Case Details
- Citation: [2017] SGCA 28
- Case Title: Global Yellow Pages Limited v Promedia Directories Pte Ltd
- Court: Court of Appeal of the Republic of Singapore
- Date of Decision: 19 April 2017
- Procedural History: Appeal from the High Court; High Court decision reported as Global Yellow Pages Limited v Promedia Directories Pte Ltd [2016] 2 SLR 165
- Appeal Number: Civil Appeal No 19 of 2016
- Summons: Civil Appeal No 19 of 2016 (Summons No 56 of 2016)
- Judges: Sundaresh Menon CJ, Chao Hick Tin JA, Andrew Phang Boon Leong JA, Judith Prakash JA, Tay Yong Kwang JA
- Appellant: Global Yellow Pages Limited (“GYP”)
- Respondent: Promedia Directories Pte Ltd (“Promedia”)
- Legal Area(s): Copyright; Authorship and subsistence; Infringement; Compilations; Groundless threats
- Statutes Referenced: Copyright Act
- Cases Cited: [2007] SGDC 216; [2017] SGCA 28
- Judgment Length: 59 pages; 18,224 words
Summary
Global Yellow Pages Limited v Promedia Directories Pte Ltd concerned competing publishers of telephone directories in Singapore and the scope of copyright protection for directory content. The appellant, Global Yellow Pages Limited (“GYP”), alleged that Promedia infringed GYP’s copyright in multiple works, including an online directory and several print directory editions, as well as “enhanced data” within those directories. The High Court had dismissed GYP’s claim, and GYP appealed to the Court of Appeal.
The Court of Appeal’s analysis focused on the relationship between (i) what copyright protects in directory materials—particularly the distinction between protected expression and unprotected facts/data—and (ii) what Promedia actually copied and how it used that material. The court also addressed the nature of copyright protection for compilations, and whether GYP’s claimed rights extended to the “data” Promedia incorporated into its own temporary database. Ultimately, the Court of Appeal upheld the dismissal of GYP’s infringement claim, emphasising that copyright does not grant a monopoly over underlying information and that infringement requires copying of protected expression rather than mere use of facts or data.
What Were the Facts of This Case?
The dispute arose from the business operations of two directory publishers. GYP produced telephone directories—both online and in print—built around a search engine and maintained online as the “Internet Yellow Pages”. In addition, GYP published multiple print directories, including the Business Listings (a “white pages” style directory with alphabetical business listings) and classified Yellow Pages directories (the Yellow Pages Business and Yellow Pages Consumer). These directories were not mere raw datasets; they were produced through processes involving verification, classification, and embellishment of subscriber information.
GYP’s production began with obtaining large volumes of subscriber information from Singapore Telecommunications Limited (“Singtel”) under an exclusive arrangement. The raw information arrived as data and then underwent computerised and manual processes. A key feature of GYP’s system was classification. Subscribers were assigned “main classification headings” (“MCHs”), which were internal organisational categories used to determine the final classifications visible to end users. The court’s description of the workflow highlighted that classification was not purely mechanical: it involved verification, matching, and in some cases manual intervention, including outsourcing parts of the classification work to a third party and retaining direct input for certain “controlled classifications”, mismatches, or unmatched cases.
GYP also operated a “free listing policy”, limiting each subscriber to one free listing in the Business Listings and in either Yellow Pages directory, with the possibility of additional paid embellishments such as more prominent listings or “cross-listings” (listings under multiple classifications). The court noted that the final directory content involved extracting, sorting, typesetting, and publishing the subscriber’s name, telephone numbers, address, and any embellishments, arranged within classifications and categories.
Promedia, by contrast, could not obtain subscriber information directly from Singtel. Instead, it gathered information from multiple third-party sources, including delivery orders, marketing collateral, field surveys, telephone calls, and company websites. However, the most significant source for the information ultimately included in its directories was the telephone directories of competing publishers, including GYP’s directories. Promedia’s internal standard operating procedures (“SOPs”) required employees to merge data from the most recent edition of GYP’s Business Listings and Yellow Pages into Promedia’s database through a form-filling process using a computer programme. In some years, this involved saving or printing Internet Yellow Pages entries and photocopying or scanning the Business Listings into what the parties referred to as Promedia’s “temporary database”.
After this initial capture, Promedia’s SOPs required further steps. Employees used the data for “market intelligence” exercises to determine the prominence of listings in competing directories, verified and updated entries by calling subscribers, and then understood the subscriber’s business offerings in greater detail to list them under multiple business types or offerings. Promedia’s approach was modelled after a well-known directory in the United States (“Thomas register”), which uses detailed and specific classifications. The evidence showed that Promedia’s final directories were visibly different from GYP’s corresponding directories, including differences in the number and structure of classifications per subscriber. Nevertheless, the court found that what was copied at the outset was essentially the data in GYP’s directories.
Importantly, the parties’ dispute was also shaped by the use of “seeds”—fictitious listings inserted by GYP into its directories to detect copying. Promedia’s employees sometimes failed to verify certain listings in the Business Listings or Yellow Pages, allowing these seeds to appear in Promedia’s CD-ROM and online versions. This evidence supported GYP’s allegation that Promedia copied directory content from GYP’s works.
At trial, GYP advanced several categories of copyright. First, it claimed copyright in “enhanced data” within its directories, meaning individual listings after verification, embellishment, arrangement, and classification. Second, it claimed copyright in each directory “in whole or in part” as compilations, arguing that the selection and arrangement of content amounted to intellectual creations. Third, it claimed copyright in individual seeds. GYP alleged infringement by Promedia’s production of its Green Book directories (print, CD-ROM, and online) and, on appeal, placed particular emphasis on Promedia’s use and deployment of material from GYP’s directories in Promedia’s temporary database.
What Were the Key Legal Issues?
The central legal issue was the nature and extent of copyright protection for directory materials. Specifically, the court had to determine what aspects of GYP’s directory content were protected by copyright and whether Promedia’s conduct amounted to infringement of those protected elements. This required careful attention to the boundary between protected expression and unprotected facts or data.
A second major issue concerned compilations. GYP argued that its directories were protected as compilations because of the intellectual creation involved in selecting and arranging directory content. The court therefore had to assess whether the directory structure and arrangement constituted protectable expression, and whether Promedia’s copying and use of directory data fell within the scope of that protection.
A further issue related to the concept of infringement through intermediate copying and temporary databases. GYP contended that Promedia’s saving, printing, photocopying, scanning, and incorporation of GYP’s directory entries into its temporary database constituted infringement, even if Promedia’s final published directories were not reproduced in the same form as GYP’s. This raised questions about whether copying for the purpose of creating a new work can still infringe copyright in the copied material.
How Did the Court Analyse the Issues?
The Court of Appeal began by situating the case within core copyright principles: copyright infringement depends on both subsistence of copyright in the claimed work and copying of protected expression. The court emphasised that copyright does not protect ideas, information, or facts as such. In the context of directories, this meant that subscriber data—names, telephone numbers, addresses, and other factual information—could not automatically be monopolised by copyright merely because it was organised into a directory. The analysis therefore required identifying what, exactly, GYP claimed as the protected subject matter and whether Promedia copied that protected subject matter.
On authorship and subsistence, the court examined GYP’s processes for creating directory content. While the court accepted that directory production involved verification, classification, and embellishment, it also scrutinised whether the “enhanced data” claimed by GYP represented protectable expression or whether it was, in substance, a compilation of factual information with organisational features that did not rise to the level of copyright-protected expression. The court’s reasoning reflected a consistent theme in copyright law: the more the claim is directed at factual content or raw information, the harder it becomes to establish infringement, because copyright cannot be used to prevent competitors from using information.
With respect to compilations, the court analysed whether GYP’s directories were genuine intellectual creations in the selection and arrangement of content. The court’s approach was to look beyond the label “compilation” and to examine the actual creative choices involved. In directory publishing, the selection and arrangement of listings may involve some skill and labour, but copyright protection for compilations requires more than routine or mechanical organisation. The court considered how classifications were determined, including the role of internal MCHs, manual classification for certain cases, and the outsourced matching process. It also considered the extent to which the final directory structure reflected creative choices rather than merely a functional method of presenting information.
Turning to infringement, the court focused on what Promedia copied. The evidence showed that Promedia’s employees copied data from GYP’s directories into a temporary database through scanning, photocopying, saving, and printing. The court accepted that this copying was not trivial: it was a deliberate process that enabled Promedia to build its own directory database. However, the court also examined whether the copied material was reproduced in the same form and whether Promedia’s final output incorporated the protected expression claimed by GYP. The court noted that Promedia’s final directories were visibly different, including differences in the number of classifications per subscriber and the overall classification structure. This supported the view that Promedia did not simply reproduce GYP’s directory as an expression, but rather used GYP’s data as a starting point for its own classification and presentation.
Crucially, the court addressed GYP’s argument that Promedia infringed by using and deploying material from GYP’s directories in its temporary database. The court’s reasoning indicated that intermediate copying can be relevant to infringement, but only if what is copied is protected expression. Where the copied content is essentially factual data, temporary storage and subsequent transformation may not amount to copying of protected expression. The court’s findings on the nature of what was copied—primarily the underlying directory data—therefore undermined GYP’s infringement theory, even though the copying process was evidenced by the presence of seeds in Promedia’s outputs.
In relation to seeds, the court treated them as a diagnostic tool rather than a standalone basis for broad copyright control over directory information. While the seeds demonstrated copying, the court still had to determine whether the seeds themselves were protected expression and whether their incorporation into Promedia’s directories established infringement of the broader claimed works. The court’s overall approach was to ensure that the legal test for infringement remained anchored to protected subject matter, rather than allowing evidence of copying to automatically expand the scope of copyright protection.
What Was the Outcome?
The Court of Appeal dismissed GYP’s appeal and upheld the High Court’s decision to dismiss GYP’s claim for breach of copyright. The practical effect of the decision is that Promedia was not found liable for copyright infringement in respect of the directory works and processes at issue, despite evidence that Promedia copied directory data and that seeds appeared in Promedia’s published outputs.
For practitioners, the outcome reinforces that in directory and database-like contexts, infringement analysis will turn on the precise identification of protectable expression and the extent to which the defendant copied that expression, rather than merely using the same underlying information to produce a different presentation.
Why Does This Case Matter?
Global Yellow Pages v Promedia is significant for its careful treatment of copyright in directory materials and compilations. The decision clarifies that copyright protection does not extend to facts and information themselves, even where those facts are organised into a directory through substantial effort. This is particularly important for industries where competitors must use similar factual datasets—such as publishing, listings, and other information services—because the case provides guidance on how courts may distinguish between protected expression (selection/arrangement that reflects intellectual creation) and unprotected data.
The case also matters for how infringement is assessed where copying occurs at an intermediate stage. Promedia’s temporary database and the acts of scanning, saving, and printing illustrate a modern pattern of copying in information workflows. The Court of Appeal’s reasoning suggests that intermediate copying will not automatically establish infringement unless the copied material constitutes protected expression. This has practical implications for litigators and counsel advising on compliance: it is not enough to show that data was copied; one must show that protected elements were copied.
Finally, the decision is useful as a reference point for arguments about compilations. Lawyers should note that courts may require a concrete demonstration of intellectual creation in selection and arrangement, rather than accepting that any structured directory is inherently protectable. The case therefore informs both enforcement strategy (what to plead and prove) and defence strategy (how to frame the copied subject matter as factual or functional rather than expressive).
Legislation Referenced
- Copyright Act (Singapore) (as referenced in the judgment)
Cases Cited
- [2007] SGDC 216
- [2017] SGCA 28
Source Documents
This article analyses [2017] SGCA 28 for legal research and educational purposes. It does not constitute legal advice. Readers should consult the full judgment for the Court's complete reasoning.