Hungary - Anonymous publisher of online press publications v Anonymous provider of an online search engine, 25 October 2023

Not publicly available.

Most likely after 25 October 2023, as the Claimant’s claim concerned the period between 1 June 2021 and 25 October 2023.

The Claimant published journalistic literary works online and monetised via advertising revenue tied to website visits. The Defendant’s search engine displayed short previews or “snippets” derived from its press publications in search results, which purportedly satisfied users’ information needs and reduced click‑through to the Claimant’s websites. The Defendant’s search results presented a hyperlink to the source page, the article title, and an automatically generated short excerpt positioned around the searched terms to enable users to assess relevance.

Via the robots exclusion protocol (REP), the Claimant could determine whether the Defendant’s search engine could perform mapping and indexing activities in relation to the website and, if so, could also determine the rules for this. It can set which pages, directories, and files the robot can request from the website and which it cannot, determine how long the robot should wait between queries, etc.; it can also create a site map containing the URLs of the most important pages on the website, thereby drawing the robots’ attention to which pages to crawl (search engine optimization or SEO). These can be configured at the webmaster level by creating a simple text file (robot.txt).

In addition, website administrators can use meta tags to provide information about the website to browsers. These meta tags are HTML elements that do not appear on the page itself, but only in its source code. It is possible to configure (i) whether search engine crawlers should store information about the website in the search index, (ii) no excerpt, index image or video preview to appear in the search results (nosnippet meta tag), (iii) the maximum number of characters in the snippet (for example, “max-snippet:250”), (iv) not to display a snippet at all (“max-snippet:0”), or (v) let the search engine to determine the appropriate length of the snippet.

Claimant’s claim

The Claimant asserted that the Defendant’s search results went beyond permissible “individual words or very short extracts” by presenting previews that substitute for reading the full article, undermine the Claimant’s investment, and thus amount to making available to the public without consent under the Hungarian Copyright Act provisions implementing Article 15 of the CDSM. The Claimant emphasized that titles may be shown free of charge, but multi‑line previews require authorization and, where applicable, remuneration. It further contended that consent could not be inferred from the absence of technical restrictions or from the Claimant’s general search engine optimisation (SEO) settings.

Defendant’s defence

The Defendant argued that search previews are automatically generated, limited excerpts that contextualise the query and do not constitute unlawful communication to the public. In any event, they fall within the statutory exception for “individual words or very short extracts” and frequently reflect unprotected facts or daily news not covered by related rights. The Defendant further submitted that publishers can unilaterally control crawling, indexing, and the appearance or length of snippets through standard technical measures. On that basis, the Claimant implicitly authorized the display of snippets by not deploying available controls, and its generalized claim is not justiciable without specific instances showing protected expression exceeding the exception.

Court’s Judgment

The Court affirmed the first‑instance dismissal. It held, first, that the Claimant’s generalized request for a declaration – untethered to specific, identified search results and concrete protected excerpts – could not succeed. Whether a particular preview infringes requires a case‑by‑case assessment of the content actually displayed. The Court agreed that the presented exemplars did not establish infringement: the excerpts either conveyed unprotected facts or daily news, or, where expressive, did not exceed the threshold of “very short extracts”, did not substitute for the original article, and did not make reading the source unnecessary.

On communication to the public, the court rejected the Claimant’s analogy between snippets and full summaries, finding that the Defendant showed only context fragments to aid relevance assessment and not summaries of the creative content. As to consent, the court endorsed the first‑instance analysis that the Hungarian Copyright Act does not impose a specific formality for a publisher’s authorization of communication to the public of press publication excerpts. In light of CJEU guidance on implicit authorization and effective technical measures, the Claimant’s deliberate choice not to deploy standard meta tags or REP to limit or suppress snippets amounted, in context, to implicit consent to the display of short previews. The court also stressed that the exception for “individual words or very short extracts” must be interpreted to maintain a fair balance: snippets may not replace the article or undermine the publisher’s legitimate interests, but they may allow users to gauge relevance and foster traffic.

Finally, the court noted that any remuneration or licensing issues arise only where previews go beyond statutory exceptions and involve protected expression. As the Claimant failed to identify specific instances surpassing those limits, no infringement could be established. The Court therefore upheld the judgment dismissing the claim.