Skip to end of metadata
Go to start of metadata

----------------------------------------------------------------------------------------------------------

This wiki was open for comments during 10 days after the Amsterdam Presidency Conference on Open Science (4-5 April 2016). The Dutch Presidency greatly appreciated all input and studied the comments carefully and with great interest.

Having studied the comments and considering the wide variety of the input, the Dutch Presidency decided not to publish a second or ‘final’ version of the Amsterdam Call for Action on Open Science. Also, composing a summary of all input was not really possible. Therefore it was decided to select a great number of highlights emerging from the comments.

Together with the Amsterdam Call for Action on Open Science, the document with the highlights will be published shortly on the website www.openaccess.nl 

Many thanks on behalf of the Dutch Presidencfor all your input and comments!

 

------------------------------------------------------------------------------------------------------------


Open science
Open science is about the way researchers work, collaborate, interact, share resources and disseminate results. A systemic change towards open science is driven by new technologies and data, the increasing demand in society to address the societal challenges of our times and the readiness of citizens to participate in research.
Increased openness and rapid, convenient and high-quality scientific communication - not just among researchers themselves but between researchers and society at large - will bring huge benefits for science itself, as well as for its connection with society. Open science has impact and has the potential to increase the quality and benefits of science by making it faster, more responsive to societal challenges, more inclusive and more accessible to new users. An example of this potential is the response to the outbreak of viral diseases such as Ebola and Zika. Access to the most recent scientific knowledge for a broad group of potential contributors, including new or unknown users of knowledge, has brought solutions closer. Open science also increases business opportunities. The speed at which innovative products and services are being developed is steadily increasing. Only companies (notably SMEs), entrepreneurs and innovative young people that have access to the latest scientific knowledge are able to apply this knowledge and to develop new market possibilities. Citizen science brings research closer to society and society closer to research.

A speedy transition is needed
For Europe to remain at the forefront and to ensure sustainable growth in the future, open science holds many promises. Reality, however, has not caught up yet with the emerging possibilities. The majority of scientific publications, research data and other research outputs are not freely accessible or reusable for potential users. Assessment, reward and evaluation systems in science are still measuring the old way. Although these issues are recognised and countless initiatives have been developed during recent years, policies are not aligned, and expertise can be shared more and better. There is a strong need for cooperation, common targets, real change, and stocktaking on a regular basis for a speedy transition towards open science.
The good news is that there is political and societal momentum. More and more researchers are supporting the transition and are moving towards open science in the way they work. Organisations from the scientific community are urging politicians to act. The European Commission and the Council of the European Union have expressed that they are prepared to take a leading role to facilitate and accelerate the transition towards open science.

From vision to action
This Call for Action is the main result of the Amsterdam conference on 'Open Science – From Vision to Action' hosted by the Netherlands' EU presidency on 4 and 5 April 2016. It is a living document reflecting the present state of open science evolution. Based on the input of all participating experts and stakeholders. Stakeholders include research funders, Research Performing Organisations (including researchers, libraries and support staff), publishers (including information service providers) and businesses. as well as outcomes of preceding international meetings and reports, a multi-actor approach was formulated to reach two important pan-European goals for 2020.

1. Full open access for all scientific publications. This requires leadership and can be accelerated through new publishing models and compliance with standards set.

2. A fundamentally new approach towards optimal reuse of research data. Data sharing and stewardship is the default approach for all publicly funded research. This requires definitions, standards and infrastructures.


To reach these goals by 2020 we need flanking policy:

3. New assessment, reward and evaluation systems. New systems that really deal with the core of knowledge creation and account for the impact of scientific research on science and society at large, including the economy, and incentivise citizen science.

4. Alignment of policies and exchange of best practices. Practices, activities and policies should be aligned and best practices and information should be shared. It will increase clarity and comparability for all parties concerned and to achieve joint and concerted actions. This should be accompanied by regular monitoring-based stocktaking.


Twelve action items with concrete actions to be taken
Twelve action items have been included in this Call for Action. They all contribute to the transition towards open science and have been grouped around five cross-cutting themes that follow the structure of the European Open Science Agenda as proposed by the European Commission. This may help for a quick-start of the Open Science Policy Platform that will be established in May 2016. Each action item contains concrete actions that can be taken immediately by the Member States, the European Commission and the stakeholders:

Removing barriers to open science
1. Change assessment, evaluation and reward systems in science
2. Facilitate text and data mining of content
3. Improve insight into IPR and issues such as privacy
4. Create transparency on the costs and conditions of academic communication

Developing research infrastructures
5. Introduce FAIR and secure data principles
6. Set up common e-infrastructures

Fostering and creating incentives for open science
7. Adopt open access principles
8. Stimulate new publishing models for knowledge transfer
9. Stimulate evidence-based research on innovations in open science

Mainstreaming and further promoting open science policies
10. Develop, implement, monitor and refine open access plans

Stimulating and embedding open science in science and society
11. Involve researchers and new users in open science
12. Encourage stakeholders to share expertise and information on open science

  • No labels

45 Comments

  1. I would like to propose to add the word 'improve' to the first science. Standing on shoulders, etc.

    Perhaps add ", and improve" and make it something like: "Open science is about the way researchers work, collaborate, interact, share, and improve resources and disseminate results."?

    Egon Willighagen

  2. Anonymous

    This could be made far more inclusive and have a far bigger impact if you exchanged the word 'science' for 'scholarship' throughout. Science to an arts and humanities scholar is an alienating term that says 'not us'. We are a stronger voice together.

    1. Anonymous

      Agree. Open Scholarship is a more inclusive term.

      Lars Bjørnshauge SPARC Europe & DOAJ

    2. Anonymous

      Without the humanities true Open Access cannot be achieved, see Berlin Declaration - Sciences and Humanities. Fortunately, the late trend of Digital Humanities is the first sign that arts and humanities are coming to the fold forced by technologies they employ.

      Nicolaie Constantinescu, Kosson

    3. Anonymous

      Disagree. To most regular people, the term 'scholarship' means funding for tuition, generally merit-based. 'Science' means the systematic acquisition of new knowledge which academics, more generally call 'scholarship'. See for example the White House Champions of Open Science, where many of the scholars recognized were from the arts and humanities for an example where the German 'Wissenschaft' interpretation of 'science' as scholarship is used in English as well. "Open Science" is a term people understand even if they are outside of academia, therefore it is more inclusive. "Open Scholarship," if you are outside of academia, sounds more like "everyone is welcome to apply here for funding." Because 'scholarship' is only used as you intend it within academia, it is the more exclusive term v. "Open Science." Similarly with "Open Research," where unfortunately 'research' is just part of the process of 'science'; Open Science is not a perfect construction but it is the best we have.

      1. Another alternative is Open Knowledge, as in the Open Knowledge Foundation (https://okfn.org/). This has an Open Science working, along with many other (https://okfn.org/get-involved/working-groups/). To me personally, Open Science is not just about natural sciences, but also humanities.

  3. Anonymous

    Hello,

    I applaud this initiative, and wish it well. We started a similar one, by producing a declaration a few weeks ago, to express the position of African researchers on this point - http://www.sci-gaia.eu/dakar-declaration/

    Perhaps we could cross-reference each other ? Instead of using a wiki, we used a Github repository http://github.com/Sci-Gaia/DakarDeclarationOpenScience

    I really like the the fact that you have a list of action items, which readers to guage whether they are indeed in support of the movement of Open Science, or whether they are paying lip-service.

  4. Anonymous

    This document has some admirable objectives and it is to be applauded for highlighting the need to change assessment, evaluation and reward systems, which is surely the key requirement for success. But I wonder if some of the goals might not be in conflict, or contradictory.

    For instance, how likely is it that new publishing models and evaluation and reward systems can be realised in any significant way if full access to all scientific publications is to be achieved by 2020?

    It seems to me that the only way that the 2020 target can be reached is by opting for the OA “flipping” strategy being pioneered by Dutch Universities, and promoted by Max Planck. While one could argue about the wider merits (and likely success) of this approach (is it even affordable?), its achievement would surely only lock the legacy publishing model (the traditional journal) into the new environment, along with the evaluation systems that delegates were so critical of (impact factor, journal prestige etc.).

    Fundamental to achieving open science is to engineer a change of culture (a word I do not think occurs once in the document). There are two ways of achieving a cultural change: carrots (new reward systems etc.) and sticks (e.g. OA policies). For the reasons stated above, the former would not seem to me to be compatible with a flipping strategy and the latter, as the Norwegian State Secretary for Education and Research Bjorn Haugstad pointed out at the event, is likely to alienate researchers rather than encourage them (and so slow progress) – a truth that the Higher Education Funding Council for England (HEFCE) will likely come to realise with its recently implemented OA policy.

    All of which is to say that what I feel is missing in this document is a clear, joined up view of how the cultural change that will be needed if open science is to be realised can be achieved, certainly in the time-scale proposed.

    Richard Poynder

    http://poynder.blogspot.co.uk/

  5. Anonymous

    One minor addition: I'd change the first goal from "Full open access for all scientific publications. " to "Full immediate open access for all scientific publications. "

    Obviously, immediacy enables participation of a broader audience, allows for better ways of collaboration amongst reserachers, and adds value in many more ways. "Delayed OA" seems to be mostly due to the idea, that a better process of sharing research results is less important than to protect certain business models in academic publishing. We should get rid of this barrier as soon as possible.

    Lambert Heller

    http://biblionik.de/

    1. Anonymous

      Agree. Immediate Open Access is the real thing.

       

      Lars Bjørnshauge

      1. Anonymous

        A real slump in the academic publishing sector is also a "real thing". Full and immediate open acces could (also) well create an outrage among researchers and many other parties. It might be more prudent to take one step after another in the pursuit of open access policies.

        Most of all, the open access debate needs more honesty about who profits from which solution and about what (reciprocal) deals can or should be negotiated between business and science regarding the exchange of data and results.

  6. Anonymous

    Kristiina Hormia-Poutanen /LIBER, president

    LIBER is an Association of European Research libraries. As an independent association LIBER is active in advocacy work, publishes policy statements, participates in the development of Open Science infrastructures, participates and coordinates EU-projects and supports LIBER libraries in fostering pen Science.

    Research libraries are not a homogeneous group. LIBER libraries are university libraries, national libraries and special libraries. Also national library associations belong t the network of LIBER .National libraries are mostly independent legal entities.

    Libraries/Research libraries should be mentioned as a stakeholder group as they have specials role in enabling Open Science

    1. Anonymous

      Totally agree. Libraries are important stakeholder and can take part in most of the proposed actions.

      Jadranka Stojanovski

    2. Anonymous

      Dear Kristina,

      Richard Poynder above and your comments resonate with two years experience with FP7 FOSTER of training the entire ecosystem if research actors that play a key role in Open Science/Scholarship implementation (DOI 10.5281/zenodo.30564):

      Open Science/Scholarship is a major CULTURE CHANGE exercise and any Call for Action need to be holistic, inclusive and take into account ripple effect up and down the research value chain.

      Librarians are the quiet and unsung heros of this culture change. One axis that drives right through all the Actions proposed is training the new generation of researchers, sitting in a Graduate School somewhere, that research excellence alone does not prepare them for the essential Research Evaluation Framework criteria changes proposed here (and research on alternative metrics funded by EC and to be implemented by 2020).

      When we approached Graduate Schools in ERA with structured Learning Objectives (http://dx.doi.org/10.5281/zenodo.15603), a taxonomy of Open Science/Scholarship so that they can use as a backbone for training graduates, their main concern was that institutional Librarians & Knowledge Managers may not have the skills and capacities to train graduates especially in the case of Research Data Management, Data Mining, Data Visualisation in the data-intensive disciplines.

      Barend Mons echoes a similar concern during the #EU2016NL Open Science meeting. This may polarise the Library community in terms of what is their role in a digital data-intensive world, but it is also a key opportunity for Librarians to be seen as key partners in opening up scholarship, co-designing research questions together, wiining proposals together and mapping the knowledge landscape together with researchers, rather than being perceived as an overhead service.

      How do we bring this on LIBER and #EU2016NL agendas and work with Librarians to boost their Data Science training and make them even more indispensable to researchers?

      Ivo Grigorov

  7. Anonymous

    The International Association for Scientific, Technical and Medical Publishers (STM) welcomes the useful contribution of the Amsterdam Call for Action on Open Science to the discussion about how to achieve Open Science. Our full statement can be found here: http://www.stm-assoc.org/2016_04_08_News_Release_STM_response_to_Call_for_Action_on_Open_Science.pdf

     

    Matt McKay, STM

     

  8. Anonymous

    Great to see the progress in the report.

    The first alinea of 'open science' could be a bit more pointly/ focussed.

    I would suggest "a priority and startingpoint of Open Science is to optimize the development, implementation and availibility of knowledge, technology and products (including medicines) to the society at a reasonable cost. "

    I include this because the starting and end point should be society: as payer and benificient. 

  9. Anonymous

    "Only companies (notably SMEs), entrepreneurs and innovative young people that have access to the latest scientific knowledge are able to apply this knowledge and to develop new market possibilities" this states that ONLY commercial enterprises and young people can do these things to the exclusion of all others. Which is simply not true. Perhaps you mean "Companies (notably SMEs), entrepreneurs and innovative young people are able to apply this knowledge and to develop new market possibilities only if they have access to the latest scientific knowledge". 

  10. Anonymous

     

    The scientific enterprise is global. Many countries at different level of scientific and economic development are currently moving forward to define and implement open science strategies and agendas, not only within but also outside Europe.

    It would be important for any action undertaken at European level to be consistent and coordinated with other international strategies and actions. In October 2015, Science ministers and senior governmental representatives from 52 countries agreed the OECD-Daejeon declaration on STI Policies for the Global and Digital Age, which includes a joint commitment to support Open Science.

    The report Making Open Science a Reality (OECD 2015) provides an overview of the areas where policy action is required both nationally and internationally, and it analyses many of the issues covered by the Amsterdam call including the need to conceive open science in a broad way beyond  access to publication and research data, the importance of skills for open science and the partial lack of incentive mechanisms to promote data-sharing practices among researchers.

    The OECD Global Science Forum is exploring policy options and business models for internationally coordinated open data infrastructure. It is important that the EC, OECD and other relevant STI policy bodies work together from the outset with a shared vision for an inclusive global Open Science enterprise. This requires coordinated policy  actions at the national, regional and global scales that complement the "bottom up" activities of the Research Data Alliance and similar international science community led actions. The proposed European Open Science Policy Platform and the European Open Science Cloud would benefit from taking an explicitly inclusive global perspective, strengthening both the European and Global Research Areas.

     

    Giulia Ajmone Marsan, OECD

     

     

    1. Anonymous

      Ivo Grigorov,

      Open Science is, as Giulia points out, about the scientific process, and that transcends Open Access to peer-reviewed publications. Open Science is, and must be, about the transparency in the entire Research Lifecycle, along each step of the research process (Open Notebook Science, Open Data, Open Research Software, Open Access to publications, Open Peer-Review of publications, Open Educational Resources based on all of the above) .

      Open Access is about one part (be it most visible part) of the research process, but Open Science DOES NOT EQUAL Open Access. OS is simply the way science should be done for optimum transparency, rigor and reproducibility, for without that tax-payer access to journal articles of less than rigorous/reproducible research is of dubious importance.

      Amsterdam Call to Actions must take this holistic thinking as a red-thread to all the actions proposed, if we are to underpin the core CULTURE CHANGE required at the knowledge generation source: the individual researchers and lab teams.

      Regards

  11. Anonymous

    Knowledge Exchange Response

    to action item number 6 (Set up common e-infrastructures) of the Amsterdam Call for Action on Open Science

    The underpinning vision of the Digital Single Market, of which EOSC is an essential part, is the free movement of goods, persons, services and capital; the access and online services available for citizens and business at fair competitive conditions; and the protection of consumer and personal data.

    Therefore the EOSC – envisioned as a federated globally accessible environment where researchers, innovators, companies and citizens can publish, find and re-use each other´s data and tools for research, innovation and educational purposes - must enable open and transparent movement of knowledge, funding of [human and technical] infrastructure to optimally leverage existing resources, and optimise future investment in a trusted, sustainable and just value for money model.

     

    In short the ambition is to close the value loop between Open Science, the Digital Single Market and the Public Interest. The value loop engages many stakeholders (e.g. research communities; research projects and H2020 projects (both data intensive and long tail); public data centres and service providers; commercial service providers, small and medium-sized enterprises; industry; society/citizens; EC, member states, councils, and more). Ideally the EOSC should meet all stakeholders’ needs to fulfil their role(s) in the research process.

    Therefore stakeholder’s demand (for required infrastructure and services), supply (of infrastructure services, research output) and the funding of these need to be given due and balanced attention if the EOSC is to be a success and become sustainable. Each required part of the value loop needs to exist, with clear responsibility, incentives, business model and visible benefits.

    In order to do so we recommend to give close attention to communication, to inform and involve all stakeholders and to target quick wins. It is important to aim to meet tangible challenges, be these contributions to new knowledge, innovation or ease of access and collaboration, or enhanced funding and business models.

    The ambition of the EOSC is to ensure that research is improved, answers to grand challenges accelerated, knowledge transfer increased, innovation and productivity sped up, and have citizens engaged in research and innovation. Europe enjoys a rich set of research infrastructures which are critical to the delivery of the EOSC. Collaboration across these parties, be they disciplinary or cross-domain infrastructures, public or private, in the traditional research domain or beyond, is crucial.

    Therefore to fulfil the ambitions of the EOSC it is essential that the development and implementation of the concept is more inclusive than it is now. We observe that currently talks and considerations on the EOSC are largely focused within the traditionally engaged stakeholder group of infrastructure and service providers that serve the needs of specific research communities. It is important to realise that the full range of stakeholders mentioned above should be engaged in establishing and will benefit from the EOSC.

    Our recommendation is that any action towards closing the value loop in a realistic and sustainable way should follow an inclusive approach in which research, SME’s and industry, public sector and society are brought in close concert.

    With a more inclusive approach the EOSC will be able to support the researcher in the entire research workflow and improve and accelerate communication of ideas and objects (inspiration!) in all phases of the research lifecycle to all - traditional and other - stakeholders and actors. Thereby fostering better research, increased collaboration, public engagement and innovation based on a sustainable financial model.

    On behalf of Knowledge Exchange

    Bas Cordewener

  12. Anonymous

    It might become confusing when the well known term "open access" is exchanged for "Full open access". Why not use the established term "open access"? /Lisbeth Söderqvist

  13. Anonymous

    First of all I’d like to applaud the Dutch government for the initiative to select "open science" as one of the priorities for the Dutch EU presidency and in particular to foster the implementation of open access to scholarly publications.   

    It will be interesting to see how this document can help to facilitate and speed up the transition.   

    In my assessment, "open science" is still a rather vague blanket term and a buzz word, at least different parties use it as a heading for quite diverse topics. Therefore it is not surprising that this document mirrors this situation: it tries to cover a broad range of topics and – at least in my view – it is missing a clear focus. The selection and the order of the 12 action items is not too convincing. In the current version of the document, some recommendations are rather ambitious and bold, while some action items (e.g. number 7) seem to follow a low key approach only. In my view, some prioritization is needed, too. In general, I share the concern expressed by R. Poynder that some of the aims might be in conflict, or even contradictory.   

    I have provided more detailed comments on the following action items, but I’d like to mention some general observations at this point:

    • Though the expression "open science" seems to be coined now, I would recommended to use the word "science" with caution in order not to exclude certain parts of research and scholarship, e.g. in the humanities.
    • I’m wondering whether no intrinsic motivation is mentioned in the very first paragraph under the headline "open science": Why/how is “open science” good for scholarship? Interestingly, in this document individual researchers are almost never mentioned in an active role. I’d like to suggest reconsidering this in order to mitigate the impression of a top-down approach.  
    • The title of this document is "Amsterdam Call for Action on Open Science", but unfortunately it is unclear to me who is calling for action. The ownership of this document should be made clear: Who is meant by "we" or "our" in this draft?

    I’m looking forward to the "final" version of this document to be used as an input for the Competitiveness Council in late May.

    Georg Botz

  14. The recommended action for research funders ignores the responsibility of funders to set priorities. The ERC is committed to a bottom-up process that treats all disciplines in the same way. That is its priority. Other Horizon2020 programmes decide priorities on other principles. In each case these may or may not align with open access principles. That is indeed something to monitor, but it does not absolve the funders from determining their own priorities

     

    Martin Stokhof, on behalf of the Open Access Working Group of the European Research Council


    Proposed changes:

    The problem

    Initiatives aiming to facilitate the transition towards open science are urgently needed. Many initiatives are already being undertaken. It is important to investigate and monitor the extent to which stakeholder actions contribute to innovations in open science. Results of such evidence-based research must be shared to show which actions we should support to move forward and which actions to abandon. Communication on successes is needed, but also on failures and actions that do not work.


    The solution

    • Facilitate evidence-based research on innovations in academic communication, while selecting and financially supporting new models.
    • Adopt an evidence-based approach for mainstreaming open science.
    • Demonstrate the benefits of opening scientific processes for scientists as well as society where such benefits exist.
    • Investigate how stakeholders can contribute to innovations in open science in ways that are in line with their mission.
    • Define and disseminate good practices and corresponding principles. 



    Concrete actions

    • All stakeholders: explore other ways of sharing result outputs, to serve the purpose of open science. Where relevant, investigate how the public could participate in the selection of scientific topics through online platforms.
    • National authorities and European Commission: actively contribute to peer learning about national policies, e.g. within the framework of the development of the European Research Area (ERA).
    • National authorities and European Commission: set up research programmes on developments in open access/open science to answer questions regarding the optimal road to open science, the advantages of open science for society at large etc.
    • Research funders: investigate how within the confines of your mission funding streams could be innovated to make science more open and innovative. Show best practices. Investigate how to best align funding schemes with open access principles. Accept uncertainty and pilots in open science research (more flexible funding, smaller scale, faster). Create a funding mechanism to explore paybacks to open science where this is in accordance with your mission.
    • Researchers and research institutes: collaborate in research into innovations in open science. 
    • Research libraries: raise awareness, participate in EU projects, collect best practices, create a forum to share experiences. 


    Expected positive effects

    • A quicker transition to mainstreaming open science;
    • An evidence-based approach that helps to make the right choices in achieving open science.|

     

  15. Anonymous

    Hello,

    what seems to be missing from the 12 concrete actions of Open Science is the actual data sharing. Nothing about sharing clinical trial data, or original research data. I wonder why?

    The point "Facilitate text and data mining of content" covers automatic reading of already published material, it is not at all about access to original research data which scientists base the texts, diagrams and tables inside their papers, but which they are too often rather reluctant to share with their peers.  

    With data sharing being apparently not that important for to Open Science policies, how exactly should "Change assessment, evaluation and reward systems in science" be done? What shall be rewarded at academic evaluation then, besides publishing in Open Access? Open Blogging? Open Tweeting? Participation at Open Access/Open Science Conferences? 

    Finally, is there any way to see the EU2016NL meeting program, with speaker list, their affiliations and talk titles/abstracts? I keep requesting it and get silence in return. Not really open, I am afraid. 

    Leonid Schneider, Science Journalist

    https://forbetterscience.wordpress.com/2015/12/07/open-science-open-scientists/

    https://forbetterscience.wordpress.com/2016/02/08/pace-trial-and-other-clinical-data-sharing-patient-privacy-concerns-and-parasite-paranoia/

     

  16. Anonymous

    See https://en.wikipedia.org/wiki/Amsterdam_Call_for_Action_on_Open_Science for the Wikipedia page about this Call for Action, with responses to the call in the references. 

  17. Anonymous

    How many and which "Organisations from the scientific community are urging politicians to act"? Could be good to be clearer about this./Lisbeth Söderqvist, Swedish Research Council 

  18. Anonymous

    I think, as others have already underlined it, that there is an implicit and 'subtextual' emphasis all through the document on the need of having a new business model with the traditional Major publishers, which would allow that all publications will be freely accessible for 2020. The risk here is IMO to reduce OA to the OA Gold APC model, while OA archiving (the so called 'Green' road) allows the dissemination of a broader (and often societally very relevant) scope of research outputs. IMO Open Science opens a more radical evolution of the science dissemination system, which will only be possible if an important change occurs in the academic incentive system.

    On the other hand, I think that there is no 'one size fits all' approach in Open Science with regard to the way to best engage with the public, or to assess the impact of science on extra-academic stakeholders. While in some cases (strategic and problem driven research in particular) Open Science practices indeed allow the researchers to be more responsible towards non-academic stakeholders and take their "societal needs" better into account, it is IMO important to underline that curiosity driven research still has an important role to play in the current ecosystem of research, and that it is not wishable to assess ex ante the "societal impact" of purely curiosity driven projects (by definition, the scientists do not know precisely what they will find.) So, the link between Open Science and "societal impact" should be presented in a more nuanced way in the document (with some "where relevant" here and there). In the same perspectice, it should be better distinguished between the quality of the scientific contents, as they are best assessed through (open) peer reviewing, the impact  of science on scientists (as indicated by citations) and the impact of science on extra-academic audiences (as indicated by altmetrics and qualitative case studies).

    Marc Vanholsbeeck, PhD
    Université Libre de Bruxelles 

  19. Anonymous

    Comments by Christoph Bruch:

    "This Call for Action is the main result of the Amsterdam conference on 'Open Science – From Vision to Action' hosted by the Netherlands' EU presidency on 4 and 5 April 2016."

    Unfortunately the host of the conference offered very little opportunity for input on the "Call for Action". In that respect the host did not head the motto of the conference "Open Science".

     

     

    "2. Facilitate text and data mining of content"

     

     

     

     

    The right to mine all content accessed legally is of fundamental importance the information society.

    This right should not be linked to any kind of use scenario such as research.

    The right to mine directly relates to freedom of information.

    This needs to be reflected in legal reforms which should clarify/establish that

    -           TDM is not a usage protected by any IPR

    -           contractual agreements to the contrary are to be considered void

    -           producing copies which are needed in the cause of TDM do not infringe copyright

    Regulating TDM via exception-like approaches linked to use-cases like research or non-commercial endeavours is a wrong approach.

    Researchers need to be able to freely mine legally accessed content but they are not the only ones.

    Freedom of information does not only imply certain access rights but also the right to freely analyse legally accessed content with any technology available and unhiderd use of the results of that mining.

     

  20. Anonymous

    AOASG response to Amsterdam Call for Action on Open Science

    Submitted on behalf of Australasian Open Access Strategy Group (AOASG http://aoasg.org.au/)  by Virginia Barbour

    General comments:

    We welcome the initiative being taken by the EU on open science.

    Like other commentators, we would urge the consideration that these principles should apply to all of forms of scholarly outputs across all disciplines, not just science.

    Specific comments below:

    We recognize the need for global, not just pan-European collaboration on many of these action items, in particular for action items1.  Change assessment, evaluation and reward systems in science and also item 6. Set up common e-infrastructures

    1. Changing the reward system, to incentivize openness and sharing is the core to any wholesale transformation of scholarly communication towards openness. This change needs to happen not just at the level of individual departments or institutions or even countries, but needs to be truly international and become part of the assessment of institutions, not just individual academics. For example, many of the current university league tables are heavily reliant on the publishing as it exists now. Unless alternatives are available there will not be any substantial buy in globally.

    6. Common infrastructure will underpin whether or not openness fulfils its promise. Rather than common "e-infrastructures" it may be better phrased as common international standards which facilitate inter-operability, reuse, citability, reproducibility and linking. Without interoperable structures globally we risk repeating the current situation with silos of open research outputs, rather than silos of closed outputs, as we have now. We note that in the detailed explanation of this point  the need is  stated to “Align practices in Europe and beyond” and we urge that this is considered early in any developments.

    We believe further clarification of the intentions and extent of the recommendations around items 2, 3 and 5 are needed. Specifically, text and data mining rights should be extended to research publications as well as data. Furthermore, reuse should extend to other uses beyond TDM.

    7. and 8. We very much support the stated intentions to “Provide a framework for developing new publishing models” and “Encourage the development of publishing models that provide free access for readers/users.” We suggest that the publication in such new models should be specifically rewarded under any new incentive structures that are developed.

     

  21. Anonymous

     Comments from Estonian Ministry of Education and Research:

    • Improve insight into IPR and issues such as privacy. We do not support Open Data to be a default principle applied to all publicly funded research. As Open Data is an expensive goal, it is important to make choices and set priorities – which data to preserve. We welcome Horizon 2020 Open Data Pilot, and believe that the experience gathered through the pilot should be used as an input to national strategies.
    • Introduce FAIR and secure data principles. Our comment is to the phrase ’educate data stewardship experts’ – which focuses on the approach that data management is something that researchers are not able to manage. We agree that in the long run we need to educate data stewardship experts; but in the first place it is important to improve the digital skills (including data management skills, new quantitative methods, data visualization etc) among researchers themselves so that they would adopt to the changing paradigm and start to prioritize and value data as such. This also includes p11 (Involve researchers and new users in open science), where topic of researchers’ digital skills is not clearly prioritized.
    • Set up common e-infrastructures. Many aspects here are unclear: how the member states are involved, what is the time frame, how it is funded, what procedures are planned for participating of different stakeholders (as well as all the MS that have to carry big part of the financial burden) in cloud initiative. It is unclear how different ongoing open cloud and open science initiatives and processes interact in parallel.
    • Adopt open access principles. In this paragraph, golden Open Access is implied. As a principle, we prefer a broader approach to Open Access. Open Science is an ongoing process, which we should not try to push inside a box, so a certain flexibility must be left for alternative developments (e.g. new publishing models, Green OA etc.).     

    Martin Eessalu

    Chief-expert 

    martin.eessalu@hm.ee

    Research Policy Department

    Ministry of Education and Research, Estonia

  22. Anonymous

    Comments of the Belgian Federal Science Policy Office, after consultation of regional Belgian authorities.

    Proper consideration is required when advocating better gearing of science towards societal and economic needs. Whereas this may be necessary for applied sciences we have to be careful not to apply this principle (at least not as a policy) to curiosity driven fundamental research which works best through serendipity, chance and failure. Reorienting fundamental research to better fit societal demand is contrary to free inquiry and is detrimental to the economy and the development of society.

    The complete absence of mentions of both pro-Green open access and pro-flipping towards academic controlled fair Gold Open Access advocacy precludes the academic option to regain control over its own output, which Belgium fears might harm research quality and funding. All too much focus on Gold Open Access - and despite calls for innovative models - on the APC journal based model - paves the way to a not so innovative 'one size fits all' solution for policy makers faced with an obligation to end all embargoes by 2020, not to speak of the looming APC crisis succeeding the serials crisis. Funding of repositories for diffusion of grey literature and post prints has to be part of the equation.

    The sustainability and economic viability of private publishers is none of our concern. Society changes and businesses adapt to them.

    The replacement of journal based impact factors by article based quality factors is a key issue to the wellbeing of science. It is not being stressed enough.

    Privatisation of data and the commercial orientation of fundamental research results through high IF journals are to be thwarted through any proposed policy. Attention should primarily go to enhanced tools for scientific exchange and openness.

     

    Eric Laureys

    Scientific Officer

    Open Access Project

    eric.laureys@stis.belspo.be

     

  23. Anonymous

    The section 'From vision to action' lists the following stakeholders: "Stakeholders include research funders, Research Performing Organisations (including researchers, libraries and support staff), publishers (including information service providers) and businesses." This list should be extended to include also learned societies, e-infrastructure providers, and relevant civil society organisations.

    Under  '2. A fundamentally new approach towards optimal reuse of research data' the text mentions that "Data sharing and stewardship is the default approach for all publicly funded research. This requires definitions, standards and infrastructures." I think to this should be added "awareness raising and training".

    Dagmar Meyer, Brussels

  24. Excellent initiative. There is overlap between actions 4, 7 and 8, and that it would be much better to have one clear thematic action on "OA to publications" (instead of trying to fit into the European OS Agenda categories) saying that 1. We need to move towards Fair Open Access, building on existing models, and 2. we need funding for flipping subscription journals to Fair Open Access on the model proposed by LingOA.

    Johan Rooryck, EiC Glossa & President, LingOA

    1. Anonymous

      While I support to reconsider the order of the 12 action items and to group some of them in a better way, I do not share the idea to promote the (widely unknown) concept "fair open access" instead of "open access" as part of this document. I simply can't see a broad consensus. (Cf. my comments on action items 4 and 8; this does not apply to the FAIR principles regarding research data.)

      And I'm not in favour of highlighting and thereby promoting special models (neither the one proposed by LingOA nor OLH) while others are not mentioned at all.

       

      Georg Botz

      1. Dear Georg

        Please consult www.lingoa.eu for a definition of 'Fair' Open access. We do not mean FAIR here in the sense of open data, but as in "just".

        If you know of other proven, working models for flipping subscription journals to Open Access, please mention them.

        I proved that subscription journals can be flipped to Open Access, if other models did as well, please mention them.

        All best,

        Johan

  25. Anonymous

    Comments from Spanish Foundation for Science and Technology FECYT

    We welcome the initiative of the Dutch Presidency to promote OA/OS and to speed up the process. There are significant actions to be implemented rightly on time, however, any advance towards the right direction requires a lot of coordinated actions concerning the adoption of interoperable standards as well as the adoption of common methodologies, monitoring system and metrics. FECYT has been working on those aspects since 2010, and in the last year great efforts have been devoted to an adequate standardized measurement and monitoring system of OA publications. We have found main methodological barriers:

    • There is a need for all repositories to include the metadata required to measure access rights / reuse of items, the project code or identification (DOI).
    • It is important to automate the matching data to avoid manual processing, which requires solving the problem of disambiguation of names, institutions, and other references.
    • Funding agencies have to collect in the proper format and exportable data on publications included in the final report of the projects they have financed.
    • Funding agencies should adopt the same mandates in relation to the obligation to publish in OA with similar requirements and deadlines; and should adopt similar standards and common fields to facilitate the development of converging metrics
  26. Anonymous

    Michael Matlosz, President Science Europe

    Many of the ideas and certainly the ideals captured in this document are commendable, absolutely worth supporting and timely to move on towards developing tangible actions.  In doing so, it’s important that policies and actions are not imposed in a rigid ‘top-down’ manner. Policy makers and institutional decision makers can best support Open Science by taking concerted action to provide framework conditions (funding, infrastructure, training, management, etc. ), support cultural change (create incentives)   and remove obstacles to the creativity of researchers.  

    I strongly support a researcher-driven development of Open Science practices that is mostly developed bottom-up within research communities in answer to their scientific needs.

    Open Science practices are not generic, but take place in the context of diverse research communities rooted in different disciplines and with diverse practices and cultures. New and innovative solutions need the time and freedom to develop within communities.

    Once practices reach a certain level of maturity or complexity or when they can no longer develop further because they encounter obstacles (for example legal barriers, dysfunctional markets, or perverse incentive structures) action by policy and decision makers becomes necessary.

    Very clearly at least two broad areas of Open Science practices have reached a level of maturity where they must be supported through policies and financially: Open access to publications and Open data (where data is openly accessible by default and closed only in justified exceptions).

  27. Anonymous

    I fully agree that it is time for action and if we want to make open science a reality we need to make steps towards implementation.  However, specific actions and continuous references to the need of a new publishing model convey implicit support to the gold vs. the green route to open access as a pre-requisite to foster open science. This is a very restrictive view that should be carefully analyzed. Moreover, in order to recommend new publishing models the document has to be more clear and open on what those publishing models have to offer to researchers, research funders and research performing organizations as well as for knowledge and society rather than to publishers. 

     

    The structure of the document should be improved to increase focus of impact; some actions are contradictory, conflictive, and overlapping. There is a substantial unbalance between specific actions covered in some sections and the vague definition of actions in other sections. There are also substantial asymmetries in the distribution of actions among different stakeholders. 

     

    The number of actions proposed (59) is excessive, and (1) all of them cannot be taken simultaneously and many of the actions require sequential planning; (2) there are not costs estimates associated with each of the measures and they should be part of the action plan in particular to assess feasibility; (3) distribution of actions among stakeholders will not ensure coordinated action; (4) coordination between the European OS Platform and other international initiatives (i.e. OECD) is also required.

     

    Concerning the objectives they are too ambitious and thus unachievable or unrealistic. In particular within the timeframe of reference (2020) if we take seriously the reforms required, in particular those concerning the adoption of new assessment and evaluation practices and metrics at a global scale, and that are at the core of the new model that will change the practice of research.

     

    Clara E. Garcia, Deputy Director General of Planning and Monitoring and NPR on Open Access

  28. Anonymous

    Comments from Universities UK.

    Universities UK welcomes the spirit of this initiative, and welcomes the opportunity to provide comment. In the UK, we have come to a policy position that embraces a mixed economy of both Gold and Green routes to open access. We believe that it is important to ensure that both routes are valid options for researchers, and therefore would suggest changes to the text to provide for this option.

     

    For example, at 10:

    “Formulate a clear pan-European target: from 2020 all new publications are available through open access from the date of publication.” – the requirement that publications are available from the date of publication leaves no scope for Green routes to OA, and therefore is not in harmony with the UK policy position. We suggest the requirement of ‘from the date of publication’ is removed.

    In addition, the degree of autonomy enjoyed by UK research performing organisations, research funding organisations and by individual researchers means that some elements of the call to action – particularly those that involve centrally mandated and coordinated activities – do not reflect the policy making process in the UK, and could potentially act to hinder innovation in this area.

    We would also suggest that any recommendations around the use of metrics – such as in 1. Change assessment, evaluation and reward systems in science – are clear that the metrics are open to all and non-proprietary.

     

    For further information, the UK has recently revised its Open Access policy position, following the completion of a comprehensive monitoring exercise.

     

    We look forward to further opportunities to engage in this important agenda, and extend our thanks to Dutch colleagues for providing a welcome and effective impetus for change.

    I can be contacted at max.hastings@universitiesuk.ac.uk and would be happy to provide further information.

     

    Max Hastings

    Policy Researcher

    Universities UK

  29. Anonymous

    Wikimedia Germany welcomes the call for action, its purpose and its approach of evolution over revolution. However, we have several suggestions and objections, the most important of which are these:

    2. Facilitate text and data mining of content

    A detailed TDM exception/limitation that sets up conditions for research is not the way forward. Not only would that produce legal complexities that drive the transaction costs up for every kind of player in the research field, private and public. To establish such a rule would furthermore in turn mean that TDM is seen as a copyright-relevant use in the first place, while it actually is by its very nature an act of reading – and thus does not require any permission. The legislative act needed in this context would rather be to clarify that TDM is not a use in the copyright sense.

    3. Improve insight into IPR and issues such as privacy 
    The statement that "private parties will still be able to profit from their investments” lacks sufficient context. It must not mean that research that was to a considerable amount funded with taxpayer money ends up as all-rights-reserved assets of any private parties. If this statement is meant to express that other value-added undertakings around such research results except monetisation of rights are still going to be possible and wanted, it should be clarified to that end. The rights themselves, that relate to publicly funded research results, should only in exceptional circumstances be a matter of private property. Moreover, it goes without saying that "researchers will still be able to use their own results". If these results are open content, everybody is able to use them, including the researchers involved in their production.
    7. Adopt open access principles
    To say that "the existing types of publishing are not necessarily conducive to open access nor to the desired degree of transparency in the science system" is nothing short of a euphemism. These types and the businesses that represent them are by far the biggest obstacle for open science, not only in Europe. The constituents of the classic business models in scientific publishing have used the more than 10 years since the Budapest and Berlin declarations to do everything in their power to stall any move towards open science they could reach. Today, entire editorial boards of well-established journals see no other way out than to quit and together move to alternative publishing houses, leaving all assets (copyright in back issues, journal titles and impact factor) behind and start new. Also, the list of stakeholders mentioned is quite incomplete. It lacks those who pay for publicly funded research, i.e. the public, and those who actually put their brains to it, i.e. the researchers.
    11. Involve researchers and new users in open science and
    12. Encourage stakeholders to share expertise and information on open science
    These are eminently important items and should not be put at the bottom of the list. While the public and other researchers are the main users of research results, their role seems to barely be on the radar of many debates around the future of science. There's already a lot of knowledge about the positive effects a user base broadened by openness of contrent can bring to the table. In projects like Wikidata it can be seen every day that more engagement leads to better data. Regarding young researchers, Wikimedia Germany is piloting a fellowship program for open science, but much more initiatives of this nature are needed.
    John H. Weitzmann
    Legal and Policy Advisor
    Wikimedia Deutschland e. V. | Tempelhofer Ufer 23-24 | 10963 Berlin
  30. Anonymous

    Natalia Manola on behalf of OpenAIRE:

    The objectives and actions are very much welcome for their focus on openness and should be decisively followed up and elaborated in the following respects:

    a)    A clear distinction between articles, books, data, software and other types of research outcomes is paramount. Each class of research outcome requires a different treatment. For example, publications (articles and books) are deeply embedded in a long established socio-economical system, while data is embedded in diverse research contexts, predominantly publicly funded by national and European investments.

    b)    We note that most proposed actions are related to publishing policies and that further refinements shall specify how data management and accompanying technologies as well as e-Infrastructures and other parts of the value chain are affected.

    c)     There is no one-solution-fits-all, especially in a diverse Europe with various research cultures and their individual needs. What is therefore important is to built in a truly participatory and democratic governance that is able to accommodate different models and readjust to emerging trends.

    d)    The proposed solutions and actions must be adaptable in the emerging vision of European Open Science Cloud as well as national and local e-Infrastructures. Repository e-Infrastructure is crucial in this respect, as this is a distributed, institutionally-rooted and community-governed backbone of e-infrastructure.

    e) Key to the success of Open Science is to utilize existing investments, institutional, national and European, which will provide robust and trusted solutions and prevent double investments. Ensuring that e-Infrastructures are driven by research interests will prevent disruption and diversion by disingenuous or inadequate financial interests.

    f)     Many of the proposed actions rely on the existing publishing model and the exploitation of IPR. Today’s networked and multidisciplinary science, i.e. Open Science, requires a free flow of knowledge and should focus on business models based on services (and not content) that process, refine and enrich knowledge and fuel new inventions as well as technological breakthroughs, thus, fostering an open innovation market.

    In the following, we comment on specific areas of interest, propose solutions and concrete actions.

  31. Anonymous

    As prominent and long-standing advocates of open access and open research data, we strongly endorse the 2020 pan-european goals and the flanking policies. We note particularly that James Wilsdon’s “Metric Tide” report for HEFCE (http://www.hefce.ac.uk/pubs/rereports/Year/2015/metrictide/Title,104463,en.html) is a powerful recent (UK) examination of the role of current and potential future indicators in research assessment and evaluation, and that there is substantial ongoing UK and Jisc work examining and aligning polices around open research and open data at institutional, funder, journal/publisher and national levels (for example work of the UK Open Research Data Forum and initiatives at the UK research councils).

    Working collaboratively to achieve open science and to prioritise it as an action area in Europe is very welcome. For some of the related latest positions in the UK see the recent support advice to DBIS, and the ministers response which advises that there should be a roadmap for more effective research data infrastructure for the UK. https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/499455/ind-16-3-open-access-report.pdf and https://www.gov.uk/government/publications/open-access-to-research-independent-advice-response

    Jisc sees “open science” as encompassing the full range of academic practice and record of research. In terms of research, we are clear that open practices can benefit research and knowledge creation outside of the sciences and technology. And we see open practices in teaching, and in training for researchers, as an integral component of what is described here as “open science”.

    David Kernohan, Jisc, UK

  32. Anonymous

    The Confederation of Open Access Repositories (COAR) welcomes the vision and objectives of the European Union to speed up the transition to Open Science, by securing full Open Access to all scientific publications and optimizing data curation and reuse in the coming years. These goals are compatible and aligned with COARs vision of a sustainable, global knowledge commons supported by a network of open access digital repositories. COAR is an international association with over 100 members and partners from around the world representing libraries, universities, research institutions, government funders and others.

    Welcoming the inititiative, we want also to provide some general critical comments and suggestions.

    Firstly, as noted in many other comments, the document seems to reflect a clear focus on promoting APC based Gold Open Access (transforming current subscriptions into publishing costs) as the preferred option to reach Open Access to publications. Being a global organization, with members around the world, COAR is aware that there is no “one sizes fits all solution”, and that stimulating an European push for APC based global OA would not correspond to other regional needs and options and would encounter skepticism, if not opposition, in other parts (North and South) of the globe.

    There is a wide variety of strategies, initiatives and infrastructures currently being used to promote Open Access. Even just considering Gold OA, not all models rely on APCs, and new models and forms of publishing and disseminating research results have emerged. On the other hand in many regions of the world, including in Europe, considerable efforts have been made, both at policy and infrastructure levels, in repositories and “Green OA”.

    Any sustainable and effective strategy for the transition to Open Access must recognize and incorporate this diversity, cannot ignore the existing policies and infrastructures (in Europe and around the world) and, on the contrary, should leverage on them.

    This is precisely the second point: while there are several references on the document to the need to “set up” or “put in place” e-infrastructures, and to the European Open Science Cloud, as “new” initiatives and infrastructures, there is little recognition of the existence of widespread (and in many cases mature) infrastructures (like repositories) at local/institutional, national and international (like OpenAIRE in Europe) levels. More than just setting up “new” e-infrastructures, what is needed, in our opinion, is to continue to support, expand and enhance existing open infrastructures, for publications and data, securing proper governance, technical interoperability and financial sustainability. Common infrastructures, like the European Open Science Cloud, should build on, and federate, the already existing infrastructures.

    The current repository infrastructure (as well as other institutional or community driven e-infrastructures) by its distributed nature, its institutional/community anchoring, funding and governance, is a fundamental component for the transition towards Open Access and Open Science. It can and it should be used to support and stimulate innovative models for scholarly communication and knowledge transfer, and fundamental changes in the evaluation and reward systems in science and academy (like repository based publication and open peer-review).

    Eloy Rodrigues

    Chair of the Executive Board of the Confederation of Open Access Repositories

  33. Comments on the Amsterdam Call for Action by Celina Ramjoué, DG CONNECT

    Structure:

    The Call for Action tries to fit 12 actions into the 5 subheadings of the so-called "European Open Science Agenda". This is a bit artificial and leads to a somewhat unclear structure with scattered information/sub-actions. It might therefore be better to group and streamline the actions thematically, for example as follows:
     

    - incentives/reward system:                                      action 1, some of 11

    - optimal re-use of research data:                             actions 2, 3, 5

    - open access to publications:                                   4, 7, some of 8, 10

    - open science infrastructure:                                    action 6

    - citizen science / citizen engagement:                      some of actions 8 and 9

    - evidence-based policymaking/monitoring:              some of actions 9 and 11, action 12

    - awareness raising/education:                                  some of actions 9 and 11, action 12
     

    Missing theme:

    Despite the very complete coverage of OS themes in the Call for Action, one very important area that is not sufficiently covered is "open science and innovation". An action should be introduced specifically emphasizing the importance of the take-use of research result for growth and innovation.

     

    Specific comments on individual actions: 

    Action 1 (assessment /rewards):

    The concrete action for national authorities and the EC seems too top-down.
     

    Action 2 (TDM):

    - First sentence: it is not the growing amount of data that enables TDM, but rather TDM that enables exploring growing amounts of data. Rephrase.

    - Solution 1: this precludes and possible goes beyond the proposal of the Commission, due in 2016

    - Solution 2: encouraging authors to keep their copyright should go into copyright/IPR in general. It is linked to TDM, but is not TDM-specific.

    - Choice of term 'machine actionable': what about 'machine readable'?

    Action 3 (IPR/Privacy):

    - IPR and Privacy/data protection (including privacy by design as a proposed solution) should be split into two actions (related to optimal re-use of research data) as these are two broad, important and different fields.

    - The description of IPR issues is limited to the issue of IPR regimes for PPPs. Commercialisation / patenting in general should at least be mentioned. And what about other form of IP protection. For example, copyright is one of them (see action 2).

    Action 4 (transparency for OA to publications agreements):

    - Transparency is one of the principles mentioned under action 7. There is no need for a separate action only on transparency as a principle.

    - If this is a proposed action on big deals, this should be made explicit.

    Action 5 (FAIR and secure data)

    - What is meant by 'secure' data?

    Action 6 ( e-Infrastructure)

    - Choice of term 'machine actionable': what about 'machine readable'?

    Action 7 (OA to publications principles)

    - One important principle is missing: the idea that ownership must remain with the researcher. This mean, for example, that authors should retain their copyright (cf. solution mentioned under action 2 on TDM) and that journal titles should remain with the scientific community / editorial board of a journal (cf. the case of LingOA à Glossa)

    - One principle does not seem to fit in: "pluralism", which is relevant for the press (media pluralism), not for science.

    - The concept of 'fair' open access publishing could be introduced here, pointing towards the idea that the scientific community should decide how it wants scientific publishing to be in the future (the market can then delivers solutions / models that comply with these  principles).

    Action 8 (publishing models/knowledge transfer)

    - This action seems to combine elements of several themes:

    * OA publishing and publishing models

    * Incentives

    * citizen science

    * innovative academic communication models

    These themes should be pulled apart and not mixed up.

    Action 9 (Monitoring/evidence base)

    - This action seems to combine elements of several themes:

    * Monitoring

    * Evidence base

    * public participation/citizen science

    * Funding

    These themes should be pulled apart and not mixed up.

    Action 10 (Open access to publications targets and monitoring, Member States)

    This action should be merged into one central OA to publications action. It could also be mentioned/cross-referenced under a monitoring / evidence-base action.

    Action 11 (Involve researches and new users)

    - This action seems to combine elements of several themes:

    * Incentives / career structure

    * Awareness raising

    * Education

    These themes should be pulled apart and not mixed up.

    - Education should receive more emphasis (training of data scientists, data skills, MOOCs, etc.).

    Action 12 (Encourage stakeholders/sharing expertise/OS roadmaps)

    This action seems to combine elements of several themes:

    - Establishing national roadmaps

    - Sharing of information on best practice

    These themes could go under 'awareness raising'.