RDA Outputs Policy

Approved by Council 28 April 2014. Formal published version at https://www.rd-alliance.org/filedepot/folder/132?fid=502

This Policy is under review as of October 2020. Licensing information changed 5 Oct 2020.


The purpose of this policy is to advance the goals of Research Data Alliance (RDA) and to encourage adoption of RDA principles and outputs. The goal of the RDA is to accelerate international data-driven innovation and discovery by facilitating research data sharing and exchange. RDA Working Groups (WG) develop, adopt, and deploy infrastructure, policy, practice, standards, and other deliverables to advance this goal. They operate under RDA principles of openness, consensus-based decision making, balanced representation, technical neutrality, harmonization across communities and technologies, and a non-profit approach.

This policy seeks to:

  1. clarify what outputs are produced by and associated with RDA,
  2. clarify the rights of RDA members and the overall organization around these outputs, and
  3. create an environment that facilitates the adoption and sustainability of the outputs.

In general, RDA encourages that any outputs associated with RDA be fully open and broadly shared in accordance with the Norms for contributing to and using RDA products. This creates a more adaptable and robust infrastructure.

This policy is approved and maintained by the RDA Council. A subsidiary document, the Process and Criteria for RDA Recommendations, is maintained by the Secretariat. Both documents are supported by the Norms for contributing to and using RDA products which, as norms not policy, are intended to be maintained as a community document.

Types of outputs associated with RDA

RDA produces diverse outputs through its WGs, Interest Groups (IGs), and other groups including the TAB, OAB, Council, Secretariat, and ad hoc Task Forces (collectively, "Groups"). WGs, in particular, produce a wide variety of documents, models, tools, prototypes, etc. We consider four broad classifications of outputs.

Discussion Documents (DD) are documents and related materials produced during the conduct of RDA activity--they might include background documents, meeting notes, initial results and prototypes, slides, and graphics, etc.. It is up to the Groups whether these documents will be made available to everyone via the RDA Web site, but they are not formal or official RDA statements. WGs and IGs must keep the records of their work open to RDA membership.
RDA Policies, Working Group Case Statements, and Interest Group Charters are official, operational RDA documents maintained by the RDA Secretariat and approved by Council (e.g., this document). These specific documents are addressed by another RDA Document Publication Policy and are kept in a special repository.
RDA Recommendations are the official endorsed results of RDA. Recommendations undergo formal phases of discussion, comment, and decision taking. Recommendations are documents in a very broad sense and may include specifications, taxonomies or ontologies, workflows, schemas, data models, etc. They must, however, meet certain criteria to be adopted and endorsed as RDA outputs including being made publicly available and usable. RDA Recommendations are distinctly labeled as such and are stored in a special place managed by the Secretariat.
Implementations are services, tools, code, registries, APIs, etc. that demonstrate and make use of RDA DDs and Recommendations.

Output status, review, and approval.

Discussion Documents need no review. Anything an RDA member has rights to and feels advances the work of RDA may be posted on the open RDA web site (e.g. within an WG or IG space). Discussion Documents are available to the entire membership under a general Creative Commons Attribution Only 4.0 license (CC-BY). Discussion Documents published on the RDA Web site need to have a value for and be maintained by a WG, IG, or other RDA group. Groups may define their own publishing criteria.
RDA Policies, Case Statements, and Charters are available under the default CC-BY 4.0 license. Formal processes are addressed in the Document Publication Policy.
Implementations are the responsibility of the person(s) or organization(s) hosting them. RDA does not maintain or formally endorse these products. They are, however, reviewed in the sense that they may have been proposed as part of a WG Case Statement, and TAB, Council, or the membership may have made suggestions to improve how the Implementations advance the work of RDA. Further, to foster open interoperability and wide-spread adoption, RDA strongly encourages that these Implementations be made available as openly as possible. RDA promotes the use and broader adoption of products that adhere to RDA principles. RDA generally recommends open source licenses from the BSD family or similarly permissive licenses where possible.
RDA Recommendations undergo formal review as endorsed Recommendations of RDA as defined in the Process and Criteria for RDA Recommendations . They are produced by WGs and IGs where robust consensus and transparency are basic principles. The Technical Advisory Board (TAB) shall have a chance to comment on draft versions and give advice on whether the content fits with the technical intentions of RDA. Finally, the Council needs to assert whether the document was developed and shared in accordance with the vision, principles, and processes of RDA. Recommendations must be open for public use and adoption. The default license should be the Creative Commons Attribution Only 4.0 license (CC-BY). RDA strongly encourages users and contributors of RDA Recommendations to adhere to Norms for contributing to and using RDA products.

RDA does not hold patents. Anyone contributing to an RDA Recommendation must disclose any known patent or any known pending patent application they hold that may restrict the open use of the RDA Recommendation. If the patent holder does not allow unrestricted use of the patented material, the material may not be part of an RDA Recommendation.



  • Mark Parsons's picture

    Author: Mark Parsons

    Date: 01 Dec, 2013

    o I assume that the intent of the commercial use restriction in the Terms of Use is to prevent someone from packaging up the material and selling it in some way. If that is the case I guess I agree with you that the language may need some work.

    o I don't think we need to make specific licensing recommendations in this document. That should be a separate document which can change while this one remains unchanged. General guidelines are a different matter.

    o I agree with Peter and Herman that the WG outputs and IG outputs are or should be quite different. If an IG needs to get specific on some topic it should form a short fused WG to do so. Otherwise I think they are approved to discuss a general area, not to take action that would result in a concrete recommendation.

    o The entire membership should be given a comment period but perhaps it goes in stages starting with TAB and WG/IG chairs, specifically to look at conflicts and overlaps.

    o Personal authorship - raises some interesting questions. If all RDA WG formal outputs (I seem to be resisting RoD - but I don't have anything better) are the product of a WG then isn't the WG the author? On the other hand some WG members may disagree with the results and object to having their name on the document. On balance I would vote for personal authorship and allow any WG member that agreed to add their name to the document. Perhaps Primary Authors and WG Members in Agreement or something like that.

    o I want to agree with your transparency recommendation -- but I doubt that "we told them the rules had changed and gave them the option to drop out if they didn't agree and they didn't respond so we took that as agreement" would pass muster. I think the whole transparency issue is one for the next phase.

  • Mark Parsons's picture

    Author: Mark Parsons

    Date: 03 Dec, 2013

    My comments center on the transparency, avaialability and certainty.  Perhaps it is too early to address them, but I thought I'd capture them here:

    1) On the WG and IG side, we need to ensure any participant has the use or ownership rights needed to make a contribution. The assurance that it is either original thought or the owning party has agreed to permit contribution to RDA's  program of work .  I suspect this can be established in a revsion of the Participation Agreement.

    2) On the output side, I suggest we use a different term for document availibility. "Public domain" has specific meanings that I think are contrary to RDA providing stable, reliable and durable documents for long-term use by the research and academic communities.  I don't think we want the selected RDA documents to be changed outside of the described process.

  • Tobias Weigel's picture

    Author: Tobias Weigel

    Date: 04 Dec, 2013

    1) What happens if after the 3+3 month process an RoD is sent back with serious comments? It is very likely the WG will already be terminated (or close to termination) at that point and thus most likely unable to redesign and re-submit it. Would there be a recommendation to form a follow-up WG specifically tasked towards improving/finalizing the RoD? I agree with the overall goal to establish some QC measures, but am wondering how this will fit within the overall WG/IG formation process. Including some clear guidelines would help to define the scope for such follow-up WGs.

    2) I like Larry's proposal regarding authorship - explicitly stating authors but also having a recognized option for people who just want to support a RoD by signing it. At least currently, there's a high number of people watching WG activities and agreement or disagreement with WG efforts is somewhat silent. We have to respect that not everyone has the time to spend on writing docs, but there should also be an option for those in favor of the general outcomes. This most importantly includes potential adopters who really need a solution but don't have resources to spend on development.

  • Walter Stewart's picture

    Author: Walter Stewart

    Date: 04 Dec, 2013

    While I understand that RDA might want to simplify things by owning little, who will own these things that RDA doesn't want to own?  I thought the purpose of the IP policy would be primarily to ensure that work of RDA remains open and freely accessible and not susceptible to being alienated by some organiziation claiming it and then making it only available commercially.  I don't see this protection in the draft policy.



  • Reagan Moore's picture

    Author: Reagan Moore

    Date: 06 Dec, 2013

    The proposed policies need to address the criteria for continued maintenance of RoDs versus the finite lifetime of a working group.

     Is there a supposition that a new working group will appear that maintains the results from the original working group?  Will there be a third category of groups (Working, Interest, Maintenance)?


    Reagan Moore

  • Inna Kouper's picture

    Author: Inna Kouper

    Date: 18 Dec, 2013

    - we might want to think about how RDA can endorse as RoDs documents created by and within other relevant organizations. Otherwise, we risk to duplicate a lot of what has already been done with regard to data. Also, will checking for duplicate efforts and overlaps among the groups and beyond RDA be part of the review process?

    - why do RoDs typically come from WGs? it seems that both WGs and IGs could generate RoDs, but WGs could also produce implementations.

    - WGs were supposed to be short-term. Once they deliver an RoD, should they disband or continue producing new deliveries? 

    - personal authorship might be challenging for RDA outputs. Some groups have 30-40 members, it'd be hard to establish author order, etc. Perhaps, a collective authorship (RDA WG) with a reference to all group members?

    - RoDs are a vague genre and RDA is too new to be an authority. What would be the criteria for RoD quality, value, authoritativeness and validity (e.g., recognized names and organizations, evidence of efficiency and good results, references to published articles, etc.?) This could also help to establish a more transparent review process (focusing on quality rather than just form).

    - There seems to be a lot of focus on documenting outputs (e.g, making them citable, etc.) rather than on disseminating / helping with adoption of their contents. Keeping an RoD current (maintenance) is not the same as making sure it's used by someone. Perhaps, a stronger connection to outreach and adoption efforts in the language of the policy?


  • Mark Parsons's picture

    Author: Mark Parsons

    Date: 19 Dec, 2013

    Hi all,
    Thanks for the feedback.  I have made various revisions to the draft in response to these comments, further task force discussion, and recommendations from Council.
    In response to a request from Council, we have made the initial default license for all RDA content to be CC-BY (probably the new ver. 4.0). Council pushed this   rather forcefully so it is incumbent upon us to argue for any alternate on grounds that the alternate is better than CC-BY.  This will involve further discussion. See the comments embedded in the policy.
    Also in response to Council, I separated the process components from the core policy an dgenerally tried to simpolify the policy.
    Regarding Walter's comments. We do promote openness and accessibility. This will be true for all endorsed content or content on the RDA web site. Council very much liked the articulation of 4 types of outputs, and they saw the "RDA Recommendations on Data" as a significant category.  RDA as a virtual organization, however, cannot truly own or control "implementations", especially if our specifications are open. We will develop guidelines to encourage these implementations to be open and will only promote those that are. I hope the two paragraphs on implementation are clear.
    Re: authorship. I clarified that Groups can be authors but that we prefer individuals. All members of a group can sign on as supporting or endorsing authors.
    Re: Maintenance and review. I added some more details as to what should be included in a maintenance plan. Also more about how WG may continue to respond to comments. We'll need to be adaptive. 
    Re: Product promotion. This was not part of our charge and given Council's plea for simplicity, I don't think it is a core part of the policy. That said, we will need a strong product promotion strategy--something for the Secretariat, Council, and chairs to all consider soon.
    Re: RoDs. I add some more "quality" criteria for the RoD. I welcome more specific criteria. RoD can be produced by both IGs and WGs. They are expected of WGs.
    Next steps are to get the legal interoperability IG to weigh in on licensing issues and then send a new revision out for comment. Keep the feedback coming.

  • Puneet Kishor's picture

    Author: Puneet Kishor

    Date: 02 Jan, 2014

    From RDA's guiding mission, "The Research Data Alliance implements the technology, practice, and connections that make Data Work across barriers." (https://rd-alliance.org/about.html)


    One easy way to make data and information work across various barriers is to eliminate the barriers in the first place. I propose RDA chooses to mark all its output with the CC0 Public Domain Dedication thereby removing the legal barriers to sharing, and maximizing legal interoperability on a global scale (http://creativecommons.org/about/cc0)


    Mechanically there is not a wide gulf between CC0 and CC BY, the major difference being that the latter imposes a legal requirement to give attribution. In a philosophical sense, however, the two are very different. CC0 indicates a philosophical position that the conversation about copyright does not belong in academia and scholarship, that attribution can and has always been given as a scholarly citation norm without imposing any legal burden or uncertainty arising therein.


    RDA wants to protect and nurture its reputation and credibility, and its community is already well accustomed to doing that. As an international, cross-jurisdictional community, RDA is influential. Using CC0 would send a signal to the world that RDA considers the Public Domain Dedication to be a worthy mechanism, and would encourage others to emulate that behavior.


    Simplicity is always good. The default position should always be toward the simplest, toward the public domain, with variance from it occurring only where absolutely necessary.

    Puneet Kishor
    Manager, Science and Data Policy
    Creative Commons

  • Jason Baird Jackson's picture

    Author: Jason Baird Jackson

    Date: 04 Jan, 2014

    This quick note simply records that I have reviewed the draft documents and associated comments. I am a IG co-chair still new to, and getting oriented within, the RDA. Thanks and appreciation to all who have worked on these important organizational and policy frameworks. Jason Baird Jackson, Indiana University, USA.

  • Mark Parsons's picture

    Author: Mark Parsons

    Date: 15 Jan, 2014

    The Task Force has completed a draft and has requested comments from the general membership: https://rd-alliance.org/request-comment-draft-rda-output-and-ip-policy.html

  • Herman Stehouwer's picture

    Author: Herman Stehouwer

    Date: 23 Jan, 2014

    Dear all,

    I wonder if we should maybe make a distinction within RoD between new standards/technologies/output and recommendation on how to use pre-existing stuff (i.e. recommendation on best practises).

    To give a concrete example, I can imagine the provenance people making a recomendation of how to use and integrate the W3C provenance standard for day-to-day automatic use. That would not be "new", but quite important.

    Anyway, just a thought. Besides that I quite like how the document is going.




  • Mark Parsons's picture

    Author: Mark Parsons

    Date: 11 Feb, 2014

    Thanks Herman. I think the definition of RDA Recommendations is sufficiently broad to cover recomendations on new and existing standards/practices/etc. I hope you agree.

  • Simon Cox's picture

    Author: Simon Cox

    Date: 21 Oct, 2014

    A very belated comment:

    Regarding RDA Recommendations, should there be any statement about whether a recommendation is expected to be unique within a particular scope? Is it allowed for RDA to make alternative recommendations to solve the same or very similar problem? For example, if a handle-based solution has been adopted at an RDA Recommendation, does that necessarily exclude URI-based solutions to the same problem being proposed? 



submit a comment