalty of a fine of two million euros per day of delay on the GOOGLE.BE and NEWS.GOOGLE.BE home page;

By submissions made on 13 November 2006, the SOFAM has changed the claim as follows:

- to establish that the activities of GOOGLE NEWS and GOOGLE IMAGES are in breach of the Act of 30 June 1994 relating to copyright and related rights,

- to order Google to withdraw all photography of the photographs from the daily press represented by SOFAM from the date of the notification of the

7

ruling under penalty of a fine of two million euros per day of delay,

- moreover to order Google clearly and without comment on its part to publish the entirety of the judgement to be pronounced for an uninterrupted duration of 20 days within ten days from the date of the notification of the ruling under penalty of a fine of € 2,000,000 [euros] per day of delay on the GOOGLE.BE and NEWS.GOOGLE.BE home page;

By petition in voluntary third party intervention submitted on 10 October 2006, the companies "Société de droit d'auteur des journalistes" (SAJ), "Société Civile des Auteurs Multimédias" (SCAM) and "Assucopie" requested to be received as third parties intervening voluntarily in the existing proceedings between the scrl Copiepresse and the company Google and requested that:

- to establish that Google could not maintain any exception as stipulated by the laws relating to copyright (1994) and related rights (1998),

- to establish that the activities of Google News and the use of the Google cache notably are in breach of the laws relating to copyright (1994) and related rights (1998),

- to order Google to withdraw from all its sites (Google News and "cache" Google under any name whatsoever), all the articles, photographs and graphic representations of the third parties intervening voluntarily from the notification of the ruling under penalty of a fine of one million euros per day of delay,

- moreover to order Google clearly and without comment on its part to publish the entirety of the judgement to be pronounced for an uninterrupted duration of 20 days from the date of the notification of the ruling under penalty of a fine of € 500,000 [euros] per day of delay on the GOOGLE and NEWS.GOOGLE home page;

By petition in voluntary third party intervention submitted on 17 November 2004, the s.a. Pressbanking requests to hear its voluntary third party intervention declared admissible and well founded and, consequently to intend:

- to establish that Google could not maintain any exception as stipulated by the laws relating to copyright (1994) and related rights (1998),

- to establish that the activities of Google News and the use of the Google cache notably are in breach of the laws relating to copyright (1994) and related rights (1998),

- to order Google to withdraw from all its sites (Google News and "cache" Google under any name whatsoever), all the articles, photographs and graphic

8

representations of the members of Copiepresse and that the third party intervening voluntarily markets from the notification of the ruling under penalty of a fine of € 1,000,000 [euros] per day of delay,

- moreover to order Google clearly and without comment on its part to publish the entirety of the judgement to be pronounced for an uninterrupted duration of 20 days from the date of the notification of the ruling under penalty of a fine of € 500,000 [euros] per day of delay on the GOOGLE and NEWS.GOOGLE home page,

- to examine whether Google acted in bad faith in accordance with article 87 of the Act on copyright and, if so, under penalty of a fine of € 2,000,000 [euros] per day of delay from 30 calendar days after the date of the notification of the judgement to be pronounced, to order Google Inc to submit the elements according to which these must be certified by a certification authority:

* the number of pages viewed GOOGLE.NEWS.BE and GOOGLE.NEWS.FR from 1^st January 2006 to 1^st December 2006,

* the number of referrals (clicks) to the websites of the Belgian publishers (with a detailed identification of the publishers and the links visited) from 1^st January 2006 until 1^st December 2006,

* the list categorised by publisher of all the articles that GOOGLE.NEWS.BE and GOOGLE.NEWS.FR copied from all the sites (i.e. not only those of the publishers represented by Copiepresse) from 1^st January 2006 to 1^st December 2006,

* the list -- categorised by publisher -- of articles that GOOGLE NEWS ARCHIVE copied from all the sites (i.e. not only those of the publishers represented by Copiepresse) from 1^st January 2006 until 1^st December 2006 as well as the statement that no articles come from the search engine GOOGLE,

* the list -- categorised by publisher -- of articles that search engine GOOGLE copied from all the publishers' sites since 1^st December 2001 (the period of the prescription by limitation of lapse of time of 5 years),

* the number of referrals (clicks) to the "cached pages" of articles from Belgian publishers that GOOGLE placed in the "cache" (with a detailed identification of the articles and the publishers) since 1^st December 2001 (the period of the prescription by limitation of lapse of time of 5 years);

CURRENT CLAIMS BY GOOGLE AND COPIEPRESSE:

Under the terms of its submissions made on 22 November 2006, the company Google changed it claim and currently requests:

9

- With regard to the claim by the company Copiepresse, to hear:

In principal, to pronounce the original claim inadmissible or at least unfounded and to dismiss the case of the company Copiepresse,

In subsidiary order:

- to limit the cessation order to the obligation for Google solely to remove the links visible in cache from the site www.google.be (hyperlink accessible with the reference "in cache") to the pages of the publishers of newspapers that are members of Copiepresse that the judgement shall identify exactly by name and the sites on which they are accessible, as well as, if applicable, the obligations for Google to remove the titles and extracts from press articles of the newspaper publishers that are members of Copiepresse from the site news.google.be, identified precisely by name in the judgement and the sites on which these are accessible,

- to establish for the remainder that the publication measure ordered by the judgement objected to has not object, the said publication having been done,

- to dismiss all other claims of Copiepresse as inadmissible or unfounded;

- With regard to the claims of the voluntary third party interventions:

In principal: to declare their claims inadmissible and at the very least unfounded;

In subsidiary order: to limit the cessation order or orders to the obligation for Google solely to remove the links visible in cache from the site www.google.be (hyperlink accessible with the reference "in cache") to the pages or documents of the authors which the plaintiffs in voluntary third party intervention justify that they hold the rights or that they have the capacity to act for the cessation of damage to these rights and that the judgement to intervene shall identify these explicitly;

By submissions made on 9 October and 14 November 2006, the company Copiepresse requested the confirmation of the decision under opposition and moreover introduced a counterclaim intending to hear Google ordered clearly and without comment on its part and at the top of the screen (i.e. before the search window) on all its home pages of all the French-speaking sites in GOOGLE and GOOGLE NEWS to publish the entirety of the judgement to be pronounced for an uninterrupted duration of 20 days within 10 days from the date of the notification of the ruling under penalty of a fine of two million euros per day of delay;

10

SITUATION OF THE DISPUTE:

Copiepresse is a management company of rights of the Belgian French and German speaking daily press publishers authorised by the Ministerial Decrees of 14 February 2000 and 20 June 2003 to exercise its activities on the national territory; its aim is to defend the copyright of its members (rights of publishers and acquired rights from journalists) and to supervise the use by third parties of the protected work of its members;

The third parties under voluntary intervention, with the exception of the s.a. Pressbanking, are also management companies of copyrights recognised by ministerial decree, including, for SOFAM, essentially photographs and expressive artists, for the SAJ, journalists, for Assucopie, school, scientific and university authors and the SCAM is the French company for copyright management;

The s.a. Pressbanking is, on its part a "pressclipping" company that electronically disseminated press articles to its customers according to the latter's requests;

The company Google is an American company, incorporated in 1998, of which the principal activity consists in developing and making a search engine available to web users on the Internet or an automated reference tool for information available on the web;

Google stipulates that its search engine is made up of indexation software or "robots" (called "googlebots") that trawl through websites moving through page after page, at regular intervals and in an entirely automatic way in order to list these in an index of all the web pages accessible to the public and the corresponding web address for each (called the URL address); Internet users can consult the index by means of keywords entered in the search bar, the search engine then displays the reference lists of pages available including the keywords searched and proceeds to an automated classification by relevance;

In parallel to this principal activity as a web search engine, Google has, over the years, developed other services; notably in 2002 it developed a service called "Google.news" qualified by it as a specialised search engine based on the indexation of press articles published on the net; this service is available in

11

Belgium under the name "Google.Actualités" since January 2006;

The scrl Copiepresse quickly responded following the arrival on the Belgian market of the service "Google.Actualités", considering that this activity went beyond a "simple" service as a search engine but acted as a "portal to the written press", Google reproducing and displaying as it saw fit a significant part of the text of the articles and this, without having obtained the agreement notably of the sites of the newspaper publishers for whom it defends the interests;

Consequently on 9 February 2006, it submitted a petition for seizure for description to the Judge of Seizures of the Court of First Instance in Brussels;

By ruling of 27 March 2006, the Judge of Seizures accepted this claim and appointed the expert [Luc] Golvers with the assignment:

• "to describe the way in which the press articles are presented and the interactivity between the visitor and the Google News website
• to describe the alert system installed on Google and Google News
• to determine whether articles that are no longer available on the websites of the plaintiff's members are still visible on Google and Google News and whether it is possible still to consult them and by what means
• to establish or try to establish the duration the articles are stored in the memory cache of Google and Google News,
• to draw up a list of articles present on Google and Google News making it possible to identify the author of the article, the periodical and its date of publication
• to explain how Google and Google News were able to obtain copies of the articles
• to identify and describe the route taken by a visitor who follows the content links on the Google and Google News websites and to compare these with a visitor who normally consults the website of the publisher of the press article concerned
• to determine the incidence of any differences ascertained in the preceding point
• to determine whether it is still possible to access Belgian press articles by no longer passing through Google.be but through Google.com and/or Google.fr
• to determine the identity of the operator of DNS Google.be, Google.fr and Google.com
• to determine whether it is possible for DNS.be to remove or make access to Google.be unavailable and -- if so -- how; or if it [is] possible to route the visitor to a page located on a different website.
• to determine whether it is possible to do the same with Google.com and to determine which technical operator is capable of realising these operations";

This ruling was served on the company Google on 27 April 2006;

12

The expert Golvers submitted his report on 6 July 2006;

By letter of 13 July 2006, the counsel for the company Copiepresse formally notified the company Google immediately to remove the newspaper articles of the Belgian press (of which the list is included in the petition and seizure description) which are present in Google News and the Google cache;

The company Google did not respond to this letter;

PROCEDURE:

1. Petition for voluntary third party intervention submitted by the s.a. Pressbanking on 17 November 2006:

Considering that in its submissions Google requested that the claim introduced by the s.a. Pressbanking by petition in voluntary third party intervention of 17 November 2006 be pronounced inadmissible or, at the very least that its examination be adjourned to a later hearing and this, in order not to delay the judgement in the principal claim;

Considering that at the hearing of 24 November 2006, the parties accepted that the claim formulated by the s.a. Pressbanking be examined later;

That they consequently agreed to separate the said claim;

2. Withdrawal from the proceedings of the parties SOFAM and SCAM:

Considering that at the hearing of 24 November 2006, the scrl Société Multimédia des Auteurs des Arts Visuels (SOFAM) and the company under French law "Société civile des auteurs multimédias" (SCAM) declared their withdrawal from their proceedings;

That the company Google declared acceptance of these withdrawals;

DISCUSSION:

1. Quality and interest to act in cessation:

Considering that the claim is based on Article 87 of the act of 30 June 1994 relating to copyright and related rights;

13

Considering that Google insists, in first order on the fact that the Chairman of the Court of First Instance is not competent, in the framework of these proceedings to establish the existence of a particular damage to the copyright itself and to order the cessation and not to establish any breach of the act on the copyright whatsoever;

That it emphasises moreover that no author is included amongst the claimants and that it consequently is their responsibility to justify their interest in acting;

Considering that the first dispute relates to the object itself of the claim; That this must be read together; that if the claim is intended to hear the damage to the copyright pronounced may appear incomplete (although it stipulates the incriminate behaviour or activities of Google.News and the use of the cache), the cessation order, as drafted, makes it possible to determine the works to which the alleged counterfeit relates (or with regard to the original claim: the articles, photographs and graphic representations of Belgian daily press publishers, French-speaking and German-speaking represented by the company Copiepresse);

Considering that with regard to the company Copiepresse's interest to act and the voluntary third party intervention of SAJ and Assucopie, these are management companies of copyrights authorised by ministerial decree to exercise their activities on the national territory;

That pursuant to Article 73 of the Act of 30 June 1994, they are authorised to act in court to defend the rights that which they are statutorily entrusted with, that it relates to the collective rights of the affiliates or their individual rights; that, if their articles of association so stipulate, the ability to act in law in the general interest of the category or categories of claimants as stated in the articles of association (F. De Visscher and B. Michaux, Précis de droit d'auteur et des droits voisins, Bruylant 2000, p. 419 and p. 512);

That the company Copiepresse is a management company of rights of the Belgian French and German speaking daily press publishers authorised by the Ministerial Decrees of 14 February 2000 and 20 June 2003 to exercise its activities on the national territory; Its aim is to defend the copyright of its members and to supervise the use by third parties of the protected work of its members; It represents the publishers who notably publish La Dernière Heure/Les Sports, L'Echo, La Libre Belgique, Le Soir (for a complete list

14

notably see the petition for seizure description of 9 February 2006);

That the aim of the SAJ is, in the framework of a collective management, to defend -- notably in the framework of court and/or administrative proceedings -- to manage, administer and operate the material and moral interests of journalists in relation to their works;

That the aim of the scrl Assucopie is to operate, administer and manage, in the broadest possible sense, the rights for reprography of school, scientific and university authors; that it may act in justice to defend the interests of its members, associates or associate members and to defend the rights which are entrusted to them by law;

That the proceedings for cessation organised by Article 87 of the Act of 30 June 1994 may be introduced at the request of all interested parties, these management companies are not required to prove any particular mandate (A. Beerenboom, Chronique de jurisprudence, Le Droit d'auteur, J.T. 2002, p. 685, no. 42);

That the original plaintiff and the voluntary intervening third parties consequently, in view of the preceding, have the capacity and interest to act for cessation on the basis of Article 87 of the Act of 30 June 1994;

That on the other hand and as Google rightly maintains, the Act of 10 August 1998 converting the directive of 11 March 1996 concerning legal protection of databases into Belgian law does not open the case to "all interested parties", nor to a management company or a professional association but is reserved solely to the holders of the right "sui generis" or the producers of the database (B. Michaux, Droit des bases de données, Kluwer 2005, p. 166); that the claim as founded on this basis must, consequently be declared inadmissible;

2. With regard to the existence of works protected by copyright:

Considering that Google accuses the plaintiff and the voluntary intervening third parties of not stipulating and a fortiori establishing, a single concrete case of damage to the copyright of which they are holders claiming moreover that they do not even give details of the authors for whom they are claiming the rights nor the works which the activity of Google allegedly damaged;

15

Considering that this claim appears inaccurate;

That in fact, in his report, the expert Golvers stipulates having carried out searches on the titles of the publishers represented by Copiepresse, stipulating that it found, on Google News, in regard to:

• Le Soir en Ligne: 1 670 articles,
• DH Net: 2 240 articles,
• L'Echo: 2 010 articles,
• La Libre Belgique: 2 360 articles;

That it results from this that the articles for these newspapers were effectively edited by the service Google.News and this, even though the expert stipulates that there were results that he qualified as false positives for L'Echo and the Libre Belgique (see page 101 of the expert report);

That it may be stated with regard to the SAJ, that it stipulates managing the copyright of numerous creators cooperating with the editing companies; that it submitted the list of its members from examination of which it is apparent that amongst its members it includes the journalists from L'Echo, La Dernière Heure / Les Sports, La Libre Belgique, Le Soir, Du Vif-L'Express, ..., or the newspapers in which editing in Google News was established by the expert witness;

That moreover, it has been admitted that article 18, paragraph 2 of the Judicial Code is applicable in the matter of the cessation proceedings, these proceedings having an essentially preventive nature (De Visscher and Michaux, op.cit., p. 510);

That with regard to the way to proceed in Google: the search engine of Google Web trawling the Web as a whole (- cf. hereafter developments relating to the cache memory) and Google Actualités stating 500 sources of information in the French-speaking press, it appears hardly disputable that there is a real risk of damage to the rights defended by Copiepresse, the SAJ and Assucopie (in relation to the latter solely in the framework of the "Cache" page);

That the law does not prevent the judge for the cessation pronounces an order for the future that intends other works or services than those actually damaged up to that point (F. De Visscher and B. Michaux, op.cit., p. 510);

16

That in effect, the aim of the cessation is rather to end a type of practice of which the infringement is established is a part; (Pres. Trib. Civ. Bxl 16 October 1996, Auteurs et Media 1996, p. 426);

That the objection formulated by Google consequently does not appear well founded;

3. With regard to respect for Article 10 of the European Convention of Human Rights:

Considering that Google considers that the Google.News service is legitimised by Article 10 of the European Convention of Human Rights which guarantees freedom of expression; That it insists on the fact that the freedom of expression protects the different aspects of the communication process including the freedom to receive and communicate information;

That if Google acknowledges that the freedom to receive and communicate information may be limited for the protection of the rights of others, including copyrights, it in any event considers that in this case, the restriction of the right of the freedom of expression claimed by the original plaintiff and the voluntary third party interventions is neither pertinent nor proportionate as Google News is a free tool for access to information which does nothing other than ensuring an indicative starting point in the search for information on the Internet;

Considering that the freedom of expression that effectively grants the freedom to receive and communicate information is not absolute, Paragraph 2 of Article 10 of the European Convention of Human Rights stipulates that "The exercise of these freedoms, since it carries with it duties and responsibilities, may be subject to such formalities, conditions, restrictions or penalties as are prescribed by law and are necessary in a democratic society, in the interests of national security, territorial integrity or public safety, for the prevention of disorder or crime, for the protection of health or morals, for the protection of the reputation or rights of others, for preventing the disclosure of information received in confidence, or for maintaining the authority and impartiality of the judiciary;"

That in view of this clause, the freedom of expression may be limited by the copyright (limitation stipulated in law for the protection of rights of others);

That this was repeated by the Court of Cassation in its judgement of 25 September 2003 which considered that the right to freedom

17

of expression guaranteed by the European Convention of Human Rights does not hinder the protection of originality according to which the author of a literary or artistic piece of work expresses his ideas and concepts (Cass. 23.09.2003, C030026N [PDF], available from the website www.cass.be [NDLR: in French and Flemish, Flash player required]);

Considering that if pursuant to Article 10 Paragraph 2 of the European Convention of Human Rights, the right to the freedom of expression may be subject to restrictions in view of guaranteeing copyright, it is not, clearly, intended to place the copyright on a pedestal, that it is notably on this ground that the law stipulates an exception;

That as stated in [EU] Directive 2001/29 of 22 May 2001 on the harmonisation of some aspects of copyright and related rights in the information society, the copyright is based on a balance between the acknowledgement of legitimate interests of the authors on the one hand and, on the other, the equally legitimate interests of the public and society as a whole; (M. Buydens, La nouvelle directive du 22 may 2001 sur l'harmonisation de certains aspects du droit d'auteur et des droits voisins dans la société d'information: le régime des exceptions, A.M. [Auteurs & Médias], 2001, p. 430);

That some exceptions to the copyright from opposing the reproduction or communication of their work to the public are based on the freedom of expression, notably such as citations (M. Buydens, op.cit. p. 431); What will be considered hereafter is whether Google may cite this exception;

Moreover that Copiepresse pertinently, raises the question of whether it is possible, in this case, to mention the exercise of a right of expression on the part of Google insofar as the system implemented by Google News is not human in any way, that Google does not employ any head editor with a view to the selection of the articles and prides itself on "trusting the judgement of the editors of information agencies to determine the articles that most merit from being included and highlighted on the home page of Google Actualités;"

That Google may consequently not limit itself to invoking Article 10 of the European Convention to justify the incriminated activities; there is consequently causes to examine concretely whether there is damage to the copyright and whether Google may cite an exception;

18

4. With regard to the violation of copyright:

a. Damage to patrimonial rights:

Considering that Copiepresse and the voluntary intervening third parties consider that by two "activities" (i.e. the "cache" memory on its site Google.be and the Google.News service), Google proceeds to the reproduction and/or communication to the public of works (or fragments of works) protected by copyright without having the authorisation of the author or its representatives;

That in its submissions Copiepresse also incriminates the Google News Archive Search service insisting on the fact that this service keeps a store of the press archived for many years; That it has not however developed any argumentation with regard to this service which, according to Google, only exists in the American version of its Google.News service; that in this context and, in the absence of any additional elements, this service, not described at any point in the report by the expert Golvers, cannot be examined in the framework of this debate;

Considering that pursuant to Article 1 of the Act of 30 June 1994 relating to copyright, the author has the sole right to reproduce or authorise the reproduction of the work "in any way and under any form whatsoever;"

That Article 1 of the Act of 30 June 1994 (as amended by the Act of 22 May 2005) also stipulates that the author of a work has the sole right to communicate it to the public "by any process whatsoever including making it available to the public in such a way that any party may have access at the place and moment he chooses himself;"

That there is cause in view of these clauses to examine whether in this case, Google reproduces and communicates works protected by copyright to the public and this, in the framework of the use of the cache on Google.be, on the one hand and in the framework of Google.News, on the other;

1. Google "cache":

Considering that Google described the operation of the "cache" as follows (see report by the expert Golvers, p. 68 and 69):

- "Google takes a snapshot of each page examined as it crawls the web and caches these which makes it possible to consult this copy at any moment [and in particular] in the case

19

that the original page (or Internet) is unavailable. If you click on the "Cached" link of a web page, Google will show the web page as it looked when it was last indexed. Moreover the cached content is the content Google uses to judge whether this page is a relevant match for your query.

When the cached page is displayed, it will have a header at the top which serves as a reminder that this is a cached copy of the page and not the original page and which cites the terms of the request which resulted its inclusion in the results of the search. (...)" (extract from the site http://www.google.be/int/fr/help/features.html#cached). [NDLR: this link is dead, try this one instead.]

- "In order to index millions of pages every day, Google uses an "army" of indexation robots, called GoogleBot (read the study on GoogleBot for more information about it). Each time one of the robots visits a page, it recovers it and stores it on the Google servers. This version of the document is called the cached version."

- "What is the Google cache used for? For the majority of internet users it has no purpose! Moreover they do not know that it exists, or how to access it. Nevertheless it may prove useful in a number of cases:

• For an internet user: a page is inaccessible or cannot be found (error 404). If Google has indexed this page, it is possible to consult it by requesting its cached version.
• For a webmaster: the cached version corresponds to the version the GoogleBot obtained when it came to index it. For example this makes it possible to see from what date the document (supposing that this page is often updated); this also makes it possible in the event of redirections to see which page GoogleBot saw.
• As proof: if you have found a site that plagiarised you but has removed its page, you may use the cache function on Google to prove this (Even if the value of this proof is relative in any event."

- "To obtain the cached version of a document: the simplest way to obtain the cached version of a document is to click on "cached copy" beside the result when searching on Google."

(extracts from the website http://www.webrankinfo.com/google/cache.php);

That in their report, drawn up at the request of Google, the professors Tiberghien and Viseur defined the cache memory as follows: "a cache memory, or more concisely a "cache" is a memory in which a copy of a document is temporarily saved in order to

20

facilitate or accelerate the access to this document during processing"; That they stipulate in more detail: "... when a robot starts to crawl a site, it starts by making a copy of the HTML codes on the useful pages in its own memory. It does this in order to make it possible to work on a trusted image of the site and to avoid imposing exaggerated traffic. Only the HTML code on the pages containing the words and links to other pages is copied. The HTML code on the pages that do not contain any images, sounds or executable programmes is not copied as this does not contribute to increasing its dictionary. The caches of search engines are not only essential for indexation but may also be useful to access pages from servers that are temporarily unavailable or to speed up access to pages from sites that are not very accessible or overloaded. (...)";

That in relation to the use of caches in Google, Messrs Tiberghien and Viseur state: "The search engine Google Web makes it possible to visit the version saved in the cache for pages indexed by the search engine if this access to the public was authorised by their owner. Google News, on the other hand, never gives access to its cached pages. It should be noted that the consultation of a page via the caches is not the preferred method for consultation. The Google layout encourages clicking on the title of each result (which links to the current page on the publishers site) rather than on the link cached link (which links to the page currently in the cache), of a smaller size, less intuitive and with a lower contrast. (...)

The Google Web cache allows access to pages to which the link was broken, in particular in the case that the web server shows that the page is unavailable. It also allows other uses, such as easily proving a case of plagiarism or to see which keywords on a page were selected by the search engine. Consequently it relates to a useful and appreciated function for users of the search engine Google Web." (Tiberghien-Viseur report, p. 19 and 20);

Considering that it is apparent from the preceding that when it indexes the web pages, the Google "robots" make a copy of each page examined, a copy which is stored in the Google memory; that it is possible for the Internet user to have access to this copy by clicking on the link "cached"; That under this hypothesis, the Internet user is not sent via a "hyperlink" to the original site but consults the copy of the page stored in the Google memory (and consequently remains on the Google site);

21

That there is consequently a material reproduction of the work and communication of this to the public in accordance with article 1 of the act on copyright;

That Google nevertheless insists on the fact that it only copies the HTML code of the page (code which only contains text elements and no image); That it moreover maintains that it is not itself but the Internet user who creates a copy of the work in such a way that only they are the author of any reproduction or communication to the public, the only act provided by Google being the supply of the installations intended to make it possible or to realise a communication to the public by Internet users;

Considering that this analysis does not appear accurate;

That it is apparent, in fact, from the aforementioned developments that Google stores a copy of the web pages in its memory, the circumstance that this copy relates to HTML code of the said pages -- or that it is converted into computer language -- appears without relevance;

That there is, in effect, reproduction in the digital field from the moment that there is a saving or storage of signals in any type of memory (F. De Visscher and B. Michaux, op.cit., p. 71), which is the case here;

That Google is consequently the author of the reproduction;

That it is this reproduction that Google makes available to the public on its own website, a reproduction accessible via the "cached" link; That in effect, unlike the hyperlinks that refer to the site of origin, by consulting the "cached" link, the Internet user is consulting the document on the Google website;

That Google's role is consequently not limited as maintains in its submissions, simply to providing the installations intended to allow or realise a communication to the public;

That there is consequently a reproduction and availability to the public by Google of the copy of the original document stored in its own memory;

That if this reproduction constitutes an integral part of the technical indexation process of the pages, it is not in any event its sole use; That in effect, it is apparent from the aforementioned developments that it is also used to allow the Internet user to consult a

22

document that is no longer available on the site of origin (either because the link has been broken or the site has been changed) directly on the Google website;

That in this way, as relates to press articles, it is apparent from the report drawn up by the expert Golvers that an article located on the home page of a site of the newspaper Le Soir of 9 February 2006 and which can no longer be viewed the day after, or at least not freely, on this site, remains viewable via the "cached" link on the site Google.be (see report by the expert Golvers, p. 35 to 38);

That Google does not dispute the fact that the press articles are protected by copyright (notably see Prés. Trib. 1^ere Instance of Brussels, 16 October 1996, Auteurs & Médias 1996, p. 426);

That the practice at Google consisting of saving works protected by copyright in its so-called cached memory and of allowing Internet users to access from the said memory itself (without referral to the original site) consequently constitutes an act of reproduction and communication to the public;

2. Google News

Considering that Copiepresse and the voluntary third party interventions consider the service "Google.Actualités" or "Google.News" to offer more than a "simple" search engine service (as provided by the "Google.be" service) and must be qualified as an "information portal"; That they emphasise, in effect, that "Google.Actualités" proposes content to Internet users without a previous search;

That they accuse Google of taking this content directly from their sites, by copying the titles of articles and the slogans, without having obtained their prior agreement, which is in their opinion in breach of Article 1 of the Act of 30 June 1994 relating to copyright and related rights;

Considering that Google disputes the fact that the service "Google.Actualités" may be qualified as an information portal insisting on the fact that at the beginning of the service offered by Google Web, it related to a search engine which is in any event specialised in news and that makes it possible for Internet users easily to identify the press articles that may interest them from amongst the titles published on the Internet over the previous 30 days and to consult them, at the source, by

23

linking with a single "click" to the sites of the publishers disseminating the said articles;

That Google insists on the fact that this service is based, like the search engine Google Web, on the automated indexation by the GoogleBot robot of the press articles disseminated over the Internet; That Google consequently considers that "Google.Actualités" is not a site distributing news but a search engine, by keywords, making it possible to search for all the index articles containing certain keywords;

That Google disputes reproducing and communicating protected works to the public, the Internet user being referred, via a hyperlink, to the site of origin with a view to consulting the article, Google limits itself to making the said article more accessible, thanks to the hyperlink;

That Google moreover disputes the fact that the elements included on the home page of its website "Google.Actualités", i.e. the title of the articles and the first sentence or sentences of the said articles, may be considered as original elements benefiting from the protection of copyright law;

That in any hypothesis, Google considers it is able to maintain the exceptions stipulated by law on copyright, i.e. the exception of citation and the exception of reporting news;

Considering that there is cause, on the previous grounds, to examine the operation of Google.News;

That in his report, the expert Golvers stipulates, in relation to the "Google.Actualités" service that "The user only needs to visit the URL address. He does not have to enter any search or selection. A page such as that reproduced in appendix 2 is displayed (see appendix 2 of the report). The content of this page develops, naturally, constantly in relation to the current news. The user discovers the articles without taking any action and is not obliged to make an exact search."; That the expert concludes: The site Google News is consequently an information portal site based on the press. The information is presented to the user without him having to do anything other than consulting the site. In any event, as we will see below, the user may adapt and personalise this page by having information appear in relation to their wishes and particular interests.

24

With regard to the information displayed, the following may be ascertained:

1. The articles are displayed in the form:

- of a title in blue and grey characters;

- follow[ed] by a line indicating the source in grey (e.g. RTBF, DH Net etc.) and in black characters the age of the information in the form "published in the last hour", "3 hours ago", etc.

- follow[ed] by an extract of a few lines of the beginning of the article as shown on the site of the original publisher of the information.

2. The titles in blue are hyperlinks that forward to the page of the original publisher of the information. By clicking on the title of the article on the Google News page, you are forwarded to the publishers site."; (see pages 5 and 6 of the report by the expert Golvers);

That the expert carried out various trials after which he drew up the following conclusion in relation to Google.News (see page 98 of the report):

"For articles that its search engine selects Google.News (news.google.be) memorises, at least the following elements:

• the title of the article;
• the text of the first lines of the article;
• the address of the page (deep hyperlink) where the article is located when it is selected by the search engine,
• if applicable, a picture associated to the article.

In other words, GoogleNews does not memorise the content of the pages with the articles of the publishers of the Belgian press in its databases but only the deep hyperlinks, which Google News memorises after having crawled over the sites of the press publishers.";

Considering that in their report, professors Tiberghien and Viseur disagree on their part of the qualification as an "information portal" of the Google.News service; That they consider that Google.News must be qualified as a "search engine" insofar as "these tools remain centred on their primary function of searching web pages and their technical function corresponds to that of a search engine (exploration, indexation, search by keywords)" (page 27 of the report);

That they continue by stipulation that "in addition to a traditional search function by keyword, Google News brings the news together on a home page per topic and subject.

25

For example Google News is able to bring together all the articles references on the web relating to the reactions of the American authorities to the nuclear testing of North Korea.
The news are moreover divided into topics: "International", "Belgium", "France", "Finance", "Science and Technology", etc.
This classification is carried out automatically, without human intervention. The classification software is based on the data collected automatically by the robot for GoogleNews that functions in a similar way as the Google Web robot.";

With regard to the "reproduction of the articles", they indicate: "It has already been stated ... that Google News does not give any access to the articles saved in its cache.
In this way a visitor to Google News may only see the title, often a short description of the article and occasional a small low resolution thumbnail (...). Clicking on the title or thumbnail systematically forwards the user to the publishers site of this article or thumbnail.";

With regard to the last paragraph, there is cause to state that the text following the title of the article does not, as written by Messrs Tiberghien and Viseur related to a short description of the article but constitute a literal reproduction of the first lines thereof, as shown on the original site (see the report by the expert Golvers for an illustration p. 7, 8 and 9: articles: "Charleroi: le PS tente de remettre de l'ordre", extract from DH Net and "Les illégaux le resteront", extract from Le Soir online);

Considering that it is apparent from the aforementioned that the homepage of the Google.News website automatically displays by the simple consultation of the site, in addition to a window that allows one to enter keywords, a series of article titles, the first title presented being followed by a short extract from the article itself, classed by topic; That the said titles constitute hyperlinks forwarding the internet user to the publisher's website where the article may be read as a whole;

Considering that in this case, it is not the hyperlink to the original site that is disputed by Copiepresse and the voluntary third party interventions but the reproduction on the Google.News site itself, of the titles of the press articles and the extracts of the press articles;

Consequently that the developments [of] Google in relation to the fact that it is admitted both by Belgian and international case law

26

and legal doctrine that a hyperlink forwarding to a work protected by copyright does not constitute a reproduction and that if there is a reproduction, it is realised by the Internet user (see pages 67 et seq. of the additional submissions and summary submissions made by Google), appear without relevance in the framework of this dispute;

That on the other hand it is unambiguously apparent from the aforementioned developments that Google.News reproduces (the expert Golvers stating that these elements are memorised by Google) and communicated to the public on the homepage of its site (these elements being accessible on the site of Google.News itself by the simple consultation of this site), the press article titles and an extract from some of these articles;

Considering that Google however maintains that these elements do not constitute works protected by copyright;

Considering that in order to benefit from the protection of the law on the copyright, a creation must be expressed in a particular form (ideas as such cannot be protected) and be original i.e. it must be marked by the personality, the stamp of the author (A. Beerenboom, Le nouveau droit d'auteur et les droits voisins, Bxl, Larcier 1997, p. 49 [New edition: Brussels, De Boeck & Larcier, 2005]; F. De Visscher and B. Michaux, Précis de droit d'auteur et des droits voisins, Bruylant 2000, p. 15);

That the length of a work is not important, a title being eligible for protection under the understanding that it fulfills the requirements of originality (A. Beerenboom, op.cit., no. 38 and 48; F. De Visscher and B. Michaux, op.cit., p. 30);

That Google considers in this regard, the titles of the press articles are not in any way original, as they are simply made up of current language, citing as an example "The King visits Sweden" or "Tom Boonen World Champion";

Considering that if all the titles of the newspaper articles cannot be considered as originals -- some appear to be purely descriptive and consequently do not bear the stamp of their author -- it may not however be considered that no titles of press articles present a sufficient originality to be able to benefit from the protection of the copyright law;

That in this way, reading the report drawn up by the expert Golvers indicates that the title of some articles stated in the said report do not appear original (e.g. "Philippe Eloy Assassinated His Wife's Friend", "Deserts At Threat From

27

The Greenhouse Effect"; "Aviation: Sowear has been dissolved") while others did (e.g. "The Illegals Remain" (p.7), "The Tax Exemption Rage of Didier Reynders" (p.89), "Music Makes School Cool" (p. 102); "Monaco Between Casino and Stratego") (p. 113);

That in the same way, it has not been disputed that articles by journalists may, in general, claim protection by copyright law insofar as the stamp of the personality of its author can be found in it (notably see Prés. Trib. 1^ere Instance of Brussels, 16 October 1996, Auteurs & Médias 1996, 426);

That if it is true that only the first lines of the said articles are reproduced on the homepage of the "Google.News" site, this does not necessarily exclude the protection by copyright;

In effect, in order to fall under the scope of exclusive copyright, the reproduction must not be total and may only be partial insofar as it is "borrowed" from that which constitutes the originality of the work, in whole or in part (F. De Visscher and B. Michaux, op.cit. p. 65);

That in a ruling of 25 September 2003, the Court of Cassation in this way considered that "the author may express their ideas in the form of a text published in a review or placed on an electronic medium; That the reproduction of elements constituting the originality of this text in a summary benefit from copyright protection; That neither the length of the summary nor the obligation for the user or reader of the summaries to consult the original summarised text after reading are determining for the assessment of the violation of the copyright; (...); That all reproduction, even partial may be sufficient to establish the infringement if it contains original elements" (Cass. 25 September 2003 C030026N [PDF] that may be consulted on the website www.cass.be [NDLR: In French and Flemish, Flash player required]);

That it consequently does not appear excluded that borrowing from the author may be marked in a short fragment of text, while in this case, it relates to the first sentences of an article, "slogan" phrases for it;

That in this way, in reference to the report by the expert Golvers, such appears to be the case in the following extracts: "Coup d'envoi des débats sur le projet Dewael à la chambre. Les sans-papiers espèrent que l'heure de la régularisation a sonné. Patrick Dewael dit ne pas avoir le même agenda." (p.7); "En plein polémique sur le "Da Vinci Code" le Pape a profité de sa visite en Pologne

28

pour rappeler aux fidèles quil leur faut résister aux tentations de relativisme" (report p. 7);

That it is apparent from the aforementioned that by reproducing titles of articles and short extracts from articles on its site Google.News, Google reproduces and communicates works to the public protected by copyright;

[4] b. Exception of citation and reporting news:

Considering that Google maintains that it can cite legal exceptions for citation and reporting news;

1. Exception of citation:

Considering that pursuant to Article 21 § 1^st of the Act of 30 June 1994 (as modified by the Act of 22 May 2005): "Citations drawn from a lawfully published work, carried out with a view to critique, polemic, review, teaching or in scientific works, in accordance with honest use in the profession and insofar as justified in the aim pursued, do not damage copyright.
The citations stated in the previous paragraph must mention the source and the name of the author unless that should prove impossible.";

Considering that the conditions stipulated by the law must be satisfied cumulatively (F. De Visscher and B. Michaux, op.cit. p. 104);

That as relating to an exception, it must, contrary to that maintained by Google, be interpreted restrictively;

That it should moreover be stated that the article of legal doctrine on which Google bases its argument to consider that the right to access to information and the transmission of knowledge may be the basis for the extension of the interpretation of exceptions, indicates that the current trend appears to be the reduction of the scope of the exceptions to copyright in the digital sector and emphasises that it is generally admitted that taking account of their nature exceptions should necessarily be interpreted restrictively (S. Dussolier, Y. Poullet and M. Buydens, Droit d'auteur et Accès à l'information dans l'environnement numérique, Bulletin du Droit d'auteur vol. XXXIV, n° 4, 2000, p. 10 and 14 [PDF]);

That as shown above (cf. p.15) it is notably in order to maintain a balance between copyright and other rights

29

(such as the right to information invoked by Google) that the exceptions have been provided;

That article 5.5 of the directive 2001/29/CE, stipulates that the use of exceptions to copyright must be realised in conformity with the obligations in force at an international level and that exceptions and limitation are only applicable in certain special cases, that do not damage the normal operation of the work, nor cause unjustified damage to the legitimate interests of the holder of the right (three stage test), which appears to be such as to confirm the restrictive nature of the exceptions (notably see consideration 44 of the directive cited by SAJ and Assucopie in their summary submissions p. 44);

That if it is correct that this test has not been integrated in the Act of 22 May 2005, this is not because the legislator considered that it were not applicable but because he considered that this concept was known and was before-all addressed to the legislator, having stipulated that this does not mean that this triple test could not equally serve as an orientation for the courts and tribunals in application of the law (F. Brison and B. Michaux, La nouvelle loi du 22 mai 2005 adapte le droit d'auteur au numérique, Auteurs & Médias 2005, p. 216);

Considering that this having been stipulated, there is cause to examine whether, in this case, the reproduction of the titles of the articles and the fragments therein on the Google.News website fulfill the different legal conditions for the exception of citation;

1° Citations drawn from lawfully published works:

Considering that this first condition is fulfilled in this case, Google.News drawing its extracts, from newspaper articles published on the publisher's website;

That the circumstance that these articles would, after a certain period only be accessible to subscribers, is irrelevant insofar as the term "lawfully published work" relates more to the right of disclosure of the author (Doc. Parl. Ch. 2003-2004, no. 51-1137/10, p. 2 [PDF]);

That the publication of the work on the Internet removes the right of disclosure, as the author decided to make their work accessible on the Internet (A. Beerenboom, Le nouveau droit d'auteur et les droits voisins, Brussels, De Boeck & Larcier 2005, p. 181) and the change to the conditions for the publication of the work not in any way cause the extinction of this right (F. De Visscher and B. Michaux, op.cit. p. 107);

30

2° Citations for the purposes of critique, polemic, teaching, review or in scientific works and insofar as justified by the intended aim:

Considering that Google maintains that the Google.New site operates as a review of the press and that the citation of the titles of newspapers and extracts from articles fulfills the aim of a review and is justifiable with regard to the said aim;

That Copiepresse and the voluntary interventions of the third parties insist, on their part, on the fact that unlike in France, the Belgian legislator did not consider creating an autonomous exception for a "review" but that this exception is subject to the system of citation so that the articles cited must be in the framework of a coherent commentary of which they only comprise an illustration, the review must also comprise other elements;

That they consequently consider that the sole random juxtaposition (as practised by Google in an automatic way) fragments of articles not being a summons, these, being by definition and accessory and that must be used in the limits of the intended demonstration;

Considering that Google.News is exclusively made up of extracts of newspaper articles groups by topic; That the reference is entirely automated;

That Google maintains that the Google News service is based on the automated indexation by a robot similar to that of the search engine Google Web for press articles disclosed on the Internet; that the classification of articles by topic is realised automatically, without human intervention;

That the Google.News site consequently does not incorporate "citations" and consequently owes its substance to extracts from reproduced works, which is contrary to the spirit of the institution of the citation law (T. Verbiest, Entre bonnes et mauvaises références. A propos des outils de recherche sur Internet, Auteurs & Médias, 1999, p. 42);

That in effect, the citation is, in principle, used to illustrate a proposal, to defend an opinion;

Moreover that it does not appear that the editing of articles carried out by Google.News may be defined as a "press review";

31

That the citation for the end of review was introduced by the Act of 22 May 2005;

That the Larousse defines a review as "Action for examining a set of elements with care and in a methodical way" while the "press review" is defined as: "Comparative report of the main articles in newspapers on the same subject";

That this definition is confirmed by the Dutch term of the act recensie or recension in French defined by Larousse as "Critical analysis and report of a work or a review";

That the object of the final conclusion of the review consequently should not be the collection of elements intended to give a general insight into a theme but the comment on a work (B. Michaux, Droit des bases de données, Kluwer 2005, p. 27);

That in this case, Google limits itself to listing the articles and classing them, and this in an automatic way; That Google.News does not carry out any analysis, comparison or critique of these articles which are not commented on in any way;

That this condition is consequently not fulfilled in this case; That it may consequently be deduced that Google cannot maintain the exception of citation without it being necessary to examine if the other legal conditions are fulfilled;

3. Exception for reporting news:

Considering that pursuant to Article 22 §1 of the Act of 30 June 1994 "When the work was lawfully published, the author cannot prohibit:
1° The reproduction to the public, with the aim of information, of short fragments of the works or plastic works in their entirety on the occasion of a report of news events";

That §2 of the said clause stipulates: "The reproduction and communication to the public of the work on the occasion of the report of news events in accordance with § 1^st, 1°, must be justified by the aim for information pursued and the source, including the author's name, must be stated unless this should prove impossible.";

Considering that Google maintains that the Google.News service fits within a framework of information and offers a report

32

of news so that the author cannot oppose the reproduction of short fragments of the articles;

Considering that this argument by Google appears to be in contradiction with that previously presented in the framework of the description of the Google.News service, Google presenting its activity as a specialised search engine service and not in any way as an information portal;

That this being the case, if one considers that the activity of Google News fits notably in the framework of information, it does not, on the other hand, appear that by listing various titles of articles grouped according to different topics Google.News offers a report of the news;

That as stated above, no comment on the news can be found, in effect, on the site of Google News which limits itself to reproducing extracts from articles grouped by topic;

That moreover, and as for citation, it appears that the protected works may only constitute an accessory to the reporting and not the principal object (In this sense: A. Beerenboom; op.cit., p. 131; T. Verbiest, Entre bonnes et mauvaise références. A propos des outils de recherche sur Internet, Auteurs & Médias 1999, p. 42; S. Hoebeke and B. Mouffe, Le droit de la presse, [Academia-]Bruylant 2000, p. 191);

Considering that there is cause, finally, to be attentive to the justification of this exception; That as stated by Google in terms of submissions, the aim of this exception is to allow the media to react quickly to current events, the speed with which the information must be reported not allowing them to request authorisation prior to the author (A. Beerenboom, op.cit., p. 131; s. Hoebeke and B. Mouffe, op.cit,. p. 190; J.P. Liège 1^st district, 30 May 1997, Auteurs & Médias 1997, p. 300);

That such is not the situation of Google; That it would, in fact, be permissible for Google if the object was to list, nearly 500 sources of information in French, the information updating every 15 minutes, to obtain the agreement from the site publishers from which it collected the said information, in advance;

That Google may consequently not maintain the exception for reporting news;

33

c. Damage to moral rights:

Considering that the SAJ and Assucopie maintain that Google also damages the moral right of the author insofar as Google discloses protected works without the authorisation of the author, which damages their integrity and omits to mention the authors name of the works used;

That Google maintains that these companies cannot claim damage to the moral rights as these are inalienable;

Considering that the circumstance of the moral rights is inalienable does not prevent management companies from acting for cessation to have such damage to a moral right cease (see Prés. Trib. Civ. Bxl 16 October 1996, Auteurs & Médias 1996, p. 426 confirmed by C.A. Bxl 28.10.1997);

That in effect and as stated above, the management companies have, in accordance with article 73 of the law relating to copyright, the right to act in court to defend the rights of those for whom it is statutorily responsible; That it is consequently not necessary for the rights which it is defending have been ceded to them (F. De Visscher and B. Michaux, op.cit., p. 512);

Considering that the right of disclosure relates to the authors right to decide, when and in what from the work would be made known to the public; That once exercised, this right is extinguished (F. De Visscher and B. Michaux, op. cit. p. 152);

Considering that in this case, Google only reproduces and communicates works already on web pages, or in other words, works already disclosed;

That this is consequently a different situation to that giving rise in the "Central Station" jurisprudence (Prés. Trib. Civ. Bxl. 16 October 1996, Auteurs & Médias 1996, p. 426 confirmed by C. A. Bxl 28.10.1997);

That there is consequently not damage to the right of disclosure;

Considering that the author has the right to respect of their work allowing them to object to any modification thereof; That this right is not subordinate in its exercise to the condition of damage on the part of the author; That the modification may not relate to the work as such but its environment, title, classification, caesura, ... (F. De Visscher and B. Michaux, op.cit., p. 157);

34

Considering that in this case, only an extract of the work is reproduced so that there is actually a modification of the work;

That Google does not dispute "the amputation" of the works as such (i.e. their modification) but considers that this does not damage "the integrity of the work in that the citation of a text is located with another cited text or photographs illustrating the report of news of another press organisation insofar as the internet user is well aware that this relates to a citation and sees the original text each time by clicking on the hyperlink in its original context";

Considering that the circumstance that the Internet user is not ignorant of the fact that this only relates to a fragment of the work appears irrelevant with regard to the assessment of the respect for the integrity of the work;

That Google moreover, implements a themed grouping of the different extracts from articles from any source whatsoever so that the editorial or philosophical line to which the author adheres may be altered (See T. Verbiest, Entre bonnes et mauvaises références. A propos des outils de recherché sur Internet, Auteurs & Médias 1999, 40; Also see C.A. Bxl 28 October 1997);

Considering that there is finally cause to conclude that the name of the author of the work is not stated on the Google News site (Prés. Trib. Civ. Bxl 16 October 1996, Auteurs & Médias 1996, p. 426) so that there is also damage to the right to paternity of the work;

5. With regard to the authorisation of publishers of sites:

Considering that Google maintains that if it should be considered that it makes copies of works protected by copyright and/or that it communicates these to the public, there is cause to conclude that this is realised by the explicit agreement, or at least the implicit agreement of the publishers of the sites concerned;

That Google maintains that some publishers of the press were in contact with them and have, in this framework, agreed to the reference to their site;

Considering that Google effectively has documents certifying that L'Echo directly contacted its services with a view to being referenced on the Google.News website (see document 1 and 2);

35

That no other publisher represented by Copiepresse appears to have made a similar approach, the circumstance that the partnerships were concluded with Google and some sites (those of La Dernière Heure and La Libre) with a view to an integration on their pages of the service modules "AdSense" (with a view to displaying adverts for the site on the Google pages) or again for numerous sites (notably those of Le Soir, La Libre, La Dernière Heure, ...) using the Google analytics service (with a view to analysing the audience on their websites) being irrelevant in this regard, insofar as these relate to a service distinct from the Google.News service;

That it may, moreover, be concluded that Google.News appeared in Belgium in January 2006 and that from 9 February 2006, Copiepresse has submitted a petition for a seizure description and served Google with an order granting this request in April 2006; That it may consequently not be reasonably maintained that all the sites listed by Google News never made the slightest reserve or least claim in relation to their inclusion in this service;

That apart from L'Echo, Google consequently has not established that the press publishers explicitly agreed to the reference to their sites;

Considering that Google insists, moreover, on the fact that the publishers had the option to parameter their sites and to authorise some actions or not in the search engine;

That Google considers that insofar as the publishers did not implement these technical means, that they have, explicitly (insofar as they use file-robots on their sites) or at least implicitly, agreed to have the pages of their sites indexed and accessible via the cached links; That Google insists in this regard on the standard and worldwide character or site parameters by means of meta-tags and robots.txt files;

Considering that as emphasises in the submissions made by SOFAM, copyright is not a right of opposition but a right for prior authorisation; That this means that the authorisation must be obtained in a certain way, prior to the intended use;

That it may not be maintained that the use on the sites of robot files implies a certain and explicit agreement of the site publishers to the use of works included on the site as concretely used by Google.News (all the more as this service only arrived on the Belgian market in January 2006

36

or, it appears, at a time when these robot-files were already integrated);

That in the same way an unconditional authorisation for referencing cannot be deduced from the absence of technical protection (C. Morlière, Les articles de presse à l'ère numérique, Le cas de Google Actualités, I.R.D.I. [Intellectuele rechten / Droits intellectuels] 2004, p. 9 and the jurisprudence cited);

That this appears to be all the more the case here, as Google.Actualités does not limit to a reference to the articles but reproduces the title and an extract;

That there is moreover cause, to state that in this case Copiepresse has, as mentioned above, responded from the appearance of the Google.News service on the Belgian market (the petition for seizure description having been submitted only one month after the arrival of the service), which appears to be of such a nature as to show any consent on its part;

Consequently that Google wrongly maintains the ability to claim the agreement of the site publishers in this case;

6. Abuse of rights:

Considering that Google maintains that insofar as it has been shown that it is easy to avoid, for companies so wishing, to appear in the cached system on Google and appearing in the Google News service while it is not possible for Google to contact all the persons referenced by the search engine, Copiepresse and the voluntary intervening third parties are guilty in pursuing these proceedings, of an abuse of rights;

Considering that Google maintains that in exercising its rights, the author must respect a code of good conduct, based notably on the legitimate nature of the interests that it intends to protect and on the reasonable character of the claim it makes;

Considering that there is cause, firstly to establish the legitimate claims of Copiepresse as well as the voluntary intervening third parties insofar as it is apparent from the aforementioned developments that the behaviour of Google incriminates them effectively relate to the damage to the interests which they are defending (i.e. the copyright of their members);

That moreover, the appreciation of any abusive attitude of Copiepresse and the voluntary intervening third parties must

37

be assessed, not with regard to the approaches that must be carried out by Google to obtain prior authorisation from the authors but with regard to the measures which must be taken by Google to cease the disputed damage;

That Google does not dispute being able, on condition of some details (see below) of implementing the requested measures, which it moreover appears to have done in execution of the ruling pronounced by default;

That the abuse of rights has consequently not been shown;

7. Violation of competition law:

Considering that Google maintains that the collective action by Copiepresse is not compatible with the prohibition of arrangements stipulated in Article 2 §1 of the Act on the protection of financial competition and Article 81 § 1 of the EC Treaty which stipulates the decision of the association of companies are intended to hinder competition;

Considering that there is cause, firstly, to claim that the action with a view to the defence of the collective interests is explicitly recognised by the law of 30 June 1994 relating to copyright;

Considering that it is moreover apparent from the aforementioned developments that there is in this case, effectively a violation of copyright by Google and consequently a matter for cessation;

That the fact that other interests may be involved and benefit from the measures that must be ordered in the framework of this instance does not necessarily cause a change in the law to other ends than those for which it was provided;

That Google has notably not shown that Copiepresse is taking, as it maintains, the Google News and Google Web cache as pretexts in order to attempt to hinder the other activities of Google (activities which are not detailed);

That this claim consequently appears unfounded;

38

8. Google's activities and the European directive on electronic trade:

Considering that in this case, it is the behaviour of Google itself that is incriminated and not the content of the sites on which Google permits access;

That the references to this directive are, consequently, irrelevant in the framework of the appreciation of the violation by Google of the copyright of Copiepresse and the voluntary third parties intervening;

That moreover, with regard to the "caching", the matter under dispute is not the temporary storage of the cached page, the storage required for the indexation of this page, but its availability (see above, p. 17 et seq.);

9. With regard to the measures to be ordered:

Cessation order

Considering that Google maintains that the judgement under opposition lacks clarity in the wording of the judgements pronounced against it notably insofar as it does not describe which activities in Google News and which use of the "Google cache" violate the copyright (Google insists in this regard on the fact that only the cache link visible to the Internet user may be concerned by a prohibition of referencing);

That Google also considers that, the measure with the effect of removing the reference to the newspaper sites concerned, there is cause to stipulate that the name of these sites in order to be able to execute the decision;

That Google moreover makes it known that the measure must be limited to the Belgian sites of Google, namely www.google.be and news.google.be, because the competence of the court is limited by national territory;

That in view of these remarks, Google proposes in subsidiary order to limit any orders for cessation to the obligation of Google:

• only to remove the visible cache links to the search engine of the site www.google.be (hyperlink accessible through the reference "cached"):

  39

• - On the pages of newspaper publishers that are members of Copiepresse that the judgement shall precisely identify them by name and the sites on which they are accessible,
• - To remove the titles and extracts of press articles from newspaper publishers that are members of Copiepresse from the site news.google.be that the judgement should identify more precisely by their name and the sites on which they are accessible;

Considering that the decision objected against has concluded that the activities of Google.News and the use of the Google cache notably violate the copyright laws and related rights (1994) and on databases;

That the decision must be confirmed in that it establishes damage, through its activities, to copyright law and related laws; That there is cause on the other hand to reform the decision as it concludes damage to the law on databases (Copiepresse not being admissible to act on this basis);

That there is no cause to complete the conclusion made, the grounds for this decision sufficiently stipulating the incriminated activities;

Considering that as regards the cessation order the judgement orders that all articles, photographs and graphic representations from Belgian French and German-speaking daily press represented by Copiepresse be removed from the sites (Google News and "Google cache" under any name whatsoever);

That by removing the Google "cache", there is cause to hear the visible cache links removed (hyperlink accessible by the statement "cached");

That contrary to that maintained by Google the said publishers are easily identifiable, as Copiepresse included them on page 1 and 2 of the petition for seizure description of the publishers it represents;

That on page 33 of its summary submissions, Google maintains moreover having removed all the results of existing searching referred to the said sites listed by Google;

That the order may consequently be confirmed on this point;

Considering that as regards the limitation of the cessation order solely to the Belgian Google sites, i.e. Google.be and Google.news.be, Copiepresse maintains that the disputed press articles are not exclusively accessible on Google.be,

40

but also on google.fr and google.com; that this is confirmed by the report by the expert Golvers (p. 122 and 123 of the report);

That it is apparent from the submission of Google that in order to exercise the order, Google stipulates of having no other choice but to exercise a complete "dereferencing" of the disputed sites; That Google stipulates that for technical reasons the "dereferencing" of these sites from the Google News engine is implemented for all the geographical versions of the site, i.e. not only the news.google.be site but also all the other sites "news.google" under the other domains (.fr, .com, .it, ...) (see page 34 of the additional and summary submissions by Google);

Consequently, that the precision Google requests be made to the decision appears without interest;

Publication:

Considering that there is cause to conclude that Google, no longer requests the reformation of the ruling opposed in relation to the order of a publication measure;

That Google solely requests that it be established that this measure no longer has an object as the said publication has already been made;

That it does not in any event appear to us, in the framework of this procedure in opposition, to assess the measures taken with a view to the execution of the ruling in opposition;

Considering that Copiepresse has, in a counterclaim, requested the publication of this order;

That Google opposes this claim, maintaining that there is no justification to order an additional publicity measure;

Considering that the publication measure in the framework of the cessation proceedings must actually contribute to the cessation and not solely to a repair measure (F. De Visscher and B. Michaux, op.cit. p. 508);

That in this case, the cessation order appears sufficient to have removed the damage;

41

That there is also cause to state that the order pronounced by default has already been the object of a publication;

That in view of these elements and also taking account of the coverage in the media of this case, there is no justification to order the publication of this decision;

Fines:

Considering that Google considers that the amount of the fines is disproportionate;

Considering that if the fines must be dissuasive, the amount must in any event remain fair;

That there is cause to reduce the amount of the fines as stated in the purview of this judgement;

With regard to the voluntary third party interventions:

Considering that in relation to the claim formulated by the SAJ and Assucopie, these request an order against Google to remove "all articles, photographs and graphic representations of the plaintiffs" from their sites, without further precision;

Considering that Google maintains that their claim should be limited to clearly identified works and authors and to the sole prohibition of referencing them in their visible "cached" links, accessible to the public, on www.google.be and to the reference on the site news.google.be;

That Google consequently proposes, in subsidiary order, to limit the cessation to the obligation to remove the visible cache links from the search engine site www.google.be (hyperlink accessible by the reference "cached") to the pages or documents of authors for which the plaintiffs in voluntary third party intervention justify that they hold the rights or for whom they have the capacity to act for cessation of a damage to these rights;

Considering that there is cause actually to state that the SAJ and Assucopie remain in default of drawing up a list stating their members; That if they have documents to this effect, this relates notably to a list of members of Assucopie that this is not exhaustive; That it is moreover difficult to find lists of members of the SAJ;

That it may moreover be concluded that in their submissions the SAJ and Assucopie have not shown any concrete

42

damage to the copyright of any of their members (which does not, as stated above, prevent them from acting);

That there is finally cause, to state that their repertory is variable;

That in this context, the detail requested by Google according to which the cessation order be limited to the pages or documents by authors for whom the plaintiffs in voluntary third party intervention justify holding the rights;

That there is cause to accompany this precision with a system of notification by the terms of which it is the responsibility of the voluntary third parties intervening to inform Google, by email at the address given by Google, of the identification of the work concerned with the proof that they belong to their list and to charge Google to remove this work within 24 hours of this notification under penalty of a fine;

That it is consequently Google's responsibility to communicate the email address to which these notifications must be sent within 8 days of the notification of this order;

FOR THESE REASONS,

We, Magerman, Judge appointed to replace the Chairman of the Court of First Instance in Brussels,

Assisted by Wansart, appointed deputy clerk of the court;

Considering the Act of 15 June 1935 relating to the use of languages in court cases;

Pronouncing judgement in a defended action;

Dismissing all additional or contrary conclusions;

State that there is cause to separate this case from the claim formulated by the s.a. Pressbanking;

Reserve judgement in relation to this claim;

Pronouncing judgement on the opposition:

Declare the opposition admissible and partially founded;

43

Consequently, confirm the ruling under opposition with the sole amendments:

1. that it concludes that the activities of Google News and the use of "Google cache" violate database law;

Declare that the original claim of Copiepresse in that it is based on the laws on databases is inadmissible;

2. that the amount of the fines is set at € 1,000,000 [euros] per day of delay in relation to the judgement against Google to remove from all these sites (Google News and Google "cache" under any name whatsoever) all the photographic and graphic representation articles and at € 500,000 [euros] in relation to the order concerning the publication of the judgement;

Set the amount of the said fines at an amount of € 25,000 [euros] per day of delay;

Pronouncing judgement on the claims formulated by the voluntary intervening third parties:

Acknowledge in relation to the scrl Société Multimédia des Auteurs des Arts Visuels (SOFAM) that they withdraw from these proceedings;

Acknowledge in relation to the company under French law Société Civile des Auteurs Multimédia (SCAM) that they withdraw from these proceedings;

Acknowledge in relation to the company Google that it accepts the withdrawal from the proceedings of the scrl Société Multimédia des Auteurs des Arts Visuels (SOFAM) and the company under French law Société Civile des Auteurs Multimédia (SCAM);

Declare the claim by the company SAJ and Assucopie inadmissible as it is based on database law;

For the remainder of the claims formulated by the company SAJ and Assucopie:

Declare the claims admissible and well founded to the limits as stipulated hereafter:

44

- Conclude that Google may not claim any exception as stipulated in copyright law and related rights;

- Conclude that the activities of Google News (i.e. the reproduction and communication to the public of the titles of articles and the short extracts from articles) and the use of the Google "cache" (i.e. the registration accessible to the public of the so-called "cache" memory of articles and documents) breach copyright law;

- Order Google to remove from all these sites (specifically from Google News as well as the visible cache links in relation to the search engine Google web) all articles, photographs and graphic representation of authors for whom the voluntary third parties intervening justify that they hold the rights;

- State that it is the responsibility of the plaintiffs in voluntary third party intervention to inform Google, by email at the address given by Google, of the identification of the work concerned with the proof that they belong to their list and to charge Google to remove this work within 24 hours of this notification under penalty of a fine of € 1,000 [euros] per day of delay;

- State that it is consequently Google's responsibility to communicate the email address to which these notifications must be sent to the plaintiffs in the third party intervention within 8 days of the notification of this order;

Order the plaintiff in opposition to pay the legal costs estimated for the scrl Copiepresse at € 121.47, for SOFAM at € 121.47, for SAJ, SCAM and ASSUCOPIE together at € 121.47 and for itself at € 295.47 + € 121.47;

Reserving the costs for the s.a. Pressbanking

So judged and pronounced at the public hearing of 13 February 2007.

[signatures]

Wansart Magerman

45 (last page)

Direct and ordain to all officers of justice charged with putting into effect the present judgement, the present ruling;

To our heads of the prosecution departments and public prosecutors in the Courts of First Instance, to uphold this judgement and all commanders and officers of the police forces to give whatever assistance may legally be required of them;

In witness whereof the present judgement, the present ruling has been sealed with the seal of the Court.

A certified true copy,

On behalf of the chief clerk of the Court,

VAN WYNENDAELE
Clerk of the Court

COURT OF FIRST INSTANCE
OF BRUSSELS

Date : 15/02/07

JBC no 7964

44 pages x 2.85 EUR

Dues paid : 125.40 EUR

The Clerk of the Court

(signature illegible)

Certified a true translation from French into English,

L. VANPARIJS, Sworn Translator.