Wheat v Alphabet Inc / Google LLC & Anor [2018] EWHC 550 (Ch)

[2018] EWHC 550 (Ch)Case No HC-2016-002540
IN THE HIGH COURT OF JUSTICE
CHANCERY DIVISION
Venue Royal Courts of Justice, Rolls Building, Fetter Lane,Date 26/03/2018
London EC4A 1NL
Before
CHIEF MASTER MARSH
Between
Christopher WheatClaimant(1) Alphabet Inc./Google LLCDefendants(2) Monaco Telecom S.A.MDefendant
Claimant in person for in personJaani Riordan (instructed by Pinsent Masons LLP) for Google LLCHearing Hearing dates: 6 February 2018
Judgment ApprovedChief Master Marsh:
[1]This judgment arises from the hearing of the claimant’s two applications for permission to serve the first defendant (“Google”) out of the jurisdiction. It should be read alongside my judgment (“the MT Judgment”) handed down on 12 December 2017 in which I set aside an order giving permission to serve the Second Defendant (“Monaco Telecom”) and declaring that the court does not have jurisdiction to hear the claim against Monaco Telecom.[2]It is necessary briefly to summarise the procedural history of the claim. The claim was issued on 2 September 2016. On the same day the claimant applied for permission to serve the claim out of the jurisdiction on Google. (He appears to have been under the misapprehension that he did not need to obtain permission to serve Monaco Telecom). On 9 September 2016, the application came before Deputy Master Arkush on paper. He ordered that the application should be served on Google. He did not make an order granting permission to serve the proceedings out of the jurisdiction on Google as the claimant appears to have believed. However, the claimant then purported to serve the proceedings on both parties by post in late December 2016. Service on that date was ineffective because, in both cases, permission was required and had not been granted. On 4 January 2017, Monaco Telecom’s solicitors wrote to the claimant notifying him of the defects in his purported service on their client. They also asked in a letter dated 12 January 2017 to be notified in advance of any steps to be taken in the claim.[3]On 16 January 2017 the claimant applied for permission to serve the proceedings on Monaco Telecom. The application was lodged at court but no notice was given to Monaco Telecom’s solicitors. I reviewed the application ‘on paper’ and made an order refusing it but giving the claimant an opportunity to renew the application at an oral hearing. In addition, the order required the claimant to “show cause why the claim and particulars of claim should not be struck out for failing to comply with CPR 16.4(1)(a) and Practice Direction 16 paragraph 1.4 and failing to show in a clear and comprehensible manner reasonable grounds for bringing the claim”. The claimant filed amended particulars of claim on 26 January 2017 and on 30 January 2017 appeared before me at a without notice hearing. An order was made granting the claimant permission to serve the proceedings on Monaco Telecom and extending the time for service of the claim to 1 September 2017. Monaco Telecom was given an opportunity to apply to set aside the order within 21 days of service.[4]The claimant did not send the claim form and particulars of claim to Monaco Telecom until 19 July 2017. The version of the particulars of claim he served was one dated 6 February 2017. Thereafter, Monaco made its application.[5]At the hearing on 30 January 2017, the claimant told the court that his claim against Google was already proceeding, permission to serve out of the jurisdiction having been granted by the Deputy Master. This was a repetition of information contained in his application notice. The information he provided was not correct because the Deputy Master had simply required that the application be served on Google. There is nothing in the Deputy Master’s order to suggest that permission was being granted to serve the claim out of the jurisdiction. The error was material because the grant of permission to serve Monaco Telecom out of the jurisdiction was made on the basis that it was a necessary party to the claim. It is very unlikely permission would have been granted had the court known permission to serve out of the jurisdiction on Google had not been obtained.[6]At the hearing on 30 January 2017, the claimant was also given permission to amend his particulars of claim and he was directed to serve the amended claim on the first defendant. On 31 August 2017, just before the extended period for service was due to expire, the claimant issued his second application seeking permission to serve Google out of the jurisdiction. It is that application, together with the application issued on 2 September 2016, that eventually came before me on 6 February 2018.[7]The claimant has made various attempts to serve the claim on Google and he has also sent Google the particulars of claim in its original and amended versions. On 31 July 2017, Google received, via the RCJ Foreign Process Section, the claim form and amended particulars of claim and the claimant has requested the court to deem Google served with the claim despite permission to serve out of the jurisdiction not having been obtained. It is beyond doubt that Google has not been lawfully served with the claim and it has neither treated service as having been validly effected nor submitted to the jurisdiction of this court by any action it has taken, including appearing to oppose the claimant’s applications. If the claim against Google is to proceed, the claimant must obtain permission to serve the claim out of the jurisdiction and an order extending the time for service, or an order deeming service as having been effected.[8]Thus far I have used the shorthand “Google” to describe the first defendant. However, as the first defendant is named in the claim form, it comprises two entirely separate legal entities, Alphabet Inc and Google Inc. Two preliminary points can be dealt with at this stage. First, the claimant has no claim against Alphabet Inc and it should not have been joined to the claim. It will be struck out from the claim form and any purported claim against it will be struck out. The claimant does not consent to this order, but he has made no case contrary to it. Secondly, Google Inc has recently changed its name to become Google LLC. My understanding is that despite the change of legal status, it is the same entity and at the hearing I gave permission to amend the claim form accordingly. The net result is that the claim is pursued solely against Google LLC (“Google”) and the applications will be treated as seeking permission to serve that entity out of the jurisdiction.[9]The claimant is the founder and proprietor of a business operated through a website named “theirearth.com”. The business is based in Redhill, Surrey and it is hosted on the internet on a server network in Germany hosted by Knipp using a single dedicated IP address. When the business was first set up, the site was hosted by Monaco Telecom. The business was created to be both an original content news media site, with original journal articles and photographs, and a directory to other popular websites. The claimant says he is the owner of the copyright work that is hosted on the site comprising 782 articles written by him and 398 photographs taken by him. For the purposes of the application, he can be taken to be the author of the website content and its first owner. He is a British citizen and accordingly the entire copyright work qualifies for protection under s.154(1)(a) of the Copyright, Designs and Patents Act 1988 (“CPDA”). This is subject to one gloss, namely that when the website was first broadcast the claimant was resident in Monaco and the works were first broadcast in Monaco. The claim does not assert that the copyright works were created when the claimant was resident in the United Kingdom or first broadcast in this jurisdiction.[10]The claimant had great ambitions for the business he set up. He says the concept and approach he adopted proved to be popular. It gained the interest of Daimler and the claimant intended it to be a rival to Google. He puts the loss he seeks in this claim at £30 million based on the potential the website had. He says his ambitions have been thwarted by the defendants.[11]The claims are not easy to discern from the amended particulars of claim. The basis of the claim against Monaco Telecom is set out in the MT Judgment and it is unnecessary it repeat it here. As an entry point to the claim, the causes of action are summarised in paragraph 1.6 of the amended particulars of claim: [It is reproduced in lower case] “A.) Copyright infringement: Google indexing a duplicate of theirearth.com on Monaco-Telecom servers and diverting google search requests to the Monaco-Telecom copy. Copyright infringement and conversion: Monaco-Telecom broadcasting an unauthorised duplicate of theirearth.com B.) Copyright infringement: Google indexing theirearth.com copyright content on connect.in.com and sending google search requests to the connect.in.com copies of theirearth.com content C.) Copyright infringement: Google indexing theirearth.com copyright content on blogspot.com and sending google search requests to the blogspot.com copies of theirearth.com content D.) Copyright infringement: Google favouring the hotlinking of website content vs. sending search requests to the original copyright content sites E.) Breach of contract: - Google AdSense advertising relationship with theirearth.com and the diversion of Ad revenue F.) Breach of a duty of care – Causing mental distress – injurious actions and omissions directed towards the claimant and committed by both defendants”.[12]In summary, the claimant has four main types of claim he wishes to pursue against Google: i. The first head of claim is based on the premise that Monaco-Telecom made an infringing copy, or a duplicate, of the claimant’s website and Google has indexed that site in preference to the claimant’s website. As a consequence, it is said Google infringed the claimant’s copyright or, in the alternative, Google is liable by having acted in a common design with Monaco-Telecom. This head of claim is dealt with in the MT Judgment. ii. Third parties such as connect.in.com have hotlinked to the claimant’s photographs on his website and Google has indexed these hotlinked images and thus directed traffic to these infringing sites and away from the claimant’s site. iii. A similar claim is made in relation to content on blogspot.com, which is a platform which allows its users to publish blogs. User content (text, photos and videos) is hosted by Google, normally on the blogspot.com domain. Unlike Google Search, the content is stored on Google’s servers. iv. Google is in breach of the terms of its Adsense On-line Terms of Service contract by diverting advertising revenue away from the claimant to other websites.[13]The claimant says, in addition to the claims in copyright and contract, Google owed him a duty of care at common law not to cause him harm and he alleges its actions have caused him mental distress. However, this cause of action does not have any legal foundation and both the hearing of Monaco-Telecom’s application and the applications relating to Google have proceeded on the basis that there is no prospect of the claimant establishing a relationship with either entity such that a duty of care not to cause mental distress could arise.[14]The claimant has provided a report from a psychiatrist, Dr Bose, and based on that report, the claimant says he lacked the mental capacity to bring this claim from January 2009 until 1 September 2010.[15]In his skeleton argument for the hearing on 6 February 2018 the claimant put forward a claim in competition law. I did not permit him to develop it because it does not form part of his claim as set out in the amended particulars of claim.[16]At its heart, the claimant complains about what he describes as the ‘hijacking’ of the claimant’s copyright content on his website, particularly his images displayed there. Examples of such images include photographs taken by the claimant of Lewis Hamilton and Novak Djokovic. The effect of this ‘hijacking’ by third parties was to considerably reduce the number of hits on his website with the consequence that advertising revenue that would have been earned by the claimant was diverted elsewhere.[17]The claimant explains his case in relation to hotlinking in this third skeleton argument: “The word “Hotlinking” appears extensively in this Claim, specifically as it relates to the claim against [Google]. By definition, hotlinking is embedding code onto an HTML website page (say hotlinkingwebsite.com) that uses absolute URLs to refer to images hosted on other servers <img src=http://www.example.com/picture.jpg/>. When the Internet browser on a user computer downloads an HTML page containing such an image, the browser will directly contact the remote server for www.example.com and request the image content and load it into the requesting HTML page content. Where copyright infringement occurs is when Google’s search engine formula algorithm routinely transposes the ownership of the copyright image from the copyright owner website example.com to the page at hotlinkingwebsite.com such that Google now shows image search results for the popular keyword associated with the copyright image as belonging to hotlinkingwebsite.com and sends search request traffic to the hotlinkingwebsite.com server. It is clear Google promotes this behavioural characteristic to transpose ownership from the original copyright owner using their search engine algorithm, and the subsequent Google creation of defective search results as they have done little since 2007 when this rebranding was first observed on Google and have raised the profile of enterprise hotlinkers like connect.in.com to Page Rank 6 and hotlinking as many as 28 million images from origin websites (sic) within the United Kingdom and in other parts of the world. Google could easily have corrected this Google Engineer programmed algorithmic characteristic but they have chosen not to in the years 2010, 2011 and 2012 when it became a substantial practice by enterprise hotlinkers as identified in the Claimant’s claim and the Claimant argues was only prolific because Google desired this business landscape in their search business and connected Adsense business.” [my emphasis][18]The language used by the claimant to describe the effects of hotlinking concerns ownership. Elsewhere he has described the effect of hotlinking as leading to a “re-assignment” of the copyright. He has said that this effect is either the result of a ‘bug’ within the algorithms used by Google or that Google knew of the effect and designed the algorithms to achieve the effect deliberately. There is, however, no evidence at all, direct or inferential, of a deliberate design to achieve the effect he describes. The high point of his case is that Google was aware in 2007 of the effect he complains of but did not fix the bug. He says this was because it worked in Google’s favour by popular images being accessible on ‘aggregator’ sites and by advertising revenue for Google being enhanced as a result. The claimant summarises his breach of copyright claim in the following way:
“… this Claim breaks new ground in a developing area of law related to what is copyright infringement of the copyright holder’s property within the United Kingdom on the internet and what is fair use of that copyright content. The Claimant would argue that a systemic practice by Google by the redirection of copyright content to copies of that content on aggregator websites is not remotely close to ‘fair use’ as it defrauds the original copyright holder of the property value of that media and its creative revenue generating capacity.”
[19]The claimant relies upon two principal sources of evidence to support his claim in addition to his own observations of the effects on his website caused by hotlinkers such as connect.in.com. He refers to entries in a Google Product Forum called the Webmasters Central Help Forum from 2011. A number of entries discuss the work of “image scrapers” and “image link farms” using “proxy and web hijacker techniques”. The forum records a post by Gary Illyes, who the claimant says is a Google employee, saying he will forward the posts to the appropriate teams. The claimant relies on Mr Illyes’ knowledge of the steps taken by hotlinkers as being knowledge by Google. Mr Illyes asks how those posting entries in the forum find the hijacked images. A contributor who is described as ‘loupiote’ says on 12 June 2011:
“the largest image link farms are sites such as connect.in.com and zimbio.com, but there are a lot of smaller image-link farm that manage to hijack a lot of image search traffic (via google-image), a least 10 on my image traffic is being hijacked by image farms that hotlink my images (I can see that from my logs) and I estimate that the traffic hijacked by copied images (ie actually copied content with my watermarking cropped off) is much higher, given the number of copies of my images that i have located … and that image search engines lists in their results I file 50 DCMA notices every week, but it’s like a wack-a-mole game, because I have 10,000 images, and the highest ranking ones have been duplicated and are hosted by about 50 sites (for each image).”
[20]In addition, the claimant relies on a blog by Thomas Kaiser of Forecheck entitled “Image Hijacking in Google Image Search: How it works and how to protect” dated 1 December 2015. Mr Kaiser describes image hijacking as an issue with a long history going back to 2007. The blog describes what happens when a third party hotlinks images and how search requests are diverted to another website. It is unnecessary to refer to the blog in detail save to remark that it illustrates first, that it is third parties, not Google, who are ‘hijacking’ images and, secondly, the problem has at least been partially solved by changes to Google’s algorithms.[21]Google’s evidence is provided in a helpful witness statement made by Robert Graham who is a solicitor with Pinsent Masons LLP. The claim has to put in the context of the explanation given in that statement about how Google Search works.[22]Mr Graham explains “hotlinking” in the following way:
“Hotlinking” (which is also known as inline linking) is a process whereby a website (“the first website”) displays a linked object, often in the form of an image, stored on the server hosting the content of another website (“the second website”). Hotlinking does not involve any copy being made of the image belonging to the second website. Instead, when a user visits a webpage on the first website containing, by way of example, a hotlinked image, the HTML code of that first website instructs the user’s browser to display the image directly from the server on which the content for the second website is hosted. The advantage of this approach from the first website operator’s perspective is that the transmission overhead needed to deliver the hotlinked image to the user is drawn from the second website’s resources, and not those of the first website.”
[23]Mr Graham sets out in his witness statement a no doubt simplified explanation of Google Search:
“How Google Search works 14. Google Search currently processes over 1.2 trillion search a year worldwide. This equates to 3.5 billion searches a day, and on average 40,000 searches every second … Google Search allows users to quickly access relevant information from billions of webpages on the web. Other search engines provide a similar service include Yahoo, Bing, Yandex and Baidu. 15. It would not be possible to navigate the web effectively without search engines. The internet is made up of over 130 trillion individual webpages and is constantly growing. According to www.worldwidewebsize.com as at 6 November 2017, the estimated number of indexable webpages on the internet was in excess of 45 billion. 16. In addition to searching for webpages, users of Google Search can undertake searches for images. Links to images published by third parties may be returned through Google Search results in a number of ways. In some instances, where a user carries out a search, thumbnail images may be returned as part of the results page, in addition to the usual text-based links to webpages. 17. When the user clicks on the “thumbnail” of an image on Google Search, this will open a further page in Google Search which identifies the domain name or website on which the image is published by the third party who operates the domain name or website. The user then has the option of either visiting the URL for the web page at which the image appears or navigating to the URL of the image file in its native size on the third party’s website. … Indexing Indexing 19. It is not possible for Google Search to search every web page available on the Internet in real time, and deliver results in a timeframe that would be acceptable to users. Google Search therefore compiles an index of the content of webpages, and it is this index that is examined during the search process. 20. To generate the index, software known as a “web crawler” (e.g. the “Google bot” web crawler) is used to find content that is on publicly available webpages. If the website owner does want content to be indexed by Google, the website owner can use standard techniques to ensure this … Google’s web crawler sends requests to servers hosting webpage content. If the website owners have configured these servers to respond to such requests, then the requested content will be sent to Google’s servers for indexing. Many thousands of requests are made simultaneously to populate Google’s index with information from the webpages that are being crawled. 21. The content of webpages examined by Google’s web crawler is saved and stored in a cache. The caching process of each webpage is automatic. In relation to images, a thumbnail copy of each image examined by Google’s web crawler is saved in a cache. The caching process of images is also automatic. The contents of webpages, including images, are stored in the cache only for a limited period of time so that they can be displayed rapidly when the web page is returned as part of Google Search results. The cache is rapidly updated at each exploration of the web by the web crawler to ensure that Google results reflect the evolution of webpages published online to provide users with an up-to-date mapping of the Internet. Google ensures the regular updating of the cached content by its web crawler in a timeframe which is consistent with the current state of the art and with industry practice and which is flexible depending on the importance of the page and the frequency with which it tends to be updated. The systems powering Google Search therefore only store cached content temporarily. 22. The web crawler does not cache content from webpages which are not generally accessible. The web crawler does not visit any webpages where the webmaster has instructed Google not to index its website as I described further below; it indexes webpages only where servers being configured to respond to its requests. Given that the web is made up of over 130 trillion individual webpages, it is evident that webpages indexed by Google which form Google’s index, a number in excess of 45 billion, represent only a proportion of the cached web. The caching is carried out with the implicit authorisation of the website publishers, since the act of publishing the content without restrictions on access implies that they agree that the information will be available to all including search engines such as Google Search. 23. The purpose of the cache storage and transmission is to facilitate the transmission of the content between the “recipients of the service” (as defined in Article 2 of the E-Commerce Directive and regulation 2 of the E-Commerce Regulations), ie between publishers of information on the internet and web users searching for content. The caching operations are necessary to optimise the transmission of Google Search results. Extracting and reviewing content in real time, from multiple sources and from servers located all over the world, for each individual search conducted, is not feasible from a technical standpoint. These caching operations therefore reduce the amount of data that must be transmitted over the web when undertaking searches using Google Search. The purpose of caching operations is to optimise and accelerate the flows of data over the relevant networks used by web users and Google, insofar as the cache copies will allow Google Search to display instantly the results corresponding to the queries of web users. 24. Once a search query is submitted by a user, Google Search results are then ranked in order of relative relevance to the user’s search query based upon the content of Google’s index. For a typical query, there may be thousands of webpages, or many more than that, with potentially relevant information. Accordingly, Google Search uses algorithms which rely on numerous signals to return results relevant to the query, ranked by their potential relevance. These signals include factors such as how often content on a website has been refreshed and the quality of user experience provided by a particular webpage. 25. One of the signals used by Google to rank potential relevance to the user’s query in the results returned by Google Search is known as “PageRank”
. Pagerank works by counting the number and quality of links to a particular webpage from other webpages, to determine how important the particular webpage is. The underlying assumption behind this signal is that more important websites are likely to receive more links from other webpages and, in turn, that the more important the webpage, the stronger its potential relevance to the user making the query, compared to other webpages returned in response to the search in question. 26. Google’s temporary caching of webpages is undertaken in a neutral, technical and automatic manner, in that Google Search’s algorithms do not modify the cached webpages in respect of their content. Google Search may re-encode the content with a view to ensuring the compatibility of the content with the web user’s display interface. This technical operation does not modify the content in any editorial sense, but simply presents the content for display as a list of results. At the same time, the algorithms access the content by following the technical instructions of the webpage in question (such as robots.txt directives and other similar instructions), so complying with the webpage publisher’s conditions of access. Control of indexing 27. As mentioned above, website publishers are able to control whether Google and other search engines index their content in this way by use of multiple technical means, including through the source code of their websites, using meta-tags or by using “robots.txt” conventions to indicate that particular content should not be crawled and indexed by search engines. Use of robots.txt and similar tools to control indexation is a standard and well-accepted practice among webmasters… 28. The robots.txt file contains specific instructions to web crawlers in the top-level directory of their web server. Webmasters can use this file to instruct any web crawlers (or specific crawlers such as Google’s) not to visit webpages within the website in question. The robots.txt file for any particular website can be viewed by entering the URL to the website’s front page and then adding the words “/robots.txt”. So, for example, it is possible to review the robots.txt file for Mr Wheat’s website by visiting “theirearth.com/robots.txt”. … the robots.txt of Mr Wheat’s website is configures as follows:
“User-agent: * Disallow” 29. “User-agent:* means that the instruction which follows applies to web crawlers visiting Mr Wheat’s website. “Disallow” has been left blank (without any text following the colon), which is an instruction to all web crawlers that all pages of Mr Wheat’s website should be indexed. 30. To prevent Google from indexing his website Mr Wheat could, at any time, configure the robots.txt file on his site to read: “User-agent: Googlebot Disallow:/”. 31. There are other methods that webmasters can use to prevent web crawlers from indexing websites, including implementing a “noindex” instruction. This instruction can either be included as a meta-tag within the HyperText Markup Language (HTML) code for a website or in an “HTTP response header” which instructs web crawlers not to index a website or parts thereof. 32. Webmasters therefore have full control over whether Google indexes their website. In general, most webmasters want to have their site indexed by web crawlers so that the website can be found by Google and other search engines. If a website is not indexed it cannot be located through search engines.”
[24]There are four further points that can be drawn from Google’s evidence: i. Google says it is an “information society service” within Article 13 of the E-Commerce Directive (“the Directive”) and regulation 18 of The Electronic Commerce (EC Directive) Regulations 2002 (“the Regulations”). This is not accepted by the claimant although he has not provided a clear basis for the position he adopts. ii. Google operates a cache removal procedure either pursuant to a court order or order of another authority, or pursuant to a non-judicial request. iii. Google is normally unable to remove webpages from the internet. iv. In the case of Blogger where content is hosted by Google, a copyright owner may request copyrighted material to be removed. Mr Graham says that in appropriate cases Google will block or remove access to the content. “Where content on Blogger is removed from the internet on a global basis, the relevant URL will not form part of the body of indexable content on the internet, from which Google’s index is formed, so the URL will no longer be returned in results on Google Search once the relevant domain has been re-crawled and re-indexed.”[25]In order to obtain permission to serve out of the jurisdiction, the claimant must satisfy the three tests set out by Lord Collins JSC in Altimo Holdings and Investment Ltd v Kyrgyz Mobil Tel Ltd [2012] 1 WLR 1804 [71]. He must show that: i. He has a good arguable case that his claim falls within a relevant gateway to the court’s jurisdiction. ii. There is a serious issue to be tried on the merits that has a real prospect of success. iii. England and Wales is clearly the most appropriate forum in the claims can be tried most suitably for the interests of all the parties and for the ends of justice, and in all the circumstances the court ought to exercise its discretion to permit service out of the jurisdiction.[26]In Four Seasons Holdings Inc v Brownlie [2017] UKSC 80 these requirements were reconsidered by the Supreme Court although all the comments of the court on matters of jurisdiction were strictly obiter dicta. There are three points of emphasis arising from the majority judgments in Four Seasons v Brownlie that are relevant to this claim. i. There is no requirement for a further gloss to be placed on the question of whether the claimant has a good arguable case. It is incumbent on the claimant to show a plausible evidential basis for the application of the relevant jurisdictional gateway and, if there is an issue of fact, the court must take a view on the material available if it can reliably do so. If no reliable assessment can be made, then a plausible evidential basis will suffice to go on to the second limb. ii. For the purposes of PD6B para. 3.1(9), damage is to be given its ordinary and natural meaning and therefore includes all the detriment (physical, financial and social) that the claimant suffers as a result of the alleged wrongdoing. This can be wider than the damage that completes the cause of action. iii. All the jurisdictional gateways require a “substantial and not merely casual or adventitious link between the cause of action and England”.[27]The claimant relies on four jurisdictional gateways: i. Paragraph 3.1(7): a breach of contract committed within the jurisdiction. ii. Paragraph 3.1(9): a claim in tort where damage has been or will be sustained within the jurisdiction or results from an act omitted or to be committed within the jurisdiction. iii. Paragraph 3.1(11): a claim relating wholly or principally to property within the jurisdiction. iv. Paragraph 3.1(16): a claim in restitution.[28]Only the first two gateways merit detailed consideration. That is because gateway (11) adds nothing to gateway (9) (for the reasons explained in the MT Judgment) and gateway (16) is not engaged because there is no claim in restitution. Although in paragraph 6.2 of the amended particulars of claim the claimant says he “… seeks restitution of at least £20,000,000 for the damages Google’s series of copyright infringements had to the theirearth.com business …”, the claim is for damages flowing from the defendants’ activities. It is not a claim in restitution in the legal sense. Breach of contract[29]Gateway (7) can conveniently to deal with it first. The claim in contract is set out in paragraph 6 of the amended particulars of claim. It is said that Google is in breach of its AdSense contract. AdSense is an advertising programme run by Google by which a webmaster permits Google to place advertisements on a website and in return the webmaster receives payments from Google.[30]Clause 2 of the contract provides: “By enrolling in AdSense, you permit Google to serve, as applicable,(i) advertisements and other content (“Ads”),(ii) Google search boxes and search results and(iii) related search queries … (each individually a “Property”). In addition, you grant Google the right of access, index and cache the Properties, or any portion thereof, including by automated means. …”. [ I remark that “serve” is not an obvious verb to use in UK English usage but it clearly means “provide” or something similar][31]Pursuant to clause 5 of the contract, the webmaster is entitled to receive payment “… related to the number of valid clicks on Ads displayed in your Properties, the number of valid impressions of Ads on your Properties, or other valid events performed in connection with the display of Ads on your Properties, in each case as determined by Google.”[32]Clause 7 of the contract is headed:
“Intellectual Property; Brand Features”
. The claimant relies on the opening words of what is a lengthy clause:
“Other than set out expressly in the Agreement, neither party will acquire any right, title or interest in any intellectual property rights belonging to the other party or to the other party’s licensors.”
[33]Clause 7 goes on to provide the claimant with a non-exclusive, non-sublicensable licence to use software provided by Google as part of the AdSense package and a similar licence to use Google’s trade names, trademarks and so on.[34]Google initially asserted that the version of the contract agreed to by the claimant is subject to the laws of California and subject to the exclusive jurisdiction of the federal and state courts of Santa Clara County, California. However, it emerged that the claimant has since 2008 agreed to different versions of the contract and at least one version provides that English law is the governing law and that the parties submit to the exclusive jurisdiction of the English courts. The remaining terms of the contract are identical in each case. Importantly, in the English law version of the agreement, the contracting party is Google Ireland Limited, not Google Inc or Google LLC. At the date of the hearing Google Ireland Limited was not a party to this claim. If the claimant relies solely on that version of the contract, the mismatch between the contracting parties and the parties to this claim is sufficient to dispose of the contractual basis of the claim. The claimant has made an informal request to join Google Ireland Limited as party to the claim. I will return to this subject later.[35]The claimant complains that hotlinking websites (he also refers to them as “aggregator” websites) presented his images to visitors to their websites with Google’s assistance and this diverted a substantial amount of traffic from the claimant’s website. This had the effect, it is said, of reducing drastically the number of hits on the claimant’s website and had the corresponding effect of reducing the advertising revenue earned from the site. The term “aggregator” website is taken by Google to refer to “… websites which typically display content on specific topics or issues, and which collate that content from multiple third-party sources using largely automatic means, rather than creating original content of their own”.[36]An example might be a hotlinking site that focuses on Formula 1 motor racing. I choose that example because the Claimant is the copyright owner of a number of images of Lewis Hamilton. The images were placed on the claimant’s website and were available, without restrictions, for search engines such as Google to index, cache and then respond to searches relating that subject matter. An aggregator site was not prevented from hotlinking the claimant’s images and Google’s web crawler would index and cache the aggregator site. The complaint made by Mr Wheat is that subsequent searches looking for images about Lewis Hamilton were likely to favour the aggregator site and thus divert, as the claimant puts it, advertising revenue away from his site to the aggregator site.[37]The claimant’s case in contract has two strands to it. First, he says that Google knew it was “… diverting the ad revenue for [his] copyright content away from the Claimant and to the third-party aggregator sites which was a breach of contract … as they [Google] were indeed acquiring an interest in theirearth.com business without paying for it”. Secondly, he says Google “… made sure the keyword “theirearth” when entered as a search request delivered “do you mean “the earth””, and hence acted to curtail the look-up of the Claimant’s website, and also to reduce his advertising revenue.” He relies on a process of “deeplinking” within the Google Image Search product that extracted full size copies of the original content from the original website without forwarding the search user to that website.[38]The claim in contract has insuperable difficulties under both the first and second limbs of the test the court must apply. Put briefly this is because: i. If the UK version of the AdSense contract is relied upon, the claim in contract can only be pursued against Google Ireland Limited and it is not a party to this claim. The premise of the application made by the claimant is that he seeks permission to serve Google Inc (LLC) out of the jurisdiction. Even if Google Ireland Limited were to be joined it would not assist. And, in any event, it may sometimes be appropriate on a strike out application to enable a defect to be cured. In this case, I am dealing with an application first made when the claim was issued in September 2016 which has already been amended once. The application must be considered on its merits by reference to the amended particulars of claim. ii. If the US version of the contract is relied upon, although the contracting parties match those who are parties to the claim, the claimant cannot go behind the exclusive jurisdiction clause and he is not entitled to bring the contractual claim in England. iii. The contractual term relied upon is in a contract that relates only to Google AdSense, not Google Search which is the subject of the complaint. iv. It is impossible to construe the contractual terms relied upon by the claimant so as to create the obligations he relied on. The terms solely concern the parties’ intellectual property rights. The words relied upon by the claimant are declaratory (“neither party shall acquire any right, title or interest”) and are intended to clarify that no transfer of rights will take place as a consequence of entering or implementing the agreement. The short answer to the claim in contract is that the claimant’s intellectual property rights remain intact and, indeed, he brings the claim on the basis that he is the copyright owner. v. It is inaccurate to describe the alleged ‘diversion’ in terms of a reassignment of the claimant’s copyright, or a rebranding. He remains the legal owner of the copyright. His complaint concerns the consequences that arise from Google indexing the images on his website and the advantage taken by ‘hotlinkers’ of access to these images. He expressly agreed to the images being indexed and cached by Google for the purpose of data that was cached being located in response to searches. Copyright infringement[39]The balance of the claim is based upon the allegation that Google has infringed the claimant’s copyright in the photographs on his website. (It is unnecessary to analyse separately the claims relating to Google Search and those concerning Blogger). There are two general points to be made at the outset. First, the claimant’s case that Google’s actions have resulted in an ‘assignment’ (or some other proprietorial effect on) of his copyright to third parties does not sit easily with his claim to enforce rights as the copyright owner. Under section 90 CPDA an assignment must be in writing and no writing is relied upon by the claimant. I will proceed on the basis that the claimant does not use ‘assignment’ in a technical legal sense. Secondly, copyright is territorial. If gateway 3.1(9) is to be engaged, the claimant must show an infringement within the United Kingdom. An infringement in, say, Monaco will not suffice. This is despite the obiter views expressed by a majority in the Supreme Court in Four Seasons v Brownlie about the type of damage that is sufficient to engage the gateway – see Lady Hale’s judgment [35] to [66] with which Lord Wilson and Lord Clarke agreed. An infringement of copyright in Monaco, does not give rise to an actionable claim for loss incurred in England because of the territorial nature of copyright unlike a claim arising from a road accident in Egypt causing economic loss in England.[40]The legal context for the claims that assert a breach of copyright can be summarised briefly. Section 16(1) CPDA provides the copyright owner the exclusive right to do the acts described in the section within the United Kingdom. They include the exclusive right to copy or issue copies of the work to the public. Section 17 provides that copying means reproducing in any material form and this includes storing the work in any medium by electronic means.[41]Section 20 deals with infringement, inter alia, by the communication to the public by the work being “broadcast”. However, broadcast is a defined term and does not include broadcasting on the internet.[42]Secondary infringement in dealt with in section 23 CPDA which occurs where a person, without the licence or consent of the copyright owner, possesses in the course of a business an article “which is, and which he knows or has reason to believe is, an infringing copy of the work.”[43]Potential difficulties that might arise for search engines such as Google are intended to be dealt with by the Directive to which Mr Graham makes reference and the Regulations. Regulations 18 and 19 provide what is described as ‘safe harbour’ protection where an ‘information society service’ is provided. Regulation 17(1) provides protection against claims where, acting as mere conduit, an information society service is provided, the service provider did not initiate the transmission, did not select the receiver of the transmission and did not select or modify the information. Regulation 17(2) makes it clear that the acts of transmission may include the automatic, intermediate and transient storage of the information provided subject to certain conditions.[44]Regulation 18 provides a defence concerning caching where the transmission of information that is subject to “… automatic, intermediate and temporary storage …” and subject to a number of a number of additional detailed requirements that are not in issue in this claim.[45]Regulation 19 provide protection from a claim for damages for “hosting” where the service provider does not have actual or constructive knowledge of unlawful activity or information and upon obtaining such knowledge acts expeditiously to remove or disable access to the information.[46]The claimant’s complaint is readily understandable in general terms. The potential value in has website has been ‘hijacked’ by sites variously described as ‘image farmers’, ‘aggregators’, ‘hotlinkers’ and so on. He says that as a result of the way in which Google Search operates, by indexing and caching images, searches have been directed away from his website and he has lost advertising revenue. However, the claimant has tried to shoehorn the matters about which he complains into a legal framework they do not readily fit.[47]The claim against Google arising from the allegation that Monaco-Telecom made a copy of the claimant’s website is parasitic on the claim against Monaco-Telecom. The alleged act of copying was that of Monaco-Telecom. This element of his claim was not found to have a real prospect of success in the MT Judgment. Leaving that aside, the claimant’s pleaded case is based on secondary infringement by Google pursuant to section 23 CPDA “… by Google possessing in the course of business, exhibiting or distributing in public using the Google Search product the infringement which is, and which the Defendant knows or has reason to believe is an infringing copy of the theirearth.com body of work within England and the rest of the world …”.[48]It is not easy to follow the claim that Google has exhibited or distributed the claimant’s copyright content in the United Kingdom since the content he refers to was hosted in Monaco. In any event, the acts referred to in section 23 CPDA are tangible acts of physical exhibition or distribution, not electronic transmission of information. None of them are engaged by the acts the claimant complains of.[49]There are numerous other difficulties faced by the claimant with the claim concerning the Monaco-Telecom copy. They include: i. It is common ground that hotlinking is not an act of infringement. In any event, it is also not clear how the claimant could establish ‘knowledge’ or ‘reason to believe’ on the part of Google in light of the evidence concerning indexing and caching. The process Mr Graham explains is fully automated. A bug in Google’s algorithms will not suffice, even assuming that such a bug exists and assuming in the claimant’s favour that Google could have resolved the bug more speedily. ii. There is no evidence of a third party accessing the alleged copy of the website in the United Kingdom. He is therefore unable to show infringing acts within the jurisdiction. iii. There is no evidential basis for the claimant to suggest that Monaco-Telecom and Google were joint tortfeasors; see the MT Judgment [34]. iv. Both versions of the website hosted by Monaco-Telecom were set up without limiting access by web crawlers to index and cache their content. Google says that the absence of any limitation can be seen as consent to search engines such as Google to index and cache the content of the site. Although the absence of any limitation is passive, it does seem to me to be right that it amounts to implicit consent in the context of the internet. This is because(a) webmasters largely rely upon the ability of search engines to index and cache the contents of their websites to enhance their ‘visibility’ and(b) the straightforward means by which web crawlers can be turned away. It is right to mention that Mr Wheat referred me to the decision of a court of first instance in Brussels in SCRL Copiepresse v Google Inc where the court reached the opposite view, but that decision is not binding on me and, in any event, the reasoning for the relevant part of the decision is unclear. v. Google is entitled to the benefit of the safe harbour caching defence under regulation 18 of the Regulations. vi. Google’s evidence is that the last links to the www.theirearth.com webpages were provided to Google Search users on 10 August 2009. Monaco-Telecom’s evidence, which was accepted in the Monaco-Telecom judgment is that the website was deleted from its server in about March 2009. Both dates are more than 6 years before this claim was issued and any claim in relation to indexing the Monaco-Telecom copy of the claimant’s website would be statute barred pursuant to section 2 of the Limitation Act 1980. I do not need to add to the remarks made in the Monaco-Telecom judgement concerning the medical evidence of Dr Bose about the claimant’s capacity.[50]There are also grave difficulties with the hotlinking claim. The real complaint lies not against Google but the hotlinkers themselves. It is their actions that have devalued the claimant’s website by taking advantage of the way in which Google Search operates.[51]The claimant’s case in copyright has not been clearly articulated. For the reasons that follow, he has come nowhere near discharging the burden on him that he has a good arguable case this part of his claim falls within gateway (9) (or (11)). i. There is no evidence that the hotlinkers have any connection with the United Kingdom. Connect.in.com, for example, is based in Mumbai. Mere accessibility of a website from the United Kingdom in insufficient to confer jurisdiction or to localise communications of that website operator within the United Kingdom – see EMI Records v British Sky Broadcasting Ltd [2013] EWHC 379 at [48] Arnold J and Omnibill (Pty) Ltd v EGPSXXX Ltd (in liq) [2014] EWHC 3762 at [11] Birss J. ii. Hotlinking does not involve an act of copying. This is accepted by the Claimant. iii. It is not open to the claimant to contend that there has been an infringement by communication to the public that is in breach of section 169(1)(d). The doctrine laid down by the ECJ in Case-466/12, Svensson, EU:C:2014:76, [24]-[28] and Case 160/15, GS Media BV, EU:C:2016:644, in [41][44] that where photographs have been made freely available with the consent of the right holder, the copyright owner cannot later complain that third parties have linked to or embedded those works from their own websites. iv. The complaint that Google has prioritised one website over another does not give rise to a claim for copyright infringement. v. There is no arguable claim for secondary infringement. vi. Equally, there is not a good arguable claim that Google is a joint tortfeasor with others. There is simply no evidence to show any connection between the automated process carried out by Google and the hotlinkers that would suffice to create a common design or give rise to procurement. vii. Google is entitled to take advantage of the safe harbour defence. Joinder of Google Ireland Limited[52]No formal application has been made to join Google Ireland Limited in order to bolster the claim in contract. It seems to me that on an application for permission to serve out of the jurisdiction, the court must take the claim in its current form and it is not the occasion for considering how the claim might be improved. In any event, if an application were to be made permission to join Google Ireland Limited would not be granted because the claim does not meet the relevant threshold of showing reasonable grounds. The claim in contract has no merit whatever. Forum[53]If the claimant were able to show he has a claim which passes the first two tests, he would still have to show that England is clearly the most appropriate forum (see paragraph 25iii above). The claim’s links with England solely arise from the claimant’s current residence and business base here. The claimant’s website is hosted by Knipp in Germany. Against that: i. The services complained about by the claimant (Search, Blogger and AdSense) are provided by Google which is domiciled in California and its UK subsidiary does not operate the Google Search service. ii. The infringing acts, as far as they can be discerned, did not take place in the United Kingdom. iii. Evidence relating to the defendants is likely to be located in Monaco and California. Documents are likely to be found in those jurisdictions. iv. Witnesses, other than the claimant, are likely to be based respectively in Germany, Monaco and California. v. One version of the AdSense contract is subject to the law of California and has an exclusive jurisdiction clause.[54]The connection with England, balanced by these considerations is slight and it is the least attractive of the competing jurisdictions. The claimant invites the court to take into account his lack of finances and his inability to afford to bring a claim other than in England. In my judgment, that is not a material consideration, but even if it were to be, it would not be sufficient to tip the balance in his favour. The court should decline to accept jurisdiction. Deemed service andExtension of time[55]The claimant’s second application also seeks orders that the claim be deemed to have been served on Google on 1 August 2017 in accordance with the Foreign Process service on that date and an extension of time in which to serve the claim form on Google. Time for service of the claim was extended on 30 January 2017 to 1 September 2017. The application to extend that period was issued the day before the period expired.[56]I can see no logical basis, having concluded that permission to serve out of the jurisdiction on Google should be refused, that an exceptional step should be taken by deeming service to have taken place. The claimant had no basis for attempting to serve the claim on Google through the Foreign Process section. Seeking permission to serve out of the jurisdiction provides an important filter to prevent claims that have no proper connection with England being pursued here. Having concluded that the claimant does not meet any of the three tests for the grant of permission to serve the claim on Google in California, it would be wholly wrong to take the exceptional step deeming service to have taken place.[57]Equally, it follows that there is no basis upon which the court should extend the time for service. The claim cannot be served on either defendant and extending the period for serving would be pointless. Conclusion[58]The claimant’s applications will be dismissed.

Cited in 3 later judgments