Aggregation: Demystifying Database Rights
Database rights are notoriously complicated. This piece seeks to demystify them and shed some much needed light on when scraping is likely to amount to database right infringement.
Many people now have apps on their smartphones which can scan product barcodes and show which shops sell the product most cheaply. This can be done both at home and, more worryingly for retailers, actually in the shops. This real-time price comparison is made possible by data aggregation technology. The technology obtains the price and product information from a range of websites and presents it in a comparison format.
In this piece, we consider one of the legal risks that these aggregators or "scrapers" face: infringement of website operators’ database rights. These rights are notoriously complicated but we seek to demystify them and shed light on when scraping is likely to amount to database right infringement.
Database rights exist in "databases" (defined as a systematic arrangement of individually accessible "independent works or data") when there has been "substantial investment in obtaining, verifying or presenting the contents of the database". As we will see, identifying the "Independent works or data" and "substantial investment in obtaining" is crucial to establishing whether a website operator has a database right to enforce against scrapers.
This right exists in all member states of the EU. The Court of Justice has interpreted the right very narrowly. The investment must relate to the resources used to seek out existing independent materials and to collect them for the database (“obtaining”). Operators of databases whose investment is mainly in the creation of the contents (e.g. price, product description) will therefore find it hard to rely on the right.
They must show, instead, that they have invested substantially in the verification or presentation of the data. To illustrate the distinction from a different context, we know that football fixture lists do not benefit from database right protection: their operators’ investment is in the setting of the fixtures (i.e. creation of data) rather than the obtaining, verification or presentation of pre-existing data. On the other hand, a carefully arranged list of football clubs in England together with, say, their dates of foundation would.
The databases in and behind websites have to be analysed to see which side of the line they fall on: is the investment made in the creation of the database, in the creation of data or in the obtaining, verifying or presenting of pre-existing data?
In order to establish what sort of investment was involved, we must first ask what the “independent works or data” contained in the final database are. For example, with retail websites the data scraped will typically include the price, product name and number, availability, photograph, product description and specification, i.e. the data contained in a typical product listing.
The best way to answer the question is to distinguish between two categories of database:
- Databases in which the data are presented in the same form as it exists outside the database (e.g. a list of football clubs). In these cases, the independent works or data will be the underlying data. There will be nothing unique about the contents of the database. The independent works will therefore not have been created by the operator but will have been obtained and database right is more likely to subsist.
- Databases in which the data are presented, grouped or arranged in a way that does not exist independently of the database (e.g. like the football fixtures) or with a different character to a list of gathered in information. In these cases the data are likely to have been created and any investment in that creation will be irrelevant as it is the wrong sort of investment. Database right is therefore less likely to subsist (although copyright might).
If the database is made up of product listings, there is a good argument that the investment lay in creating those listings (i.e. the listings are like football fixtures) and not in obtaining the data contained in the database. Similarly, it is likely that the website operator will have heavily invested in setting the prices of a product. Prices, therefore, are likely to have been created, rather than obtained, by the operator. So, unless the website operator can show there has been substantial investment in verifying or presenting those listings, database right is unlikely to subsist and a scraper might be able to take created listings without infringing database rights.
It may therefore be possible, on this view, to scrape product prices without infringing database rights. The difficulty comes when, as will usually be the case, more data is taken. Product names, descriptions and specifications will usually not be created by a website operator.
A decision of the French Supreme Court illustrates the problems which retail website operators may face in showing that they have database right in their product listings (at least where they originate those listings). The case involved the aggregation of real estate advertisements. The target website operator was responsible for collecting the information which formed the adverts and this involved a good deal of investment. Nonetheless, the Court held that the investments were combined in creating the advertisements and integrating them into the database. Without any real verification of the advertisements, there was no database right as the investment was only in the creation of the advertisements.
Given the prevalence of this kind of aggregation technology (which will only increase as smartphones becomes more prevalent and powerful), it is surprising that its use has not yet been the subject of any litigation in England. However, given the restrictive interpretation of database rights, it would be no surprise if website operators were to find it difficult (relying on database rights alone) to stop all activities carried out by aggregators. They may have more success with other intellectual property rights (such as copyright) or other rights (such as contract) discussed on this site.
"Database rights are useful for website operators but may not always give the strongest protection against aggregation"
"The databases in and behind websites have to be analysed to see which side of the line they fall on: is the investment made in the creation of the database, in the creation of data or in the obtaining, verifying or presenting of pre-existing data? "