User Oriented Business Models in Peer-to-Peer Services

User Oriented Business Models in Peer-to- Peer Services

Lars-Erik ERIKSSON

, Olle FINDAHL

, Håkan SELG

1

Royal Institute of Technology, School of Information and Communication Technology, Kista, S 164 40, Sweden

Tel: +46 70 340 10 94, Email: leeri@kth.se

2

World Internet Institute, c/o Dep. of Information Technology, Uppsala University, Box 337, S-75105, Uppsala, Sweden

Tel: +46 70 332 2252, Email: olle.findahl@wii.se

3

Royal Institute of Technology, School of Computer Science and Communication, Stockholm, S 100 44, Sweden

Tel: +46 70 630 0085, Email: selg@kth.se

Abstract: A limited number of business and legal solutions have been used over the decades to solve the content industry’s “problems” arising from new technology.

Content industry has been lagging behind technology developers especially when it comes to different file-sharing (Peer-2-Peer) technologies and systems. The paper shows that P2P technology will be necessary in the future. Research has shown that it is possible to group file-sharers in three major gropus – “free riders”, “samplers”

and “squirrells”. Studying the properties of these groups a number of business models can be defined. This study show that business models such as licensing, subscription, retail and e-commerce models can be derived from these groups of users. This study also concludes that new underpinning technologies such as DRM and a natural language approach to content mark up are necessary to be able to develop solutions to make these business models satisfactory for both users and content owners.

1. Introduction

Since the introduction and growth of Peer-to-Peer (P2P) services (file-sharing), the music and film industry has shown no desire to legalise this widespread activity. The industry’s strong message has been that file-sharing is piracy (stealing), that every download amounts to a lost sale of a unit of physical product, and that file sharing allows consumers to get content for free. Provision of alternative legal downloading services has been slow to say the least, a fact that is readily admitted by the industry.

Today we start to see several legal downloading services making use of file-sharing technology but still locking in the user to consume content how the content owner prescribes by using DRM (Digital Rights Management) systems. On the other hand the users have clearly demonstrated they do not want to be locked into the business models of the media industry. Users want to consume content “on demand”. In addition, users want a more diverse offering of cultural content than the media industry want to provide – something that the web can fulfil through P2P services. The users have also clearly demonstrated that they are willing to add value to the services – in general they pay money for bandwidth and equipment and spend time to maintain the content of the database.

The paper addresses two issues: the arguments and rationale for P2P services and new business models for P2P services in the context of users’ needs; and content owners’

requirements.

2. Objectives

2.1 A Capacity Rationale for P2P Services

P2P services are important if wide spread “on demand” services are to be realised. Today operators report unofficially that ~80 percent of network traffic is P2P. This traffic is increasing and sometimes temporarily goes down when the debate on file sharing and IPR is high. ~20 percent of traffic is client-server oriented according to “old” Internet paradigm.

If all P2P traffic was transformed to the client-server paradigm then the server capacity on the Internet must be increased with a factor 4 (80/20). A general estimation is that about 7-8 percent of the population are active file sharers [1]. For young people the figure is close to 50%. Over time it is not unrealistic to assume that young peoples habits will be the normal behaviour and thus at least 50 percent of the population would like to retrieve content like young file sharers do today. One argument for this is that the diversity of the content in file sharing networks is much larger than elsewhere. This would mean that another factor 5 of server capacity would be needed. Furthermore, when taking into account that a broad implementation of HDTV is “just around the corner”, another factor 4 in server capacity is required. Increasing the Internet server capacity with a factor 80 seems rather unrealistic and thus there is a capacity argument that P2P services should be pushed.

2.2 Users Add Value to P2P Services

P2P services have a lot of interesting properties to take account of when discussing new business models. Some of these are:

1. Decentralised services

2. Cost effective (~0 cost to add a new user)

P2P services are software based services on top of traditional operators business models. The operators add very little value to this – users are in the best position to add value than the operator and in certain cases the content owner.

3. The user participates in the development of the service by creating catalogues and databases of content and information.

4. The users pay for most of the service by paying for broadband access and server capacity in their home terminals and on top of that they spend management time. Some users (maintaining super nodes) take a larger “responsibility” by having very large server capacity and very high access bandwidth.

5. “New” search methods/business models developed for search engines such as Google and its predecessors are technologies that support the consumer in finding information/content. The search engines are methods to present the most relevant information to the question posed by the user. Some of the methods are quite simple but the technology has developed over time. Taking into account that you know from the question what the user is interested in it is possible to present “extra information”

(advertisements) to the user. The accuracy that the advertisement will “hit” the user is so much larger in this case in comparison with “broadcasting” that the advertiser is prepared to pay much more. This advertising model, sometimes also called the

“Amazon logic” is one that has shown to be successful and vigorous on Internet. The business model exists also in P2P based applications.

When discussing new business models points 3 and 4 are very important. The users must get revenue back for his/her contribution.

2.3 Users are Willing to Pay for Content

The industry’s strong message that file-sharing is piracy (stealing), that every download

amounts to a lost sale of a unit must be reconsidered in light of a 2005 study by World

Internet Institute (WII). WII interviewed file-sharers about how much music they buy now compared to when they were not file-sharers. The answers were 3% buy much more, 7% a little more, 55% as before, 25% a little less and 10% much less. The diagram below shows the relations. An important conclusion is that there is a willingness to pay for content.

Table 1: File-Sharers Willingness to Pay for Music

3% 10%

25%

55%

7%

much less much more

A little more

a little less

as before

3. Methodology

3.1 Categories of File-Sharers

It is generally assumed that the listening to music is the primary goal of the file-sharing activities. Investigations [2] indicate that it is not necessarily the case. Among the heavy down-loaders in particular, with P2P traffic as a part of their daily activities, music seems to be a means rather than a goal. Huge amounts of content files are swapped between individuals who appear to derive great satisfaction from activities involving the handling and management of computers and their content. A university survey [3] shows that a disproportionate share of heavy users are students in computer and engineering sciences, giving reason to believe that the preferred hobby is computers and not music. Since this category down-loads more than they can consume, we call them “squirrels”.

For a majority of users though, the main purpose of file-sharing activities is listening to music. Reference [2] shows that two different categories can be distinguished:

Samplers, who are acquiring and listening to MP3 files in order to decide what records to buy. For these people the MP3-format on Internet has replaced, or is complementary to traditional methods like listening to music on the radio or at record shops, borrowing records from friends, etc.

File-sharers

Those with music as a hobby

Samplers Free riders

Those with computers as a hobby

“Squirrels”

Figure 1: Categories of file-sharers

Free-riders, who download and consume music without the intention to buy the corresponding physical products. This music corresponds to background music, but with a far more diverse repertoire than what is available on public broadcast channels, all which contain almost the same music. In conclusion we can define three categories of file-sharers

4. Technology Description

There are four important technologies causing a technology shift in the media industry giving the user such lead that he/she can bypass content owners. The technologies are:

• Signal processing and in particular new formats such as MP3 and MPEG2.

• Storage technology that becomes cheaper and cheaper

• Broadband (symmetric)

• P2P technologies supporting users to build databases of huge amount of content

Note that these technologies have been around and constantly developed for many years – digital signal processing of sound and images (moving and still) since early 1970 and memories for more than 40 years. P2P systems are also old technologies with roots in the early developments of Internet. Broadband networks are not that old – only around 20 years (ADSL was “invented” 1987) but new application start to appear with high penetration of broadband in combination with flat rate billing and the three other technologies.

Analysing the three groups of file-sharers mentioned above in the in the context of these technologies it is possible to find a way forward to design future business models for P2P services that satisfy both users and IPR holders.

From an historical perspective, the convulsions caused in the music industry by the development of signal processing, storage systems, digital networks and file-sharing is hardly surprising. This should be a temporary phase while the positions of different parties are established, new actors are positioned, and financial/legal rules of play agreed [4].

Indeed, technology shifts follow logical patterns. Studying technology shifts in several industrial areas the following patterns can be observed [5] Analysing these statements a little closer in the present situation the following comments to the points are relevant:

Investments in almost all distribution chains are becoming obsolete since physical transportation can be substituted by cheaper and faster digital transportation of content. The investments in making CDs also start to become obsolete since with new cheap memories types new physical bearers can carry a lot more content than any CD.

Selling CDs and other read-only hardware type with stored information/content bits in stores is threatened, since sales can be done over Internet with payment methods that are convenient and more and more reliable.

Cheap memories – i.e. the prize for storage of information/content goes quickly towards zero – supports the separation of the information/content bits from the physical bearer.

Together with new coding methods for sound, music, images, film and TV there is strong prize erosion. The coding format MP3 and cheap memories are out competing CDs. How many CDs can you usefully own? The MP3 format allows you to put 10 000 songs on a laptop which you easily can take with you. This is impossible with CDs.

With P2P systems the user can build databases of far more content than the content industry can (is willing to?) and thus threatening the industry’s present control of content.

What we experience today is the transition from an “old” technology to a “new”.

The content industry has been lagging behind the technology developer for quite some time, but has, in different ways, started to show a deep concern and interest in the new technology. Especially when it comes to different file-sharing technologies and systems.

For example, the industry has used file-sharing networks and”sniffing” tehnology to find

out what consumers’ interests are [6]. The music industry has also been active in

reformulating contracts with artists.

Table 2. Logical Patterns of Technology Shifts

Destructive Elements for Existing Companies

New Branches Start to Emerge

1 Investments done become obsolete Many new actors – most of them knocked out or bought

2 Existing payment streams are threatened Successful companies start to set new business standards

3 Cannibalisation and prize erosion – delaying tactics

Many stable companies are knocked out – others survive

4 Old technology develops to perfection New branch structures which promote growth start to develop

5 Mergers and acquisitions as preventions Gradual consolidation to global oligopoly markets

5. Results

Summarizing some conclusions so far

• File sharing networks are probably necessary to cope with expected volumes of content transported over Internet in the long run.

• Presently we can see three categories of file sharers: samplers, free riders and those interested in technology.

• The coding format MP3 in combination with storage in cheap random access memories is a complement or even a substitute of CDs.

• Digital transportation as a substitute for physical transportation.

• “Sniffing” as a method to understand consumer’s curiosity, interests and preferences for marketing and/or a method to estimate consumption as grounds for assignments could possibly develop into measurement methods to estimate content traffic.

• Qualities in some file sharing networks such as consumer’s possibilities to create common databases for information/content substituting new types of EPGs.

• File sharing networks generally exhibit similar qualities as search engines – knowledge of consumer’s interests, possibility to advertise to loyal users.

Looking at “free riders” listening to MP3 files and downloading other content with no intention to buy we find that it is related to the situation of background music on the radio, but it happens on Internet instead. Downloading MP3 files to listen is a substitute for “free riders” to background music on radio. Interesting content in this situation could lead to that content is bought in shops or even more likely on Internet. In the long run we believe that a licensing business model is relevant and will be developed. Todays “sniffing” on the Internet could develop to a method for estimating traffic for assignments in the licensing model. Since this could develop to a massmarket we expect that file sharing will be a relevant technology to use. Advertising in this kind of network is probably benefical and possible and thus an important source for income.

Looking at samplers’ activities today they are going to the shop and listen to CDs and then buy the CD. Most likely they listen to music that has been recommended by a friend, a reviewer on radio/TV or in the press. Selecting what to buy from listening to MP3 files on Internet complements this behaviour. Business models such as subscription, promotion/advertising and mail order are relevant here. Most likely P2P systems in this case will have a community character.

Looking at the “squirrels” we find the technology developers. They are interested in

developing the P2P methods and protocols. Here it is also likely that we find the technology

“crackers”. If DRM (Digital Rights Methods) go too far and are too intrusive or too controlling [7] they participate in ways to work around the DRM methods [8].

Following the conclusions above we can sketch the following broad picture:

Technology as hobby Music as hobby

Squirrels Free riders

Listen to background music on the radio.

Buy CD in shop

Samplers Listen in shop.

Buy CD in shop

Complemented by Substituted by

Developers Listen to MP3 on the

net

With the intention to buy CD on the net.

Listen to MP3 on the net.

Maybe buy CD on the net.

Advertising Promotion

Business model Business

model

Subscription Retail e-commerce Licensing models

Crackers

Underpinning technologies

for measurement or

“sniffing”

Queering technologies

Measurement methods of content

in the network.

In Figure 2 there is a list of underpinning technologies necessary for business models.

Figure 2: User Oriented Business Models

For licensing models to meet content owners requirement it is necessary to have some monitoring technology in the operator network or related to the P2P system to be able to judge how the collected licensing fees shall be distributed among the rights holders. Today such technology is not readily available. The amount of file transfer traffic can be measured but this is not enough – you need to know the type of content i.e. the payload must be monitored. A tricky question to solve since it means intrusion of privacy.

One can expect that a lot of content – not only music – will be made available users with file sharing technology. To be able to find content queering technologies are necessary. In well defined areas in normal life e.g. in a library, in a video store, in a music store, in a phone book you find pieces of information – metadata – that you can use to find the real information you are looking for. The web today is not like a very big library.

Hardly any metadata is available. To find information brute force methods with search

robots is state of the art on the web. In file sharing networks the situation is a little different and especially in structured file sharing networks where you can search for titles, directors etc. One can expect that when the average user will start to put content on P2P networks – which happens already today – the amount of information will increase drastically. There exist a lot of document mark-up technologies (SGML, HTML, XML, TEI, RDF …). A natural language approach would be more convenient for the average user. There are some approaches for this [9] but this is not a subject for study here.

Another important underpinning technology is DRM (Digital Rights Management). We shortly discussed DRM above. DRM methods can be used for measuring the amount of files transferred of a particular piece of content. The content industry regards DRM as an mandatory technology – the question is how intrusive and how controlling to satisfy both users and content industry.

6. Conclusions

File-sharing is a disruptive technology. The media industry dislikes it, to say the least. The users love the technology to such extent that they do not respect the new IPR law. More than 2/3 of the artists love it because they have new ways of reaching new audiences they never had a chance to meet before causing new revenue streams.

Presently it is a very turbulent period and there are many new business approaches utilising P2P technology to provide services to the users, however still on terms dictated by content owners.

This study has shows that:

• P2P technology is necessary in an “on demand” mass market.

• It is possible to formulate a number of business models in line with users preferences.

• New technology needs to be developed to make these business models satisfactory to both users and content owners.

Acknowledgement

This study was carried out under EU

MUSICLESSONS.

References

[1] Blomqvist Ulf, Eriksson Lars-Erik, Findahl Olle, Selg Håkan, Wallis Roger (2005) Trends in downloading. EU-FP6 MusicLessons project. Deliverable D5, www.musiclessons.se

[2] Blomqvist Ulf, Eriksson Lars-Erik, Findahl Olle, Selg Håkan, Wallis Roger (2005) File sharing in peer- to-peer networks – actors, motives and effects. EU-FP6 MusicLessons Project. Deliverable D4.

www.musiclessons.se

[3] Selg, Håkan (2003) Internet i den svenska högskolan våren 2003, SUNET Rapport 2003-10-29

[4] Blomqvist Ulf, Eriksson Lars-Erik, Findahl Olle, Selg Håkan, Wallis Roger (2005) Report on technology versus usage and effects. EU-FP6 MusicLesson Project. Deliverable D1. www.musiclessons.se

[5] Professor E. Giertz. KTH Executive School. Royal Institute of Technology, Stockholm. Private communication.

[6] Edström Frejman, Anders (2005) Spoofing, sniffing and suing – the major record companies’ response to file sharing in Cunningham, P. et al. (eds.): Innovation and the Knowledge Economy: Issuses, Applications, Case studies, IOS Press, Amsterdam, 2005, pp. 933-940.

[7] Blomqvist Ulf, Fritzell Michael, Olofsson Marcus (2005) DRM – Intrusion or Solution? in Cunningham, P. et al. (eds.): Innovation and the Knowledge Economy: Issuses, Applications, Case studies, IOS Press, Amsterdam, 2005, pp. 925-933.

[8] Doctrow, Cory (2004) Microsoft Research DRM talk. http://craphound.com/msftdrm.txt

[9] Schwitter Rolf, Controlled Natural Language as Interface Language to the Semantic Web, in: B. Prasad (editor), Proceedings of the 2nd Indian International Conference on Artificial Intelligence (IICAI-05), Pune, India, December 20-22, pp. 1699-1718, 2005.