Browse By

Is Caching Copyright Infringement?

Update: The Copyright Amendment Act, 2012 (PDF), which is now in force, has introduced Section 52(1)(b) and (c). These two sections provide exemptions to intermediaries for the storage of transient copies of information. Clause (b) protects Internet Service Providers, while clause (c) protects ‘information intermediaries’ such as Google, Facebook etc. Therefore, liability for ‘caching’ should be analysed in the context of these new amendments.

‘Caching’ is a technical process which essentially involves the storage of information so that future requests for the same information can be performed faster. In relation to Google, this can be illustrated as follows: Google crawls as many websites as possible so that it can serve results from these pages when a user searches for certain keywords. As Google crawls these websites, it picks up necessary information about the site (size, title, URL etc.) but more importantly, also creates a temporary copy of the webpage, which is called a ‘cache’. This is an exact replica of the page and is temporarily stored on Google’s servers so that it can be displayed to users in case the original webpage is taken down (or other reasons as explained below). The cache is refreshed approximately every two weeks.


Reasons for Making Cached Copies

Although Google serves the original webpage in its search results, it sometimes becomes necessary to serve the cached copy in case the underlying webpage has gone offline because of server load, the link is broken or has changed, or the webpage may have been deleted. Very often, gaining access to a cached webpage is crucial, for example in the course of a criminal investigation.

Opting Out of Caching

Google allows the owner of a website the option to prevent Google from caching that website. This can be done by means of a simple HTML tag, as discussed in the Google FAQ page. This is an important fact that had a bearing on the case discussed below.

Legality of Caching

In the case of Field v. Google, Blake Fields, who is an attorney and a poet, published certain poems on his blog. Google, in the course of indexing this blog, created a cached copy of his poems. In September of 2004, he filed a copyright infringement suit against Google claiming that caching of his poem “Good Tea” involved the unauthorised copying and distributino of his work. He claimed that when Google served, and users clicked on the cached copy of his poems, Google was not only ‘distributing’ unauthorised copies of his work, but ‘creating’ an unauthorised copy as well. It is also relevant to note that Field registered his poems with the Copyright Office before publishing them on his website and he did not employ the feature that prevented Google from caching his website.


The court framed two issues to decide the matter: (1) Does the creation of a cached copy constitute unauthorised copying? 2) When Google serves the cached copy to the user as a search result, does it amount to unauthorised distribution? The court found that no copyright infringement had occurred and delivered the judgement under the following broad heads:

Direct Infringement: There was no direct infringement since the entire process – starting from the display of search results and the subsequent viewing of the cached page – was a non-volitional act on the part of Google. This is consistent with the prevailing law on intermediary liability, which states that service providers cannot be held liable if the allegedly infringing act was an automated process. Bear in mind that Field did not accuse Google of infringement for creating the cached copy in the first instance or that users who viewed the cached copy were violating his copyright, and therefore Google was thereby liable for secondary infringement. Field appears to have taken the most tenuous defence – that the ‘creation of new copies’, when a user clicked the cached link, constituted copyright infringement. 

Implied License: This is crux of the decision, with ramifications for future Internet-related copyright cases. The court found that in the age of the Internet, there is an ‘implied license’ to view pages containing copyright protected content. Consequently, Google had an implied license to store and display cached copies of Field’s poems. Although this may seem unusual at first, it is consistent with the need for an implied license to view content across multiple online services through hyperlinks  Every time a user clicks on a link (say an editorial piece of a newspaper) the webpage is stored in the user’s computer memory (also called ‘RAM’) as well the browser’s history (think ‘temporary internet files’) and possibly also the service provider’s servers. If an there was no implied license do make such temporary copies, it could be deemed as infringement.

The court relied heavily on the opt-out option granted by Google to all website owners, including Field. Since Field was aware of this option, the court attributed prior ‘knowledge’ to Field’s actions, and concluded that Field ‘made a conscious decision to permit’ his poems being cached. With respect to the fair use analysis, the court remarked on the four factors that come into play:

Purpose and character: Since Field claimed his poems were ‘artistic works’ under copyright law, the court held that Google, in creating cached copies of his poems and making them available, were increasing access and stimulating creativity for the enrichment of the general public, which was the underlying purpose of copyright law. Further, it was held to be ‘transformative’ as it permitted access when the original page was inaccessible and allowed changes to the content to be detected. The fact that Google is a commercial enterprise was held to be insignificant.

Nature of the copied work: Since Field’s works were creative in nature and already freely available on his website, the court was in favour of permitting greater access to the poems.

Amount of copying: Since all of the material was freely available to begin with, the court held this factor to be neutral.

Effect upon original work’s value: The court found no evidence that Google’s ‘cached’ links had any impact on the potential market for Field’s work, since the works were available for free and so could not produce any commercial impact.

On its part, Google invoked the statutory defence provided under Section 512(b) of the DMCA, and the court agreed that it had satisfied the tests: the storage was temporary; the work had been initially transmitted by Field to Google and he chose not to employ the preventive measure made available to him; and Google’s act of serving the cached page was on the basis of a user search query and therefore non-volitional.

Proposed Amendment in India

While the ‘implied license’ ruling is important, online service providers can take resort to safe harbour provisions to avoid liability (the DMCA in the US). In India, it is relevant to refer to the proposed amendment to the Indian Copyright Act, which adds Section 52(1) that states “the transient and incidental storage of a work or performance purely in the technical process of electronic transmission or communication to the public” shall not be treated as copyright infringement. The clause enumerates other conditions to be satisfied – if the copyright owner has expressly prohibited access to that link, or if the service provider is aware that the content is infringing. then the immunity will not be available. Further it creates a system wherein the notice has to backed up by a lawful order within a prescribed period. If such order is not produced, the content can go back up on the website.

In the context of caching, there are some important differences between the DMCA and the applicable Indian laws. The DMCA explicitly uses the word ‘cache’, while the Indian Copyright Act or IT Act does not.  Most importantly, the safe harbour provisions in India, contained in Section 79 of the IT Act, are overridden by Section 81. This has produced extreme difficulty in interpreting the scope of immunity for intermediaries in the context of copyright infringement. Therefore, only with the reform of the IT Act will the proposed amendment to the Copyright Act have any teeth.

Current Position in India

The crucial question is the legal position if not for the propsed amendment. In the absence of any clear judicial precedent, it appears the reasoning in Field would apply in India as well. Th fair use provisions under U.S. law are admittedly more detailed, but Indian courts are known to recognise the idea of ‘transformative value’ and may apply that to the act of caching. The proposed amendment to the Copyright Act, if passed, would be an additional ground for defence.

Does posting your content online give others the right to copy?

If you are an independent artist or creator, you must be wondering how the ‘implied license’ affects your content online. For example, if someone were to copy or share this post without permission, would it amount to copyright infringement? The implied license argued could flow from the fact that I have:

1. Posted this work online for free 2. Allowed others to share this content 3. Not prevented search engines from indexing this content

However, I would argue against such logic for two reasons – the first argument is technical, since Blogger does not, to my knowledge, have an opt-out option for caching content. Secondly, while Google’s act of caching was a non-volitional or automated process, the act of re-publication or copying by a human would be considered voluntary. With that said, it is important to understand ‘implied license’ in a narrow context, applicable in certain specific situations where the technical process involved in transmission or hosting requires such a license. It cannot be taken as a defence when there is an intention to infringe or there is active human intervention.

Image from William Wallace Denslow.

This post was first published in a modified form on SpicyIP

About Amlan Mohanty

Amlan is a lawyer based in India. He engages in research, writing, speaking and teaching at the intersection of technology, law and policy.