Select Page
What is good data governance?

What is good data governance?

Governance is a hashtag that has been growing in popularity.  So it’s great to see the coop sector responding with the BCCM’s impending release of Co-operative and Mutual Enterprise Governance Principles for the Australian co-op and mutual sector. And folk like Coop News dedicating a Governance Edition to analyse the subject.

We’re interested in how this emerging thinking applies to data.

Data governance is the mechanism through which values and expectations with respect to data can be translated into effective management practices.

Good governance is a function of the clarity of shared intent and trust in expected behavior. As Dee Hock writes: “This is not to say that contracts, laws, and regulations do not serve a purpose. Rather it is to point out that…rules and regulations, laws and contracts, can never replace clarity of shared purpose and clear, deeply held principles about conduct in pursuit of that purpose.”

By these definitions, good data governance is broader than legal frameworks or codes of practice – it bridges the gap between the expectations and values that folk have with respect to their data, and the mechanisms by which these are made tangible. It encompasses legal documentation and decision-making protocols that are agreed between the actors involved. And it includes the social interactions that enable these to be made real.

The Cambridge Analytics seraglio exposed the under-belly of the market that preys on our data. And the GDPR has done wonders for focussing attention on data management practices. Next step is to get some clarity around the governance practices that will lay the foundation for better matching of expectations with practice.

PS – An example of a emerging model for personal data, Sovrin have a pretty comprehensive Trust Framework

 


 

My non-fungible token is my home

Tokens are migrating to the mainstream. CryptoKitties has demonstrated how the blockchain can extend beyond money into real-world ‘things’. it doesn’t sound like a big leap to enable me to own my own digital cat, but it has the potential to be a whole lot more useful than digital currencies alone.

With CryptoKitties, I can have a cat and that cat belongs to me. If the developers stop supporting the front-end that enables me to look after my cat, it doesn’t change the immutable fact that the cat exists and is mine. I can use someone else’s front-end and my cat will waiting for me. Even better other developers can build other ways of interacting with my cat – my cat can have a hat and that hat belongs to the cat and not me – and the original developers can’t stop this from happening.

This conversation with one of the creators of CryptoKitties Dieter Shirley is good – he is a lucid thinker. Really liked the way he described the potential future for ‘non-fungible tokens’ (and his self-confessed struggle with coming up for a new word for ‘thing’ which must be one of the oldest word in any language).

Once we have things that can exist in their own right on a blockchain, then we have the basis for a whole heap of applications. Dieter talks about something that you earn that contributes to your status – for example, only people who attended a Kanye concert can get a certain token. Collect 10 of these tokens and you graduate to a gold token. This mechanism can send a strong social signal – it’s a form a digital self-expression the way our vinyl record collection used to be. This is exactly the type of mechanism that we are hunting for in regards to ‘reward-for-effort’ tokens in the cooperative space.

The biggest hurdles to mass market solutions are solving the currently very high technical barriers around on-boarding and know-your-customer compliance. And beyond that the scaling problems with ethereum. As Dieter explains ‘we’re running the network on the equivalent of an Apple IIe’… Ah, brings back memories, I loved my IIe.

A coop’s competitive advantage

A coop’s competitive advantage

We’ve been using this diagram to help people understand where coops fit in the world as we know it. It neatly describes how coops slip in between your run-of-the-mill-profit-driven-firm and not-for-profits.

There is often a light-bulb moment around what this is really describing – that there are three different ways that value can be distributed:

  1. In a for-profit company, return-on-shareholders-capital is the sole focus
  2. In a not-for-profit, the division of value is determined by its pre-ordained mandate
  3. In a cooperative, it’s the members who decide how value is distributed

It also helps to makes sense of why coops are democratic organisations, where every member gets one vote, as opposed to for-profits where decisions are based on proportionate holdings of voting shares, and NFP’s where the beneficiaries typically have very little control.

We particularly like the implication that coops take the best out of both worlds!

 

Tech and design workers making moves

Tech and design workers making moves

A surprisingly well attended meeting at the Trades Hall last night given the low key publicity. Developers, designers and folk from 5 different trade unions were there. Interesting perspective on the potential power of developers to influence the shape of new tech to meet privacy-by-design and other yet-to-be-defined principles. Only one developer raising the blockchain flag as a solution in waiting. But perhaps most encouraging, the trade unions showed real willingness to engage on the issue and explore how things like platform coops can help change the balance in favour of workers from all industries – particularly designers.

When tech meets organised labour then the pendulum starts its swing back from current economic extremes.

A meeting for workers interested in collective action & unionism

We believe in workers’ rights, social justice, diversity and equality. We want to challenge corporate control over our technology.

We share a vision for an inclusive & equitable technology industry. We want to collaborate with workers and friends to build tech worker power, organise on workplace issues and create a space for educating ourselves and exchanging ideas.

Technology for the many, not the few!

 


 

When algorithms rule the world

When algorithms rule the world

 

When will homo sapiens be usurped at the top of the food chain? Or has it already happened?

The transition from big data to algorithm-driven AI is happening frightening quickly. In fields from medicine to movies to mass surveillance, smart machines have already proven to be quicker and more reliable than their human peers. It won’t be long before they know us better than we know ourselves…

The need to align data management with our collective and individual interests is urgent. Left to their own devices, government and private capital are ill-equipped to balance the common good with individual agency. We need to take ownership. Co-operative governance and ownership of the technologies that underpin the algorithm economy are key…

 


 

The pros & cons of ratings

We’ve been studying the various ways accreditation can assist farmers in the migration to smart machines and edge computing.

One of the barriers to the adoption of data sharing across the agricultural supply chain is the relative immaturity of the technology that is used on-farm. While there is currently an explosion of new devices being developed and offered to farmers, the inchoate mess is difficult for farmers to navigate – it’s simply hard to know where to start. On the other hand, existing systems are ageing and are not generally fit for API integrations while the reliability of sensors can be patchy leading to questions about data accuracy. In short, data lacks the reliability, accessibility and inter-operability to be really useful.

Accreditation offers one way we can create a bridge between the current environment and the next, where smart farming is commonplace. The idea is that better farming practices will be a key driver towards the adoption of smart machines – whether it is to promote precision farming or to support growing consumer awareness around the foods they consume. And accreditation around those farming practices may provide a way to solve the data quality issue. In effect, we can look to leverage the business processes of the various certification agents to filter data quality.

Accreditation has the potential to be quite diverse. It can include certification around organics. It can include verification of farming practices, application of fertilisers or chemicals in water runoff. It could look at measures of biodiversity on the farm, regeneration practices, and covenants on land preservation. By attaching independent third party accreditation to farm produce, this data becomes valuable in a way that can’t be achieved in the absence of deep integrations between systems.

Of course, the risk is that accreditation agencies can be compromised if their interests are not aligned correctly. We’ve seen this time and again in financial markets. If the gamekeeper is getting paid by the poachers, then they are only too willing to see things from the poacher’s perspective. Either the measures need to be so blatantly objective that they cannot be compromised or we need to ensure that the accreditation providers are not financially beholden to those that could benefit from misplaced confidence.

So while we stumble headlong towards a hyper-connected farm to table, we can perhaps begin to capture the benefits of data through accreditation…

 


 

The power of co-operatives

 

“The leader is driven by a desire for power, while the organiser is driven by a desire to create.”

Coops attract organisers as the structure naturally diffuses the power of any single person. Of course the risk is entropy where coordinated action dissolves into a cacophony of voices. The aim of a coop is then to cultivate an active membership which focuses energy towards a common goal – following the adage that ‘power is focussed energy – that enables the ability to do a thing’…

 


 

Not Godwin’s law

 

We were brain-storming ideas for injecting some energy into the 888 Brewery Coop. It was on the long drove back from Horsham, the sun had gone down and we were listening to Radio Birdman descend into the maelstrom… and within mere minutes the discussion had gone from community engagement ideas to issuing ICO’s backed by kegs of beer.

So if Godwin’s law states that the probability of comparisons to Hitler rise to 100% the longer an online debate continues (which a study of Reddit discussions seems to confirm), perhaps crypto-currencies offer a greed-driven equivalent. Something like:

There are no more than six-degrees of separation between a good business idea and a get-rich-quick-crypto-currency-backed-ponzi-scheme.

 


 

Co-operatively growing farmer data

 

Very productive day with the good folk from Birchip Cropping Group yesterday. It was our second workshop in the Farming Together sponsored project being undertaken by BCCM to help grain growers from the Wimmera-Mallee explore a data coop.

In a new land speed record, by the end of the workshop we had crossed everything off our wish-list in terms of things we needed to draw up a draft set of Co-operative Rules – once again proving that farmers and nurses are the most likely to survive the zombie apocalypse.

The data coop can be a big concept, and there are some key pieces of the puzzle that are still formative in their shape and readiness, so it was really encouraging that the farmers present were beginning to have concrete ideas about how it may work. There were some cracking specific examples that are likely to prove fertile ground for testing of the data coop mechanics and economics.

Perhaps more encouraging still was that we got the sense that the farmers were starting to take formal ownership of the project, thinking that they could build it from the ground up (rather than having a structure imposed on them). Notably, it was suggested that the data coop could leverage the BCG infrastructure to provide a springboard to get started.

Also, really interesting discussion about farming practices, quality assurance and compliance. Noting that the thinking to date has been that precision agriculture would lead the use cases for the data coop, we discussed the rising cost of compliance and increasing accuracy of testing and standards and whether this pushed ‘traceability’ higher up the priorities.

We were also lucky to have Leanne Wiseman and Jay Sanderson from Griffith University so we could untangle some of the trickier questions around rights to data. We sketched out some of the key structural elements that could be developed for the coop – such as members agreeing to an “Ethical Good Governance Code” that would be updated every year and would require things like disclosure of conflicts of interest, disclosure of 3rd party data sharing arrangements and agreement to minimum standards around data being shared, accuracy and types.

Finally, the emerging view seemed to be that the data coop would be akin to social infrastructure and would be best structured as a non-distributing coop. This would better align member and community interests and also enhance the collaborative business model with customers.

 


 

Why haystacks need needles

 

One of the challenges of a co-operative is balancing the interests of the group with those of each individual member. Unlike a proprietary limited company, stakeholder interests are not beholden to the profit motive. Rather the primary focus is on return to members – which can be delivered in a variety of ways including via dividends, rebates and patronage. The difficulty is that without a return on capital signal, the cooperative itself has to determine how value is distributed among its members.

A data co-operative provides a simple illustration of this challenge. On the one hand, members come together to aggregate market power – as individuals they are weakly positioned to bargain the terms on which they share their data, while as a collective their individual control is amplified. On the other, one of the principal objectives of a data coop is to enable individual members to have more control over their data. For this reason, a data coop will create permissioning mechanisms that enable the individual to differentially share their data with others. The data coop leverages group market power to deliver value for individual members.

But that is only part of the story. By aggregating control over data, the data coop will also enable sharing of data in aggregate. And there are likely to be circumstances where it is in the collective’s interests to share aggregate data. This is after all why Big Data and the Algorithms have been dominating the pop charts over recent years. For this reason, it may be a condition of membership that coop members agree to authorising the coop to share their data. You can see the challenge here…

For this reason, it is really important that cooperatives are very clear about how a cooperative distributes value amongst its members. If interests are not aligned, fractures across the points of difference are likely to occur when the structure is stressed.