iridium – JISC MRD02 monthly updates to January ’13

Progress to date

Workpackage 0 (Project Management)

  • 4th Steering Group meeting took take place on 08 January 2013, with two further planned for April and June 2013

Workpackage 1 (Requirements)

  • activities completed

Workpackage 2 (Policy)

Workpackage 3/4/5 (Tools, Systems, & Implementation)

  • Research Data Catalogue (RDC) user testing data loaded
  • DCC DMP Online tool used to host draft iridium institution-specific post-award DMP template
  • local CKAN test system established
  • e-Science Central Shibboleth authentication added, with SWORD protocol functionality being investigated

Workpackage 6 (Human Support)

  • additional good practice guidance content and revised draft policy principles added to developing RDM support website
  • meetings with Netskills on 09 Jan and 13 Feb 2013 to outlined RDM workshop and online training requirements
  • documentation for RDM tools authored
  • productive meeting with Staff Development Unit on 14 February 2013

Workpackage 8 (Evaluation)

  • draft RDM policy principles for available for open consultation from institutional website
  • testing fitness for purpose of identified RDM tools underway (see https://iridiummrd.wordpress.com/?s=tools+evaluation)
  • RDC minimal addition metadata entry testing and user acceptance will take place from February 2013, with 12 PIs invited to record data location (representing 39 research projects and 474 publications)
  • iridium support team have evaluated project RDMP template implementation within DMP Online system, to be followed by research funding support staff and researchers

Workpackage 9 (Dissemination)

  • Niall O’Loughlin (RES) gave a talk to research office representatives from Brunswick Group on 25 January 2013 on ‘Open Research Data’.

Activities in the last months

  • Dr Ben Allen and Dr Simon Kometa registered to attend ‘CKAN for RDM’ on 18 February 2013
  • Paul Haldane registered to attend ‘RDM Storage Workshop’ on 25 February 2013
  • Niall O’Loughlin registered to attend ‘Research data in the Visual Arts’ event on 06 March 2013

Risks & issues

Risk/issues likely to present in the coming 3 months are:

  • project still needs to be very clear that the iridium project is not about data storage space provision
  • managing researchers’ expectations about provision of RDM tools
  • REF 2014 will be a major priority for researchers and institution in coming months

Milestones & challenges

In the coming 3 months, identified milestones are:

  • project outputs (policy, tools and training) aligned into pilot infrastructure
  • pilot infrastructure evaluated against research projects’ needs
  • monitoring institutional IT re-structuring for embedding of project outputs
  • promoting uptake of pilot infrastructure and outputs
  • business case authoring
Advertisements

iridium JISC MRD02 – monthly updates to November ’12

Progress to date

Workpackage 0 (Project Management)

  • draft proposal for extension of project submitted
  • full-team meeting took place on 12 November 2012 to review project against JISC Progress Meeting RDM ‘components’

Workpackage 1 (Requirements)

  • requirements gathering outputs disseminated locally and externally to JISC MRD Programme
  • resultant project actions dissemination locally to stakeholders through Registrar’s regular update to Heads of Departments
  • anonymised requirements data shared with local projects and initiatives (Computing Science data security Choice Modelling, Digital Campus Initiative Infrastructure Discovery and ISS Information Security projects)

Workpackage 2 (Policy)

  • working group met on 06 September and 27 September 2012
  • policy principles and Code of Good Practice have been revised and are in a mature draft form
  • briefing of senior project advocates is planned ahead of tabling of draft policy document at URC
  • permission will be requested from URC for draft policy principles to be published on project website for open consultation

Workpackage 3/4/5 (tools, systems, & implementation)

  • research data catalogue will be tested with subset of Changing Age researchers
  • draft Newcastle-specific post-award DMP template will be tested within DCC DMP Online system from Decemebr 2012
  • SWORD2 protocol investigations blogged
  • CKAN platform to be tested

Workpackage 6 (Human support)

  • RDM support materials writing sessions continued on 08 and 18 October 2012
  • RDM support website has being populated with FAQ, draft policy principles and good practice guidance content
  • documentation of selected RDM-specific tools continuing
  • Dr Simon Kometa (ISS) attended JISC ‘Research Data Management Training’ workshop on 26 October 2012

Workpackage 8 (Evaluation)

  • draft policy principles reviewed in light of comments received from stakeholder to date
  • evaluation to date reported internally on tools (i.e. e-Science Central and DMP Online)
  • Lindsay Wood attended JISC benefits meeting in Bristol 29-30 November 2012

Workpackage 9 (Dissemination)

  • newsletter-style project update disseminated on 05 October 2012
  • project disseminated at JISC Progress Meeting on 24-25 October 2012
  • Niall O’Loughlin (RES) gave a talk to AFRD Research Committee on 09 November 2012

Activities in the last months

  • EPSRC-funded FRICCTT report published that referenced JISC iridium project
  • local EPSRC-funded Cyber Security project awarded that referenced JISC iridium project requirements gathering data in case for support
  • Lindsay Wood attend JISC IRIOS-2 RIM meeting in Newcastle on 21 September 2012
  • Suzanne Hardy (MEDEV) attended the DataCite citing sensitive data workshop on 29 October 2012 at the British Library
  • Dr Ben Allen (ISS) attended RDMF9 workshop on 14-15 November 2012 in Cambridge

Risks & issues

Risk/issues likely to present in the coming 3 months are:

  • project still needs to be very clear that the iridium project is not about data storage space provision
  • managing researchers’ expectations about provision of RDM tools
  • degree of policy revisions post-URC, that might be required, is unknown
  • maintaining good communication links with the URC during project team staffing changes
  • REF 2014 will be a major priority for researchers and institution in coming months

Milestones & challenges

In the coming 3 months, identified milestones are:

  • draft policy approved by URC and then by Executive Board
  • project outputs (policy, tools and training) aligned into pilot infrastructure
  • pilot infrastructure evaluated against research projects’ needs
  • monitoring institutional IT re-structuring for embedding of project outputs

iridium – workshop talk and dissemination at JISC Progress Meeting, Nottingham

The iridium project presented at the JISC MRD02 Progress Meeting in Nottingham. The two day schedule from the event is here, together with the Programme introductory/close slides.

Workshop topics were:

  • Institutional RDM policies; developing an institutional strategy and an ‘EPSRC’ roadmap
  • Managing active data: storage, access, academic dropbox services
  • Data management planning: developing good practice and providing effective support
  • Data repositories and storage: options for repository service solutions
  • Training & guidance
  • Triage and handover: what to keep and where to entrust it? Selection and appraisal, deposit and handover
  • Business case: covering roles, responsibility, costing, sustainability, advocacy etc
  • Data catalogues: metadata profiles, identifiers

Individual projects were encouraged to contextualise presentations around the following themes:

[1] “what has worked/is working”
[2] “what lessons you have learned and how generalisable these may be”
[3] “what challenges remain”
[4] “how such challenges may be approached and what your institution/project intends to do”
[5] “what DCC / MRD activity you think may help make the challenge more tractable”

iridium ‘support’ presentation within ‘Training & Guidance’ session:

iridium presentation thumbnail

iridium presentation

iridium_JISC_Progress_25_10_2012_v4_web_sml_LW [.pdf]

We also presented two posters, one on the research data catalogue proof-of-concept and the second on our thematic analysis requirements gathering.

Other project presentations from the Programme are available here.

iridium – summary of thematic analysis of RDM researchers’ requirements from interviews

iridium thematic analysis – summary of qualatative interviews

Process

  • A two stage thematic analysis was conducted on the interview data generated by the Iridium Project.
  • 29 Interviews underwent Analysis.
  • The first stage of Analysis was deductive, and conducted by interviewers on the transcripts of interviews.
  • The deductive analysis was categorised into 5 initial themes:
    • Perception – To ascertain interviewee’s concepts of data and data management.
    • Purpose – To ascertain interviewee’s data usage and destination.
    • Process – To ascertain interviewee’s data lifecycle.
    • People – To ascertain the people involved in the data lifecycle.
    • Provoking – A Catchall to gather any other salient points expressed by interviewees.
  • Interim Categorisation Results:

iridium thematic analysis figure

  • The second stage of the analysis was inductive.
  • Uses results of deductive analysis as a starting point – attempts to build meaningful themes from whole data corpus.
  • Analysis done by a single researcher.
  • Generates a new set of themes

Final Themes

Diversity

Informed by elements from across the deductive categories, in particular perceptions of data usage, longevity and security, purpose of data, the people involved and the processes required.

The overarching theme is that across many aspects of data management there is a great deal of diversity amongst users and any policy should enable users to achieve best practice rather than apply a one size fits all “solution” to data management.

Data Analysis

Generated mainly from the process and perception of data and its life-cycle, a Data Analysis theme was apparent across many interviewees.

The consensus was that much of the processing of data that currently takes place on local machines would be more efficiently accomplished on larger scale servers, but that users are largely unaware that such services may exist in the university.

Longevity / Life Cycle

The longevity of data was a strong theme from across all categories. There was a strong consensus from interviewees that data should never be thrown away. There should be a separate system for archiving data and current data.

Any policy should attempt to support long term storage of research data, and the access to it, as well as current data.

Responsibility

The strongest theme coming from the people category was that of responsibility – who should be doing what with the data, with storage, with security and with access. Many interviewees were unclear about what falls to them and what the responsibility of the university is.

The recommendation to take forward is to make the situation clear, and provide training if needed for users.

Sharing and Collaboration

Another strong theme that emerged from across the initial categories is the concept of data access, in particular sharing data with collaborators, both internal and external. This becomes problematic with very large data sets, or with collaborators insisting on using “their” systems.

The university should provide a method of sharing data post publication, which should be linked to publications and to researcher’s profiles, and provide flexible guidelines on sharing with collaborators.

The full thematic analysis report is available to download: http://research.ncl.ac.uk/media/sites/researchwebsites/iridium/iridium_interview_thematic_analysis_5_7_2012_v1_PH.pdf

iridium – summary of online RDM requirements gathering survey findings

Quantitative online survey –  summary

Key Findings:

  • One hundred and twenty eight projects completed the online survey and over half of the projects are from the Faculty of Medical Sciences (nearly 52%).
  • Over 97% of projects’ data is in digital / physical format.
  • Generally projects have many files e.g. 23.4% of projects have between 100- 1000 files, 28.1% between 10-100 files and 19.5% have between 1000 – 10000 files.
  • Thirty one percent of projects’ files take up to 4GB of space. Just over 11% of projects’ files take up between 64 – 0.5TB. One project had more than 100TB, but no project required more than 1 PB.
  • For nearly 54% of projects space required at collection is greatly different from space required after processing / analysis. However for nearly 29% of projects the space was not greatly different.
  • There was lots of variability in data file format amongst projects. Some projects used many different file formats indeed. However Excel was the most common file format.
  • About 30% of projects said they store their data on ISS managed systems; about 18% of projects used academic unit managed systems and about 21% used personal systems. A small number of projects used external systems / services such as cloud and SurveyMonkey.
  • The majority of projects (nearly 43%) intend to keep data for 5 to 10 years. Just over 29% intend to keep data for 10 – 25 years and about 17% intend to keep data for more than 25 years. Eleven percent intends to keep data for 1 – 5 years.
  • 93% of projects have multiple copies / partial back up of data.
  • Nearly 49% of the projects have tested how successful it will be to retrieve backed up data.
  • Just over 73% of projects share their data with others within the University.
  • Just over 50% of projects share data externally.
  • Of 64 projects (50%) who share data external to the University nearly 30% don’t have any agreement in place while nearly 33% have other types of agreement not specified.
  • About 56% of projects have a data management plan or partial / informal plan.
  • Nearly 65% of projects don’t have any specific tool for RDM.
  • The majority (46%) of projects said that they do not have any deletion policy and just fewer than 17% have a deletion policy.
  • For over 75% of projects data have to be quite secure or very secure.
  • 84% of projects said they store their data quite securely or very securely.
  • Over 90% of projects used password, anonymisation or physical measures for data security.
  • Not many projects are aware of the policies and legislation that applies to their data e.g. only 63% of projects are aware of DPA and only 40% are aware of FOI.
  • 64% of projects said that the PI should have the primary responsibility for RDM support. Next in line was the Research Associate with just 11%.
  • 55% of projects believe that going forward the PI should still have the primary responsibility for RDM support. All the other officers got less than 10% of the vote except for computing support officer where just over 17% of projects think they should have the primary responsibility for RDM support.
  • Only 5 projects said they are aware of training sessions and materials on RDM.
  • 60% of projects gave a positive response to make their research data publicly available at the end of the project.
  • 73% of projects have not deposited any of their data in a data repository.
  • Nearly 60% of projects are willing to submit data to a data repository.
  • An overwhelming majority of projects (nearly 80%) are happy to submit data to a repository at the publication stage.
  • Nearly 41% of projects are willing to make data supporting any publication available immediately.
  • Nearly 73% of projects are willing to share data if they have control over who can access the data.
  • There is no clear consensus from projects on intellectual property rights (IPR), just over 30% of projects believe that it is owned by the University and about 17% of projects do not know; just over 19% think it belongs to their research group and 10.4% said other. For about 7% and 15% of projects it belonged to the funder and the researcher respectively.
  • The majority of projects were funded by either charity or research council 35.5% and 31.4% respectively, that is, a total of nearly 67%.

For more details, see full survey report: http://research.ncl.ac.uk/media/sites/researchwebsites/iridium/iridium_online_survey_report_17_8_2012_v2.1_SK.pdf

iridium – poster dissemination at JISC Progress Meeting, Nottingham

Janet Wheeler and I will be attending the JISC Progress Meeting/DCC event in Nottingham and are presenting a poster on developing an RDM tool to support implementation of policy principles.

JISC MRD Progress Meeting poster

JISC MRD Progress Meeting poster

iridium – poster presentation of RDM thematic analysis at Digital Research 2012, Oxford

iridium project team members from the Digital Institute presented a poster on the JISC-funded thematic analysis work carried out with local research and related staff to gather and understand RDM requirements. The work was present at the recent Digital Research 2012 conference in Oxford.

iridium DI Digital Research 2012 poster

iridium DI Digital Research 2012 poster

%d bloggers like this: