STARDIT and Wikimedia Australia
Last year Wikimedia Australia agreed to support STARDIT, an innovative development in online open knowledge sharing. WMAU will provide hosting and technical support for STARDIT on our servers, alongside backing the ongoing development of the project. Jack Nunn is Director of the not-for-profit education organisation Science for All, and developed the concept for STARDIT from his extensive experience and research on equitable and ethical ways for all people to actively engage in science. In this insightful guest blog post, Jack generously shares his perspective on STARDIT and its envisioned role as a trusted entity in the open data ecosystem.
What on Earth is going on?
What on Earth is going on? It’s a common question, and the truth is that often, no one knows.
As a species, I believe we can do better at collecting, reporting and learning from our collective human actions.
Standardised Data on Initiatives (STARDIT) is an attempt to build a way of answering this question, with anyone on Earth able to access, contribute, edit or verify information about collective human actions (known as ‘initiatives’). The basic idea is that there should be a standard way to describe human actions, and any consequences, in a way that works across languages and cultures. Currently, we don’t even have a standardised way to spell standardized in English.
How can you use STARDIT?
Think of STARDIT as a way of creating a Wikipedia page about something that wouldn’t normally have an entry. This could be anything from data about a clinical trial, environmental research, a community arts project, a car, or the very device you are reading this on.
STARDIT provides a way for anyone to collaborate on describing collective action. Crucially, it can be used across different areas of human knowledge, from health research, environmental research and education, to government policies, manufacturing, or the arts. It can report data on who did which tasks, where resources came from and any impacts or outcomes. STARDIT enables multiple categories of data to be reported in a standardised way across disciplines and languages. Because the data is organised into common shared data fields across disciplines, it makes it easier to compare initiatives in a way which has not been possible in the past.
This can enable well-founded comparisons of the effectiveness of different methods, including the most effective methods of involving different stakeholders. In this way, we hope STARDIT can support participatory ways of working and help improve the equity and quality of initiatives worldwide.
“All major problems, including complex global problems such as air pollution and pandemics, require reliable data sharing between disciplines in order to respond effectively. Such problems require evidence-informed collaborative methods, multidisciplinary research and interventions in which the people who are affected are involved in every stage” Taken from the peer-reviewed article ‘Standardised data on initiatives—STARDIT: Beta version’
If we’re serious about preventing irreversible climate change, pausing mass extinction, or preventing sea levels rising any higher than already projected - we need evidence informed action. The solutions will involve research, government policy, education interventions, manufacturing and the arts. People working in each one of these disciplines need to be able to communicate with each other, understand who did what, and understand any reported impacts or outcomes. For example, educating women and girls has been highlighted as one of the most effective ways of preventing irreversible climate change [Ref]. Imagine the complexity of trying to report how education has led to the outcome of reduced climate change across languages and disciplines. STARDIT can be used to report data in all these areas, and can be used as a tool to help us understand what the most effective methods are for education in different settings, and any impacts.
How does STARDIT work?
STARDIT uses Wikidata as one of the ways of structuring the data. Wikidata is ‘a free and open knowledge base that can be read and edited by both humans and machines’. By using Wikidata, STARDIT works across human languages for both humans and machines. By working across languages, free, open access, editable by anyone, STARDIT provides standardised, verified information which can be read by both humans and machines. By ensuring all the data is open access, we can help ensure more people can be involved in sharing knowledge and learning.
So why was STARDIT built, and why is it urgent to build on this project?
Trust in information underpins human civilisation. During the public consultation on STARDIT, one person said ‘most of our decisions are based on trust’. Millions of people around the world are trusted to grow and make our food, manufacture our medicines, clothes and other products. Millions are involved in research, data sharing, government policy, education and environmental management. We trust that these initiatives, or data about them, are of a certain quality. But often it is blind trust. We have limited ways of understanding where things have come from, how they were made, how decisions were made, how resources were sourced, if the people involved were enslaved, how humans and other animals involved were treated, and any environmental impacts. We often don’t have access to this raw data to help us make informed decisions, and to help us trust it is of ‘good quality’, or aligns with our own values.
In addition, we also know that there are often powerful people with vested interests funding misinformation campaigns which contradict everything that the data and evidence tells us we should be doing, if we want to prevent things such as irreversible climate change and mass extinction. Similarly, in health research, industry funded research is more likely to have outcomes favoring those with financial interests who are sponsoring the research. While it's easy to attack or blame individuals or even organizations and countries for sharing incorrect or false data, we all need to be involved in improving access to reliable data; to help uphold the United Nations human right to “seek, receive and impart information and ideas through any media and regardless of frontiers”.
"Information warfare" presents an even more alarming reality in an age of increasingly accessible machine learning and AI. Divide and conquer is an old tactic for a reason, and we need to empower people to connect and share reliable data across the world.
What if humanity could unite around shared values which are codified in multiple languages? What if we had open data about all data, products, media and policies? What if we had evidence-informed methods of achieving the United Nation’s Sustainable Development Goals, which everyone can understand and act on, regardless of location, income or spoken language?
In a world where machine-generated content is getting more sophisticated, and harder for most people to identify, it is essential that we humans have tools to be able to collectively share information about who (or what) was involved in the creation of information such as media (including videos), algorithms, policies, medical guidelines, molecules (such as drugs) and larger objects (like cars, airplanes and satellites).
We can only do this by working together as a species. We can only work together by understanding what on Earth is going on, and what, collectively, we think we should do.
There will be no right or wrong answers, but with STARDIT, the intention is that at least there will be data. And data about the data, or the lack of data. This data can help us make more wise decisions, and be used by the self-correcting lens of the scientific method to save us from the greatest threat facing humanity and life on Earth, ourselves.
It was five years ago that Science for All started hosting the co-design process for STARDIT, and in that time we’ve co-created Alpha and Beta versions with over 100 people around the world, published peer-reviewed articles about the project and had the project endorsed by Australian Genomics. Having Wikimedia Australia as a partner organization is a wonderful evolution for the project.
By working with Wikimedia Australia and others to build Version One of STARDIT, we hope this will itself build trust in this project, and help realise some of the more ambitious goals of the project.
Anyone in the world can get involved in this project. The next step is to involve as many people as we can - as equitably and inclusively as possible.
To help do this, we are actively seeking funding from sources which align with the co-created values of the project. At the moment this project relies on volunteers.
I hope that the values and aims of the STARDIT project resonate with some of you reading this.
You can learn more about the project and get involved at ScienceForAll.World/STARDIT
As well as reading, editing and creating STARDIT reports, you can get involved in the following ways:
- Join our online community which is used to make decisions about the project transparently and collaboratively
- Volunteer your time to help code, and use machine learning to create reports, or manually create a report
- Get involved in any aspect of the project including governance and co-authoring publications
- Become a partner organisation and publicly sign-up as a supporter or partner of the STARDIT project
Conclusion from WMAU
In closing, Wikimedia Australia is excited to support STARDIT's journey, a groundbreaking fusion of innovation and transparency in the open-knowledge landscape. We encourage anyone interested in open data to connect with Jack at Science for All and STARDIT. The importance of partnerships in building a culture of openness, transparency and accessibility is absolutely crucial. Through our collaboration we’ll continue to advocate for fair access to knowledge and through practical initiatives such as STARDIT, contribute to a more informed and digital future.