top of page
Search

The Revolutionary Microscopes Powering Global Tech Equality with Richard Bowman

Why do we so often underestimate the power of open licensing as a tool for global justice? In this episode, Eleanor Drage speaks with Richard Bowman about how open-source innovation and "frugal tech" can dismantle barriers to scientific discovery and help build a more equitable world.


Their conversation traces Bowman’s journey into developing high-performance scientific hardware rooted in accessibility and the "Right to Repair." He shares how projects like the OpenFlexure Microscope—a 3D-printed, laboratory-grade device—can reshape global health by allowing local communities to manufacture and maintain their own diagnostic tools. From reimagining the role of the researcher to questioning the myth of neutrality in technology, Bowman demonstrates how open hardware operates as both a scientific and political force. The discussion highlights how community-driven design strategies can empower marginalized regions, particularly through partnerships in places like Tanzania, to create spaces for collective innovation and care.


Bowman also reflects on the broader implications of embedding open-source principles into scientific education and practice. By centering transparency, collaboration, and accountability, he points toward more inclusive ways of making and thinking about tech ownership. This episode ultimately reveals how technology—far from being a "black box" controlled by corporations—can serve as a powerful framework for liberation, resistance, and systemic transformation.


Richard Bowman is a researcher, physicist, and advocate for open-source scientific hardware whose work focuses on making high-quality instrumentation accessible to everyone. Through his research and the OpenFlexure project, he explores how open licensing can confront systemic inequities, amplify local expertise, and contribute to building more just and inclusive global systems.


Reading list:


Transcript:

Eleanor Drage (00:53)

So, I'm thrilled to be speaking with you today. We're going to be chatting with Richard Bowman, who is a Royal Society University Research Fellow at the University of Glasgow. I initially got in touch because I had seen something that you had created at a conference called re:publica in Germany, in Berlin, and thought it was fantastic, just the most innovative, interesting idea I'd seen for a long time. So, I'm really happy to be speaking with you.


Can you first tell us what you do and what brings you to the topic of justice and technology?


Richard Bowman (01:35)

Sure. So, I'm a researcher in a physics department, mostly interested in microscopy and in building and automating laboratory instruments. About eight or nine years ago, I got interested in whether you could 3D-print a microscope. That led me down a rabbit hole that is now my main piece of research, which is focused on making a scientific instrument that is meaningfully complicated and useful, but that you can manufacture anywhere. I think that whole endeavor gets more interesting if you release it under an open license that gives the world the freedom to do what they like with it.


Eleanor Drage (02:12)

And we'll talk about that open license in a bit because that's really, really crucial and totally under-talked about in this field. But to start, I'm going to ask you our three "Good Robot" questions: So, what is good technology? Is it even possible? And how can ideas about justice help us work towards it?


Richard Bowman (02:32)

Yeah, those are three big questions. Good technology, that's a big one. I guess the technology itself is perhaps fairly neutral, and what we do with it and how we do it perhaps has more of a moral dimension, at least for me.


Eleanor Drage (02:48)

Do you really believe that? I mean, given how technology is usually created, whether we're extracting lithium from, you know, huge fields that mean that people in local areas can't access water, surely the technology itself, its materiality, how it's built, and the labor that goes into it... all those things are political; they can be extractive.


Richard Bowman (03:08)

In my head, I would file that into how we do it rather than the technology itself. An atom of lithium doesn't exercise moral judgment. Someone who is digging the lithium out of the ground, or indeed running a corporation that is digging the lithium out of the ground, for sure is part of humanity and has to make the moral judgments there.


I think for me, and for the OpenFlexure project, a lot of that has arisen around the question of technology ownership. That comes back to the open licensing question about how we work together with people in very different contexts. For example, between a university in the UK and a small engineering company in Tanzania, there's a very tricky paradigm there. There is a lot of thought needed to get that one right, and it's definitely a work in progress for us, as for everyone else.


Perhaps there are also questions around sustainability and how technology can be used and controlled. Talking through this, I'm seeing there's a recurring theme here, which is control. Our open and distributed structure is very much intended to put control of the technology in the hands of the people who are making and using it. As much as possible, we want to avoid centralizing that control, whether that is avoiding a UK university holding all the keys to the intellectual property or avoiding reliance on a cloud server that you may or may not be able to reach in a rural area. It's a kind of central theme that we really want this technology to be useful to the person holding it and not make them beholden to some external agency.


Eleanor Drage (05:01)

So, tell us more about that then. What happens usually at a UK university? You create a technology or some sort of device, and then the university claims ownership of it and wants to do something with that intellectual property. Just explain that process, and then how you had to come up against the university to do something different and more distributive.


Richard Bowman (05:26)

Sure. So, the default position of a university in the UK, but to be honest, it's fairly global, is that the intellectual property generated by its employees is owned by the university. Now, there's a push in the UK, and again in most of the rest of the world, to release work done in universities in a way that the taxpayers who fund it are able to access it. That's led to the rise of open-access publications, open-access data, and open-access software.


Hardware has lagged here because universities are also tasked with ensuring that their developments make it into the real world. There's a well-trodden path for getting a hardware gizmo out of my lab and into a company, and thus into the hands of paying customers. The way that works is: I come up with a good idea; I disclose that idea to my university's intellectual property team; I then work on developing, implementing, and testing that idea in secret, in confidence. Assuming everything's gone well and I've made a good thing, I then go back to my intellectual property office and try to persuade them to patent my invention.


That then assures me that I will always be recognized as its inventor, and it provides a legal mechanism for the owner, which would be some combination of me and the university, depending on various things, to control how that intellectual property is used. Usually, what it means is that you patent the technology, you then find or create a company to make a thing based on that intellectual property, and then the university will license the patent to that company and enable them to make the product, usually with some financial return to the university.


Although I think if you talk to most universities, they would be fairly comfortable with the statement that this is not done primarily as a money-making exercise. It's done because the university wants to maximize the impact of its work, and they feel that retaining some level of control over their intellectual property helps them to do that.


Eleanor Drage (07:36)

Interesting. It's something that I had to deal with a little bit when I produced this EU AI Act compliance toolkit. It's an open-access tool; it's free. We had the IP people ask us, "You know, what's the… it doesn't belong to you?" We're like, "Not really." You can see them quite unhappy at the other end of the Teams call.


But what is it about having, you know, other universities or other companies across the world being able to use this tool, so it in some way belongs to them too? They can see how it's made; they can take it apart and put it back together; they can fix it themselves. None of this involves having to pay for a warranty, sending it back to Glasgow to get it fixed, and then getting it sent all the way back. What are the benefits of all of those things that come with an open-access or open source tool?


Richard Bowman (08:35)

The OpenFlexure route is quite different from that because we have released everything under an open license. Partly, that was actually made much easier by the fact that there are lots of different people involved. So, this is not just a University of Glasgow project. It started when I was at the University of Cambridge, then I was at the University of Bath, and there was a team there. Now there are people in Bath, Cambridge, Glasgow, and a number who are elsewhere, including a charity that we founded: the Humanitarian Technology Trust.


The fact that there are lots of contributors all contributing to something which is under an open source license means that, actually, nobody anymore has the right to say, "This is ours; we're restricting it." The university only has to say, "We will respect the terms of the license of this thing that is already in the open."


One of the terms of that license is that if we create a derivative work, meaning if we continue development on it, then the derivative work also has to be covered by the same license. I've talked with several projects where they kind of did all of the development in secret, as you would if you were going to patent it, but with the intention that they would release it under an open license at the end. That, I have found, is a much harder conversation to have with the university because, at that point, you're saying, "Look, I've made this shiny, valuable thing and I want to give it away for free."


Now, my hunch is that most universities would still accept the argument that we'll achieve a greater impact by giving it away openly than we would by trying to patent and license it. That's certainly a very clear case for OpenFlexure. But actually, by releasing it a little bit by little, there was never a point in the development where I had a fully-fledged, ready-to-license project that I then decided to give away for free. What I did was release lots of little incremental updates. So, there are lots of advantages to working openly and within a community, but actually, that was one that I never really appreciated until I saw other projects take a different route.


Eleanor Drage (10:48)

So interesting. It's the process that has to be open and all. And this is really important in AI, where things are constantly developing. It's something that I see a lot from friends who work for AI companies or, you know, work on their own to create more open source practices and ways of working to just further the field, you know, in the pursuit of a better way of doing things. I mean, this is an amazing endorsement of the open source process.


It changes the world, you know? And this is why I think the article I wrote in The Guardian got so much traction around "frugal AI." People are so captivated by these ideas of, you know, collaborative technologies and ways of working that can move between different countries and don't make some countries beholden to others, or, you know, don't locate AI infrastructure or tech infrastructure in high-resource, privileged nations and universities, which just further accumulates power. Like, this is an amazing way forward.


So, the next thing I want to ask you is about the internet, because when we think about technology, we tend to think that it's connected to the internet in some way. And this is a fairly recent phenomenon. But what's the problem with this? What's the problem with seeing technologies as "internet devices"?


Richard Bowman (12:13)

Well, I mean, I guess the obvious one is that if you don't have an internet connection, you're suddenly excluded. And it's not even quite as simple as "if you don't have an internet connection." I guess I'm very much in a place where I assume I have internet connectivity 24 hours a day. A 10-minute outage to the internet in one of our practical labs is an absolute disaster; much of the technology that we have basically stops working when your internet connection goes down.


There are many places in the world where you just can't make those assumptions. You may well have internet much of the time, but it's probably on a mobile connection. There are lots of reasons that it might be briefly interrupted. And the problems that we faced were that even if the interruptions were only fairly brief, if you're relying on that connection persisting so that you can spend an hour scanning a microscope slide, then any interruption during that hour is a problem. So making systems that work reliably offline is really important; otherwise, you exclude a lot of people.


This is much more of a problem outside of somewhere like the UK, where internet connections really are pretty ubiquitous. But that's part of managing the challenge of working in different contexts. The assumptions that we make are just not reliable in other places. I think it wasn't until my team and I started regularly traveling to visit our colleagues in, for example, Tanzania, that we really appreciated quite how that works. If you ask, "Do you have reliable power?" the answer is "Yes." But the answer is "Yes" because there's a backup generator. So there is power 99% of the day, but there are probably a couple of brief blips when it switches from mains power to generator power. If you're building a sensitive piece of equipment and not taking account of that, that's easily enough to make things not work properly.


Eleanor Drage (14:12)

I'm really glad you're talking about this because I'm going to give a talk next week to industry students at Central Saint Martins, which is, for those who don't know, a big design university in London. Some professors there were kind of upset that, you know, they're teaching their students all this amazing stuff about a new way of creating tools and products that really think deeply about the infrastructure, about the way that they're made and not just the final thing and yet a lot of people then, you know, run off and go work for Google or whatever and actually don't question the... not that, you know, there's anything... You know, someone has to.


But there are so many other interesting ways of producing technologies that are beneficial, that consider the internet as one reason why people might not be able to access a good technology, or the expense of repairs for a microscope that might already be a million quid plus. So, for you, when speaking to students you work with at a university, about different ways of creating technology, does the idea of "frugal tech" work for you? What is it about? What is frugal technology, perhaps?


Richard Bowman (15:31)

Again, I suspect that's another one that means different things to different people. Where I came from was not actually focusing on "How can we make something that is really useful if you're way off-grid in rural Africa?" And indeed, we're still some way from being useful there. I think I've always been curious about how you can do science without having to completely throw caution to the wind in your budget.


It's very easy, if you look at a scientific catalog, to think that you're being taken for a ride. There's just an extra zero on the price of everything. I think it took me a long time to realize that, actually, this isn't because we're being hung out to dry by the science suppliers; it's actually because selling a scientific instrument is harder because you sell fewer of them. If you're doing that commercially, your fixed costs are still quite high. And so, if you're only going to sell 10 things this year, they'd better be quite expensive things.


On the other hand, as a scientist, you can do quite a lot of science using hardware that is not science-specific. A lot of the time, you don't actually need a science-grade camera; you just need a webcam. Or you don't need £20,000 worth of ultra-precision translation stage; you just need something that can waggle back and forth a bit. And so, particularly with the advent of "maker culture" and people playing with Arduinos, Raspberry Pis, and 3D printers, it's become much more accessible for many more laboratory scientists to make their own equipment. And to do that in a way that is not, kind of, top-flight engineering it's "good enough."


By "good enough," I mean good enough for a high-quality research lab in a prestigious institution. I think one of the things that we have had to try very hard to manage is this suggestion that somehow you've made something that's "good enough for Africa." I think we have to push back very hard against that idea.


Eleanor Drage (17:30)

Yeah.


Richard Bowman (17:33)

Because it would be wrong to say that you can use lower-quality equipment for medical diagnostics in one place versus another place. Actually, the standards are the same. It's not ethical for us to offload slightly rubbish technology on somewhere that doesn't have the really expensive stuff.


I think, instead, being frugal about it is not about cutting corners and doing something that's not good enough. It's about doing something that's good enough but isn't over-engineered, perhaps something that pays a bit more attention to the resources you're using and what's available in the context where you're working. But ultimately, the aim is not to compromise on quality, or at least to be honest about the quality that you're achieving. Clearly, an OpenFlexure microscope that represents a couple of hundred pounds' worth of parts is not the same as a £200,000 confocal. It's not even the same as a £50,000 motorized brightfield microscope, but for a huge number of applications, it does hit the spec that it needs to hit. I think it's about making that distinction for me. That's what frugal technology is good for.


Eleanor Drage (18:49)

To me, having all the money in the world and trying things that aren't even useful and don't speak to different contexts is not clever. That's just... anyone can give that a go. What's really ingenious are these experiments that really consider what needs to be done, particular kinds of work, and respond directly to them. That doesn't take a huge amount of cash, but it does take a really interesting knowledge about the technology itself, of what fundamentally that technology needs to do.


I love this kind of resurgence in maker spaces and tech craft that we're seeing because it's great for us to reclaim agency. And by "us," I mean everyday people who might not be involved in building technology. We must be able to repair our own devices. I mean, it's madness that we use stuff that we don't really understand. When there are so many great people working in civil society trying to improve competence-building around technology and specifically AI, what I mean by that is not just understanding how things work, but getting involved: being able to tinker around, to fiddle with things, to hack them and jam them, whatever.


People get such great joy in being able to fiddle around with their own devices in a way that, you know, they might have been able to 100 years ago, but certainly not today. I feel like our agency is being taken away from us. We're so aware that we just don't understand how things work, which is why even Instagram channels that tell you how different bits of equipment work are so popular; it gives us a better understanding and attachment to the tools that we're using day to day. So, you were saying that these microscopes that you're creating can be repaired locally? Can you tell us about that?


Richard Bowman (20:56)

Sure. So, to build an OpenFlexure microscope, you need a few off-the-shelf electronic parts, some nuts and bolts, and so on. Then the kind of complicated, custom mechanical bits are all 3D printed. These days, a 3D printer doesn't really count as an advanced technology; they are not expensive, and they're pretty ubiquitous in terms of availability.


They still have some challenges. For example, if your power tends to blip, then that makes it hard to do a 12-hour 3D print. That's something we had to work with the engineers that we were talking to. They ended up buying a UPS for that one because that was what it took. Sorry, an "uninterruptible power supply." So, a big battery, basically. It runs on mains power most of the time, but it means that if your power dips out for five minutes, your printer doesn't have to stop.


Here is the final part of your transcript. I have corrected the grammar and punctuation while keeping your original wording exactly as requested.


Eleanor Drage (21:38)

A UPS?


Richard Bowman (21:50)

And some people've worked on printer firmware that will let it stop and then pick up where it left off if there's a power outage. Though, to be honest, the UPS is much more convenient.


Eleanor Drage (22:02)

Tell us about the way that OpenFlexure is contributing to women's health.


Richard Bowman (22:08)

Sure. So, it wasn't sort of our key aim when we started, but actually quite a few of the medical applications that the microscope has been picked up for are in women's health. And so there are various cancers, for example, looking at cervical smears, where the microscope is under evaluation as a diagnostic tool.


It's been picked up by a couple of studies. There's one in the UK looking at persistent UTIs, which is the wonderfully named EAT-UP study led by the Bladder Infection and Immunity Group at University College London. There's another much bigger collaboration led by the University of Birmingham and the Netherlands Cancer Institute, which is looking at developing open source AI analysis algorithms to detect cancer. But one of their aims is to make sure that it runs both on the kind of hardware you'd find in a hospital here, a big hospital here, which would be a slide scanner costing hundreds of thousands, probably, but also that it runs on hardware like OpenFlexure, which is clearly accessible to many, many more places.


That's not just countries where healthcare budgets are smaller than ours, but it's also smaller healthcare facilities. So, if you're in a rural place in the UK, if you're on an island that's part of the UK, traveling to a big city hospital is already a big effort. How can we push diagnostic services like this closer to patients so that we're less heavily reliant on these big centralized facilities? Better access to digital pathology is a useful part of that.


Eleanor Drage (23:47)

Apparently, in China, they're now training hundreds and thousands of people, not as doctors, but to provide medical care in regions that are inaccessible to hospitals. I mean, these are people who don't need serious medical attention, but need some medical attention that can be performed elsewhere. Anyway, the joys of having enough capital to roll out those kinds of initiatives.


I'm assuming it's not a coincidence that OpenFlexure has found applications in women's health. Is this, from my very skeptical understanding of things, because women's health is underserved in mainstream research, or why is this the case?


Richard Bowman (24:30)

I suspect there is an element of that. It wasn't something we set out to do. A lot of things have been colored just by the people who have joined the open project. For example, we have a couple of pathologists that we've worked with quite closely for a few years whose interests are in cancer, but specifically breast cancer or cervical cancer.


But I think also, for example, the trial on persistent UTIs is probably an area that's been neglected a bit. And I think their main interest in OpenFlexure isn't that it's cheap, but that it could enable diagnostics closer to the patient. I think it definitely is the case that open tools mean that grassroots organizations are a bit more able to access the kind of research that would otherwise be the preserve of well-funded labs.


There's an example that isn't women's health of a community group in Argentina that was looking at soil health and biodiversity. That's an area that's very hard to find funding for because there are big commercial pressures that want to keep farming in the ways that we're farming. And so it's up to these small community groups to then say, "Actually, we're going to go out, we're going to take microscopes, we're going to do soil microscopy, we're going to talk to farmers, and we're going to try and build some evidence that better farming practices improve the health of the soil." But also, we're going to involve the farmers in this, and they can see firsthand what's living in the soil. Perhaps that helps drive change at a local scale, if you like. But making the tools of experimental science available to groups like that is, I think, a really wonderful thing to do.


Eleanor Drage (26:16)

Yeah, it's phenomenal. In our last minute, is there anything else you want to speak about regarding projects that OpenFlexure is doing that you think are really cool?


Richard Bowman (26:27)

I mean, I'm biased; I think everything's really cool! But our big push at the moment is really towards nailing down all of the engineering details, refining things, and improving the quality and reliability of the code as we push towards something that might be taken up for medical use. There's lots of excitement around the medical evaluations we've done so far, but I always have to be careful to say that they're evaluations; it's not approved yet.


What we would love is to see some companies around the world be able to market an OpenFlexure microscope for medical use. That's probably still a couple of years away, but that's something that we're really pushing towards. Founding the Humanitarian Technology Trust is our first step towards putting the project on a sustainable footing that isn't just relying on stuff coming out of university research projects, and is saying, "Actually, probably, aspects of this are not research anymore, they are development." They're really doing something. And that, I think, is super exciting.


Eleanor Drage (27:30)

What can listeners who are interested in the project do to help, or how can they follow you and see what you're up to?


Richard Bowman (27:38)

Head to openflexure.org and get involved. You can build one; you can buy a kit from one of our vendors. You can get in touch if you want to support the charitable work and enable us to help some companies make medical devices in places where there aren't many medical device companies yet. Pop up on the forum, talk to us, build a microscope, and help us make it better.


It really is a two-way conversation there. The joy of an open project like this is that anyone can take it, improve it, and share that improvement back with the original project. Whenever that happens, we all sort of do a little dance. It's so nice to see the project come to life and really grow beyond the confines of what you do in a university research lab.


Eleanor Drage (28:28)

Yeah, absolutely. And you do terrific stuff. I love the idea that capital stays in the country that is using the device. It's back to that when people thought charity meant doing silly things that actually didn't produce a sustainable benefit to the country where the charity was taking place. This is not that. This is creating a new way of developing technology for a sustainable future that brings out the best of the talent across the world and can create great devices that are ingenious, that actually do good, and are useful. And that people can take apart, not like, you know, my iPhone, which is notoriously impenetrable.


So, thank you so much for coming on the show and talking to me.


Richard Bowman (29:13)

Thank you.

 
 
Join our mailing list

Thanks for submitting!

Reversed white logo CMYK.png
  • Amazon
  • kradl, podcast recommendations, podcast discovery, interesting podcasts, find-podcast reco
  • Spotify
  • Deezer
  • Twitter
bottom of page