CURE-FAIR WG Case Statement

15 May 2020

CURE-FAIR WG Case Statement

The goal of the working group is to establish guidelines and standards for curating for reproducible and FAIR data and code (Wilkinson et al., 2016). The ultimate objective is to improve FAIR-ness and long-term usability of “reproducible file bundles” across domains.

When we think of specific research outputs, we might think of data, software, codebooks, etc. These individual outputs may have inherent value. For example, a set of observations that is very costly to produce, or that cannot be repeated, or a script that can be used by others for computation. Traditional curation has considered these outputs as its core objects. But in the context of empirical research, these outputs interact with each other, often to produce specific findings or results. Nowadays, the process by which results are generated is captured in computation. Our approach to curation takes into account this process and focuses on computational reproducibility.

 

Computational reproducibility is the ability to repeat the analysis and arrive at the same results (National Academies of Sciences, Engineering, and Medicine, 2019; Stodden, 2015). It requires using the data and code used in the original analysis, and additional information about study methods and computational environment. The reason to pursue computational reproducibility is to preserve a complete scientific record , to verify scientific claims, to do science and build upon the findings, and to teach (Elman, Kapieszewski, & Lupia, 2018; Resnik & Shamoo, 2017; Stodden, Bailey, & Borwein, 2013).

 

In this framework, the object of the curation is a “reproducible file bundle” and its component parts, including the files and their elements (e.g., variables), with the goal of enabling continued access and independent reuse of the bundle for the long term.

The CURE-FAIR WG is focused on the curation practices that support computational reproducibility and FAIR principles.

By curation we refer to the activities designed for “maintaining, preserving and adding value to digital research data throughout its lifecycle” (Digital Curation Center, n.d.).

The WG will deliver,

  1. A snapshot of the current state of CURE-FAIR practices drawing upon community surveys and reviews of practice.
  2. A synthesis of practices relating to curating for computational reproducibility and FAIR principles.
  3. A final document outlining standards and guidelines for CURE-FAIR best practices in publishing and archiving computationally reproducible studies, including the associated computational methods and materials.
Review period start: 
Friday, 15 May, 2020 to Monday, 15 June, 2020
Documents : 
AttachmentSize
PDF icon CURE-FAIR WG Case Statement_May2020.pdf152.05 KB
  • Paula Andrea Martinez's picture

    Author: Paula Andrea Ma...

    Date: 15 Jun, 2020

    This comment is on behalf of the FAIR4RS steering committee. 

    We aim 

    • to clarify and differentiate the scope of the groups CURE-FAIR and FAIR4RS to avoid duplication of work.

    • address that these two groups are complementary.

    While FAIR4RS aims at adapting and extending the FAIR data principles to better cover the research software case, the CURE-FAIR aims at providing guidelines to digital curation of “reproducible file bundles” focused on computational reproducibility from a FAIR perspective. The two groups are however related as they both use the FAIR principles as a basis for their own work.

     

    We would like to invite the members of the two groups to collaborate with each other and allow engagement and milestones’ work to be complementary.

     

    • We would suggest that the CURE-FAIR includes the FAIR4RS WG to the list of RDA groups with goals that complement those of the proposed CURE-FAIR WG (Point 3.1). This to clearly express the collaboration between both group’s members.

    • Point 4.3 Milestones and Deliverables. For instance, the first deliverable of FAIR4RS (A review document identifying challenges in defining FAIR for research software) would be useful to the CURE-FAIR WG, particularly towards their second deliverable (synthesize input from their representative communities about current practices in curating for computational reproducibility). While the third deliverable of CURE-FAIR (document outlining standards and guidelines for CURE-FAIR best practices in publishing and archiving computationally reproducible studies) would be useful to FAIR4RS (but the times would make it challenging for the FAIR4RS WG to fully take it into account). 

    • Point 4.5 Community Engagement and Participation, we would like to encourage CURE-FAIR to seek input from participants and institutions currently developing computational reproducible workflows, such as those listed in the first part of Point 3. Engagement with existing work in the area (of the FAIR4RS).

     

    An additional last comment, is to bring to the attention to the community that all of the Chairs are from one region in the world and also from one science domain, we would encourage the group to think of ways of diversifying and allowing for diverse input during the course of the working group activities. 

     

    To close comments, the Chairs of the FAIR4RS would be happy to schedule an online meeting to further detail the provided feedback for incorporation. 

  • Limor Peer's picture

    Author: Limor Peer

    Date: 17 Jun, 2020

    Dear FAIR4RS steering committee,

    Thank you very much for your excellent comments. We very much share your goals of differentiating the scope of CURE-FAIR and FAIR4RS to avoid duplication of work and clarifying the complementary nature of the groups’ work. We’re looking forward to learning from the FAIR4RS WG and expect it will inform our work on curating for computational reproducibility from a FAIR perspective.

    Regarding the specific suggestions to revise the CURE-FAIR WG case statement, we agree that,

    • Point 3.1 should include FAIR4RS WG on the list of RDA groups with goals that complement those of the proposed CURE-FAIR WG.
    • (Point 4.3) A fruitful exchange of deliverables, and possible cross-WG collaboration, would be beneficial to the WGs and to RDA. We are happy to discuss at an online meeting.
    • (Point 4.5) It is advisable for CURE-FAIR to seek input from participants and institutions currently developing computational reproducible workflows, as listed in the FAIR4RS case statement Point 3.

    With regard to allowing for diverse input during the course of the working group activities, we agree that regional, domain, and other diversity are of utmost importance. We are pleased that the CURE-FAIR initial membership list includes regional and disciplinary diversity. This initial membership list represents a diverse group of researchers, information professionals, and other stakeholders from various geographic regions and disciplinary communities. In addition, we can include more details on how the CURE-FAIR WG plans to engage a diverse community (e.g., encouraging global participation by accommodating various time-zones). Finally, while the current CURE-FAIR co-chairs come from one region, we represent diverse backgrounds: social science, information science, and epidemiology. We are open to expanding regional representation among co-chairs and encourage WG members from all regions and backgrounds to take on leadership roles.

    CURE-FAIR WG co-chairs

submit a comment