To save our planet, we need to work together

Posted at — Oct 15, 2020

Even before you go to university, we are told to pick a single field of study. While fields are beneficial for ease of discussion, they limit our view of what we can do with our knowledge and skills. A good example of rejecting these boundaries is Cambridge, at which you do not study undergraduate Physics or Biology, you study Natural Sciences. This is because there are no hard boundaries between the sciences. Physics effects Chemistry which effects Biology. Similarly, there aren’t hard boundaries between studying materials in Engineering versus Chemistry — it is arbitrary where we draw the lines. By default our thinking is interdisciplinary, but we box our skills into distinct fields.

Why are most degrees aimed at a single subject, if it is not beneficial for research? Because it is beneficial to the job market looking for new graduates. It helps companies to know that incoming grads will have a certain knowledge base. They can then tailor their onboarding to the graduates without having to know each employee individually. However, having cookie cutter grads with the same malleable expertise, does not produce people with the diverse understandings needed to tackle global issues like the Climate Emergency. For such a problem, having the same knowledge base is a detriment to generating ideas outside of that framework. As people to work together across diverse fields, they will apply their knowledge and skills in ways that haven’t been imagined yet. We need people that will bring their knowledge of how ants work together to Computer Science; we need people that will bring their knowledge of weathering houses to Geography. Radical knowledge sharing enables us to actually think outside of the box.

Promoting interdisciplinary research encourages life-long learning. We will always need people with deep theoretical knowledge of their area but learning in different areas does not sacrifice knowledge in another. The human mind has no capacity and, in fact, it seems that diverse ideas strengthen knowledge overall. Case in point, a rising trend in knowledge management systems is the idea of a Zettelkasten¹. This is a wiki-like system in which every note is linked to another. This linking help develops deeply interconnected thought which can surface links that otherwise would be hidden. And the more nodes in the web, the more connections that can be made.

A network diagram of interconnected notes — An example digital Zettelkasten showing the interconnection of notes. Each node in the graph is an idea, and over time the network of ideas becomes more and more complex. Source

On the other side, interdisciplinary collaboration encourages life-long teaching. Teaching is one of the best ways to ensure your knowledge is complete and to challenge it — questions that your colleagues ask may unearth ideas that you haven’t yet considered. For instance, Feynman tells a story² of being invited to a philosophy debate about “essential objects”. They explain the concept and ask Feynman, a physicist, what he thinks of it. He simply asks “Is a brick an essential object?” intending to then ask if the inside of a brick is an essential object. Instead of an answer, this sparked a debate that lasted to the end of the session. Often it is simple manipulations or questions that can bring great advances in the subject. Another example is in artificial intelligence, in which it took 19 years to develop multi-level neural networks because no one thought to stack the layers. In short, there may be obvious things that have yet to be considered that can be unearthed through working with researchers in other areas.

A diagram of a multi-level neural network — A multi-level neural network can solve entire new classes of problems that were previously out of reach. Source

Recontextualising information you already have will also allow for new perspectives — if your mental model is different, you will see different solutions. For example, biological methods to process genetic codes came directly from sub-string algorithms which are common knowledge to computer scientists. Someone from a maths background however may see this is constraint solving problem. On the other hand, it is possible that two researchers are working on the same problem from different directions which can then join forces. To illustrate, Chomsky was working on language classification while Schützenberger was working on grammars in formal language theory. They realised that they were both creating a hierarchy of languages, and together they created the Chomsky hierarchy³. While either could have discovered it on their own, the collaboration doubtlessly helped validate and strengthen this foundational idea.

Sharing knowledge shouldn’t just be between rich, white, able-bodied scientists. We need to open the gates and allow research from all areas and people. For example, Wanda Díaz-Merced lost her sight and thought her career in astrophysics was over. However, she pioneered sonification which turns large data sets into sound and allows for previously unseen trends to be heard instead. She has since been crowned one of the seven trailblazing women in science by the BBC⁴. Similarly, colonisers must respect the knowledge and oral traditions of the indigenous people of the land they are occupying. For example, indigenous people were caring for the Californian land practising groundfires which prevent build-up of flammable materials. However, this practise was ended when the land was stolen and occupied by the American settlers. Consequently, the area has uncontrollable wildfires⁵. By respecting indigenous knowledge and allowing sovereign nations to control their land, we can help protect the biodiversity of the earth. These examples show us that allowing more people to contribute to science, whether in data collection, analysis or otherwise, allows us to contribute much more than if we only allow researchers with PhDs and Master’s degrees. Breaking down these institutional barriers and creating space for all, goes hand in hand with sharing knowledge between established faculty.

Another example of the effect of volunteers over experts is the website Nupedia, launched in 2000. The concept was an open online encyclopedia which could only be edited by subject matter experts. Editing was slow though, and Wikipedia was created in 2001 as a feeder project to provide a skeleton for Nupedia. The rest is history, Wikipedia is the most successful online collaborative project of all time and has article quality on par with expert written encyclopaedias⁶. How is this possible with no academic bar to clear, no money motive, and with anonymous contributions? The answer is simply guidelines, peer review and consensus. These work together to ensure that the encyclopedia improves overtime and reacts to new knowledge. By emphasising a neutral point of view and citations, edits generally escape bias that would derail similar projects. Finally, at the beginning of the Coronavirus pandemic, Wikipedia was the watering well at which everyone, from scientists to everyday people, could find the most up to date information from all countries. Imagine if we could apply this effect to the Climate Emergency?

In this vein, advocating and contributing to open knowledge is some of the most important work that you can do. Contributing cutting edge research to Wikipedia allows more people to view this information than any other venue. Or, breaking down vocab and concepts in a blog post of “X field for Yists” creates a snowball effect, allowing other people to learn faster and better than you did. It’s not necessary that you are an expert, someone will be thankful for your contributions. Furthermore, this is a great way to develop your science communication skills. Being able to explain complex ideas is a key skill necessary for developing an educated population that is able to understand and help tackle the Climate Emergency.

If we don’t share our knowledge and instead compete, we will produce countless over-lapping and duplicate trials, finding the same dead-ends that others before us have run into. It is only through open publication of all successful and unsuccessful trials that we can avoid this and find where the gaps in the research is. And open publication requires that data, models, code, papers, etc are all open. This can be through well maintained Jupyter notebooks⁷ or by making code available in public repositories and publishing open preprints. Alternatively, the journals have the power to waive access costs, as they did for the Coronavirus pandemic. When will the climate emergency be urgent enough for similar open access policies?

In competitions, like the free market, there are winners and losers. However, we can’t afford to have losers; we can’t afford to save one country or one hemisphere. Our planet knows no borders and our skills know no field. To tackle the Climate Emergency, we must unite so that we all have a chance of winning.

Further Reading:

https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1004214

References: