Version classiqueVersion mobile

White Paper — Open Science in a Digital Republic

Scientific and Technical Information Department - CNRS

Overview: “science in transition”

The legal vacuums

Texte intégral

1The emergence of digital technologies in STI practices is creating a discrepancy between law and practice. While the French Research Code organises public research and defines its objectives, it does not at any time affirm the common values of science.

2STI is managed by multiple platforms with a non-existent legal model. These offer data-processing tools of questionable legality.

3In addition, the contractual practices are not aligned with the practices of the research communities, to the detriment of science.

Lack of a legal framework for science

4No law for science. There is no legal provision, no text that reflects the values of the scientific communities and affirms the best interests of science.

5This legal vacuum is increasingly felt in the framework of the digital transition and the development of value-added services for scientific data.

Law on platforms: Developments in progress

6The concept of “platform” has no legal status or regime. This legal vacuum entails a certain legal insecurity that has already been highlighted by the French Digital Council in its Opinion of 13 June 2014, as well as by the Council of State in its 2014 report Le numérique et les droits fondamentaux (Digital technology and fundamental rights).

7Article 22 of the Digital Republic Bill provides for the introduction of a definition of the concept of “platform”:

  • activities consisting in classifying or referencing content, goods, or services offered or placed online by third parties, or putting several parties in contact with each other, by electronic means, with a view to selling goods, providing services, including non-paying ones, or exchanging or sharing goods or services.”

8A duty to act in good faith is also imposed on the platform operator:

  • any operator of an online platform is required to give the consumer fair, clear and transparent information on the general conditions of use of the intermediation service it offers and on the terms for referencing, classifying and dereferencing the content, goods or services to which this service provides access.”

9Although the text of the Bill seems to govern relationships with consumers, Open Science platforms will need to apply these principles and provide researcher-users with fair information with respect to their conditions of use.

A right to TDM: An absence with serious consequences

Challenges of TDM

  • 1 In 2001, the Massachusetts Institute of Technology presented data exploration as one of the 10 emer (...)

10Major challenge. TDM is a major issue for science, research and innovation in that it enables scientists to identify new research subjects, produce new knowledge and address economic, social and societal issues.1

11It also creates opportunities in terms of the exploitation of this new knowledge with all this entails for innovation, growth and employment.

12The scientific and economic issues are especially important in that TDM is practised around the world and is governed by different standards in different countries, including within Europe. The United States and the United Kingdom have affirmed the right of researchers to conduct TDM; Germany has introduced a right of secondary exploitation of scientific publications.

13A draft revised text of the InfoSoc Directive2 (Directive 2001/29/EC of the European Parliament and of the Council of 22 May 2001 on the harmonisation of certain aspects of copyright and related rights in the information society) was expected for the end of 2015 but had not yet been circulated on the date this White Paper was published. The process for accepting a draft directive is long: at least two to three years are needed before this directive is accepted by the Member States, and an additional two years for it to be transposed into French law.

14French research cannot afford to suffer discrimination with regard to its European neighbours and allow an unbridgeable gap to open with multiple harmful consequences: delayed development of digital research techniques, delay in emerging research subjects, loss of partnership contracts at European level, fall in the number of patents filed, risk of privatisation of data-mining techniques, etc.

15Draft Bill V.1. By inserting provisions in the first version of the draft Digital Republic Bill allowing researchers to carry out data-mining operations, the government seemed to have understood what is at stake regarding TDM for research, and especially public research. However, these provisions were removed and are no longer included in the Bill as sent to the Council of State.

16The removal of these provisions from the Bill was raised in many of the contributions by the scientific community as part of the public consultation on the Bill.

17The government offers two arguments to justify this removal:

  • the proposed legislation permitted an exception to copyright; however, the text of the InfoSoc Directive regulating copyright between the members of the European Union prevents Member States from creating exceptions not provided for by the texts;

  • the InfoSoc Directive will undoubtedly be revised, and the reports submitted in this regard are virtually unanimous in advocating the introduction of this TDM exception to the copyright provisions.

18Bill – adopted text (TA) No. 663. As part of the parliamentary debate on the Digital Republic Bill, deputies of different political persuasions supported the introduction of an amendment creating an exception for text and data mining. An exception to copyright and the right of the database creator was introduced in the text of the Bill (Article 18 bis (new) of the adopted text No. 663).

19Fleur Pellerin, Minister of Culture and Communication and Thierry Mandon, Secretary of State for Higher Education and Research, together entrusted Charles Huot, President of the French Professional Group for B-to-B Information and Knowledge (GFII), with a “mission of consultation and proposal to facilitate the development of the use in France of text- and data-mining technologies”.

Text and data mining with regard to copyright

20Copyright includes a monopoly on reproduction, including the adaptation of works. The publisher owning the property rights for the scientific literature it publishes can therefore prohibit third parties, as well as authors, from making any full or partial reproduction as well as any translation, adaptation or transformation, arrangement or reproduction by any technique or process (Article L.122-4 of the Intellectual Property Code).

21TDM does not have its own legal status and this lack is a source of legal insecurity. Indeed, data-exploration services reflect multiple technical operations, including:

  • operations to analyse or process knowledge alone: these are acts undertaken freely;

  • technical operations involving the full-text reproduction of data that can be protected by copyright, and their modification in forms such as sections, extracts, mergers, compilations, etc.: some of these acts may concern author monopolies and consequently require the prior authorisation of the copyright owner.

22The absence of a legal status for TDM and the lively doctrinal debate on the incompatibility of these exploration techniques with the copyright provisions are a source of legal uncertainty and call for rapid legislative clarification.

+ The absence of a legal status for data exploration is a source of legal insecurity that the law must address.

Text and data mining with regard to the right of the database creator

23Principles of the sui generis right. Although, in principle, data cannot be individually protected (except in the event that the data are protected by a private right, intellectual property right, right of personal data, right to privacy), the aggregation of a significant amount of data can, where appropriate, be protected under the sui generis right of the database creator.

24Databases are defined in the French Intellectual Property Code as a “collection of independent works, data or other materials, arranged in a systematic or methodical way, and capable of being individually accessed by electronic or any other means”.3

25The legal framework for the protection of data is defined by the provisions of the Directive of 11 March 1996 on databases4 (transposed in France by the Act on the Legal Protection of Databases5), which create a “sui generis” right in favour of the database creator.

26The database creator is defined as the person who takes the initiative and the risk of the investment. The creator may prohibit:

  • the extraction of all or a substantial part of the content of the database;

  • the reuse of all or a qualitatively or quantitatively substantial part of the content of the database;

  • and/or the repeated and systematic extraction or reuse of qualitatively or quantitatively non-substantial parts of the content of the database when these operations clearly exceed the normal conditions of use of the base.6

27Research database. Digital STI is accessible from the databases of scientific publishers that have, little by little, mainly replaced their print editions by online access to their journal via their platform. As such, the publisher is the database creator and can therefore prohibit any qualitatively or quantitatively substantial extraction of its base.

28Digital STI is also available from institutional databases, overlay journals and open access databases. The creators of each of these databases are also holders of the sui generis right and may prohibit any qualitatively or quantitatively substantial extraction from them.

29Sui generis right versus TDM. To perform TDM on corpora of data, the following operations are necessary:

  • extraction from databases covered by the exclusive right of the database creator;

  • technical operations not covered by the regime of the database creator.

30Some argue for a revision of the right of the database creator, maintaining that the original text corresponds “to a static vision of data processing that is now giving way to dynamic processing”.

  • 7 La nécessaire évolution du droit du producteur de base de données pour permettre son adaptation à l (...)

31On the basis of a review of the legislation, the existing case law and the issues facing the actors in the data production, processing and analysis sector, proposed changes can be made to adapt the right of database creators to its new technological and commercial environment, transforming it into a right of data and database producers and operators.”7

+ The absence of a legal status for data exploration and the unsuitability of the right of database creators to the dynamic processing of knowledge are sources of legal insecurity that the law must address.

The need for reformed rights in scientific digital publishing

32Scientific publishing practices are fundamentally different from those of literary publishing:

  • the purpose of the publication is different: unlike a literary work, in scientific publication the informational content prevails over the form of expression, which may be incidental;

  • the content of the publication is different: a literary creation is specific to its author while scientists, whose “raw material” is science, mostly exploit the work of their predecessors;

  • the work provided by the publisher is different: the literary publisher assists their authors in the drafting of the book, encourages them (including financially by the provision of credit), involves them in sales, works on formatting and presentation, organises publication, distribution and promotion, supports the authors in these promotion activities, etc. Conversely, the scientific publisher receives only completed articles (after they have undergone peer review), does not involve the authors in the selling of journals or subscriptions, and works on layout and online distribution.

33Digital scientific publishing is leading to practices being transformed and necessarily imposes a revision of the contractual rules of the game:

  • in the relationship between publisher and researcher;

  • in subscription contracts;

  • in public–private partnership contracts and the organisation of publication of articles resulting from the partnership.

Publisher–researcher contract: Contract of transfer of copyright

34Articles by researchers are published by publishers in the framework of a publishing contract. In addition, the publication is especially important to the researcher, who is evaluated mainly on the basis of this indicator.

35Publishers are responsible for several major tasks:

  • dating of the article;

  • ensuring that the article is reread and validated by an editorial board consisting of specialist researchers (peer-reviewing), usually for no payment;

  • possible page layout of the article;

  • possible correction of the language;

  • dissemination of the article through its own channels;

  • registering the article in a database or assigning a DOI (Digital Object Identifier);8

  • archiving of the article;

  • management of copyright;

  • publicity for the journal and the article.

36The publishing contract between a researcher and a publisher most often takes the form of an adhesion contract. It provides for a transfer of the researcher’s copyright to the publisher, most often on an exclusive basis and free of charge, for exploitation worldwide and for the entire legal term of copyright. Many of the testimonies mentioned the practice of publishers of getting researchers to sign a “copyright transfer form”. This contract is “written in a way that only a specialist in ‘copyright’ law can understand”.

  • 9 Contribution of Marie Farge, Senior Researcher, CNRS.

37Researchers often sign these even without reading them because they often lack support from their establishment in offering a reasoned opinion and the means to defend their interests as creators. Lastly, the speed of publication is often an important point in the context of international competition, and researchers are rarely given the time to implement a procedure to validate the appropriate contract.9

38Hybrid model. Authors making their work freely available in hybrid digital scientific journals (free access and subscription access) must generally pay a fee (article processing charges) to cover the journal’s costs.

39The CNRS Ethics Committee (COMETS), in an Opinion “on the relationships between researchers and scientific publishing houses” of 31 January 2011,10 describes this situation as follows:

  • [t]he transfer of copyright for an article accepted by the editorial board of a journal, which may be based in one country or another, on the recommendation of one or more reviewers, is most often requested by the publisher free of charge. If an author refuses to sign the copyright transfer form, their article, despite having been accepted by the editorial board, will generally not be published. If on the other hand they sign this form, in principle they surrender the right to disseminate their article themselves and to use the figures and tables of data it contains, since the publishing house has become the owner, and in most cases has not even paid the author for the loss of these rights.”

40Indeed, entering into a publishing contract with exclusive transfer prevents researchers, in particular, from:

  • placing their article online on the institutional platform of their employer organisation, which funded the research that led to the article;

  • sharing the article with other researchers interested in the work;

  • disseminating the article on the researcher’s website;

  • reusing the graphics and media in other publications or oral presentations;

  • depositing the article in an open archive.

41Some publishers, aware of the importance to research of making articles available and the trend towards Open Science, authorise articles to be placed online on an open archive after an embargo period has been respected (post-print). The HELOISE site (​) is an information service on publishers’ policies with respect to the filing of articles. This service concerns only articles deposited on the websites of the scientists themselves and of scientific institutions.

Publisher–institute contract: Licence contract

42In order to gain access to scientific journals and books, the institutes enter into subscription contracts with publishers, which make access available to the publisher’s journals and online services.

43This subscription relating to access to the publisher’s current collections and/or archives, is:

  • either entered into directly between the institute and the publisher;

  • or entered into via a group order (as part of a “national licence”).

  • 11 Analysis of contracts:
       - Elsevier – ABES of 31/01/2014, subscription to the Freedom Collection;

44The licence agreements contain different provisions depending on the publishers but generally provide for a common set of rights in favour of the institutes:11

  • a right to access, consult or display the collections;

  • a right to print or make an electronic copy for the subscriber’s own use;

  • sometimes a right to practice TDM via the API (Application Programming Interface) of the publisher, which then retrieves the data on usage; the dissemination of the result of the TDM may include extracts from the full text limited to a certain number of words or a percentage of the text, under the Creative Commons CC BY-NC licence, and with a link to the full text of the article on the publisher’s site.

45Users are not permitted to extract, modify, translate, or create any derived work of any kind from the data made available by the publisher as part of the subscription.

Publication in industrial exploitation contracts

  • 12 Regime that emerged from the Decree of 3 July 2012 relating to the protection of the nation’s scien (...)

46The provision of research data must be organised and must take into account the nature of the data. Data from a restricted regime area (ZRR)12 or from ongoing research, as well as data related to know-how or secrets, or to industrial property titles, must not be made available systematically.

47Moreover, in the framework of research collaboration contracts and public–private partnership contracts, the terms of publication of research results are covered by specific provisions, especially when the contract leads to the filing of a patent or is subject to a duty of confidentiality.

48Patent clause. The contract may stipulate that if the research results are patentable, the partners shall file a patent.

49The contract must then stipulate the terms for filing the patent (single or joint ownership), and the terms for exploiting the patent (exclusive operating licence, operating licence by business sector or by geographic area, conditions for sharing of royalties, transfer).

50An invention is patentable if it constitutes a novelty with regard to the state of the art. The latter is understood to mean anything that has been made accessible to the public before the patent filing date, by a written or oral description, a usage or any other means, including the content of French, European or international patent applications, provided that they designate France, were filed earlier, and were not yet published on the date the patent application in question was filed.13

51Thus, if the invention has been made public in any part of the world, not only by a prior patent but also by a publication, a public exhibition (at a trade fair for example) or even a simple oral disclosure, it is no longer new.

52However, if the disclosure of the invention resulted from wrongdoing,14 it does not affect the novelty of the invention if it took place in the six months preceding the filing of the patent application.

53Disclosure may be regarded as improper if it occurs:

  • without the agreement of the inventor (theft of the invention, industrial espionage);

  • in violation of a secret (trade secrets, for example);

  • in violation of a contractual duty of confidentiality.

54When a contract provides for the filing of a patent at the conclusion of the cooperation, the researcher(s) responsible for research and development are prohibited from publication as this risks destroying the novelty of the innovation and preventing any patent from being filed.

55Confidentiality clause. A growing number of research contracts provide for clauses governing the confidentiality of the “own knowledge” of each of the parties and the conditions of publication of common results derived from the research.

56The following partial clauses can for example be included in research contracts entered into by the CNRS with industrial companies:

  • Example 1 is drawn from a standard research collaboration contract between the CNRS and an industrial partner;

  • Examples 2 and 3 are derived from framework contracts between the CNRS and industrial partners.

Example 1

Example 2

Example 3

Key definitions

Confidential Information
Own Knowledge
New Knowledge

Confidential Information
Common Results

Confidential Information (list)
Business Data from X
Exploitable Results 


Duty of confidentiality concerning the Confidential Information during the term of the contract and in the 5 years following the termination or expiry of the contract.

Duty of confidentiality concerning the Confidential Information during the term of the contract and in the 5 years following the termination or expiry of the contract.

Are considered as Confidential Information:
   - another Party’s Own Knowledge and the Business Data from X for a period of confidentiality of 10 years after contract termination, for whatever reason;
   - the Exploitable Results.


Written agreement of the other Party for any publication about the New Knowledge during the term of the current contract and in the 6 months following its expiry.

Any draft publication or communication shall be subject to the consent of the other Party, which may delete or modify certain statements whose disclosure could be detrimental to the industrial and commercial exploitation of the new knowledge under favourable conditions. Such deletions or amendments must not affect the scientific value of the publication.

If the information contained in the publication or communication must be protected as Industrial Property, one of the Parties may withhold publication or communication by a maximum period of 18 months from the date of the corresponding request.

Publications or communications relating to the derived Common Results shall make reference to the cooperation of the Parties.

Any draft publication or communication relating to the Common Results must, during the term of the specific agreement and in the 24 months following its expiration or termination, receive the prior and written agreement of the other party.

They may decide:
   - to accept without reservation the draft publication or communication; or
   - to request that the Confidential Information belonging to it be removed from the draft; or
   - to request that the Common Results that have been the subject of a secret technical file be removed from the draft; or
   - to request changes, especially if some of the information contained in the draft communication could be detrimental to the industrial and commercial exploitation of the Common Results; or
   - to request that the communication be deferred, if genuine and serious reasons seem to warrant it, in particular if an application is to be made for protection of the information contained in the draft communication as Industrial Property.

A prior request in writing, by registered letter with acknowledgement of receipt, must be made by one Party to the other Party for any draft publication or communication, regardless of the form or media, relating to the Research Programme and the Results, for the period of confidentiality of the Confidential Information.

The other Party may request the deletion or modification of certain elements of the publication whose disclosure it believes could be detrimental to it, or detrimental to the industrial or commercial use of the Results of the Research Programme or to the protection of an intellectual property title.

In particular, a Party may request that the publication or communication be delayed, for a maximum period of 18 months, if certain information is to be protected under an intellectual property right.


Free for use in researcher activity reports and thesis defences.

Free for use in researcher activity reports and thesis defences.

Free for use in researcher activity reports and thesis defences.
Free for use in communications and filing of patent application on own results.

57This growing contractual practice, increasingly involving negotiations, whose aim is to regulate publication by patent and/or confidentiality clauses, must be taken into account for the provision of research data, in order to preserve the balance necessary for the exploitation of innovations.


1 In 2001, the Massachusetts Institute of Technology presented data exploration as one of the 10 emerging technologies that would “change the world in the 21st century”. (Stéphane Tuffery, Data mining et statistique décisionnelle – l’intelligence des données [Data mining and statistics for decision-making], Editions Technip, 2012).


3 CPI Art. L. 112-3.

4 See on the legal protection of databases.

5 Act 98-536 of 1-7-1998 on the transposition into the Intellectual Property Code of Directive 96/9/EC of the European Parliament and of the Council of 11 March 1996 on the legal protection of databases.

6 CPI Art. L. 342-3.

7 La nécessaire évolution du droit du producteur de base de données pour permettre son adaptation à l’émergence du Big Data (The necessary change in the rights of database creators to enable their adaptation to the emergence of Big Data) by Nicolas Courtier for La propriété intellectuelle & la transformation numérique de l’économie, Inpi, 10/9/2015.


9 Contribution of Marie Farge, Senior Researcher, CNRS.


11 Analysis of contracts:
   - Elsevier – ABES of 31/01/2014, subscription to the Freedom Collection;
   - Elsevier – ABES for ISTEX 2013-20 contract;
Elsevier – CNRS 2010-09 contract.

12 Regime that emerged from the Decree of 3 July 2012 relating to the protection of the nation’s scientific and technical potential and the Inter-Ministerial Circular on establishment of the scheme for protection of the nation’s scientific and technical potential of 7 November 2012


14 Article L. 611-13 of the CPI


Open access


Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search