Day 3: Long-term ethical, legal and policy changes needed to move from the current scenario to the desired scenarios

Hello Everyone

Welcome to Day 3 of this E-Consultation on the ethical, legal and policy aspects of data sharing affecting farmers. I am about 10 hours ahead of many of you so I thought I would being our discussion a little earlier with a few thoughts.

My name is Leanne Wiseman  I am also excited to accept the challenge of moderating  Day 3 of our discussion. This discussion will begin on Wednesday the 6th June.

Today we would like to focus on the  Long-term ethical, legal and policy changes needed to move from the current scenario to the desired scenarios.

We have been provided with some guidance of some points for our discussion so I look forward to hearing your thoughts and suggestions as to how we address some long terms goals.

• Ethical: what are the ethical questions we should make to contribute to increase food and nutrition security and better livelihoods of the poorest and most vulnerable farmers around the world by harnessing their access to and use of agricultural data? How should our mindset change in the long term about data rights? Should data rights be recognized as human rights? Should we still consider receiving a service in exchange for surrendering data an ethically acceptable business model?

The issue of whether data rights should be recognized as human rights is an interesting notion. Certainly the introduction and impact of the GDPR in the EU and also the soon to be introduced Consumer Data Right (CDR) (see http://dataavailability.pmc.gov.au/consumer-data-right) that will be introduced into Australia, sector by sector, in the near future  reinforces the notion that individuals should have the right to data that is generated about them or by them. 

There is a useful discussion here of data as a human right http://www.ohchr.org/Documents/Issues/HRIndicators/GuidanceNoteonApproachtoData.pdf

• Legal: what do you think are the best legal approaches to achieve the desired scenarios? How to recognize and implement farmers’ intellectual property rights over their data and traditional knowledge? How should the licensing scenario change?

When thinking about the best legal approaches to solve problems,  the law can be a very blunt and inflexible tool when trying to encourage behaviour change. It is often 'soft' law or policy driven changes that are more influential in influencing and shaping behaviour. How can the current data licensing practices be changed to ensure a more equitable outcome for famers? How best can traditional knowledge, particularly in relation to farmining practices be safeguarded from exploitation?  Empowering our farmers and their communities with the tools to realise the value of their information is one positive step. How can this be done?  Should not the onus of changing the current licensing practices come from regulators who can shape the conduct of multi-nationals by requiring more transparency and openess about their data management practices?

While we have seen the adoption of data principles and codes of conduct around data management, which have resulted in the raising of awareness of the issues of data sharing and the potential for misuse, there has been little  uptake by multinationals. How does 'best practice' of data management and use be embedded into businesses' social conscience?

• Policy: what strategies and priority areas of intervention would you recommend at the policy level? Do you think there should be more governance and by whom? Do you think international agreements could help?

 We have seen the positive impact of international agreements such as the Nagoya Protocol on Access to Genetic Resources and the Fair and Equitable Sharing of Benefits Arising from their Utilization to the Convention on Biological Diversity (CBD) so there is clear evidence that international treaties can make a positive change to behaviour.

An interesting question is whether agricultural data/ farm data is sufficiently meritorious of a new independant treaty or whether the international shift, such as the GDPR etc around the need for better data managemetn practices around consumer data are sufficient to provide coverage for  the issues arising from the increasing collection, aggregation and dissemination of agricultural data? Some would argue that agronomic and agricultural data has a significantly different value proposition to that of consumer data ? others disagree ?

Looking forward to your contributions to what long-term ethical, legal and policy changes that need to happen

A strategic insertion of digitized agriculture data can best be explained, perhaps, by way of example specific to nutrient management.  Nutrient management is extremely complicated to implement in the best of conditions given the rate and timing of nutrient application, its placement as well as the measurement and assessment.  These data points also need to be algorythmically evaluated together with other variables like soil condition and weather.  It seems undeniable to many that leveraging digitized and artificially-intelligence enabled supply chain data flows from sensors and satellites and more, on behalf of the very best nutrient management practices, will have the largest most immediate impact on lives of farmers due to higher quantity and quality of production.  A key challenge is that in order to reach the fullest potential for nutrient management the agriculture sector must become more aware of the technology, business model and economic dynamics of blockchain, sensors, UAV's,  wires, satellites, controllers, monitors, computer programs and more.  How do we respond to this key challenge?  I just read a wonderful report by FAO on unmanned aerial vehicles (UAV's) with multiple case studies on their use in rice and other crops.   

Meanwhile, there is rapidly emerging the formation of cross sectoral partnerships to deliver agriculture digital solutions.  It includes agribusinesses, mobile network operators, financial institutions, civil society, etc.  From Ghana to Kenya and points in between agribusinesses, MNOs, donors and others have partnered and are continuing to partner on multiple initiatives.  A key challenge here is how can we begin to codify a strategic approach to the formation of agriculture digital solution alliances?  This codification can be best served by first having this discussion about ethics, legality and policy.  

To the ethics question for today:  There is signficant investment by global, regional and local agribusiness and other private sector players to deliver digital products and services to farmers.  What is missing from their investment decisions is any real farmer perspective or voice.  This discussion we are having is vitally important so that we can consider the ethics, legality and policies in order to help the farmers share their voice with the supply chain from their cell phone.   Key ethical considerations include data ownership and data privacy.  I recently wrote about the General Data Protection Regulation (GDPR) and its contrast to blockchain technology here:

https://medium.com/@leebabcock/digital-version-2-0-for-poor-farmers-1d97b8384a27 

To the legal and policy questions for today:  What kind of policies do we need?  An analogy is the European 'principles-based' International Accounting Standards (IAS) compared to the U.S. 'rules-based' Generally Accepted Accounting Standards (GAAP).  This is like comparing 15 pages of IAS 'policy' guidance to 5000 pages of GAAP 'policy' guidance.  When we think about policies and standards for agriculture data are we thinking about 'principles' or 'rules' or both?   When we kinda have an eye on what type of policies and standards this will best inform our thinking about legal issues.  

Since we're supposed to think of long-term changes to move towards the desired scenarios, for those who didn't follow the discussion yesterday here is a recap of some common points in scenarios envisioned by participants (you can also look at the more detailed summaries by Manuel Ruiz yesterday, from which I borrowed something, thanks Manuel!).

Many of Leanne's suggestions can be usefully discussed in relation to these scenarios.

1. There will be cooperation /collaboration not competition, trust, free flows and exchange of data: several participants envisioned an ecosystem where different actors share data for mutual benefit

2. Common data platforms, data shared in the cloud, "organized data communities": all farmers' data shared with other farmers but also platforms where all necessary data from outside the farm is shared (weather, financial, market...); farmers organized and registered in a platform that brings in also big service providers; some participants envisage centralized platforms.

3. Organizational aspects: associations / cooperatives / "virtual aggregations" allowing for larger scale planning of farmland and cropping cycles, more efficient access and use of farm inputs, more precise forecast, better marketing. 

4. Specificity: data, products and processes for data driven agriculture must be context specific, designed to suit the conditions of farmers; weather data will be broadcast in a language understood by farmers, market data will be available via messaging updates and through tradition media; technologies co-developed with farmers; traditional knowledge recognized

5. "Success": improving not only production but contributing to overall welfare and and improved livelihoods; success interpreted in a social way (environment, food security, health, equality, transparency) or from the point of view of the farm as a business?

6. Related to point 2, data protection and transparency: data shared with appropriate licenses; through blockchain technology for maximum transparency.

7. Envisioned roles of actors: government ("our government will have repositories for farmers data and data related services available all the time"), private sector (providing technologies and services, using farmers' shared data with no unfair advantage, but still with the comparative advantage that comes with better processing and application of the data), intermediaries (enabling smallholder farmers to structure, organise and make use of any data that might be available), farmers organizations (aggregating farmers' data, negotiating data access on their behalf). Everything to the benefit of the farmer.

Dear all,

Sorry I was not able to participate in the first two days of the consultation due to an overbuderden agenda. I have read the contributions of day 1 and 2 and may in the course of this discussion  come back to some of the issues discussed previously. Let me start by stating my interests in this topic:

1. I am big data focal point at CIMMYT

2. I chair our Internal Research Ethics Committee

3. I coordinate the community of practice on socio-economic data, which is part of the CGIAR Platform for Big data in Agriculture

4. I lead the foresight ex-ante impact assessment and targeting research at CIMMYT, hence a keen interest in all things future.

Prior to the 20th century, scientists guarded their findings meticulously and only expanded on them after they were many steps further and considered their previous scientific discoveries less valuable. Today scientists rush to publish their findings to be the first. While till recently research data was guarded by scientists, it is becoming common practice in many settings to share data sooner than later. I don't think anybody has any qualms about bringing their old newspapers to a paper recycler, knowing that the recycler is making some money. So let's not get too hung up on data sharing.

Having said that, data that contains personally identifiable information is a different matter. Its use may pose risk to the subject of that data. It is therefore useful to distinguish between data with and without identifiers. Note that high resolution satellite images fall under the heading of personally identifiable information.   

The main question at stake here is farmer level data and how that can benefit those farmers, especially the poor farmers in resource scarce environments who have not been benefiting from global economic development. The promise of data-driven agronomy (not necessarilly precision-ag) holds in store for accelerating productivity for ppor farmers requires some level of data-sharing.  GDPR offers some useful pointers in a direction that it may evolve. 

1. Consent: The data subject should provide unambiguous positive consent regarding the processing of his/her personal data for specific purpose(s).
2. Performance of contract: The data in question should be processed in order to allow/facilitate the performance of a contract that the data subject is part of. Data processing can also be a way for the data subject to initiate a contract agreement.

A service contract between a farmer and a service provider, whereby the service provider must guarantee the privacy and data confidentiality. Unless explicit consent is given the service provider cannot (legally) share that data with third parties.

Hyper ledger technology (HLT, better known as blockchain) is one way of dealing with personal data sharing since the authorization granting rights can stay with the original data provider (farmer)  if a system is set up that way. Think a few generation further than the current best HLT solutions and don't worry about the current drawbacks of, for example, bitcoin blockchain or Ethereum. 

Since technology is still developing and evolving rapidly, any policy such be principle based and not rule based, because rules are likely to be obsolete before they are published.

I'll chip in more later.