Open Data Publication - Requirements, Good practices, and Benefits

Download Open Data Publication - Requirements, Good practices, and Benefits

Post on 05-Dec-2014

204 views

Category:

Data & Analytics

1 download

Embed Size (px)

DESCRIPTION

Presentation by Guntram Geser, Salzburg Research EAA 2014 session: Open Access and Open Data in Archaeology Istanbul, Turkey 13 September 2013

TRANSCRIPT

<ul><li> 1. ARIADNE is funded by the European Commission's Seventh Framework Programme Open Data Publication Requirements, Good Practices, and Benefits Open Data Publication Requirements, Good Practices, and Benefits EAA 2014 - Istanbul Session: Barriers and Opportunities: Open Access and Open Data in Archaeology 13 September 2014 Guntram Geser Salzburg Research EAA 2014 - Istanbul Session: Barriers and Opportunities: Open Access and Open Data in Archaeology 13 September 2014 Guntram Geser Salzburg Research </li> <li> 2. ARIADNE Advanced Research Infrastructure for archaeological Dataset Networking in Europe EU FP7-Infrastructures project, type Integrating Activity Runs 4 years, 02/2013-01/2017 Focus on archaeological datasets help overcome data fragmentation and foster a culture of sharing and re-using 23 partners of 18 European countries Open to other participants, e.g. Transnational Access Programme Special Interest Groups Website: www.ariadne-infrastructure.eu Advanced Research Infrastructure for archaeological Dataset Networking in Europe EU FP7-Infrastructures project, type Integrating Activity Runs 4 years, 02/2013-01/2017 Focus on archaeological datasets help overcome data fragmentation and foster a culture of sharing and re-using 23 partners of 18 European countries Open to other participants, e.g. Transnational Access Programme Special Interest Groups Website: www.ariadne-infrastructure.eu </li> <li> 3. Main topics of this presentation Open Data expectations, criteria, drivers Current data (non-)sharing behaviour Reasons for lack of open data sharing ARIADNE survey results on data publication How to benefit from open data publication Issue of actual data re-use Takeaway points Open Data expectations, criteria, drivers Current data (non-)sharing behaviour Reasons for lack of open data sharing ARIADNE survey results on data publication How to benefit from open data publication Issue of actual data re-use Takeaway points </li> <li> 4. Open Data expectations The scientific method: provide evidence for knowledge claims (show us your data) Output of publicly funded research should be openly available, and properly preserved More open data sharing =&gt; better analysis =&gt; better decision-making / solutions for critical issues Better return-on-investment, e.g. no duplication of data collection better exploitation of available data therefore emphasis on re-use Innovation through data-intensive, data-driven, big data (incl. archaeology?) The scientific method: provide evidence for knowledge claims (show us your data) Output of publicly funded research should be openly available, and properly preserved More open data sharing =&gt; better analysis =&gt; better decision-making / solutions for critical issues Better return-on-investment, e.g. no duplication of data collection better exploitation of available data therefore emphasis on re-use Innovation through data-intensive, data-driven, big data (incl. archaeology?) </li> <li> 5. Open Data criteria Accessible online not necessarily without registration Reusable not summarized data (i.e. figures, charts, etc.) canned in publications state: raw, cleaned, normalized, (accord. to practice) open format (e.g. not PDF documents) Openly licensed (e.g. CC-BY, if other no NonDerivative!) For free yes, but somebody has to pay to ensure sustainability Accessible online not necessarily without registration Reusable not summarized data (i.e. figures, charts, etc.) canned in publications state: raw, cleaned, normalized, (accord. to practice) open format (e.g. not PDF documents) Openly licensed (e.g. CC-BY, if other no NonDerivative!) For free yes, but somebody has to pay to ensure sustainability Publishing data in a reusable form to support findings must be mandatory one of six key areas for action highlighted in the The Royal Societys report Science as an Open Enterprise (2012) </li> <li> 6. Open Data drivers /1 High-level policies &amp; initiatives OECD Declaration on Access to Research Data from Public Funding (2004; Principles and Guidelines, 2007) EC Communications, Open data (2011), Towards better access to scientific information (2012) Many others, e.g. Research Data Alliance international initiative (launched in March 2013), various working &amp; interest groups (archaeology not represented yet) High-level policies &amp; initiatives OECD Declaration on Access to Research Data from Public Funding (2004; Principles and Guidelines, 2007) EC Communications, Open data (2011), Towards better access to scientific information (2012) Many others, e.g. Research Data Alliance international initiative (launched in March 2013), various working &amp; interest groups (archaeology not represented yet) Neelie Kroes, EC Vice-President, 2012: Taxpayers should not have to pay twice for scientific research and they need seam- less access to raw data. We want to bring dissemination and exploitation of scientific research results to the next level. </li> <li> 7. Open Data drivers /2 Research funding agencies Open Access mandates extended to data Mandatory data management plans, i.e. data sharing must be considered already at application stage Data archiving &amp; access infrastructures put in place Data centres / repositories General : zenodo (related to OpenAIRE), Figshare, Archaeology: e.g. ADS (UK), eDNA (NL), IANUS (Germany, in preparation), MAPPA (IT), OpenContext and tDAR (USA) Data registries / catalogues Data citation standard, e.g. DataCite Research funding agencies Open Access mandates extended to data Mandatory data management plans, i.e. data sharing must be considered already at application stage Data archiving &amp; access infrastructures put in place Data centres / repositories General : zenodo (related to OpenAIRE), Figshare, Archaeology: e.g. ADS (UK), eDNA (NL), IANUS (Germany, in preparation), MAPPA (IT), OpenContext and tDAR (USA) Data registries / catalogues Data citation standard, e.g. DataCite </li> <li> 8. Open Data drivers /3 New publication format of data papers Describe a dataset/database, kind of extensive metadata record or documentation, also including examples of use / usefulness for research Examples of data journals in archaeology Journal of Open Archaeology Data, started 2012 Internet Archaeology, started a series of data papers in 2013 New publication format of data papers Describe a dataset/database, kind of extensive metadata record or documentation, also including examples of use / usefulness for research Examples of data journals in archaeology Journal of Open Archaeology Data, started 2012 Internet Archaeology, started a series of data papers in 2013 </li> <li> 9. Barriers to open data sharing Many obstacles to providing open access to reusable data Priority of published papers Little academic reward for development and sharing of datasets/DB Required effort to share re-usable data (incl. formatting, metadata creation, licensing etc.) Existing copyrights, confidential and sensitive data Concerns that data could be scooped, misused or misinterpreted Potential reputational risk (e.g. data quality, errors,) Overall a bad ratio of additional effort &amp; risks to potential benefits Many obstacles to providing open access to reusable data Priority of published papers Little academic reward for development and sharing of datasets/DB Required effort to share re-usable data (incl. formatting, metadata creation, licensing etc.) Existing copyrights, confidential and sensitive data Concerns that data could be scooped, misused or misinterpreted Potential reputational risk (e.g. data quality, errors,) Overall a bad ratio of additional effort &amp; risks to potential benefits </li> <li> 10. Current data (non-)sharing behaviour Contrary to what advocates of proper management and sharing of data would like them to do According to representative surveys (PARSE.Insight 2009, Science 2011): Most data remains locked away On personal computers Portable storage carriers Restricted access servers Eventually discarded as obsolete or lost otherwise Mostly not considered: Potential value of the data for alternate and new uses by others Only 6-8% of researchers sometimes deposit data in a community archive Contrary to what advocates of proper management and sharing of data would like them to do According to representative surveys (PARSE.Insight 2009, Science 2011): Most data remains locked away On personal computers Portable storage carriers Restricted access servers Eventually discarded as obsolete or lost otherwise Mostly not considered: Potential value of the data for alternate and new uses by others Only 6-8% of researchers sometimes deposit data in a community archive </li> <li> 11. Archiving/storing data for future use /1 PARSE.Insight survey 2009: 1202 respondents from different research domains and countries </li> <li> 12. Archiving/storing data for future use /2 Science journal 2011 survey of peer reviewers: 1700 responses, (international and multi-disciplinary Where do you archive most of the data generated in your lab or for your research? Science journal 2011 survey of peer reviewers: 1700 responses, (international and multi-disciplinary Where do you archive most of the data generated in your lab or for your research? Note: archived curated 50.2% in our lab 38.5% university server 7.6% community repository 3.2% other 0.5% not stored </li> <li> 13. ARIADNE online survey on user needs Participants (881 questionnaires received, 692 with sufficient inform.) o586 archaeological researchers o 54 directors of research institutes o 52 repository managers Organisational context (640 responses) o54% Universities o16% Governmental organisations o12% Private companies or institutes o13% Self-employed Gender (482 responses) o57% male o43% female Geographic distribution (482 resp.) o83% EU (UK: 79, France: 51, Italy: 47% NL: 35,) o 5% other European o12% non-European November-December 2013 </li> <li> 14. Data publication </li> <li> 15. Barriers to data deposit/publication </li> <li> 16. How would you rate the importance of the following potential barriers to enhancing access to research data? European Commission: Online survey on scientific information in the digital age (2012); 1140 participants from around Europe. Need to make clear benefits of open data publication! </li> <li> 17. Authors benefits focus Goal = recognition and academic reward for data providers at least same as for other publications Core mechanism = citation of published data/set Confirms value of the data contributed Indicates providers of good data Promotes further use of the data (i.e. more citations) Allows the use and impact of the data to be tracked and measured But data citation metrics not implemented yet Some indications of higher citation rates of publications that make underlying data available Goal = recognition and academic reward for data providers at least same as for other publications Core mechanism = citation of published data/set Confirms value of the data contributed Indicates providers of good data Promotes further use of the data (i.e. more citations) Allows the use and impact of the data to be tracked and measured But data citation metrics not implemented yet Some indications of higher citation rates of publications that make underlying data available </li> <li> 18. How to reap the benefits? / 1 Deposit data that underpins your research results in a reliable, community recognised repository See: Data Seal of Approval; Trusted Repositories Audit &amp; Certification (TRAC) and other checklists Should provide unique persistent identifiers (e.g. DOIs) Require following citation standard as part of user agreement (e.g. DataCite; citation in reference list) Provide good metadata no pain, no gain Key for data re-use without direct contact with creator Costs of preparing data and metadata for publication should be included in project funding Deposit data that underpins your research results in a reliable, community recognised repository See: Data Seal of Approval; Trusted Repositories Audit &amp; Certification (TRAC) and other checklists Should provide unique persistent identifiers (e.g. DOIs) Require following citation standard as part of user agreement (e.g. DataCite; citation in reference list) Provide good metadata no pain, no gain Key for data re-use without direct contact with creator Costs of preparing data and metadata for publication should be included in project funding </li> <li> 19....</li></ul>

Recommended

View more >