Tuesday, 16 July 2013

UK Government promotes Open Data - a public good but where is the money?

A debate was held at The Royal Society on 10th July, 2013 titled Maximising the use of public data – should research and publicly acquired data be made more accessible? At this debate David Willetts MP and Minister of State for Universities and Science gave his strong support for Open Data and stated that "the public have a right to access publicly funded research".

This blog post gives details of the debate with links to the resources available from the organisers The Foundation for Science and Technology. It also asks the question: is this an unfunded mandate?

Debate Summary

The debate was an invitation only event held at the The Royal Society on 10th July, 2013 and organised by the The Foundation for Science and Technology.

The participants were:
Chair: The Earl of Selborne GBE FRS Chairman, The Foundation for Science and Technology

Speakers:
  • Professor Geoffrey Boulton OBE FRS FRSE Chair, The Royal Society Inquiry into Science as an Open Enterprise
  • Professor Sir Nigel Shadbolt FREng Chairman and Co-Founder of the Open Data Institute
  • The Rt Hon David Willetts MP, Minister of State for Universities and Science, Department for Business,  Innovation and Skills
Panellist: Professor Sheila M Bird OBE FRSE  Programme Leader, MRC Biostatistics Unit, Institute of Public Health, Cambridge.

The following are my notes and takeaways from each speaker. In every case I link to the audio of the debate posted on the Foundations website. Any errors or misrepresentations are thus my own and you can cross check if you desire...


Professor Geoffrey Boulton

The full slides and audio is available here

















  • The processes through which science is done is now acquiring extraordinary volumes of data - Question is what impacts that data has upon how we do our science?
  • Cited Henry Oldenburg – openness of scientific correspondence in mid-1600’s
  • Openness one of the most fundamental and critical factors for peer review in science
  • Cited Tim Gowers – crowd sourcing mathematics. 32 days = 27 people = 800 substantial contributions that resolved a mathematical problem.
    “it’s like driving a car whilst normal research is like pushing it”
  • Addressed the changing social dynamic of science – the public demand the evidence upon which scientific claims are being based.
  • Openness of data per se has little value. Open Science is more than disclosure, it is Intelligent Openness.
    • For effective communication, replication and re-purposing we need intelligent openness. Data and meta-data must be:
      • Accessible
      • Intelligible
      • Assessable
      • Re-usable
    • Only when these four criteria are fulfilled are data properly open.
  • Action on open data – international, European and UK (JISC gets a shout!)
  • This "is a process not an event, will we ever arrive there, no we wont" – there is more to do!
My favourite slide from Prof Boulton - a taxonomy of openness















Professor Geoffrey Boulton was to my mind the most informed and helpful speaker of the night. He grasped all the intricacies of the subject whilst communicating them in a clear and unfussy manner.




Professor Sir Nigel Shadbolt



















  • Open Licences + Open standards + Open data + Open participation + Open Source software = Open Innovation
  • Is Open Innovation a new paradigm - the fifth paradigm?
  • Open Data is information that is available for anyone to use, for any purpose, at no cost. has to have a licence that says it is open data. Without a licence, the data can’t be reused.
  • Open data challenges:
    • Infrastructure – what we store and how we store it
    • Quality – how do we improve the data quality attributes – what does OpenData 2.0 look like?
    • Data literacy – build capability to understand data and information – how do we engage broadly in a world where data matters to our daily life?
    • Security and privacy – e.g. ship positioning in real time- great for insurance companies but also for Somali pirates...
    • Elites – gatekeeper attitudes to data on basis that others wont understand it
    • Incumbents – strong data position so why share?
    • Legislation – Non delivery of Open Govt Data – can we make money from the data so why share?
  • Open innovation is when everyone can engage in this process because the data is intelligently open.





















Fascinating cheerleading for Open Data and exciting insights. However, I did not buy the 5th paradigm idea he proposed, I felt this was throwing terms around for the sake of it...




Professor Sheila M Bird

The Minister was late so Professor Bird made some comments at this point to fill time. She did so very entertainingly and incredibly intelligently. Her comments upon privacy and informed consent were especially fascinating. I recommend you go listen to her responses in full as my notes do not do them justice.

  • 5 paradigms do not necessarily occur in the order given – some recurse and will need to be subject to further experimentation.
  • Scientific method has nothing to fear from openness.
  •  Delayed discoveries – costs of buying informatics data has gone down in recent years but the costs of longitudinal studies and record linkage studies has risen in costs.


The Rt Hon David Willetts MP

The Minister's talk is available in full audio here.

UK Government sees 3 levels of openness to consider:

  • Sharing research findings
  • Securing the data and making it available
  • The role of the public in securing the data through citizen science
  • Open Access to research finding – “the public have a right to access publicly funded research”
  • Willetts prefers Golden Open Access as a model but approves of Green OA as well.
  • Negotiating a national licence for walk in access to libraries
  • Questions whether OA can just be achieved with Green OA in a very short timescale – believes the debates includes longer timescales and costed models
  • Knowledge exchange and impact – how does the research impact on the wider world
  • Struck by the internal tensions in UK Univs are revealed by this policy – for instance Universities wont pay for Gold publishing models for their own academics – suggests disconnect between University and their own academics interests in seeing research published.
  • Open Data – important that the data behind research be openly accessible. Dream must surely be that the incredible database of all the funded research everywhere in advanced Western countries should be available for full use and re-use.
  • Privacy regime needs to be done right – deeper you dig the harder it becomes – are people entitled to a complete opt out of all their medical information for research purposes without individual consent for instance?
  • Security issues – dual use of research, i.e. publishing research which could be used for other purposes such as the H5N1 research possibly being re-purposed for terrorism etc.
  • Usability and capability – modern librarianship is challenged by the size and extent of the data.

Unfunded mandate

During the debate I asked the following question of the panel to gain their insights into the economic issues behind Open Data. I also got in a plug for our Masters in Digital Asset and Media Management at King's to address the skills gap identified during the evening.

We have at present an unfunded mandate – open your data, make it meaningful and keep it open for perpetuity – but the funding models are based on 3-5 year funding cycles often referred to as fund and forget. What fiscal models do the panel suggest should be adopted by Universities and funders to resolve this presently unfunded mandate?

Prof Boulton responded that the creation of Open Data should be an integral part of doing the research in the first place and thus the costs are front loaded into that research. Noted that universities tend to be bad at long term thinking and suggested that we might very well need national infrastructures.

Prof Bird responded that the economic issues are important. For instance in linked data, to keep answers current you have to go back to basics and relink the data – so the storage of the software and programmes to run the updated analysis is vital not just the data.

David Willetts responded that is was an important question, but the fiscal issues were not the vital aspect, rather how we store data is the key. In my opinion he avoided the fiscal issues and did not answer the question.


Updates:

The Foundation for Science and Technology has added their own summary here.


3 comments:

  1. Thanks for this very useful summary, Simon.

    ReplyDelete
  2. This blog post gives details of the debate with links to the resources available from the organisers The Foundation for Science and Technology. It also asks the question: is this an unfunded mandate? English Songs Download

    ReplyDelete
  3. With Open data has the potential to improve the economy, environment and our society, but there's plenty of room for improvement before it is actually able to achieve those changes.

    ReplyDelete