Computers’ ability to effectively simulate a huge variety of functions has allowed them to replace hi-fis, cameras, maps, DVD players, calculators, typewriters, cash registers, musical instruments, watches, TVs, newspapers, books, shopfronts, money and more. Although the changes caused by the general-purpose power of computation are widely observed, it is arguably less discussed than the role of data and/or the costless copying of information.[1] Perhaps due to the high visibility of the large data-driven US tech companies and because talking about costs fits more readily with the world view of policymakers and economists than general-purpose machines. [2] It is though no less important.

The versatility of computation unifies many seemingly disconnected phenomena. It is not just a neutral technological change; it has economic and political implications too. Spotting early on that many devices could be replaced by software on premium hardware is a key part of Apple’s success and other companies’ decline (Nokia, Kodak). That computers can simulate other computers - virtual machines - is central to cloud computing. The largest cloud provider, Amazon, leverages the spare capacity of its huge computing resources by virtualising and reselling it to provide on demand digital infrastructure for many companies. [3] The Raspberry Pi has sparked a wave of inventiveness as, within the constraints of its hardware, it can do what any other computer can at small scale and low cost. A future of self-driving cars steered by software is highly concerning to any country with large traditional car industries such as Germany and Japan. The great public interest in Artificial Intelligence (AI), and on the automation of work are, in many ways, concentrations on specific manifestations of the general phenomenon that lots of things can be simulated by software on the right hardware.

Some things follow from this:

1. If software is systematically going to be embedded across the economy and society then programming skills and computer security become very important.
2. If many roles are going to be automated through software, then we should focus on developing the related technologies that are driving the automation such as AI and robotics (or risk losing out).
3. That as computers get deployed in ever more areas, converting domain knowledge into software-based business models, and the interface with users i.e. design become ever more important. As this requires new ways of working, which is often harder for incumbents, this creates more opportunities for start-ups.[4]
4. We should also focus on activities that are less likely to be automated, such as creative activity.[5]

Computation is however having its own impact on the creative industries.

The ability of computers to perform many activities is having several effects on the creative economy; the creative industries and wider sectors that employ creative skills.

These changes are well known within the sector, but are less known outside it.

Computation democratising creation That computers are enabling individuals to undertake a wide range of tasks is democratising creative potential. It enables people to programme, to create music, games and graphic designs from just a laptop. This is having artistic impacts. A reason for electronic music’s increased popularity is advances in what can be produced with a laptop and software such as Ableton and Pro Tools.[6] This change is also contributing to the structure of labour markets. The ability for many creative activities to be produced from a laptop, combined with the ability to distribute it electronically, has enabled people to work more flexibly encouraging freelancing in the creative industries.

Games technology being used more widely Computer games are, perhaps, the ultimate example of the computer’s power of simulation – the creation of an interactive virtual environment in real time. Games technology is being deployed in more areas. It is also expanding into more immersive experiences with Virtual Reality and the real world with Augmented Reality. In the recent TV adaption of Phillip Pullman’s His Dark Materials the Unreal engine (A programme used to create computer games) was used to create a replica of the film sets and the animated characters. An animated game simulation of the visual effects was then superimposed on the real set using Augmented Reality, linking what could be seen through the viewfinder to digital animation in real time informing the final version. [7] The Weather Channel uses the Unreal engine to visualise extreme weather events, such as a storm surge. Self-driving car companies are using environments from driving games to test drive their algorithms. Games are also being used to train AI algorithms more generally. For example there is a partnership between the AI company DeepMind and the games engine Unity to develop game environments for AI.

Computation powering creation by producing new techniques There are many digital creative software tools: ranging from Instagram filters to dedicated software like Photoshop, Blender (3D animation), Renderman (Rendering), Processing (programming to create images), TidalCycles (programming to create live music) to name but a few not already mentioned. Recent developments in AI such as Generative Adversarial Networks (GANs) and style transfer are starting to be used creatively, such as in the Aphex Twin's T69 collapse video.[8] There are also other approaches such as generative design, where algorithms are used to produce designs that are becoming more influential.

The convergence of creative domains Digital convergence has created a universal medium for production and distribution bringing different parts of the creative industries closer together leading to a common skill set – a trend that is likely to continue in future.[9] There are a range of technical skills relating to 3D modelling such as rendering, photogrammetry and motion capture that are used in multiple areas. For example motion capture, where people’s movements are tracked and then used to move a digital avatar (perhaps most famously Andy Serkis’ Gollum in Lord of the Rings) is deployed in both visual effects and computer games.[10] [11] The 3D digital models that represent characters inside computer games can sometimes be downloaded and 3D printed.

What this implies:

Although these trends are common knowledge within the sector, institutions and policies are still catching up with them. Several implications follow from them:

1. Skills: Recognising the development of an ecosystem of technology and artistic skills spanning different domains In the commercial sphere boundaries between disciplines are already fusing, with common skills being used across multiple domains, and we should see this reflected in the higher education system and policy. That technology is producing a range of connected skills which can be used artistically in multiple domains both inside and outside the creative industries means there are more opportunities in this area. This is something that can be seen from our work at Nesta into online job adverts and the future of skills.[12] It is being recognised with the development of new institutes such as the University of the Arts London Creative Computing Institute. However, it is not fully reflected in the higher education system more widely.
2. Integration: Integrating artistic and technical skills more closely That software will be increasingly embedded across both the distribution and production of creative content means we will see greater need for the combination of artistic and technical skills. Apple’s core skill is in many ways how it integrates both design and technology. Economic development is likely to lead to a richer, more highly educated and demanding global market. As hardware is often commodified with relatively few chip and memory manufactures, the integration of software and design is likely to provide a way to stand out in the product and services spaces – digital and physical - creating new opportunities. This is therefore at least as much about the integration of such skills as it is about the skills themselves.
3. Ecosystems: Supporting skill development in the freelancing ecosystem Just because technology has enabled more people to work on their own creatively does not mean it necessarily results in an ecosystem that produces optimal outcomes.[14] With accelerating technological change, learning has never been more central. Also a side-effect of being able to do more at an individual level, is that things become more competitive increasing the returns from collaboration. Companies are less likely to invest in the skill developments of staff who are not permanent employees. Time spent learning by freelancers is time that could be spent earning, but which has wider benefits across the companies that they work for. Finding ways to support skills development in freelancing therefore becomes more important.
4. Business models: Being clearer on what the effects of platform models are These technological changes mean more people can produce creative content. It does not necessarily mean demand for that content increases - all other things being equal it increases the competition. However, the supply of content will be more heterogeneous and so perhaps matches individual tastes better. In practice, although computers general purpose power has a decentralising effect, networks effects, data, the volume of content and low-cost scaling push towards concentration of distribution in a small number of platforms such as Spotify and YouTube based on algorithmic recommendations. This raises the question of how the new digital gatekeepers affect outcomes and what the policy implications are.[14] This is an area we will be working on as part of a new research programme on platform regulation in the Creative Industries Policy and Evidence Centre (PEC).

The pervasiveness of digital certainly does not mean we will see the end of the many traditional creative artforms. As happened with painting and photography, the development of new media does not necessarily eliminate existing ones, although it may lead to a change in focus. However, there has never been a creative tool with the computer's range of possibilities and we need to think harder how to get the most from it.

Acknowledgements: The second stage of this post draws on the findings from a report that the author wrote with Georgia Ward Dyer for the EU Parliament

Davies, J. and Ward Dyer, G. (2019), ‘The relationship between artistic activities and digital technology development’, European Parliament. https://www.europarl.europa.eu/RegData/etudes/STUD/2019/634440/EPRS_STU(2019)634440_EN.pdf

[1] It’s not that it’s never discussed. The most famous statement of this being Andreessen (2011), ‘Why Software Is Eating the World’ https://a16z.com/2011/08/20/why-software-is-eating-the-world/, Wall Street Journal. In economics there is a growing field of work algorithmic game theory that combines elements of economics and computer science, and work on the effects of digitisation, but it is probably fair to say that computation itself is not something that is fully embedded in the subject.

[2] The versatility of computation should not be a complete surprise. When Alan Turing first proposed a mathematical model for the computer he, in an abstract mathematical sense, showed that given a few set elements a programmable computer was in principle capable of doing what any other form of computational device could do: a universal machine. Due to huge technical progress in both hardware and software this theoretical potential is being brought into ever greater fulfilment. This does not mean computers can do everything. Turing also showed there are things it is impossible for computers to do. For example it is not possible to programme a computer to generate certain kinds of number (so called incomputable numbers) or to develop a general procedure for deciding whether a given programme when presented with a particular input will terminate or not. Whether human intelligence is computable i.e. can be simulated by computer has been a long running debate ever since.

[3] Coyle, D. and Ngyen, D. “ Cloud Computing and National Accounting” (ESCoE DP 2018-19). Black, B. EC2 Origins. https://blog.b3k.us/2009/01/25/ec2-origins.html Barr, J. ’Amazon EC2Beta’ https://aws.amazon.com/blogs/aws/amazon_ec2_beta/

[4] Bornschein, C. (2015),’Talk at the digital society conference’, D21-Fachkongress.

[5] Bakhshi, H., Frey, C. and Osborne, M., 'Creativity vs Robots. The creative economy and the future of employment', Nesta, 2015.

[6] Milner, G. (2009), ‘Perfecting Sound Forever’,

[7] BBC Click (2019), ‘His Dark Materials: Behind The VFX - BBC Click’

[8] Bailey. J. (2020), ‘The tools of generative art, from flash to neural networks’, Art in America.

Dark Matter Laboratories, Open Systems Lab, Centre for Spatial Tech (2020), ‘Design Tech Digital technology is on the brink of changing how we design cities’, Connected Places Catapult.

https://readymag.com/yard/designtech/

[9] Easton, E. and Djumalieva, J. (2019), ‘Creativity and the Future of Skills’.

Bakhshi, H., Djumalieva, J. and Easton, E. (2019) ‘The Creative Digital Skills Revolution’, Creative Industries Policy and Evidence Centre.

https://pec.ac.uk/research-reports/the-creative-digital-skills-revolution

[10] Radke, R. (2013), ‘Computer Vision for Visual effects’, Cambridge University Press.

[11] Davies, J. and Ward Dyer, G. (2019), ‘The relationship between artistic activities and digital technology development’, European Parliament. https://www.europarl.europa.eu/RegData/etudes/STUD/2019/634440/EPRS_STU(2019)634440_EN.pdf

[12] Bakhshi, H., Downing, J. Michael A. Osborne, M. and Schneider, P., 'The Future of Skills: Employment in 2030', Nesta, 2017.

[13] Creative Industries Federation (2017), ‘Creative Freelancers’. https://www.creativeindustriesfederation.com/publications/creative-freelancers

[14] Finlayson, A. (2019), ‘The changing economics of electronic music’, Resident Advisor. https://www.residentadvisor.net/features/3511