1 of 25

Calling all research data publication stakeholders:

researchers, repositories, publishers, and funders…

We need a workflow model!

​

Thursday, January 28

4:00pm - 5:30pm

Session Chairs:

Danie Kinkade (BCO-DMO); Shelley Stall (AGU); Lesley Wyborn (ANU)

2021 ESIP Winter Meeting

2 of 25

Why Do We Need a Workflow Diagram?

  • Increasing recognition of the importance of supporting data
  • Act of publishing data supporting scholarly publications grew organically but is becoming more formalized
  • Domain and institutional repositories are challenged to fit into publication workflows
  • Bottlenecks are arising in touchpoints between stakeholder workflows
  • Little awareness and communication across relevant stakeholders
  • Need consensus and common frame of reference to understand issues
  • Serve as framework to collaboratively address problems and improve efficiency of data flow from researcher to repository to publisher

3 of 25

History of the Data Publication Workflow Diagram

  • Sprang from conversations within Council of Data Facilities Webinars in Feb 2020
  • Concept socialized at ESIP Summer Meeting 2020 COPDESS session
  • Presented at RDA VP16 in the ESES IG Session Breakout on repositories
    • Recognition of RDA/WDS Publishing Data Workflows WG (Dallmeier-Tiessen, et al., 2016 http://doi.org/10.15497/RDA00004)
  • Refinement and continued sharing at AGU FM
  • Further refinement now - hope to share among all stakeholders

4 of 25

Considerations to keep in mind...

  • Focus on geoscience observational data (vs. samples or code)
  • Focus on new data submissions vs. reuse
  • Long tail vs operational/streaming
  • Probably incomplete!

http://bit.ly/dataPubWkflo

5 of 25

The Rest of This Session...

  • Speakers providing context on current challenges
    • repository perspectives
      • BCO-DMO (Domain, long-tail)
      • Deep Blue (Institutional)
    • Publisher perspective (AGU)
  • Breakouts to review and refine workflow
    • Gather comments/alternate flows
    • Identify barriers/challenges throughout
  • Help identify next steps

6 of 25

BCO-DMO Data Curation and Publication Workflow

submitsDataTo

providesDataTo

primarilyFundedBy

locatedAt

scientists

public

conductsResearch

ocean ecosystem dynamics

General: info@bco-dmo.org

contactUsAt

7 of 25

  • Engages researchers early in data management planning
  • Applies domain knowledge to increase interoperability: harmonizes formats, observational precision and metadata; performs gross quality control; creates domain-appropriate metadata
  • Applies licensing, DOIs, creates citations, and documents provenance and versioning
  • Employs emerging technologies, community best practices and standards to improve discoverability, access, and reuse
  • Educates our research community on domain data best practices

BCO-DMO Data Curation and Publication Workflow

8 of 25

Hi BCO-DMO

Please find attached the data for my next paper that has just been accepted at <JOURNAL X>. In the Zip file you will find:

​

• draft of the paper that has been accepted

• Excel file of data and the metadata

• bco-dmo form

​

I imagine we have quite a few weeks before anything will happen with this draft at <JOURNAL X>. Historically they have been slower than <JOURNAL Y>. If possible, it would be lovely to publish the paper with data DOIs. I can keep you posted as to deadlines.

​

7 January 2021

BCO-DMO Data Curation and Publication Workflow

Pain Point!

NOTE:

  • Only data relevant to the paper (not full project output) is getting submitted
  • Data are submitted *after* drafting and submitting the manuscript
  • Time requirements of the journal for data access and DOI need are not known

​

Hi BCO-DMO — ughh. The proofs have arrived this week, much to my surprise. This does not seem feasible to add the DOIs to the paper, unless you advise otherwise. I am sorry this is so hard to tie the threads together.

9 January 2021

​

… no worries. It is not a reasonable request.

​

What would happen if I requested data doi when I submit the paper and then deal with the risk that the paper might get rejected?

​

​

​

10 January 2021

9 of 25

BCO-DMO Data Curation and Publication Workflow

10 January 2021

How Do We...

  • Encourage (researchers, funders) to
    • Submit full project output
    • Submit data once ready for analysis (i.e., as early as possible!) not at time of scholarly publication
      • How to better align what’s needed for publication vs full project output?
  • Manage needs of publishers while providing robust curation of data?

10 of 25

The University of Michigan’s Data Institutional Repository

deepblue.lib.umich.edu

​

Jake Carlson

Director of Deep Blue Repository & Research Data Services

jakecar@umich.edu

​

11 of 25

Who we are

Deep Blue Repository and Research Data Services (DBRRDS) oversees and provides supporting services for the Library’s Institutional Repositories:

  • Deep Blue
  • Deep Blue Data

Jake Carlson

Director of DBBRDS

Susan Borda

Data Workflows Specialist

Jim Ottaviani

Deep Blue Coordinator

Rachel Woodbrook

Data Curation Librarian

12 of 25

What do we do?

Provide a means for a means for the U-M community to make their work openly accessible to anyone in the world with an internet connection.

13 of 25

What do we do?

Publish content of scholarly or educational value, research data in particular, that may not have a defined path for dissemination and might otherwise be unavailable or lost.

14 of 25

The Deep Blue Repositories

  • Every deposit is assigned a unique identifier
  • Indexed by Google and other search engines
  • Basic preservation for all format types; works in open formats remain readable by future applications

15 of 25

Why is sharing data hard to do?

  • It may not be easy for people outside of the research project to understand, trust and use the data
  • It may not be clear how to prepare the data for others to use
  • It requires an investment of time, attention and effort over the course of the data lifecycle

​

16 of 25

Data Lifecycles

17 of 25

Your Data Lifecycle

Every data set has its own path to follow which is influenced by:

  • Questions being researched
  • Community norms
  • Tools & Resources
  • Personal Practices & Preferences
  • Funder, Institution, Publisher Requirements

18 of 25

DBRRDS Services

19 of 25

Deposit

Our Docs and Data repositories have somewhat different deposit processes.

20 of 25

Data Curation Review

https://datacurationnetwork.org/

21 of 25

Negotiation and Publication

22 of 25

Services & Deposit Workflow (Full)

23 of 25

Journals need to better connect and communicate with repositories

Shelley Stall, AGU

27 January 2021

24 of 25

Publishers Connecting Authors to Repositories

  • Helping author select the appropriate domain/institutional/general repository for the data (if they have not done so yet).
  • Connecting the authors to data managers at the repository for support.
  • Giving the author a sense of timing that the repository needs to curate and preserve the data to better plan their publication time frame.
  • Information on the repositories use of share links for use during paper peer review prior to paper publication. Note: links that resolve prior to the DOI being fully registered to the dataset.
  • Details about when the repository makes the data public, the paper being accepted, the DOI for the data, and timing in general specific to the repository process.

25 of 25

Publishers Needs from Repositories �(not covered in CoreTrustSeal)

  • The ability for the paper peer reviewer to confidentially access the data specific to the paper. This can be done by a share link, or other means.
    • Some journals have double-blind peer review (T&F) and can only use limited number of repositories that support that.

​

  • The assurance and commitment from the repository that the peer reviewer identity will be kept confidential. A statement pertaining to this on the repository page is helpful here.