Independent consumer research
The European Consumers’ Organization (BEUC) commissioned an independent institute, Intertek Research and Performance Testing, to perform an analysis of some of the UK-based music services and widely available digital music players and find out how interoperable they are, in other words the limitations, and "how these limitations restrict the consumer’s traditional ability to transfer their music between platforms and players". This has been an intriguing question for consumers, most of whom have heard about the issue, but in the end they have had to accept the present situation of non-interoperable music download services.

Intertek chose four portable music players:

  • an Apple iPod Photo was selected for its compatibility with Apple iTunes Music Store (AAC format files),
  • a Creative Zen Micro was selected for its compatibility with MSN Music and Windows Media Player 10 (WMA format files),
  • a Rio Carbon was selected for its WMA support (second player with WMA format files chosen because of WMA popularity, and also in order to test a second music store, HMV, using Microsoft’s format) and
  • a Sony Network Walkman for its compatibility with CONNECT Music Store (ATRAC3 format)

Each player was tested with the corresponding service, and players were also cross-checked to find out what level of interoperability exists, all of this from the layman’s point of view.

Technical report
The result of the analysis was a technical report, which is now available for the public on a new web site titled Consumers Digital Rights (cf. sources), created as an information source for a wide range of readers, from politicians through journalists to consumers of DRM-protected content. "On this website we invite you to discover everything you always wanted to know about your consumer rights in the digital environment" – they declare.

Accordingly, the technical report is concise and easy to read, also providing basic information to those not really familiar with digital audio, DRM and related services. The report starts by giving a background to compressed audio, which, as opposed to traditional digital audio like CDs, makes new usages possible: "individual tracks or whole albums can easily be downloaded from the Internet where they can be purchased at lower cost and where new music can be discovered. Also, entire music collections can be copied and stored on a home computer/laptop or portable hard disc based audio file player" (p. 4).

In the following, different compressed audio formats are explained:

  • mp3, as the most widespread format does not support DRM, and therefore it is not generally supported by major record companies
  • WMA, Microsoft’s file format does support DRM. Most music web sites have music available to download in this format, and it is also very popular with the manufacturers of portable players. The reason for the latter is that most modern PCs will already have the Windows Media Player (the player for this format), as the report says, though I have to disagree with this: I think that the main reason is that this is the only widespread technology that is free for everyone to license.
  • ATRAC, Sony’s file format, also supports DRM. This is said to provide the best sound quality for a given bitrate, but Sony so far has not licensed it to anyone, therefore it is a very proprietary format.
  • AAC, the choice of Apple, is employed in the iTunes Music Store. While AAC is an open standard free to implement and use for everyone, Apple coupled it with its proprietary FairPlay DRM system, which makes it inaccessible for any of the few players that manage the compression format itself. (At this point the Technical Report is a bit confusing, saying that "AAC files can only be purchased through the iTunes web site" and suggesting that AAC is a file type supporting DRM. AAC in fact is just the compression method, unlike Microsoft’s and Sony’s compression-protection formats.)

Actual tests
The report moves on to the actual testing done at the research institute. They created accounts at the mentioned music stores, and bought a couple of songs, trying to play, burn, transfer (copy to a different computer) and transcode (convert to a different format) them. Importing songs to a different media player framework from where they were purchased and loading to portable devices were also main points of investigation.

To cut the long story short, each music store was quite comparable in terms and offerings. Each needed a special media player framework (Apple iTunes, Microsoft Media Player 10, HMV’s own software and Sony SonicStage) to handle the music. They could be used for discovering new music, buying, organizing and playing songs and for transferring them to portable devices and finally exporting (burning to CD). Each compression method used about the same compression ratio (around the same file size for the same track). Apple and Sony provided only "permanent purchase" models with unlimited plays for a one-time payment and export options, while the two Microsoft DRM-based systems additionally allowed monthly subscriptions where an unlimited number of songs can be played, but only as long as the subscription is maintained and after this period the songs become unavailable (the exporting option is in this case disabled).

The report analyses individual terms: number of PCs where purchased songs are playable, number of portable devices they can be transferred to and also the number of CD burns. There were some differences, but to me it seemed that the offerings were all liberal enough to not disturb the ordinary user. Terms were mostly correctly displayed before purchase, but in two cases the testers reported unexpected anomalies: in the MSN Music store "tracks were time restricted to 31/12/2099", which is a bad thing, but personally I do not consider this restriction very limiting to myself, the other was in the HMV store where the DRM system allowed a lot more than was stated before purchase – a nice surprise.

The authors of the report gave this chapter the title "File Compatibility" (page 11), but personally I would have preferred "system compatibility". They tried to import music bought from one system to another system’s music library (the collection of music handled by the media player framework). It turned out that the two Microsoft DRM-based systems were (apart from one glitch) compatible with each other, but taking these two as one (since the employed technology was the same) protected music could not be transferred to other systems. This means complete lack of interoperability. Unprotected WMA files can be imported to iTunes and SonicStage, and thus transferred to an iPod, and a Network Walkaman successfully, but AAC and ATRAC files can not be transferred between systems. This is due to DRM-incompatibility rather than file-incompatibility in the case of AAC (since it is an open standard), and due to the incompatibility of both in case of ATRAC (since Sony uses a closed proprietary format) (pages 12, 13 and 14).

What I missed here was the analysis of whether MP3, OGG Vorbis, or other unprotected formats could be imported or exported to and from the respective systems.

When it comes to the analysis of DRM systems, the report becomes rather speculative. What is checked carefully is in particular the contracting terms. Not surprisingly the report discovers that different music stores have different conditions in terms of number of CD burns, portable players, etc.; usage restrictions are not clearly labeled, information on the web sites is not transparent and inconsistent across different music stores and licensing terms are difficult to understand; and by using proprietary formats download web sites can control what one can do with the music and the devices they will play on (cf. Summary, p. 2).

What is more astounding is that "the terms and conditions on these music stores allow the service provider to unilaterally change the terms", and "this would not even break the contract". (cf. Summary, p. 2) On the other hand, technically, it is also possible to change "limitations to a consumers existing collection", which means that in the future there is a possibility for music stores to retrospectively further restrict our purchased music – however, this would be technically challenging and highly unlikely (p. 16).

Yet what made me really wonder was that at the end of the report, in the Appendix, a detailed description is given about how to achieve artificial interoperability between the incompatible systems. More precisely, I was surprised to see this information made publicly available by a high level interest group. Burning the songs to CDs, and then re-ripping them with the target systems’ media player frameworks might be a slightly inconvenient, but certainly effective way of lifting the DRM from the protected music (p. 19). And while the report says that this method is "time consuming", my opinion is that it is possible to create tools (and will therefore be such tools) which automatically do this.

Bottom line
The report talks about a media consultant, who said "My only confidence is that sooner or later the consumers will prevail by voting with credit card against the worst systems" – the same conclusion which INDICARE has drawn in its State-of-the-Art Reports. Therefore beside the experiments carried out underlying the findings, there is nothing really special in the report that DRM experts were not aware of.

So I consider the main value of the research is that it is easy to read for the public, and comes to the right conclusions, therefore educating consumers about today’s DRM systems limitations – and also on how to exercise their wish for interoperability by circumventing content protection.


About the author: Kristóf Kerényi is a researcher at Budapest University of Technology and Economics in the SEARCH Laboratory. His interests include mobile and wireless IT security, as well as technological aspects of DRM. He received a MSc in computer science from BUTE. Contact:

Status: first posted 21/12/05; licensed under Creative Commons