I’ve had the exceptionally good fortune of working with a considerable number of extremely talented, tenacious, and insightful scientists over the years. One of those was Julian Stirling, whose PhD I ostensibly supervised. (In reality, Julian spent quite some time supervising me.) Julian is now a postdoctoral researcher at the University of Bath and is involved in a number of exciting projects there (and elsewhere), including that he describes in the guest post below. Over to you Julian…
Universities love spin-offs — they show that research has had impact! — but does the tax payer or the scientific community get good value for money? More importantly, does spinning off help or hurt the research? I fall strongly on the side of arguing that it hurts. Perhaps I am ideologically driven in my support for openness, but when it comes to building scientific instruments I think I have a strong case.
Imagine a scientist has a great idea for a new instrument. It takes three years to build it, and the results are amazing; it revolutionises the field. The scientist will be encouraged by funding bodies to make the research open. Alongside the flashy science papers will probably be a pretty dry paper on the concept of the instrument; these will be openly published. However, there will be no technical drawings, no control software, no warnings to “Never assemble X before Y or all your data will be wrong and you will only find out 3 months later!“. The university and funding agencies will want all of this key information to be held as intellectual property by a spin-off company. This company will then sell instruments to scientists (many funded by the same source that paid for the development).
The real problem comes when two more scientists both have great new ideas which require a sightly modified version of the instrument. Unfortunately, as the plans are not available, both their groups must spend 2-3 years reinventing the wheel for their own design just so they can add a new feature. Inevitably both new instruments get spun off. Very soon, the tax payer has paid for the instrument to be developed three times; a huge amount of time has been put into duplicating effort. And, very probably, the spin-off companies will get into legal battles over intellectual property. This pushes the price of the instruments up as their lawyers get rich. I have ranted about this so many times there is even a cartoon of my rant…
We live in a time when governments are requiring scientific publications to be open access. We live in a world where open source software is so stable and powerful it runs most web-servers, most phones, and all 500 of the worlds fastest supercomputers. Why can’t science hardware be open too? There is a growing movement to do just that, but it is somewhat hampered by people conflating open source hardware and low-cost hardware. If science is going to progress, we should share as much knowledge as possible.
In January 2018 I was very lucky to get a post-doctoral position working on open source hardware at the University of Bath. I became part of the OpenFlexure Microscope project, an open-source laboratory-grade motorised 3D-printed microscope. What most people don’t realise about microscopes is that the majority of the design work goes into working out how to precisely position a sample so you can find and focus on the interesting parts. The OpenFlexure microscope is lower cost than most microscopes due to 3D printing, but this has not been done by just 3D printing the same shapes you would normally machine from metal. That would produce an awful microscope. Instead, the main microscope stage is one single complex piece that only a 3D printer could make. Rather than sliding fine-ground metal components, the flexibility of plastic is used to create a number of flexure hinges. The result is a high performance microscope which is undergoing trials for malaria diagnosis in Tanzania.
But what about production? A key benefit of the microscope being open is that local companies in regions that desperately need more microscopes can build them for their communities. This creates local industry and lowers initial costs, but, most importantly, it guarantees that local engineers can fix the equipment. Time and time again well-meaning groups send expensive scientific equipment into low resource settings with no consideration of how it performs in those conditions nor any plans for how it can be fixed when problems do arise. For these reasons the research project has a Tanzanian partner, STICLab, who are building (and will soon be selling) microscopes in Tanzania. We hope that other companies in other locations will start to do the same.
The research project had plans to support distributed manufacturing abroad. But what if people in the UK want a microscope? They can always build their own — but this requires time, effort, and a 3D printer. For this reason, Richard Bowman (the creator of OpenFlexure Microscope) and I started our own company, OpenFlexure Industries, to distribute microscopes. Technically, it is not a spin-off as it owns no intellectual property. We hope to show that scientific instruments can be distributed by successful businesses, while the entire project remains open.
People ask me “How do you stop another company undercutting you and selling them for less?” The answer is: we don’t. We want people to have microscopes, if someone undercuts us we achieved this goal. The taxpayer rented Richard’s brain when they gave him the funding to develop the microscope, and now everyone owns the design.
The company is only a month old, but we are happy to have been nominated for a Great West Business Award. If you support the cause of open source hardware and distributed manufacturing we would love your vote.