Introduction to Data Licences

Introduction to Data Licences

• 1.
  INTRODUCTION TO DATA LICENCES Textis licensed under a Creative Commons Attribution 4.0 International License. Other licences apply to images (see slide 19) Gareth Knight gareth.knight@lshtm.ac.uk 26th May 2015 Disclaimer: I am Not a Lawyer. This is not legal advice
• 2.
  Overview • Reasons toapply a data licence • Database rights • Expectations for data licences (LSHTM, funder, journal) • Licensing your research data (CC, ODC, DTAs) • Working with 3rd party data • Conclusion
• 3.
  Data Licence • Datalicence motivated by need to share data • Establish conditions under which data may be accessed, used & cited • Many rights apply to data – Copyright – Moral – Database • Rights issues vary between countries • Ensure that your project has clarified rights issues before sharing
• 4.
  Database Rights • Doesnot require 'creative' aspect defined by copyright to be protected • ‘Copyright and Rights in Databases Regulations’ 1997 applies it to: – DBs created after 27 March 1996 – Created by EEA nationals, EEA residents & businesses with central operations in EEA • Last for 15 years from completion or publication (whichever is longest) • Any substantial update causes the 15 year period to begin anew Recognise investment made into compilation of database
• 5.
  LSHTM Expectations • Rightsinformation should be: – Clear and unambiguous – Documented at an appropriate level of granularity • Recommend a non-exclusive licence that allows many people to access and use data • Should not assign exclusive rights to 3rd party, unless it is a condition imposed by contractual or other obligations. “Rights assigned to research data should not unnecessarily restrict its management, sharing, or use” RDM Policy, Principle 2 LSHTM RDM Policy http://www.lshtm.ac.uk/research/researchdataman/rdm_policy.html#principle02 LSHTM RDM Policy http://www.lshtm.ac.uk/research/researchdataman/rdm_policy.html#principle02
• 6.
  • Many statedata should be: "made openly available and accessible with as few restrictions as possible" • Encourage FAIR principles - Findable, Accessible, Interoperable & Re-usable • Expect consortium agreements to enable data sharing • However, recognise legal, ethical & commercial constraints may limit access and use • Encourage licences that encourage uptake in some form (Open, controlled, commercial) Funder Expectations
• 7.
  Journal Expectations • Growingnumber of journals that: – Specialise in open data – Expect data to be made available • Expect openness by default, unless good reason can be provided – Creative Commons Attribution (CC-BY) – Anonymous download from a public repository • Recognise open sharing not always feasible for health data – Controlled access – Data sharing agreements
• 8.
  Licensing your researchdata 1. Identify rights holders – LSHTM – Collaborators – Participants – Data providers & others 2. Determine terms to be applied – Policy, Consortium agreement, consent forms 3. Review licence models – Creative, Commons – Open Government Licence – Data Transfer Agreement 4. Select appropriate licence
• 9.
  Creative Commons ATTRIBUTION NON-COMMERCIAL NO DERIVATIVES SHAREALIKE + : Permits open reuse - : ‘Attribute stacking’ – distant authors - : Unable to control how data is used + : Permits academic & other NC use - : Cannot be used by commercial collaborators + : Allows analysis - : No cleaned, remixed, or other derivatives allowed + : Prevents future users re-licensing derivatives under a restrictive licence - : Difficult to combine with other licences
• 10.
  Open Data Commons Sharedata Create works Adapt & modify ODC Open Database License (ODbL) http://opendatacommons.org/licenses/odbl/ Attribution, Share alike & keep open ODC Attribution http://opendatacommons.org/licenses/by/ Attribution ODC Public Domain Dedication http://opendatacommons.org/licenses/pddl/summary/ None
• 11.
  Data Transfer Agreement Tailoredlicence form that defines conditions such as: • Use for specific purpose • Handling of participant information (no attempt to re-identify) • Storage in secure environment • Retention period (e.g. 1 yr) • Administration charges Signed agreement that specifies storage & use conditions Contact the Knowledge Transfer Manager (deborah.carter@lshtm.ac.uk) for guidanceContact the Knowledge Transfer Manager (deborah.carter@lshtm.ac.uk) for guidance
• 12.
  Working with 3rdparty data Myth: • Data has no licence: It can be analysed for any purpose • Data is free to download: It can be re-published in any form Review dataset description for licence information: 1. What purposes are you allowed to use data for? 2. Are there any expectations? – e.g. citation, costs 3. Are there any limitations? • If there's no licence, contact the creator/publisher. Large amount of data available, but can you use it?
• 13.
  Dealing with multipledatasets Potential for licence conflict when working with multiple datasets Licence compatibility definitions: • Exact match: Both licences use terms that have the same purpose, meaning and effect • One-way: Data made available under a permissive license can be combined with data that has a more restrictive licence Does source licence allow mashups to be shared? http://creativecommons.org/compatiblelicenses http://opendefinition.org/licenses/ http://opensource.com/law/11/9/mpl-20-copyleft-and-license-compatibility
• 14.
  Creative Commons: No Derivatives Instructionsneeded to explains how dataset may be merged for validation purposes Based upon scenarios by Leigh Dodds "Oil and Water: When Data Licences Don't Mix“ http://www.slideshare.net/ldodds/oil-and-water-when-data-licences-dont-mix
• 15.
  Creative Commons: Share alikeConflict Each update can be published separately & instructions provided on how to merge for validation Based upon scenarios by Leigh Dodds "Oil and Water: When Data Licences Don't Mix“ http://www.slideshare.net/ldodds/oil-and-water-when-data-licences-dont-mix
• 16.
  Creative Commons: Mixing openlicences If licences are compatible, least open licence wins Based upon scenarios by Leigh Dodds "Oil and Water: When Data Licences Don't Mix“ http://www.slideshare.net/ldodds/oil-and-water-when-data-licences-dont-mix
• 17.
  Conclusion • Licence modelsare a key component of data sharing • Researchers must consider data licences when – Reusing existing data – Creating own data (particularly when there is a need make data available) • Rights information should be clear, unambiguous and documented at appropriate granularity • Recommend use of a non-exclusive licence that allows many people to access and use data • Should not assign exclusive rights to 3rd party, unless it is a condition imposed by contractual or other obligations.
• 18.
  Resources • LSHTM. Choosea Data Licence http://www.lshtm.ac.uk/research/researchdataman/share/choose_licence.html • Digital Curation Centre: How to License Research Data http://www.dcc.ac.uk/resources/how-guides/license-research-data • Korn & Oppenheim (2011). Licensing Open Data: A Practical Guide http://discovery.ac.uk/files/pdf/Licensing_Open_Data_A_Practical_Guide.pdf • ODI. Publisher’s Guide to Open Data Licensing https://theodi.org/guides/publishers-guide-open-data-licensing
• 19.
  Images Slide 3: “That’sRight” by Kaytee Riek (CC BY-NC-SA 2.0) https://www.flickr.com/photos/riekhavoc/4813140176/ Slide 4: “WordPress 2.7 Database Schema” by Rafael Poveda (CC BY-NC-SA 2.0) https://www.flickr.com/photos/bioxid/3640432505/ Slide 8: “Reuse” by Steev Hise (CC BY-NC-SA 2.0) https://www.flickr.com/photos/steev/39393264/ Slide 11: “Permit Holders Only A” by Gregory Wake (CC BY-NC-SA 2.0) https://www.flickr.com/photos/gregwake/2301264039 Slide 12: “Sharing” by ryancr (CC BY-NC 2.0) https://www.flickr.com/photos/ryanr/142455033/ Slide 13: “Death By…” by nataliej (CC BY-NC 2.0) https://www.flickr.com/photos/nataliejohnson/2296566285/