Version classiqueVersion mobile

The Digital Public Domain

 | 
Melanie Dulong de Rosnay
, 
Juan Carlos De Martin

III. Developments and Case Studies

11. The Value of Registering Creative Works

Roland Alton-Scheidl, Joe Benso et Martin Springer

Texte intégral

  • 1 The authors are affiliated with Registered Commons, a service initiative launched in 2006.

1In this chapter we present good practices for online registration services. We will be asking the following questions: is reliable and simple registration of works the right way to improve confidentiality and trust? How could rights collecting societies benefit from such registries? What kind of governance is required to run such registries? And do they conflict with public patent laws or authorities? We will conclude with a proposal for either adapting the Digital Media Project (DMP) authority scheme or establishing registration peering and using existing namespaces.1

1. Improving confidentiality and trust

2Copyright is an automatic right; works are protected by copyright across the world from the moment they are created. As long as a creator of a work does not care about his or her copyright, registration should not be necessary. The purpose of registering works is so that an author can produce a verifiable record of his or her work as it existed before it was published. This record can serve as a proof in case somebody infringes copyright, for instance by publishing the work as their own.

3Users in the digital media value-chain frequently have to accept licensing conditions before they can access and use copyrighted content. A certificate that a given piece of content is really the content that has been registered with a trusted agency can improve a user’s confidence that he or she may use this content according to the license terms.

4Even where license information is available, for example through a web link to a Creative Commons (CC) license, the relation to a given work is not guarded. Companies who may be interested in the commercial distribution and other uses of the material require legal security for their business, notably in the form of reliable authorship information. This reliable link to the author is missing from many websites that offer content for re-use. Even if the work has assigned a license with a weblink, or the work is said to be in the public domain, how can one know that the relation is correct and authorship or the freedom granted is as claimed?

2. Emerging content registries

5Registering creative works implies that content must be reliably and unambiguously identified. This task requires specific capabilities, as identification constitutes a key element of trust establishment. The task of content identification needs to be carried out by organisations that are properly accredited with a trusted authority.

6Registries are following various goals. First, they provide trust among parties who would like to use or share published works. Secondly, registries are being used to furnish evidence of a work’s authorship. A typical use case is to prevent stealing ideas or concepts in the creative sector. People tend to lock their own work in a registry under full copyright and do not necessarily have the intention of publishing it. They register the work in case a similar idea is evolving, for example, after they showed the artwork at a design pitch.

7Additionally, many registries offer value added services, such as paper certificates, storage of the work or offering a point of sales for usages beyond a sharing license. CC+ allows people to provide a link, where conditions for commercial use are described.

3. Survey of registries supporting CC

  • 2 Eric Steuer, ”Creative Commons Announces Major Funding Support from Omidyar Network”, Creative Com (...)

8Creative Commons has started to put a focus on providing value added services through registries, funded by the Omidyar Network. Joi Ito, Creative Commons CEO, stated in a press release: ”…the grant will allow us to explore providing fee-based, value-added services, which can benefit our community and help support the organization financially. The registry is our first big project in which we plan to explore these possibilities”.2

9Registration of intellectual property has long been a service reserved only for authorities like the US copyright office and the Library of Congress. When authors want to provide proof of ownership of a work, proper registration with a trusted party can be a valuable resource. A traditional registration process can, however, be costly to an artist. One piece of work can cost US$45, and registering may be a time consuming process. Now, in an era of digital distribution, the ability to quickly recognise ownership of content is becoming increasingly important. Licensing models like CC have been adopted by many artists as the preferred licensing solution to their work online. However, with the popular ”some rights reserved” CC licensing, there is a growing need to provide users of this content a proof of ownership. This proof is important for all content creators as the copying and distribution of their content becomes increasingly easy.

10From the perspective of CC, it would be in their interest to support a trusted authority for registration of CC-licensed material. As the field of CC registries grows, the benefits to the artists are being communicated, and possible relationships between registries and the US copyright office or other design, trademark or patent offices worldwide may emerge. CC started collecting profiles for service providers of registries that support CC licenses. The following table is an early snapshot.

  • 3 See http://oami.europa.eu.

11In the following sections, we will have a closer look at some of the registry services listed above. Still, this list is not exhaustive. There are other ways to register creative works such as national patent offices or the European Office for Harmonization of the Internal Market, which provide services to protect trademarks and design.3

3.1 Registered Commons

12A creator who registers a work with Registered Commons (RC) benefits from two important advantages. First, a certificate is issued which allows the creator to provide evidence for his or her intellectual ownership of a work. Secondly, and almost as important for evidence of authorship as a certificate, RC digitally records the exact time of a registration with a timestamp, obtained from a trusted third party. Typical users are musicians or photographers who are keen on posting individual works on the Internet, but who wish to retain control over them; or bloggers and even agencies who, prior to giving client presentations, wish to protect their work from plagiarism using the timestamp. These new licensing models render the free distribution of artistic works on the Internet considerably easier, as they free the works from the tight corset of traditional copyright management. However they also exhibit their own significant defects. Firstly, they cannot prevent the unlawful use of works. Secondly, the user’s rights of such works for commercial purposes (for example, the use of a piece of music under a non-commercial CC license for the credits of an independent film) are increasingly more difficult to clarify, as many of these works are published without reliable information on the author and the user’s rights needed for commercial uses.

  • 4 See http://osAlliance.com.

13RC is a service established and provided by a public-private partnership. The public partner, the Vorarlberg University of Applied Sciences, located in the Austrian Alps close to the Swiss and German border, has been providing infrastructure in the start-up phase. Private partners are organised in the International Media Association, osAlliance.4 Registered users of the service may also become shareholders. Members of the association keep control over the service through their vote in the general assembly. A trademark agreement has been signed with CC in 2007. The transparency of its governance is a core strength of RC which earns the trust of the users.

14RC uses Open Source software and runs with a PostGres Database, and the SpunQ database modeller on a BSD platform. Hosting is secured by standardised control and surveillance mechanisms and frequent backups. In 2008, RC could establish agreements with the Austrian Chamber of Commerce, for which the branded version ”Creativdepot.at” is being provided. The business model has been slightly adapted in a way that partners may obtain a voucher for their clients or members. The number of free registrations per user has been restricted, unless they use a voucher code. In 2009, RC introduced revenue-generating models for their users. This model aims to incorporate a value-added service that allows a user’s work to be licensed for commercial use as a separate contract. Copyright owners can set a price tag—for commercial use of a work, for example—and RC is offering to handle the clearing.

3.2 SafeCreative

15SafeCreative is a registration service provided by an investment company called AAR Futuro. They keep a record of work registered for identification purposes. This allows interested parties to consult the registry and obtain information with respect to the rights of use or the distribution. In their service description they want to dissuade third parties from plagiarism or falsely claiming the work as their own. While basic registrations are free, they charge a fee for multi-authorship, extended storage or time-limited custom licenses.

16Their experience is that only one out of four of those who register chooses options that are not full copyright.

Table 2: Distribution of Licenses used at SafeCreative (http://en.safecreative.net/​)

17According to SafeCreative’s analysis in 2008, the percentages of licenses used in 18,000 registered works are:

1872.8 % - All rights reserved

1912.9 % - CC Attribution Non-commercial No Derivatives

206.3 % - CC Attribution Non-commercial Share Alike

213.3 % - CC Attribution Non-commercial

222.1 % - CC Attribution

231.2 % - CC Attribution Share Alike

240.7 % - CC Attribution No Derivs

250.4 - GNU LGPL

260.3 % - GNU GPL

270.1 % - GNU FDL

28This seems to be a clear indication that there is a demand for online registries not only for people who are aware of license templates, such as CC, but also for commercial licensing.

3.3 RightsAgent

29Registrants should be cautious on the ownership structure of the service provider. RightsAgent promised unified feeds to text, photos or videos, which allowed tracking uses of works. Value could be gained by building a reputation score and by collecting fees for commercial license agreements or when using the CC+ option. However, the service disappeared in 2009.

3.4 Public Domain Registry in Canada

  • 5 See Mia Garlick, ”Canadian Public Domain Registry Announced” (3 March 2006), available at http://c (...)

30The Canada-based clearing service Access Copyright had announced a partnership with CC and the Wikimedia Foundation on the creation of a Canadian Public Domain Registry.5 Access Copyright’s role in this project was to provide bibliographic information on Canadian published works contained in its Rights Management System (RMS). This list of over 300,000 works, including works that are currently in the public domain and those that will enter the public domain in the future, was provided to CC and the Wikimedia Foundation in September 2007. Since that time, both organisations have used this list to build the registry’s database. A Beta version of the Public Domain Wiki is still being tested by all three partners.

  • 6 See http://www.accesscopyright.ca/docs/Public%20Domain%20Registry%20Update%20 Jan%202008.pdf and h (...)

31Once testing is complete, the registry should look similar to the online encyclopaedia Wikipedia. It will allow users to search and edit records in the registry to provide additional information about individual creators, the history of the registry’s works and to add additional works to the system. However, their hope according to their public announcement in January 2008—that it will be a model for similar public domain registries around the globe—is still far from becoming realised.6

3.5 Open Knowledge Registry (CKAN)

  • 7 See http://www.ckan.net/.
  • 8 See http://thedatahub.org/.

32CKAN intended to develop a registry of open knowledge packages and projects, ”be that a set of Shakespeare’s works, a global population density database, the voting records of MPs, or 30 years of US patents”.7 As of 2011, the project has been merged with their hub of open data sets.8

  • 9 See Open Knowledge Foundation: http://www.okfn.org/.

33CKAN is looking for people to register ”packages”, that is, collections with some kind of structure rather than individual items. So a substantial set of photos, datasets of all kinds and the writings of Shakespeare are allowed, but not an individual blog, or your Flickr photo collection (unless it is very big). The material should be free to use, reuse and redistribute without major restrictions, referring to common open definition guidelines.9 Even if search engines could also be restricted to open content search results, CKAN offers extended metadata on the collections and reuse of the material.

4. What kind of authority is required to run such registries?

  • 10 See http://docs.oasis-open.org/xri/xri-resolution/2.0/specs/cs01/xri-resolution-V2.0-cs-01.htm.
  • 11 ”Proposal for ‘Approved Document No. 5, WD1.1 – Technical Reference: Certification and Registrati (...)

34In order to guarantee that content identifiers are unique and thus content could be mirrored between registries easily, there needs to be one and only one root authority—called the Registration Authority (RA)—which may have responsibility for many Registration Agencies (RAgs). The RA is responsible for allocating namespaces to RAgs, and it will appoint RAgs on the basis of general rules. The primary role of RAgs is to provide services to registrants—allocating identifier name prefixes, registering identifier names and providing the necessary infrastructure to allow registrants to declare and maintain metadata and state data. In general, identifier management policies can be defined on a community-by-community basis. The OASIS committee proposes that a resolution community chooses to create a community root authority.10 When a community changes the root authority, it should define policies for assigning and managing identifiers under this authority. Furthermore, it should define what resolution protocol(s) may be used for these identifiers. The Digital Media Project has published Approved Document No. 6, which proposes a list of procedural and operational responsibilities for RAs and RAgs.11 This list can serve as good starting point for the steps that need to be performed for setting up and running content registries.

  • 12 See http://www.ifpi.org/content/section_resources/grid.html.

35As an existing example in the music industry, the Global Release Identifier scheme has been established by rights collecting societies.12 A GRid consists of 18 characters, made up of an Identifier Scheme element followed by an Issuer Code, a Release Number element and a Check Character as follows:

  • Identifier Scheme element (2 characters)
  • Issuer Code element (5 characters)
  • Release Number element (10 characters)
  • Check Character element (1 character)

36When a GRid is written, printed or otherwise visually presented, the four elements of the GRid shall be separated from each other by a hyphen. The hyphens do not form part of the GRid. It is recommended that when a GRid is visually presented, the
used should clearly distinguish between the number 1 and 0, and the letters I and O.

37Example: A1–2425G-ABC1234002-M

38A1 = Identifier Scheme element

392425G = Issuer Code element

40ABC1234002 = Release Number element

41M = Check Character element

42The Identifier Scheme element distinguishes the GRid Identifier Scheme from any other standard identification scheme which adopts the same or a similar structure. Registered Commons has adopted this approach by simply issuing codes on the identifier scheme RC; thus their identification scheme should be called RCid. An administrative authority to manage namespaces for the Identifier Scheme Element is missing, and a simple directory service would be sufficient. A peer-to-peer protocol combined with globally unique IDs may also be a good approach to identify any registration process as a singular, retraceable activity.

5. How could rights collecting societies benefit from such registries?

43Traditionally, rights collecting societies of Collective Management Societies (CMS) managed the exploitation of any of the rights-holders rights. According to the principle of territorial exploitation, the applicable law is the law of the place of exploitation. For some forms of exploitation (for example, cable retransmission, the making of sound recordings, etc.), collective rights management is compulsory but, in principle, an author can choose between individual and collective management of his or her rights. CMS usually administers, monitors, collects and distributes the payment of royalties for an entire group of right holders, on the basis of the national law of its territory, with respect to that territory.

44In order to be compensated by CMS, authors must register works in the country of their residence. Across Europe more than 70 different CMS manage authors’ rights. Within the European Union, most CMS are allowed to retain a monopoly status or have conferred on them a monopoly position in relation to their specific fields of activity. The reason is economically persuasive. For users, it is simply more expedient to be directed to one collective body that manages one specific type of right. For decades, CMS have been criticised for lack of transparency (they are not really controlled by their members) and for their unbalanced distribution schemes, favouring blockbusters and top ten-charting music over independent and less commercially successful work.

45CMS currently use different schemes to identify works and store the corresponding data in proprietary systems and databases. They wish to protect their assets and defend their monopolies to register and manage works. Many European CMS oblige their members to collectively license all forms of distribution for a certain category of works and thus prevent new forms of individual licensing.

46While some years ago legitimate commercial media offers were absent from the Internet, and media content was shared via peer-to-peer (P2P) file-sharing networks, companies today appear to have figured out new business models taking advantage of the ”long tail”, user-created content and micro-advertising. Apparently this ”new deal” is that consumers (users) generate content and the commercial companies use these data for their web services. Consumers can use these services either for free together with advertising content or without ads and with additional features when paying a subscription fee. As the ”freemium” model usually compensates neither the user’s nor the artist’s works, we have clear evidence that an interoperable platform for the management of copyright, the rights-clearing and the micropayment of services is still missing.

47Network operators, CMS and providers of Web2.0 services appear to be quite happy with the status quo, which preserves their monopolies. Therefore they maintain technical and legal obstacles, for example, by operating proprietary subscriber management/billing systems or granting access to their services through end-user license agreements (EULAs), which are a mystery to the average consumer. The interesting thing is that in P2P time (Napster and before), there was still some money flowing back to the creators. With Web2.0, there is none. Service providers control the business. Essentially they do not intend to prop up creators by providing them with the means to benefit from their creations, they simply subscribe to the idea that there is ”money to make” from those who create. We can ask ourselves why the demand for rights clearing and micropayment has decreased in the past. If the only viable business models on the Internet are flat rates for services and advertising for content, it is clear that users/ operators neither require micropayment nor rights compensation.

48Meanwhile bis service providers contract artists or labels directly, or creators simply publish content on their own websites. Theoretically, end-users who want to access this content could be charged directly through the publishers’ web services. Still, no generic solution for the rights clearing and payment of content is on the horizon, which is cross platform and cross vendor compatible. After all, from the perspective of creators and end-users the best solution would be a decentralised system where individual peers (creators and end-users) can register content and negotiate with each other the usage terms and conditions of content.

  • 13 See http://flattr.com/.

49With content registration services, converging networks and next-generation P2P environments in place, there will be a new market for micropayment. New concepts of micropayment services, such as Flattr, shortcut the circuit to rights collecting societies, who still could benefit by adapting their terms and allowing CC licensed and voluntary payment schemes on their repertoires.13

6. A registration service only for Creative Commons licensed content?

50A creator of a work is free to choose the licensing conditions. The decision will usually depend on several factors, for instance if the licensee will use a work for commercial or non-commercial purposes, or if the licensee obtains an exclusive license to use the work. One can also imagine that licensors may want to change conditions after a time, because they find out that another license is more suitable for their business model. The decision about the licensing follows the initial registration of a work. Since content registration precedes the licensing, a registration authority exclusively for commercial content or only for CC content would not make much sense. Therefore, we believe that the setup of a CRA/CRags should be independent of the license.

51This raises the question of whether a service like RC, being a service that promotes the publishing of works under a CC license, would be in competition with a service that allows many different licenses. One approach could be that the registration of works which will be released under an open license (such as CC-by) is free of charge, whereas registrants would have to pay a fee for the registration of their works if they intend a commercial deployment or want to reserve the right to decide on the license at a later date. A common technical platform for content registration and licensing of copyrighted content could be developed as an open source project. This would benefit public interests through non-commercial dissemination of content as well as commercial interests. Any co-branding would be neutral in the sense that it would not favour or penalise digital media business models.

52A registration service for commercially deployed works would contend with existing service the CMS are offering to their clients. Actually, some CMS, such as the Spanish Sociedad General de Autores y Editores (SGAE), are providing online services for the electronic registration of works.

  • 14 The Management Agreement is available at http://www.sgae.es/resources/pdf/5/4/1180694570945.pdf.
  • 15 Florian Philapitsch, ”Die Creative Commons Lizenzen”, Medien & Recht, 2 (2008), 82–97.

53However, the SGAE Membership Agreement implies that members could not publish their works under a CC license, since SGAE owns the exclusive rights for the reproduction and distribution of their members’ works.14 On the other hand, with fresh ideas and new registration services, the existing CMS monopolies could be stimulated to innovate their activities and business models. The CMS Buma/Stemra in the Netherlands had a one-year test phase that allows musicians to publish their work under a non-commercial CC license, whereas KODA in Denmark was the first to allow CC licensing to all their members.15

7. What are the challenges going forward in the digital registry space?

54The first challenge is to develop policies for CRA/CRags that are accepted in many countries, by all value-chain players, including the major rights-holders and CMS. Small and independent digital registries could issue different identifiers for copyrighted works. This is not a problem as long as the identification of content is unique and identifiers are accepted everywhere. Large registries operated by major rights-holders or CMS could misuse their monopolies by requiring proprietary technical platforms. As a consequence not every value-chain player could process content identifiers managed by CMS. National societies could decide to refuse content identifiers issued elsewhere.

55As mentioned above, RC has prepared the technical means to incorporate other registries that could support the RCid code.16 GRid provides a system for the unique identification of releases of music over electronic networks. The RC service provides an application programming interface (API) to exchange queries for GRids. Fasttrack, an alliance of major music CMS provides tools to exchange information about their members’ works by means of a global decentralised database network.17 A web interface to their ISWC Network provides a search tool for ISWC identifiers.18 It is noteworthy that GRid only identifies the electronic release of a work whereas ISWC identifies a work independently of its manifestations.

  • 19 See http://chillout.dmpf.org.

56The second challenge is to develop a technical infrastructure based on open standards for the protocols and interfaces to communicate with content registries. In the Digital Media Project we have contributed to the development of technical specifications for content registration and management services. Chillout is a reference implementation of the DMP specifications.19 Some of the DMP specifications have been standardised in MPEG (for example, ISO/IEC 23000–5 Media Streaming MAF, ISO/IEC FDIS 23000–7 Open Release MAF). We believe that it should be common sense to implement the technical infrastructure for copyright registration and management services using open source software to minimise security risks and gain transparency.

8. Heterogeneous business models

57The third challenge is to develop business models to finance registration services. Digital content registries may offer value-added services to users in the digital media value-chain. Primary services may include the secure and reliable storage of digital content for rights-holders. If necessary, the registry must be able to prove that a first fixation (a ”digital original”) of a work has been deposited at a given time.

  • 20 See http://creativwirtschaft.at.

58We also found that there is an interest of creative sector promoters to offer secondary services. Incremental access restrictions with advanced access for owners of a voucher are attractive to various stakeholders. For example, the Austrian Chamber of Commerce and the City of Linz, both promoting the Creative Industries, have agreed to issue vouchers to their members for the content registry CreativDepot.at.20 The service includes a personalised feed of registered works from individual artists.

59The following use case describes a secondary service of content registries which is being offered to users who intend to re-use existing works for their own creations at OneLoudr.com. In order to produce a remix, sampling artists conduct an exploratory creative process, copying different samples from different musical works, adapting and merging them. All this is legal as long as it is kept private. Samples clearance must only take place when the creator wants to make the resulting creation public, that is, to publish it online or to produce and distribute copies. The business model is based on a small fee for collecting and redistributing payments of value-added services, such as bonus tracks with remixes produced by fans.

9. Some conclusions on trust and governance

60As different registries emerge, the challenge of separating practices and maintaining quality standards and trust with different registries approaches. Registry services should certainly not undermine copyright in the sense that only registered works are protected. As soon as we publish a work it shall be copyrighted, unless we say that we grant specific usages, expressed by licenses like one of CC. However, we want to be able to define usage rules for each work or for sets of works, and those rules need to be described somewhere. For example, automatic enclosure of license-related metadata for pictures would be used by many photographers if it were easy to handle. Still, the photographer needs to sign her works in a trusted way when publishing the pictures.

61How can we achieve trust? Registration services need to be able to verify that a person’s online identity can be tracked down to his or her real identity. This can be assured by a ”web of trust” such as CAcert or by testing the user’s postal address and bank account with a pro-forma credit card payment. However, most registration services also offer a simple check of the existence of an email address. This method of identifying a creator may be efficient for the publisher, but not for the commercial user, who needs full trust that the work is by the author that is claimed and nobody else. Fraudulent claims of copyright could not be solved effectively either, and a conflict resolution procedure needs to be offered. Better user identification would result in fewer frauds. Following the guidelines of implementing digital timestamps (RFC 3161), a revocation procedure needs to be provided for objects which have been signed erroneously or by fraud.

62Which organisational structures are providing trust? We need to ensure technical operation for decades and we need to show transparency for the processes. According to economic theory, long-term institutions are either public or have incorporated democratic control elements. Purely profitoriented enterprises may be sold sooner or later, refocus their business or just go out of business. In that case, their limited liability will not be of any use for registrants or users of registered works, unless another registry is taking over the data set, which may be a cumbersome issue for personal data protection. National registries tend to be rather inflexible and nonprofit organisations may not take the opportunity to do business.

63For RC, a public-private partnership between an academic institution and a for-profit business co-operative of design and IT experts was found to be an ideal structure for running a registry. The co-operative is being steered by the general assembly, a board of directors, a controlling board and an agency for co-operatives. Additionally, an audit on public welfare (Gemeinwohlökonomie) has been conducted in 2011. All five levels guarantee high performance, effective conflict resolution and reasonable quality control.

64Independently of the organisational structure, any registry must work out a shutdown scenario, which needs to be agreed with the users in the terms of use. An authority to issue global identifier scheme elements would probably be an independent organisation such as International Organization for Standardization (ISO) or the World Intellectual Property Organization (WIPO). That way, registries would efficiently back up confidentiality for all partners involved in the creative sector’s value chains, no matter which way copyright regulation evolves.

Notes

1 The authors are affiliated with Registered Commons, a service initiative launched in 2006.

2 Eric Steuer, ”Creative Commons Announces Major Funding Support from Omidyar Network”, Creative Commons Press Release, San Franciso (28 May 2008), available at http://creativecommons.org/press-releases/entry/8322.

3 See http://oami.europa.eu.

4 See http://osAlliance.com.

5 See Mia Garlick, ”Canadian Public Domain Registry Announced” (3 March 2006), available at http://creativecommons.org/weblog/entry/5809.

6 See http://www.accesscopyright.ca/docs/Public%20Domain%20Registry%20Update%20 Jan%202008.pdf and http://wiki.creativecommons.org/PDWiki.

7 See http://www.ckan.net/.

8 See http://thedatahub.org/.

9 See Open Knowledge Foundation: http://www.okfn.org/.

10 See http://docs.oasis-open.org/xri/xri-resolution/2.0/specs/cs01/xri-resolution-V2.0-cs-01.htm.

11 ”Proposal for ‘Approved Document No. 5, WD1.1 – Technical Reference: Certification and Registration Authorities, Version 3.0’” (15 July 2007), available at http://www.dmpf. org/open/dmp1015.zip.

12 See http://www.ifpi.org/content/section_resources/grid.html.

13 See http://flattr.com/.

14 The Management Agreement is available at http://www.sgae.es/resources/pdf/5/4/1180694570945.pdf.

15 Florian Philapitsch, ”Die Creative Commons Lizenzen”, Medien & Recht, 2 (2008), 82–97.

16 See http://en.wikipedia.org/wiki/Grid.

17 See http://www.fasttrackdcn.net/.

18 See http://iswcnet.cisac.org/ISWCNET-MWI and http://en.wikipedia.org/wiki/International_Standard_Musical_Work_Code.

19 See http://chillout.dmpf.org.

20 See http://creativwirtschaft.at.

Table des illustrations

URL http://books.openedition.org/obp/docannexe/image/562/img-1.jpg
Fichier image/jpeg, 123k
URL http://books.openedition.org/obp/docannexe/image/562/img-2.jpg
Fichier image/jpeg, 20k
Légende Table 2: Distribution of Licenses used at SafeCreative (http://en.safecreative.net/​)
URL http://books.openedition.org/obp/docannexe/image/562/img-3.jpg
Fichier image/jpeg, 25k

Acheter

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search