Thursday, October 22, 2020

Digital Sovereignty.

It has been a long time since I've posted and there is so much to discuss. The continued industrialisation of the technology stack, the automation of radicalisaton online, the impacts of physical isolation in accelerating adoption, the ethics of choice against the ethics of care or the entire question of how to balance "Me" vs "We" in a modern society? There is so much to choose from that it's difficult to know where to start. But start I will and I suppose a bugbear is as good a place as any. My bugbear for today is digital sovereignty.

I'll need to explain what digital sovereignty is and that task is about as easy as explaining what culture is, something that anthropoligist have failed to do in over one hundred years of concerted effort. To make matters worse, I'm going to have to use culture to explain digital sovereignty. I'm already starting to regret my decision to leave my self imposed exile on twitter to return to a bit of blogging.

Anyway, enough dallying. Let us begin with the culture map (see figure 1). Now, you might not have seen this before but that's ok. The map itself is simply a representation of what culture is and the details are not important. What matters is to realise that culture consists of many components including values, behaviours, memory and other concepts linked to our collective. We belong to many collectives from nation state to church to family.

Figure 1 - The Culture Map

Within the map, there can be found numerous feedback loops. These loops can be both stabilising and destabilising (see figure 2) for the collective. For example, it's the visible success of our collective in spreading its values through our behaviour that makes us feel safe with a strong sense of belonging to our collective. For example, our economic success in the West and our values of democracy are intertwinned with our fervent belonging to our collective (notions of the West is Best etc) and the sense of safety that we gain for being "right" or at least more powerful than those other collectives which we might fear. Of course, if our economic success stumbles or other collectives become seen as more successful than us (in whatever pursuit) then our sense of safety and belonging might decline.  We might put up barriers to those dangerous "others".

Figure 2 - Feedback loops











Again the mechanics of this are not important, well not for now. What matters is that the major feedback loops are connected to landscape (see figure 3). 

Figure 3 - Landscape








 


So, what do I mean by this? Well, we live in a context i.e. a physical landsdape that we live within. We use maps to describe this context, "our land", "our borders" and the bit that "our collective" occupies (my nation, my home, my church etc) and where we impose our values, behaviours and even have collective memory. It could be UK, it could be France (see figure 4), it doesn't matter. The collective memories include heroes and rituals, for example "Remember, Remember the 5th of November" really only has meaning within the context of Great Britain. 

Anyway, in the land that my collective rules then we have physical sovereignty. 

Figure 4 - Map

Now, we don't just live and compete in a physical landscape. There are other landscapes that we and our collectives are involved in. For the example, the competitive landscape of business. This can also be mapped. However, we need to be careful here. Most of the things that we call maps in business have one thing in common - they're not maps, they're graphs. To understand the distincton (see figure 5), in a map space has meaning i.e. if you move things it changes your representation of the space. In other words, take an altlas, shift Australia next to the UK and you've got a "different" atlas. The same is not true for most business "maps" - mind maps, business process maps, system maps - because they're "graphs".

Figure 5 - Graph vs Map


In order to create a map, you need three basic characteristics - an anchor (i.e. north), position of pieces relative to that anchor (this is north or east of that) and consistency of movement (i.e. south means south). For reference I have provided a map of an industry that has those characteristics in figure 6. The map was produced over five years ago and was a projection of the future of the automotive industry from its current state at that time to 2025. Maps are often used for anticipation of change and other areas that are beyond the scope of this discussion.

Figure 6 - A Map of Business










Well, the map has many components most of which are becoming commodity-like according to this projection. Firstly you need to know that all maps are projections. They are imperfect representations of a space (they have to be in order to be useful) and being models they're all wrong. But despite being imperfect and wrong, they happen to be useful. Secondly, maps are excellent forms of communication, challenge and learning. According to the map which was produced in UK Government around 2014 then much of the industry would start to head towards utility like models with self driving cars and little differentiation. Hence there would be pressure in the market to find new models to recreate difference and status. The possible options included digital subscription models linked to route management (see figure 7) and sure enough, four years later, BMW was talking in such terms. There are a lot of negative consequences of this path including embedding social inequality into transportation system but that's another discussion.

Figure 7 - Anticipating with Maps.








 


If you look closely at the map, the anchor which it is based around is the user. However, users are members of collectives (nation, family and otherwise). Those collectives have value and we embed those values in our technology systems and choices. In this case, through training data used in  AI and simulation models (see figure 8)

Figure 8 - Collectives, Automotives and Simulation







 

 


You can see this yourself by asking the Trolley Question. If the self driving car has to kill someone would it be the impoverished family of four or the wealthy industrialist? The answer will depend upon what you value in your collective and those answers will vary between collectives. Regardless of your answer, self driving cars will have values embedded in them through training data and those values may well not be our own if we don't produce the vehicles. This is already becoming clear through the Beijing AI Principles. There are many principles within that declaration that other collectives might not agree with, for example "be designed to benefit as many people as possible" might not chime well with those looking to sell exclusive products to a select few.

Digital sovereignty is all about us (as a collective) deciding which parts of this competitive space that we want to own, compete, defend, dominate and represent our values and our behaviours in. It's all about where are our borders in this space. It's no different to physical sovereignty (see figure 9)

Figure 9 - Digital Sovereignty

Unfortunately, the field (in the West) seems to be dominated by management consultants and other gurus telling stories and trying to define what "digital sovereignty" is as though the general who wins the war is the one who comes up with the best name for it. Our repsonses all seem to include a slide into protectionism with claims that we need to build our own cloud industries. We seem to have decided to forget that we don't produce all our own food and cross border trade is an important part of life. Lastly we do like a good moonshot and yes, an artillery barrage can do wonders but it's a really good idea to look at the landscape before you press fire.

China's play on the other hand remains sublime as they go from strength to strength building on the work of Deng Xiaoping in the 1970s. The remarkable feat of constantly climbing up the more industrialised components of the value chain (something that Amazon has also done) from luggage to high end technology (or books to everything) has been spectacular to watch in both in its skill and co-ordination of directed investment (see figure 10). Obviously as China's industries have succeeded and brought their values into areas once dominated by the West it has created tensions particularly as our own collective slowly start to question our success. It's hard to maintain our fervent value that "our form of democracy is the answer" when you're losing ground to others  even if we can't see the ground that we're losing. It's a bit like that moment in 2014 when IBM and others had finally started to realise what dreadful mistakes they had made in 2007 by letting Amazon run freely.

Figure 10 - China









Anyway, that's what Digital Sovereignty is all about and yes, you need a map if you want any chance of  playing in this game. 

So why a bugbear? These games take time, China has been operating its play for 40+ years but we are rushing to be seen to do stuff, powered by storytelling, cheered along by management consultants and without a map in sight. 

We're not in a good place and we're doing little to help ourselves.