Monday 25 June 2012

OpenAIREplus workshop - notes from the breakout session


One of Copenhagen's bridges being opened, so we can sail through!

1. Funders and data policy
 * Lots of interest in the data value checklist - compare UK and Australian data value checklist
 * It's cheaper to keep data rather than recreate it
 * Can you require open availability of data brought into a project? Case by case negotiation
 * Multiple funders - which data policy will be applied?
 * CODATA preparing a toolkit for funders about open data policy
 * Role of institutional repositories? Data centres are good places to handle data pools
 * Need clear metadata!
 * How to handle data management plans once the project is over? Fund data management post-project. Should remain institutional responsibility.
 * Identifiers - need researcher identifiers, funder acknowledgements, DOIs - all to pull together project information and data
 * Are there international approached in data management plans?

2. Institutional policy
 * Most institutions don't have a policy yet because they're not easy to create
 * What other steps need to be done before policy?
 * Hierarchy - who to get involved - academic champions
 * Broad overview - what are the needs of researchers -  don't want extra admin
 * Don't contradict other policies or legislation
 * Smaller institutions don't have monet or effort to get into big data infrastructures
 * policy can guide researchers on what to do with their data
 * What should be deposited, what should be kept
 * How can we help insitutions develop data management plans?
 * Guidelines on developing data management policies
 * What kind of questions do we need to know before drafting policy?

3. Researchers and publishers
 * Current examples are life and environmental sciences
 * we need other examples in other fiels
 * Researchers need acknowledgement for their work on data - not having it stops them shring
 * Quality issues are important - need principles for peer review of research data
 * Users of data are candidates to review it
 * there are varying degrees of openness in peer-review - which will be appropriate for data?
 * What stopes researchers sharing data? Quality, promotion, confidence in the value of the data
 * We can give researchers more confidence in their data by promoting community standards
 * Change beahviour so that data management is done every day, instead of just at the beginning and end of the project.
 * Publishers can influence researchers when it comes to data management.
 * Metrics are needed, data citation, but also alt.metrics
 * Need for good examples of data management to educate researchers
 * Need a list of trusted databases/repositories
 * URLs aren't trusted, because they break!

4. Technical
 * Finland is constructing a national data catalogue, containing a mix of metadata records and data
 * OpenAIRE data model and services are using trust levels for entities and (automatic and man-made) relations
 * Need to guarantee long term data availability for enhanced publications to be trustworthy, or at least know what bits will last for how long
 * Level of trust needed to develop services to show levels of preservation
 * Services should still exist for low trust objects - e.g.g use a robot to check if the object is still there, and if not, drop the connection.

5. OpenAIREplus
 * Are there licensing restrictions for metadata?
 * Case studies of scientific communities should be published as soon as possible
 * Credit for researchers is important
 * Libraries have a role too - even if there is a fear of data management
 * Universities are very disparate - makes it hard politically to agree on data policy.


2 comments:

  1. Sarah - what exactly are the "data value checklists" for the UK and Australia that you are referring to?

    ReplyDelete
  2. Hi Triangle,

    The data value checklist is something that colleagues of mine have been working on for quite a while now - basically, it's a standardised method for determining if data is of long term value to the community and therefore is suitable for archiving and curation. I haven't actually seen it yet, but there's a lot of interest in it already.

    I believe the Australians have been working on something similar.

    ReplyDelete