Digital Thinking
seeing the world with digital eyes

Alan Dix

Computational Foundry, Swansea University, Wales

Keynote at CHI Greece 2021, 26th November 2021

Download draft paper (PDF, 234Kb)

Digital technology is ubiquitous and has transformed many aspects of domestic and business life. At a personal level there is an 'app for everything', in commerce banks are shifting online and even the heat and oil of the factory floor is being transformed by industry 4.0. In some cases, the changes are incremental, simply making existing process more efficient, or allowing online access to previous face-to-face services. However, there are also more radical changes. Some of these are within the methods of digital production from the perpetual beta of Web 2.0 and A-B testing of user interfaces to agile software development. Other changes are enabled by digital technology, such as more flexible industrial processes due to digital fabrication and applications of AI in medicine. There is a distinctly digital eye that allows us to think differently about the world, for example greater levels of personalisation in consumer products, or more dynamic sensor-rich industrial processes. Sometimes these innovations happen by accident, but we can explicitly adopt this viewpoint to prompt more radical design practice. In this talk I will draw out some of the facets and design heuristics of this new mode of digital thinking.



Digital Thinking: seeing the world with digital eyes from Alan Dix


Quaestion and answer session

Some of the following are transcribed from live discussion at the event, some answered afterwards based on questions asked on Discord


(In reaction to washing machine example)

It’s a bit like virtual manufacturing in Industry 4.0, where the actual manufacturing and assembly is distributed across many collaborating partners!

I have been thinking about cars this way. The chassis could be kept but newer, cleaner engines and other bits could be changed, even body panels to change the shape. Hopefully electric conversions will become a big part of the industry in the near future. Cars are a huge huge waste...

I'd like to try to link these points you've made back to the presentation by Neurial Oliver that we had yesterday. She gave us an excellent talk about how they handled COVID in the region of Valencia and all the work they did on data science and AI. They would not have been able to do it if it weren't for the for the politicians who were prepared to accept the important role of AI and digital technology.

I was wondering if you could comment on why we are not at this stage yet where we have these modifiable washing machines or modifiable cars is it a political thing, is it manufacturing thing. is it a technology thing? What do you think needs to happen so that we can actually see this this future happening.

I think it's a combination of existing entrenched interests and lack of imagination.

Back in 2000 I used to talk about shoes of different sizes. I thought that there would be fundamental changes in the logistics of shops because of Internet shopping (e-shopping as it was called then). I thought this would happen online first. We all have feet of different sizes, different widths, etc., but when you go to the shop you can only have so many different sizes. This is largely because the physical storage in shops there's only so many shoe boxes you can stock, which restricts the range of sizes and this creates the need to have fixed sizes and standard shapes of shoes. There's no reason why you shouldn't have a larger range of shoe sizes in a warehouse. It would even be possible to have left and right shoes stored differently so you can bring them together for an individual, perhaps a left shoe that is wider and shorter and a right shoe that's that's narrower and longer. Based on your order the picking system can put them together in a box and send them to you.

Why hasn't this happen. Well partly this is about imagination. In fact we do see some of this kind of thing start to happen for some high-end production but there's no reason why the same process shouldn't work for volume production also. It's just that people and manufacturers are just so fixed in the idea of a small set of “shoe sizes” even though this disregards the diversity of real feet.

I also thought that because there would be changes in Internet shopping this would also transform the logistics of physical High Street shops in fact going back far further way before 2000 in fact before the before the web I imagined the idea of the electronic village shop you'd go in in the morning say I'm going to cook such and such this evening and and by the time you came back that evening to the shop they would have the precise ingredients you need for what you want to cook just waiting ready for you um this is happening to some extent now with pure Internet deliveries where you get packs of ingredients sent to you but is not happening in physical shops even when the same shop has an intellect delivery arm.

One reason this hasn't happened is that many companies, when they started Internet selling, did so as a completely separate exercise. They didn't reimagine the logistics of their whole company; they just created a separate Internet businesses in parallel to their existing physical businesses. Their High Street businesses continued as normal with no transformation whatsoever.

Problem is it's easier to see new points new innovations new visions than it is to see paths from where you are now to the new point one of the reasons for using the fashion story is that the existing logistics the fashion industry would allow that delivery of part made garments and you already have local manufacturing so it's somewhere where you can see the path to a new kind of industry from where we are now.

In general, you need to both imagine the path and you also want it to happen, which involves understanding the values and risk of the transition.

Sadly at the natural thing to happen with digital technology is that one has an existing logic structure and the digital technology is added in a patchwork fashion and with no thought to radical transformation. Think about digital manufacturing and the way that it is moving. The existing logistics of spare parts procurement for whether it's for washing machines of cars is that a local engineer orders a part, the order then goes to the manufacturer who will deliver it from their parts warehouse wgich are supplied from central factories. The digital manufacturing change is that they might put a metal printer in that parts warehouse so instead of it coming off the shelf it is manufactured ‘just in time’, but the fundamentals of the underlying logistic process have not radically changed; you've just substituted one element without any radical reimagination. It doesn't lead to a fundamentally new kind of process.

Turning now to cars. If you want to design an electric car, why would you want a transmission system? It's inefficient and complex. However, most conventional manufacturers just take the electric engine and place it where they would have had a petrol engine and the rest of the car functions the same – there is no fundamental re-imagination of the nature of the car. In contrast Tesla did what, in effect. is the obvious thing because they designed from scratch. Instead of a central motor they put a motor in each wheel. This gets rid of the whole transmission system, especially the complexity around the steerable front wheels. This is simpler, requires less materials, is easier to maintain and, in the long run, will be cheaper to produce.

The existing manufacturers don't do this because their ways of thinking about design are embedded in the past and constrained by manufacturing processes framed around the internal combustion engine. It's a bit like the earliest cars, “the horseless carriage”, which were steered with a rudder rather like a ship. There's a similar lack of imagination when most manufacturers create electric cars that means they just recapitulate the old petrol designs. Crucially this does not give you a good electric car.


Your talk makes me think the way computation have created new art practices and genres (from algorithmic art to choreography and AI art), I would like to hear your thoughts about the implications of “deconstructing behavior and experience”, such as reducing the complexity of the human lived experience.

The arts comment certainly makes me think about some of the transitions we've had in the past in the arts; for instance from theater tosilent films and then from silent films to the talkies. The genres of production and audience reception had to change – initially you had films that were simply mimicing theater production,but over time new genres emerged.

As you alluded to in your question I've worked in the past on this idea of `deconstructing experience’ and certainly I've found that a helpful way to address these kinds of things trying to tear apart what is it that we really want, understanding what's going on in the existing patterns of behaviour, digging back from the superficial ways there they are done to ask more fundamental questions about why they are done at all. I’ve already talked a little about that in manufacturing, thinking about end-to-end production and consumption not the way it is currently executed.

You ask about the ‘complexity of the human lived experience’ and certainly one of the problems with digital technology is that it often makes life more complex rather than less so. We are at a point of digital maturity when we should be looking at our day-to-day lives and asking what's it really about. There have been many ethnographies of the home, but really get into the heart why we do the things we do, and we want out of our lives. We can then look to digital technology, and instead of asking, “what can we do with digital technology?” instead seek to address these fundamental questions about human lives, about our day-to-day mundane existence, using technology to help rather than add meaningless noise.


In a time where we, as technology designers, have the tools to mix the physical and the digital in mixed or extended reality systems, we are often wondering by which principles or methods may we make these design decisions.

Well that's a good question as it just so happens that after a long, long period my cauthors and I just delivered the manuscript of TouchIT to Oxford University Press, which is our book about designing products that blend the physical and digital, from simple device controls to smart spaces. The design of these hybrid objects is not the same as for pure digital products. When we design for the physical and digital world together, we must understand both. In fact, one of the things that has come out of this book and over many years thinking about these issues, is that the use digital technology, including its problems, acts as a sort of mirror casting light on the physical world, so that we understand physical phenomena better or even seem to see them clearly for the first time. This increased understanding of course feeds back and puts us in a better position to design digital interventions


Do you think COVID 19 has boosted digital thinking or are we drowning in trying to use different software?

There's been some tremendous improvements in video technology in particular. However, I’ve not felt any of the more fundamental changes I’ve described, except maybe in some hyperlocal food delivery services. However, the big transformational effect of COVID has been to throw us back and start to ask us fundamental questions, so that maybe post COVID we can start to ask we want the world and our lives to be like. COVID is acting as a forcing function that shows we can make fundamental changes and therefore it is possible and reasonable to ask the hard questions about the kind of world we want to live in.


It seems that humans do not understand the geology around them and we are stuck to habits even though these habits might be irrational? Doesn’t this go beyond technology? Is it human nature?

You're absolutely right most of us are completely unaware of the details the geology beneath our feet. We might see a rock sticking out of the ground but we don't know instinctively that that comes from an old volcano or ancient sea bed. However that doesn't mean professional geologists cannot understand and certainly the whole of the global oil industry and of course the coal industry in Wales depends on that uderstanding. Similarly when we think about the way digital technology affects our day-to-day lives, we do not instinctively understand what's the right thing to do, but as human–computer interaction specialists, as designers of user experience, as designers of industrial process, and of healthcare and education systems, we can, in the same way as professional geologists, use that other core aspect of human nature. We can analyze and to make sense of things in order to create technology that fits those parts of human nature that are hard to change, to make the technology that fits with people and enables them to live fulfilling lives.


Would personalization be a central element in digital thinking? …I was also thinking how this can affect art and creativity… Can we somehow order the preferred artwork like lords used to do a few hundred years ago? Or as rich people do today (e.g. their portrait)? Will this democratize personalized art and will it affect the quality of the produced artworks?

Absolutely personalization is one of the things we expect of purely digital products and examples like the fashion case study or the washing machine where you could perhaps change the fachia suggest ways we can personalize manufacture.

For art it gets a little bit more complicated. If you are happy with purely digital art, this is relatively easy, you could take your photograph and have it rendered as a portrait in the style of Rembrandt. That's possible already and will get better over time. However, when we say ‘art’ we are usually thinking about the personal element that comes from the human artist. This is then more the way digital technology can bring artists with those who want art, which we see already with platforms such as Folksy see an Etsy although I know not without problems.

However, this also makes me wonder about this relationship between those who can pay and the artwork. A small number of artists get very large amounts of money for their works but there are vast numbers with very little income. Is this the way things have to be? Maybe artists would be prepared to be part of collectives, not necessarily sharing equally, but that perhaps earlier on in your career you might get subsidised to help you make a small, living but perhaps later on in your career accept that some percentage of your million-pound painting gets distributed. This sounds a little like internal taxation among artists but not as taxation, while for the public good, always feels imposed. Can digital technology facilitate ways of doing economies which are less based on competition and market forces and more on giving and sharing? There's so much talk about the sharing economy but that's usually means sharing in the sense of a few small number of people taking a huge profit from vast numbers of people. Perhaps there are truly sharing ways of connecting, rather like the FreeCycle network that allows you to give things away, or Olio app for sharing unwanted food. Although these are the ways things are and have been for a couple of 100 years, but not the way they have always been and certainly don't need to be the way they always will be.

Alan Dix 19/11/2021