How do the creative industries use data?
Welcome to Edinburgh Culture Conversations Welcome everybody and thank you for joining us. My name is Janet Archer I'm the Director of Festival, Cultural and City Events at the University of Edinburgh, and I'm your host for this evening. So a few thanks before we begin, thank you to everyone at the University who has helped make this series happen, especially the Festivals, Cultural, and City Events team and also Catie Cundell from the Research Centres Support team at the Edinburgh College of Art who's helping us out tonight. And a special thank you to the artists, creative academics and cultural leaders who are contributing as panelists across the series, which forms part of the global debate on how to make sense of culture in this time of Coronavirus. Thank you too, to the Edinburgh Futures Institute, who are our partners for this event. EFI is all about bringing people together to solve global challenges and build a sustainable future.
Vitally important before Covid, but even more essential now. And thank you to Donna Jewell and Greg Colquhoun from Just Sign, who have joined us for the series as our BSL interpreters. All of the previous events have been uploaded to our website if you want to watch them, and we'd be interested in your feedback, you can contact us on firstname.lastname@example.org The series is taking place against the backdrop of the world's biggest festival city.
Successful because of its extraordinary community of festivals large and small. 2020 is the first year since 1947 that the spring and summer festivals haven't been able to take place. It's felt important to mark this moment and capture how the arts and creative sectors can help society recover from the effects of Covid19 This year, people from all around the world have engaged with the festivals in a different way through an impressive range of online work.
While it definitely hasn't been the same as the live experience, audience members have been, audience numbers have been impressive and there's much to build on. I want to give you a heads up that the Edinburgh International Festival's platform will feature two newly commissioned EFI artworks by Anna Ridler and Caroline Sinders, and Jake Elwes The programme's been curated by Chancellor's Fellow Drew Hemment and explores the boundary between the real and the artificial. The work will be launched on the 17th of October and available to view until the 30th of November. So tonight's conversation focuses on data. Questions we're going to discuss include how do the creative industries use data? Data is much more than just technology. How is data being used imaginatively by people working in culture, and the creative industries? Are there new ways that data could be used to open doors to future opportunities? This context is important to us at the University of Edinburgh, together with Herriot Watt University we're delivering data-driven innovation as part of the Edinburgh, and South East Scotland City Region Deal.
We're using world-leading research in data and analytics capabilities in partnership with public, private and third sector organisations to improve, products and services, transforming the city region into the data capital of Europe. Alongside this, creative informatics is an ambitious research and development programme which aims to bring the city's world-class creative industries and tech sectors together, providing funding and development opportunities that enable creative individuals, organisations to explore how data can be used to drive groundbreaking new products, businesses, and experiences. I'm delighted that Professor Crisp Speed who has successfully established this programme is here with us today. And I also want to give a plug to the University of Edinburgh's Business School, which runs courses such as Developing a Data-Driven Creative Company, a bespoke course for the creative sector, which is running from the 23rd of September, 11th of November.
And it's designed to build your data knowledge and develop competence in handling your own data. So I'm joined today by an inspiring group of experts who will explore the role that data can play and culture and creativity. Wayne McGregor is unable to be with us this evening due to a last minute change in his schedule.
However, I'm delighted that Ben Cullen Williams, who's a longstanding collaborator of Wayne's, also an Alumnus of the University, has been able to join in his place. And we've got a special treat this evening is the rest of our panelists to share a glimpse of their practice with you. Before I ask the rest of the panel to introduce themselves and want to flag a couple of housekeeping points.
I'd ask you to save your questions until about halfway through. I'll do my best to take as many as possible, but I'm sure I won't get them all in. Please do forgive me, if I don't manage to answer your question.
To ask a question, use the Zoom Q&A function and you'll find it to the bottom of your screen. We're recording the event. So please take that into account in asking and answering questions. And if you wanted to tweet, use the hashtag #uoecultureconversations.
that's #uoeculture conversations. And be aware that the University's approach to ethics means that we are guided by the principles of dignity, respect for others, integrity, objectivity, and openness. And I'd ask you to present questions in this context.
We will reserve the right to remove questions if they don't fit within this. So now I'd like to welcome my guests for this evening. Ask each say a few words about themselves. And I'll start with Professor Chris Speed.
Professor Chris Speed - Hello, good evening folks. My name's Chris. I lead the Institute for Design Informatics in which creative informatics is a project and, simply put, it's a collaboration between a Design School and a world-class School of Informatics where the study and the flow of data is all part of how we understand what we do and what we mean in a digital economy. Thanks, Janet. Thank Chris. I'll move on to Mike Chappell who's joining us from Perth, Australia. Michael Chappell - Yes, that's right, thanks so much for having me. I'm Mike Chappell, I have a company called Culture Counts which evaluates the various types of impacts of arts and cultural activity on the communities that they serve.
And it's a great pleasure to be joining you from Perth, where it's dark outside as you can see. It's a great pleasure to have you with us, Mike. And let's move on to Bev. Beverly Hood - Hello, good evening and thanks also for the invitation to join you tonight.
So I'm an artist and I'm a Reader in Technological Embodiment and Creative Practice and also Director of Research in the School of Design Edinburgh College of Art. My artworks that I may use. Digital media and performance to try and explore technology, science and the body and the relationship between those and the lived experience of those. And I do lots of collaboration with scientists and technologists and medics. Thank you Bev. And let's move onto Asad.
Asad Khan - Yeah, it's so great to be here amongst you all. I'm a computational designer and architecture researcher. Currently I'm pursuing a practice lead Doctorate at the Edinburgh School of Architecture and Landscape Architecture. I've got some abnormal curiosities for quite some time. now, I'd be like exploring a specific cause of scientific statements turned as existential risks or ex risks. Extra statements indexes the extinction or like planetary originating intelligent life or the permanent inhalation of it's potential for desirable future development.
I used remote sensing, astronomical imaging, machine learning, forensic science, computation simulation, robotic fabrications. I usually create data visualisation, cinematic installations and a document tree animations. My projects are like, they isolate this sort of synthetic world, a moment where we can explore in different forms in which humanity can contemplate its possible extinctions. Thank you. Thank you, Asad.
And I'm going to ask you to begin in a moment when I've been, when we've introduced Patricia and Ben with showing your work. But let's move on to Ben because you've got something in common with us. Ben Cullen Williams - Yeah, I studied on my undergraduate architecture degree at Edinburgh University and now I'm an artist based in London. My practice consists of sculptures, kind of large-scale installations, photography and video. And in my work I try to kind of explore mankind's relationship to the world in this kind of increasingly changing environment. I'm especially interested in kind of issues such as the in-between on a global scale and local scale and also on a personal scale.
Also recently I've been investigating more about how related spatial typologies can be understood as a physical manifestation of our own human condition. Often I like to do collaborations with a whole range of people and see one of my collaborators who invited me a few years ago to work on a piece of his is Wayne McGregor who I'm standing in for this evening. Thanks Ben. And interestingly, this whole concept of the space in between has come up at least once before in this series. It would be good to unpick that a bit when we get to our conversation.
But's let's introduce Patricia Wu Wu - Thank you Janet, Hello everyone. First of all, I'm very glad to be joining you here. So I'm fashion designer, researcher, and I'm currently undertaking a Doctorate at the Edinburgh College of Art.
And in both my research and practice, I'm looking at the concept of the Anthropocene, the current geological era in which human activity has been the dominant force on the planet. And so I designed conceptual and material fashioned objects using data, computational simulation, digital fabrication. And I navigate with their procedures and their demands to create special forms of design protocols.
And a lot of my work is in the form of fashion, wearables, hybrid installations, and moving image, which for me serve as a kind of site for thought to challenge local constraints on our knowledge about our bodies in the Anthropocene. Brilliant. So there's a lot to process there in all of your presentations. But I'm going to start by letting us look at some practice. Asad do you want to introduce your piece that you brought to show this evening. Asad Khan - Yep. Sure. So. I think it
was. I'm sorry, I'm just going to.. It was like 2005, August. What are the deadliest cyclones in the history of like America? Hurricane Katrina kind of like swept away across the Gulf coast and causing like a cataclysmic catastrophes. And I think like three weeks later there were like several teams. One was from like NASA the other was from EUSGS. And they started using this sort of apparatus which is called a lie detection and ranging lidar scanner, which uses a kind of a beam of laser that computes and calculates distances so they wanted to see how much damages were caused by the cyclone in several suburbs in New Orleans.
So they created this sort of archive and they deposited this archive online. And in 2017, I guess I started requesting them to be able to see what was in it. What kind of conditions if the cyclone guys see or images of annihilation destruction away. And I started working with this, and converted it into a moving image using like artificial intelligent algorithms to detrimental camera pods and cinematography, cinematographic roots to kind of narrate and curate this archive, but also from the point of this like in human eye that it's not, there is no longer human and that there is kinda oscillating the dense efficiency between the human and the inhuman in a way. So I think, yeah, you can play the moving image and I can join in the conversation a bit later.
But usually like what I do is I reused like remote sense digital archives that are retrieved from this accidental mega structure, which is comprising of orbiting satellites, synthetic constellation, terrestrial sensors and so on and all of these like remote sensing technologies they use like the sort of envoys, which are amplified photons and infrared waves in Sonic emissions. And it kinda quantizes or world into this finally grained informational geography. And if we consider like every surface in the universe as kind of like it has this affordance a receptive sensor which can register and collate traumatic events in the form of insights cause or physical changes. And then these remote sensing technologies can actually equate them into numerical particulates and can suspend them in a limbo free from the ravages of time or weathering and stuff.
So I treat these sort of archives as like data fossils more generally. And maybe it's not this one. I think we've got Patricia's work haven't we? Is that right? Shall we pause and bring Asad's work into play.
And we'll look forward to that one. Yeah. So we got onto data fossils. I love that idea. Yes, I kinda treat them like data fossils which we can mine and sometimes I exhume them like an archeologist and sometimes I work like it's a practitioner, or indeed a surgeon who is performing autopsies in the sort of like you know, numerical bones. And I'm kind of like curating their stories what actually happened. And whenever like and sometimes I like induce virtual time into these data fossils to unwrap them from this, this warped state, to create new forms of digitals, especially temporalysis. And this sort of like weight abducts or creative free sense towards the sort of like alien in warmance which are foreclose to our perceptual understanding of how we see catastrophe.
Yeah, so this is a moving image I created as a result of this process. That's beautiful Asad Thank you so much for sharing that. That's an incredible sense of trauma, leaving a mark on landscape and the fragility and vulnerability of our environment.
That's very beautiful and moving. Thank you. very much. Just tell me before we move on to the next piece. Talk us through how you used data to create that work.
Asad Khan - Yeah. So, what I did was I has access to these large, big data archives which were in the form of like files. And it had like numerical coordinates like XYZ values. Some of them were collated on huge like USG has United, Goldman, several teams and some of it was taken by satellites. So what I did was I can mapped all of those data together to create like a huge 3D model, which was at the size of like a 100 gigabytes or even more than that Then I programmed this sort of shortest path algorithm, which is where Google Maps use this to calculate the shortest distance from one point to the other. And, and I trained like I usually work with AI in terms it's like I treat these algorithms kind of like children, so I've named them.
One of them is called Kansai and the other is called Hitoshi and Akira and all that. I don't know why they are Japanese, but then I treat them like I give them, like I treat them, feed them with catastrophic archives to see what kind of intelligence breathes on them and how they navigate to a landscape to sometimes like the very points, the very datasets, detriments the path of the the camera movements in a way that the data sets is kinda co- contingent upon how it wants to be seen, how it wants to be curated. And this sort of like the process, is more to do with the fact that I'm more generally looking into how these AI algorithms and machine learning algorithms precisely dreams or hallucinate or how disobedient they are towards the artists agency, but are more in-frame towards the datasets themselves. Which is very interesting for me because it overthrows my my, my rational agency as a designer, my sapiens creativity in a way. And I think this is something to do with how every creative thought labours to liberate itself from the tyranny of the designer or the artist.
And that's why it always takes these sort of like an ontological different forms like it's sometimes like in the work of a painting. Sometime it comes up in in film or dance. But in order for creative thought, liberate itself from its material entrapment, from the human finitude. It must first, I think, manipulate the very material that hold sway over it, which is the human condition, which is this sort of like carnal slash and these the brain, the by-posture of the human animal.
So that's why I decided to look at AI de-dream these sort of archives and rate differential rule of catastrophes outside the purview of human perceptual cognition. Yeah. Thank you. Did you know that you're a poet as well as everything else? You're use of language is fantastic! Yeah. Yeah, it's really hard to be a poet and a researcher at the same time. So I'm going to move us on. We've had a glimpse of Patricia Wu Wu's work.
So, Patricia, will you just give us a few words about your practice building on the things that you said. Also focused on the human condition. And and then we'll watch your excerpt. Yeah. So the work I'm about to
show is called The Dust Enforcer and, first there's a philosophical counterpart to the work which derives from philosopher Reza Negarestani's book called Cyclonopedia. I won't go into too much detail, but overall, he's also kind of reimagining the landscape of the Middle East through the lens of fictional entities. And so I was particularly interested in the use of fiction as a form of speculation about the body and how I could re-imagine what we see or think about the body beyond the lightness of the physical appearance.
And how I could use that as a gateway to stimulate new conceptual material manifestations. And there's two components to this work, first, there's a 3D printed mask which I have made from using LiDAR, then I have retrieved that data simulated that into a 3D printed mask, and then there's a performative element where I'm also wearing the mask on using LiDAR again to kind of capture my choreographed movements. And there's also different materials that I was exploring to kind of refracts LiDAR's emission of light into disoriented positions. So usually the purpose of using LiDAR is to capture its target through precise measurements.
But for me it was more not to reproduce the data, but to kind of deviate from the original function. To really kind of explore different layers of obstruction that are sent to the path of the scanner. And also how to harness different contingencies in working with such technologies in a subversive manner that can allow me to create this emergent co-composition of forms between me, the scanner and the materials at play.
Brilliant, thank you, Patricia. So Catie, can we have a look at Patricia's work? Beautiful, that's extraordinary It's just an incredible sense of humanity and there's a wonderful sense of how the translucency ...is that a word..of your sculpture just draws you in. And quite extraordinary to witness. So let's move on.
We've got a theme running tonight, clearly now, which is all about the body and that push and pull between technology and physicality. Beverley, can you talk a little bit about the work that you are going to show this evening? Beverley Hood - Hi there, yeah, the work that I'm showing is an extract from is called Immobile Choreography and how it was initiated with a commission through Grampion Health Arts Trust and the University of Aberdeen biomedical physics department. So they were working on a large scale Horizon 2020 funded project which was developing the next generation MRI scanners, FFC, which is fast field cycle in which changes, normally with MRIs, the magnetic field will be fixed and with FFC they realised you can get more data if you could actually change the magnetic field. So this was as part of their public engagement funding that I was then commissioned to make this new piece of work. And it was going to be located specifically within the City Art Space, which is a really interesting, unusual gallery that's cited in the middle of the Aberdeen Royal Infirmary.
So it's also got specific sensitivities to the space as well in that the vast majority of the people who go there are either patients or people visiting patients. So it's a very unique context-free gallery. And so I, the way I approach the project was how I often would so it's very much in collaboration with the scientists, in this case, the biomedical physicist.
I'm actually spending quite a lot of time observing their processes and the technology that they're working with so in this case it involved me going through the MRI process as well. And what I was very interested in was the experience of these contradictions inherent with it, which is that there is a stillness required as anybody who knows who's been in an MRI scanner. There's a stillness that's required... you really must be completely still in terms of it.
But actually the amount of movement that's happening within the body at that moment, that the process of the magnetic field that happens basically aligning all the protons in the body. So actually that your whole body becomes into alignment at the time and you're completely still. And I was very interested in the idea, I mean, this is very much following on from what Assad and Patricia was saying.
It's really about making and immaterialising the material body into a dataset so that process of the MRI. But also there was lots of layers that I was interested in in how it did that. So the actual physical cavity itself that you located in when you experience which some people find very problematic and has a very limited space and also the way that the MRI itself works and the relationship with the body, which is completely imperceptible, which is the way that the radio frequencies then are pulsed within the body. And then the cells are then it's the principle of the relaxation time, which is the time it takes for the radio frequencies to return back to the equipment.
That the different tissues and different pathologies in the body take longer. But I thought this was a really interesting choreographic potential process and principle, both with the physicality of the space, its limited space. The idea that you're immobile, but actually your body is not immobile, it's completely in action at that moment. And this pulsing relaxation time. So what I did was I made a piece which it's not so much in terms of what Assad and Patricia have done where it uses the data directly.
But what it does is it refers to the principles of that. And what I thought about was also with MRI's there's a lot of work that exists that already use a lot of really fascinating work that artists have made over the years that use MRI, the look of it. That's a very sort of established look, and I was actually looking for something different that would refer to the scanning process, but would look not directly as MRI scanners. And also for me there was an element of trying to make it more in some way poetic in a different way and separating it from a direct medicalisation of the body because realising as well the audience going into this space, no matter how attentive the health system staff might be around you, you go into a hospital space and you're medicalised. So it's trying to create something that was more poetic, space for people to go in that would refer to those processes but take it somewhere else as a point of reflection.
So what I did was I worked with a dancer and a dancer that I've worked with for a number of years, Freya Jeffs. And we then developed three sequences of choreography which were based on the parameters of the space. This relaxation principle, which was basically we developed movement sequences which would in theory fit within the cavity of the MRI scanner. And that then they would work in pulse motion which fit with the relaxation time principal.
So there were three pieces of movement we developed, filmed, and then processed which refer to the three developed for different forms of coils which are the coils you wear when you're in the MRI. There's the head coil, the breast coil and the knee coil and each of the movements are hinged to the coils. They are actually then the crucial part of your body that has to be completely still. Whereas really you do have to be still. But in an imagined way, it was what was possible as imaginative, poetic, potentially, what would it be possible within the parameters of that space? So the choreography was developed as a combination of movement that was developed specifically with Freya, edited together, and then sort of extended and pushed further through a process of animation.
And the final insight - it was developed as an installation piece. What you'll see is one of the sequences of the video footage, which was projected on the floor and the walls. So that was the three the head, the breast and the knee coil footage projected around the space. The sound you hear is the sound and it's the pulse, the beat of the FFT MRI, which is reassuringly for anybody who's been an MRI, is much, much quieter, infinitely quieter than a normal, the normal very, very loud mechanical noise.
And then what was also exhibited in the space was also the actual coils themselves that the biomedical physicists had developed for the MRI as well. So it was a combination and I quite often do that. I quite often use some of the actual scientific real-world materials within exhibitions as well.
So that's what the head coil sequence that will be coming up. Thank you, I like it that you use that word reassuring. I had a little look at your work this morning and it just seemed to me that it gave a nice sense of reassurance about that experience and a poetic experience that was very reflective. So let's, let's have a look at Beverley's work.
Brilliant. Thank You. Really good. So I'm going to move this conversation on to some of the other things that I've been thinking about at least and Chris, I'm gonna ask you, what do you think the opportunities and challenges for creative businesses are through data-driven innovation? What are your thoughts on that? Yeah, well, I'll try do a segway as well because I think what you've been watching is artist creatives, designers designing with data. And I say with because a few years ago, an old pal, the late Professor John Oberlander wrote a paper with myself around a way of helping us all think about working with this material called data.
And I think I used the word material because some people treat it as a material. Wood, metals, ceramics, plastics. And clearly we can see three artists really getting stuck in with a sense of materiality. But John was very sharp on Latin and always wanted to ask questions using the ablative framework, which asked us to say, when are we designing or creating from data, when are we designing it with data, and when are we working by data. From might be when we get to talk to Mike actually because in some ways, I think Assad certainly went from data.
A data set which was by the look of it, 15 years old old, dove into that data and it was, it was fixed. It wasn't live in any means. And then by mediating it, he makes it live. And of course, the archaeological piece ask questions around the trauma as we look back over time. I think on other occasions there's evidence that we're designing with data through all the three people's work. And I suspect when we get to Ben's work, will find the with, because it remains highly live in real time.
That by is interesting though, the by also speaks of the algorithm by something as opposed to, you know, by Janet, by Chris, or by the algorithm that is performing actions within a data set. So I think - from - with - by - really helps me because some artists don't ask questions, forgive me, where the data's from and what the ethical implications are, they grab, go back to the studio and we do need to be sensitive. But when we think about companies, they really do need to design from and where possible with, because it allows them to change their business models, adapt to what audiences are thinking. And it's in this continuum, we decide the time from, could be quite old, right? It could be, I'm going to design from last year's festival data.
It could be with and very fast and I was minded to think of the comedians and the theatre folk who couldn't come to Edinburgh this year, but who would've been working through the spring on rehearsing, trialing out when an audience laughs, reconfiguring their show all the time, designing from the audience feedback, but with it. And then of course, a 20-day run in Edinburgh. So all of the time I think creatives are very good at designing from more-so width when it's live. And then we're just beginning to learn how to use the bi.
And how we start thinking about that. If we acknowledge that's a condition we know quite well. The question then is how do we change the business models? Because to find success and survive post Covid, or during Covid we're going to have to use those three phrases very carefully to find the new models really that capture value. If that makes sense, that I would concur with that.
And just for those people in the audience who aren't wholly clear what we mean by data, what would you, what constitutes data in the arts and creative industries, is everything or are there definitions that categorise different types of data? What's your view? Professor Chris Speed - Yeah, I think we, I mean, there's lots of forms of data and I think what we're talking about here is ways of, ways in which instruments have gathered insight from the world and placed it within data sets. Data in some ways it is just a way of knowing the world. And it informs feedback loops to allow us to make informed judgments. Much of those for our lives have been analogue. But once we introduced, introduce digital instruments, then we were able to store them, use them as memories, use them at vast scale.
And of course, when you apply an algorithm amongst a dataset, which might be as simple as numbers, but it might be pictures, it might be faces, then we ask questions around who wrote it, what that dataset is doing, what biases are in that data set. And then we begin to understand its potential to offer insight if we asked it certain questions. Or again, we have to acknowledge its, its problems because not eh, very few data sets are wholly inclusive, if that makes sense.
But for me, it's datasets, it's collections of data, and whether those are streamed in real time. Whether that perhaps gathered through questionnaires, through interviews, through photographs which might be frozen and then brought into studios. I hope that makes sense. It's a hard question. I like hard questions. No it's a simple one. I'm going to play that question out a bit actually, and just ask Mike, in your world what constitutes data.
And then I'm Ben, I'm going to ask you the same question in a minute. Mike Chappell - We'll, I'm an economist, so I have a different but not entirely unrelated view of what data is. The data context that I'm going to present in my short video, really talks to the integration of audience feedback into programming decisions and other information sets that arts administrators, arts organisations, and artists have at their disposal to understand and really create narratives to explain the impact of their work on their communities of interest.
So it's about language really. There are algorithms that we use in our work. Excuse me, to interpret open-ended text, comments that people provide through, through surveys and identify themes within those and then cluster the pieces of feedback around those themes. And the interaction between the AI and the researcher is, the researcher can select certain things and then the AI will go and find terms that are synonymous with, with that particular term. So it enables you to view the feedback from multiple different lenses. Investors in arts and culture.
find this work very interesting because it enables them to guide their funding contributions by the outcomes that it might present. And then those outcomes are not just cultural outcomes, they're social outcomes, they're economic outcomes. And in this time of 'otherism' is the way I explain it.
Where there are these divides between different groups of people artificial or real. Funding out what draws us together in community is increasingly important and then being able to articulate that. Many artists find it useful to have this feedback because that they need to understand the connection. That they are actually achieving with their audience. And it's useful for them in artistic collaborations as well between communicating between themselves, other artists, boards or management that might have those creatives if they're in large organisations. And finally, I think the audiences appreciate the language of evaluation because it gives them a platform for communicating what can be some quite difficult and nuanced language.
I've just been so impressed by the way that each of the artists who have presented their work today is able to articulate what the work means to them and what their motivations were and their methods. But often audiences in the general public don't have that, that language. So we like to provide them with a platform and introduce terms that can help them to explain to themselves and to their friends how they experience the work. That's really interesting. So that's linguistic analysis as opposed to statistical analysis. Mike Chappell - Well it's both. Yeah, yeah, fascinating.
There's normal Survey Monkey type analytics in there as well. But we find that the most use of the data is in helping to elaborate and give foundation to the narratives that people use to describe their own work and the narratives as audiences and their investors understand about the, the outcomes, the multiplicity of outcomes that their work can generate. Interesting, so Ben, moving onto you. What constitutes data for you in your world? Ben Cullen Williams - Well, I guess I see data as kind of kind of any input really, if you think, you know, artists have always been working with some sort of data. We have a, our kind of natural receptors, our eyes, our ears, our hands, and these are our ways of seeing the world naturally. And then from that, those, those kind of inputs, then we've kinda used, found different tools, be it paint brushes or, or sculpting tools in order to kind of express that data we had collected and we kind of output that data through our, through our human lens to our kind of our way of interpreting the world.
And. As we've come to see, as we kind of developed and kind of moved into the digital age our tools have moved much more into the digital world and our datasets are ways of seeing the world are not just limited to the receptors we have with the human body. We can macro data. We can see huge kind of, you know, huge kind of epochs of time that will see, would never have been able to see before.
We can see data from under the sea, from space. So kind of I guess as an artist is kind of how do you then use this data in order to see the world in a new way, kind of embody that data and then output that in a whole series of different ways. And I guess the tools to output that data can be, can be digital or you can output that data as, obviously within this project with Wayne McGregor. Wayne McGregor output to that data as dance.
So we kind of have a whole series of different tools and you know things that we can, we can kind of work that they are at the moment. So the project with with, with Wayne McGregor, Wayne McGregor had been worked with Google Arts and Culture for, for some time to create a, to create an algorithm that was able to create AI generated choreography. And what that did, that projected. It analysed all of Wayne's archive over about 25 years and analysed the video. And it learned how to dance, kind of in the, from the language that Wayne McGregor created for such a over the last 25 years of his career. Then he worked with that, he worked with that data, the output from the tool that was created to generate new choreography.
So the new choreography feels created in collaboration with the tool. And then I guess for me the way I worked with that data, when Wayne called me up and he said, you would you like to work on a project with me for a performance in LA, of of some of the kind of experimental work he'd been creating in collaboration with the tool. So obviously I said yes, and then I went to Google Arts and Culture Lab in Paris, and spent some time with them. And I said to them, Can you give me like like half an hour of AI generated choreography that's been created from the, that's been created from the dataset of Wayne's work over a period of time. The tool, the new tool, not a hammer, not a paintbrush, the new tool that Google created in order to generate this choreography. And then what I did after this, with the code from that, with Google Arts and Culture, hijack that code and kind of outputted in a whole series of different ways.
So, you know, as a person we only have a kind of arms, legs, and you know our body to express ourselves with so but obviously using digital means we could output that in a whole series of different ways. So we can see dance, and see movement in a whole spectrum. Fantastic. And I'm really interested to see this, this work I presented Wayne and commissioned Wayne in the nineties when I was a dance presenter.
So I've witnessed his evolution from the very beginning. And of course what always fascinated us at the very beginning, is that his work emerged out of his own physicality, out of his own body, which he then gifted to other dancers. And over time that's, that's evolved, so I'm curious to know whether the AI has managed to pick that up and make sense of that. Shall we have a look at the work. Let's see it play.
How is it that we could work with machine learning or artificial intelligence to amplify aspects of the process that allows us to do things we could never have done before. Started, first of all, with a trip to the lab in Paris to meet Damien and his team. That was amazing that first conversation because it was really exploratory.
We explore with artists and cultural institutions the intersection of art and technology. So when we started working with Wayne and we were really interested about movement prediction, we had to gift the archive, and it was a huge archive, thousands of hours of video to be analysed. For that tool to work, you have to have a body, first of all, inspiring it. It was phenomenal, we were able to have this performance in LA. What was interesting to me is once we'd gone through this process, and we're building these phrases of movement. We were building these packages of choreography.
Is they inherently look different from things that we've made before. This was a technological marriage, if you like, it was that we were working together to inspire something really new. Ben Cullen Williams is an artist I've worked with several times before. And when I realised that I wanted to find some other translation of the archive in another discipline I felt Ben would be the ideal collaborator for that, this project was very fascinating for me because it was a brand new type of collaboration. Having the chance to work with AI to create a brand new piece of abstract video.
We made that movement interact with a whole series of different environments. Now the archive is going to be open to the public. So all of this private stuff that we've done to generate this new piece of choreography is going to be blown open and anybody can access it. We made an online experiment which is really a fun way to navigate through this map using your body. It gives you an atlas or a map of all of the moves in the archive. You get a sense in which there are hundreds of thousands of unique moments of movement.
So movements that never used to go next to one another, are all of a sudden next to each other. And you click them together and make a little phrase, a physical action. And you can play it. And now you have a unique version of a phrase that nobody else has ever made. It speaks to this idea that everybody has the opportunity to see movement uniquely and everybody can actually make choreography. Amazing so Ben, we can find that on Google Arts is that right? Yeah, so you can, you can go into the Google Arts and Culture Experiment and you can see the, you can see the project that Wayne made with Google Arts and Culture, and then alongside that you can actually see the process that Google Arts and Culture and myself, how we worked together.
We worked together on Google Chrome platform. We, we basically created these kind of different environments which fed the code into it. So it was all made through Google Chrome. And you can actually go on to those kind of experiments and you can kind of play around with different parameters that we, we kind of, we put on there in order to change, manipulate the different dance code.
I should say that for me the most kind of, I guess the interesting thing about that was the, was the choreography, the AI generated choreography that was kind of outputted from there. It was, it was the kind of, some of it was very kind of, I guess traditionally beautiful. Some sort of stuff you would naturally. see as ok, that's kind of very appealing dance.
And some of it was like super ugly and kind of weird and strange. And my natural inclination when making the video installation was kinda to edit these bits out, but I was like, after, you know I've got to stop myself and I'm going to leave those in. Because those things, those kind of things that AI will do, they're kind of the things that will knock us side ways and be able to think us in different ways, are the things that don't privilege our own human lens. We like to kind of treat AI as a kind of equal if it's going to kind of stimulate progress and stimulate change in order to get us out of our kind of preconceived habits. Thank you. And I mean, there's lots of links to be made across everything that you're saying, but linking back to Patricia's work and that reinterpretation of the body and giving us the opportunity to almost get inside of a physical, a physical entity.
And I'm just thinking about what Mike said about bringing people together is so important in our world. And the interactivity of, of, of this project is incredible in giving people the chance to really get under the skin and get inside work and understand it from a completely different perspective in a way that I wouldn't I wouldn't even have imagined would have been possible. Back in the day when I was presenting dance before any of these new forms were even invented. So really, really interesting. I'm going to move us back to Mike and we're going to have a look at your film and understand a little bit more about the really interesting work that you're doing with Culture Counts. To use data in creative ways, but also in very hard ways of driving business practice and opportunities for businesses to reach their goals.
Shall we see the film? Yes please. And then we can learn a wee bit about how you've approached that. We work with vulnerable people, people who experience disability or disadvantage and enable them and empower them to take part and excel in the arts. When we took part in culture counts, what that actually gave me is much more than I thought because what it looks at is the artistic vibrancy, the social impact, and the organisational ability of how the project is delivered. So it's those three key areas which I get phenomenal feedback on. With our work.
with Culture Counts, we were able to consult with an external body on what would be good current, relevant questions to ask. How are we going to benchmark our organisation against other organisations in the arts or other disability organisations. What Culture Counts gives, it gives the external evidence.
This is making a difference to these people's lives. People can also see ... Mike, do you want to come in and talk about this particular project? Sure I just took this.
This was one of the early organisations who adopted the Culture Counts platform. It's an organisation in Queensland that works with people with disabilities. It's obviously a participatory practice that they have. And they used the platform to measure the impacts that participation of their clients had on their socialisation, their engagement with community, and then use that information to apply for grants and basically validate their achievement of the mission.
I chose that because a) it was a video that I had ready to hand. but b),that it demonstrates the holistic impact that arts and cultural activity has. It's not just the cultural impact of cultural activity that matters. And it's not just the audience perspective that matters. It's, it's a very participatory way of viewing the world. And that, that we saw, I thought that was a fantastic way of engaging participants digitally with the ability to, to choreograph the different aspects of a piece of dance work.
So it's just a snapshot of the way that we do where we work with local authorities, we work with place managers. We work with arts and cultural organisations to explain the holistic, the social, the cultural. The community connection aspects and the cultural and economic aspects of all different types of artistic work. And of course Culture Counts is being used in England at the moment for a very specific purpose with the Arts Council England, which we can perhaps talk about more if we have time.
Yeah, and measuring value is it's fundamental to everything that everybody's doing in the room. Being able to find the right means by which we can measure value both through narratives and storytelling, as well as statistical evidence and data. is crucially important to justify public investment in the arts. so... Mike Chappell - the two have to to go hand in hand? It can't be just a subjective narrative alone and it can't just be data in a statistical sense. It has to be the one supporting the other.
And in one of our sessions we talked about the concept of the triple bottom line or the quadruple bottom line and the interface between social, cultural, and economic value. Is that something? Well I think I know that it is something that is in discussion and very developed in an Australian context. Yes. So that's, that's been our work for
the past 7.5 years actually, and we collaborated when this platform was created, we collaborated with an excellent colleague in England by the name of John Knell, who was instrumental in the formation of the initial measurement framework. And we continue to work with him on the Arts Council England work. Great. Thank you. So how transformative can working with data be for the arts and creative industries, just going back to that question. We're clearly looking at this question through many different lenses.
How transformative is it from a business perspective? And how, how can we get people to recognise and realise that data is something that can bring real meaning to the work that they want to pursue. Chris do you want to come in on that question again? Professor Chris Speed - Sure, I mean, we've been working with the startup community in Edinburgh, it's quite, Well, we've got a couple of Unicorns and we've got Codebase, terrific partners. And some learning from those folks, they tend to talk about Build Measure Learn. And I'm curious about how the arts begin to adopt Build Measure, Learn.
The phrase simply talks about we build something. We measure how it's landing with its audience, with users. And then we learn from it. And that cycle keeps turning as we begin to tune its value. So I guess Mike is using natural language processing to pickup feedback from his audience. He's finding out where it lands appropriately to build value, social, cultural, economic.
And then he's informing people who use that platform. About that and they're benefiting, right? So I do, I'm curious about how the Build Measure Learn is implicit probably in practice. In academia, I think ironically, we tended to be very learned. And then we will measure something with a view to a proposition to building it. And we did it once. And we most PhD, sorry, PhD students, it's the one thing. And I'm fascinated in picking up the speed a bit now by using data in a build, measure, learn you're adapting far more iteratively and using it alongside.
And we'll see how many of the old guard institutions want to do that. It's a very startup kind of idea, but it does allow you to pivot and navigate toward the space where value is co-created because you're listening all the time. So build, measure, learn, works for me at the moment.
Great, and how would you, how would you go about persuading people that data is, is something that they should draw from in that build, measure, learn framework. Professor Chris Speed - Well, another little bit of jargon, if you like, is that Code Base whispered in my ear as we went for an interview for the £6 million, and said, look, remember, creatives are the best defense against the dark arts. And what they meant was, this was just after Cambridge Analytica. If we're going to enter a period where we're designing with data and data sets are deeply challenged, probably skewed with limited diversity in them.
That actually if we're going to get ourselves out of a hole where data-driven technology is owned by the darks. Then, then let's get, let's get the creatives in there. So it's more of a call to arms actually that let's attempt to pivot and turn those datasets which perhaps be biased and begin to think about how we represent the values that we really want for our community. So I would treat it as a call to arms.
But of course, Build, Measure, Learn is something we do as humans, we're just listening. The way that we adapt to a conversation. It's building a conversation, throwing a word out there. Measuring how it lands, and then modifying.
And very few of us as humans want to lead a conversation, we often want to stay in it. So, I think it's relatively intuitive for creatives actually. Yes, I suspect we're all going to want to stay in this conversation well beyond the deadline that we've set ourselves of half past seven however, I am going to try and stick to time. But I just wanted to say to the audience, if anybody has any question, they want to send through now's the moment. We've got nothing yet but if audience members want to send questions through, then please do start to do that. And we'll try to take them.
Before we go on. I just want to check with Mike, do you want us to have another go at sharing your film or are you content that we showed enough. Mike Chappell - that's fine. I think, I think you got you got the idea that the platform can be used in any, any high arts context or all or in a community working with people with disabilities type environment as well. They're equally meritorious for achieving the outcomes that they set out to achieve. OK, thank you. And I'm just gonna Before we go on to our next round of questions, I just wanted to ask all of you what drew you to wanting to work with data, particularly the artists in the room? Patricia, What was it that was seductive about moving into this, this form of work as opposed to other forms of fashion.
So that's a very interesting question actually, because also when I think about fashion, it kind of also begins with data in the sense of measurement. How are they start with the mannequin, and there's precise measurements and how it constructs fabric is all about different calculations. But I think it's just going back to what Ben was talking about earlier is just the means of how we capture and work with data has also changed the way we see things.
Particularly when we move from the macro to the micro. And so for me, it's very interesting to kind of challenge those boundaries, particularly in the standardised domain, in the fashion industry in terms of the aesthetic sensibility. And so a lot of my work, for example, I use remote sensing methods, such as LiDARs we have seen to capture my body and I use that as a design material. That I also work with large-scale data sets, which derive from biological behaviour or actual physical phenomena. And more recently, I was working on a project which looks at Covid 19 mortality rate data sets. And I was looking into how I could convert them into 3D printed face masks.
So it was also in a way, I'm quite interested in kind of looking at the global and the local and those shifting boundaries. But also I'm very interested in the mutations that take place when I'm translating the data because I'm looking to materialise that. And there's that loss of translation that I'm looking for.
So I talked about previously that for me, so about reproducing that data. So it's not about looking for a predetermined outcomes, but more about the different contingencies and disjunctures and how I can use that as a design process to unfold new morphological potentialities. So in a way, it's also going back to the build, measure, and learn that sort of iterative cycle. But for me, I think I'm looking for the unexpected outcome in that process of kind of re-learning from the build and measure. And I'm also kind of interested in how during this physicalisation process, there are fallacies in between computation and fabrication process and how this also kind of challenges the aesthetic implications in fashion. So in a way, it's also to talk about, it's also a conceptual way of understanding and.
kind of redefining what fashion is. Brilliant. Thank you. And I loved that...pulse around changing the way we see things. I can really see that in the little glimpse of the work that you've shown me which really, made me want to look at more. So thank you for that.
We do have some questions that are coming through now from our audience. So I'm going to take them from the top. So question from Amanda Grim. Can any speaker talk about client the climate impact of using a large datasets? And is there any way to use, use large amounts of data and a truly sustainable way? So I'm going to ask, who would like to take that particular question. And if nobody puts their hands up then I will pick on someone and see if I can get you to answer it. Nobody has put their hand up so, Mike, I'm going to ask you to have a go at that particular question.
and then maybe back to Beverley. Would you like to have a go. Mike Chappell - I'm not sure that I understand the meaning behind the question. Is there any way to use large amounts of data truly sustainably? Data is a renewable resource because it's constantly being generated, whether or not we capture it and use it in a meaningful way is probably questionable. But I don't, I don't see that it has any other environmental impact in the way that I understand that term, perhaps Beverley as a point of view. Beverley Hood - So I think I mean, I don't have a direct answer to it, but it would make me think about, I'm reading The Data Feminism book at the moment and it makes me think a lot of the conversation is really a follow on from what Mike was talking about as well.
I think it's actually the data that we use. It's what data that is and what it's representative of, and who it's representative of, so I think that's that's to me some of the bigger issues and what I'm interested in, I suppose if you talk about, I mean, I'm using like medical data a lot of time and some of those datasets will be huge. But actually they're meaningless, although they do have meaning. But it's also, there's a big challenge which is about actually connecting the scientific, and the technological with the personal, with the social, with the political, with the cultural. And they're to me some of the key questions that I mean did feminism as an approach now is really trying to do look at proper representation within the data. Who is the data taken from? What is the agenda that has been drafted? And there are multiple ways people can either be included or not included for political and useful reasons.
And I think these are all the questions. So Big Data to me can become, still depends on its use because sometimes that can be problematic because it can just, it can disguise some of the issues that are actually going on within the integrity of the data itself. So it's not really an answer it's more my current, and what is the current thinking around really from that techno- feminist point of view around data. Ben Cullen Williams - I can maybe chime in from perhaps a more literal Interpretation of the question, I guess my interpretation of the question is, is more about E-waste and digital pollution and algorithms and big data set use huge amounts of server resources in order to, in order to function.
So we do have, is a kind of a larger question of how do we process big data? How do we use algorithms in order to process big data which are held in servers kind of all around the world, disconnected from us. And uses and draws huge amounts of energy and resources in order to do that. So a project I'm working on at the moment which I have a large preoccupation with. Antarctica - a lot of, some of my projects stem from that, from an expedition I did. And, uh, for that project, we're using AI technology and for that, I'm making sure that the servers that we use, the server bases we use in America, close to hydroelectric dams. So using renewable energy in order to, we're using renewable energy in order to process the algorithms because it's such a large draw on data and also the servers themselves because you know, that generates so much heat, they need to be, they need to be cooled constantly. That's the main thing.
And I think there's a if we talk about emails as well, just in terms of kind of e-waste, I think like the average business user generates... I don't have the figure to have, but it's astonishing. Amount of CO2 per year is equivalent to x amount of transatlantic crossings for a flight.
So I don't know if I'm addressing that, I think that is a question that needs to be talked about in terms of sustainability of data and AI. Asad, you wanted to come in there. Asad Khan - Yeah - it's a very paradoxical situation because like planetary scale computation itself, like a climate change itself is the epistemic achievement of planetary scale computation. So the very problems that enable the kind of triggered climate change is the very thing that is diagnosing it back. So for example, if you take the, the Earth as a spherical body and then suddenly you have got synthetic satellites and constellations. Literally kind of changing the morphology of the Earth, which is what we call the geological epoch of the Anthropocene.
But with the thing with the digital data is very interesting because people deem data as a kind of immaterial thing, but I don't think, I think of it is more of a daringly imminent material entity that's incarnated and processes which induces all a lot of energy, a lot of oil, a lot of like you know, literally using a lot of other things at the same time. So if climate changes is pervasive, entity is called. This philosopher Timothy Moore calls it a hyper object, which is like objects and distributed in space-time locality. So at the same time. these hyper arteries are made visible manifested by the very big data sets that we generate to capture them. But the machinery that we require to capture these big datasets are coming from geologically mining, like you know re-org mental summary in like you know in Africa somewhere like in the Middle East and all that.
So this is kind of like a paradoxical situation and this dance of agency where the plan is reacting to how much or how much if we are containing it in the ontology of signs or numbers in quantities. And I think it started all the way from enlightenment. But like you could say with Galimir Copernicus ? , who was like, you know, kind of shifting humanity off from the, from the centre of the universe.
And, and, and it was the basis for the autonomy of reasons. I think it's good, at one instance to like we've literally terraform the planets using the very datasets that we are trying to map and terraform our own world with it. But at the same time, it also allows us to kind of like to epistemically desert mind what is, what is causing all of it. I think that would be my response and key asset.
And I am going to move us on to the next question. Otherwise, we're not going to get them in from audience members. If we have time, Patricia, we'll come back to you on that one. But Mike, I'm just going to come to you.
Jessie Zagar is asking about Build, Measure, Learn. Particularly she's particularly interested in examples if the measure and how measure is perhaps analytic, but also narrative so that sense of weave across the social, cultural, economic. Do you have another example? You've given us some examples of how you've captured the value. through, build, measure, learn through your work.
Mike Chappell - Well, I think the important distinction is what is being measured traditionally in arts and culture, particularly in publicly funded arts and culture, there's been a focus on inputs and outputs rather than outcomes. So first of all, there's the distinction between purely activity by some measures and the outputs that they produce. Numbers of performances, numbers of works commissioned, whatever those metrix might be. And outcomes which are the more holistic social, cultural, personal and societal level outcomes that, that, are achieved by the, by the funding and production of work. As I said before. And I'll rewrite the point again. The statistical analysis and demographics and such like has a use and it's, it's very important.
But really it's only to explain the richness of the narrative, to support the richness of the narrative. And we found that by popularising the vocabulary, if you like, particularly in Australia, can I say that there's a general mistrust of high art. I hope I won't get into trouble for saying that, but giving giving the language of explanation in the questions that they're asked. sparks and it's very satisfying seeing people being interviewed audience members being interviewed, perhaps perhaps after a performance and playing with the words, with the words like connectedness and belonging and injecting those words into the narrative which might have been quite simple for them. It's wonderful to see them interpreting what that means for them, discussing it with their friends, and sort of learning and growing through the development of language for them. And it demystifies the feelings that they might have where it helps him to explain them.
So that's, that's a very gratifying part of the exposing the narrative behind the work itself. Thank you, Mike, and I am going to move us on to our next question from Bryce Zidalles. So Patricia, I'm going to ask you to respond to this. And then maybe Bev so, does a data-driven approach to design possibly encroach on underm