About this blog

Featured

I’ve been interested in what people are interested in for years. I was the customer satisfaction manager, in Europe, for the part of Motorola that Google bought. Then I researched customer loyalty and how value is created for a decade as an academic.

Now I still teach managers, research and consult in this area. And through my research and work with different brands I’ve uncovered some startling new insights on how firms can create huge commercial value by getting a deeper understanding of their customers’ interests  – using some up and coming technologies.

I talk to a lot of managers and firms in my work at Nottingham University Business School but I wanted to share these ideas with a wider audience. Also new technologies like Big Data, the cloud and mobile phones are presenting some game changing opportunities for firms and private individuals.

So I started this blog on how to take advantage of these cutting edge ideas.

AIs need to be accountable when they makes choices.

pexels-photo-277593One type of AI software uses neural nets to recognise patterns in data – and it’s increasingly being used by tech firms like Google and IBM. This type of AI is good at spotting patterns but there is no way to explain why it does so. Which is a bit of a problem when the decisions need to be fully accountable and explainable. 

I could have called this post ‘One thing you cannot do with AI at the moment. There are many things that AIs are helping businesses with right now. But if your firm is going to use them then it’s important to know their limitations.

I remember doing my maths homework once and getting low marks even though I got the right answers. The reason I lost marks was because I didn’t show my working out.
Sometimes the way that the answer is produced needs to be clear as well.

It is like that with some AI technologies right now. There are types machine learning AI that are amazing at recognising patterns but there is no way to explain how they do it.

This lack of explainability can be a real barrier. For example, would you trust a military AI robot armed with machine guns and other weapons if you weren’t sure why it would use them?

Or in medicine, where certain treatments carry their own risks or other costs. Medics need to understand why an AI diagnosis had been made.

Or in law, where early versions of the EU’s General Data Protection Regulation (GDPR) introduce a “right to explanation” for decisions based on people’s data.

The problem is that for some types of machine learning, called “Deep Learning”, it is inherently difficult to understand how the software makes a decision.

Deep Learning technology uses software that mimics layers and layers of artificial neurons – neural networks. The different levels of layers are taught to recognise different levels of abstraction in images, sounds or whatever dataset they are trained with.

Lower level layers recognise simpler things and higher level layers recognise more complicated higher level structures. A bit like lower level staff working on the details and higher level managers dealing with the bigger picture.

Developers train the software by showing it examples of what they want to it recognise, they call this ‘training data’. The layers of neural networks link up in different ways until the inputs and the outputs in the training data line up. That’s what is mean by ‘learning’.

But neural network AIs are like ‘black boxes’. Yes, it is possible to find out exactly how the neurons are connected up to guess an output from a given input. But a map of these connections does not explain why these specific inputs create these specific outputs.

Neural network AIs like Google’s Deep Mind are being used to diagnose illnesses. And IBM’s Watson helps firms find patterns in their data and powers chatbots and virtual assistants.

But on its own a neural network AI cannot justify the pattern it finds. Knowing how the neurons are connected up does not tell us why we should use the pattern. These types of AIs just imitate their training data, they do not explain.

The problem is that lack of accountability and explainability. Some services need proof, provenance or a paper trail. For example, difficult legal rulings or risky medical decisions need some sort of justification before action is taken.

Sometimes transparency is required when making decisions. Or maybe we just need to generate a range of different options.

However, there are some possible solutions. Perhaps a neural network AI cannot tell us how it decides something. But we can give it some operating rules. These could be like the metal cages that shielded production workers from the uncertain movements of early industrial robots. As long as a person did not move into the volume that the robot could move through then they would be safe.

Like safe paces to cross a road. Operating rules would be like rules of warfare, ground rules, policy and safety guidelines. Structures that limited the extent of decisions when the details of why the decisions are made are not known.

A similar idea is to test the AI to understand the structure of what sort of decisions it might make. Sort of the reverse of the first idea. You could use one AI to test another but feeding it huge numbers of problems to get a feel for the responses that it would provide.

Another idea is to work the AI in reverse to get an indication of how it operates. Like this picture of an antelope generated by Google’s Deep Dream AI.

The antelope image that was generated by the AI shows a little about how the AI software considers to be separate objects in the original picture.

For example, the AI recognises that both antelopes are separate from their background – although the horns on the right hand antelope seem to extend and merge into the background.

Also, there is a small vertical line between the legs of the left hand antelope. This seems to be an artefact of the AI software rather than a part of the original photo. And knowing biases like that helps us to understand what an AI might do even if we do not know why.

But whatever the eventual solution, the fact that some AIs lose marks for not showing their working out highlights that there are many different types of AIs and they each have their strengths and weaknesses.

A blueprint for your Internet of Things ecosystem

blueprintpexels-photo-239886

My new article on Linkedin Pulse suggests how firms that make IoT products and services could build their own IoT ecosystems and how they could persuade other firms to join.

How the car industry is building Internet of Things ecosystems.

DOG IN CAR 1pexels-photo-236452

My new article on Linkedin Pulse looks at the implications of how the sensors on modern cars can tell manufactures where you drive, how you drive and when it might be about to breakdown. This data could be a massive help to car repair garages, insurers and other firms. But manufactures might want to charge for it and this might increase the cost of servicing your car. So how will firms partner together?

My research suggests strategies for building Internet of Things (IoT) ecosystems in the car industry and other IoT product industries.

 

Smart contracts using blockchains could help IoT devices to share data appropriately and securely.

The success of the Internet of Things (IoT) depends on sharing data between different devices but it needs to be done appropriately. But there are some brand new technologies that might support appropriate data sharing and they can do it on a massive scale. 

Schengen Agreement

Sharing (data) is good for us. The more we share then the more perspectives we have on any given consumer or any given problem. In retail this means getting a better understanding of what customers need even before they do. For the IoT this means combining the sensors and abilities of many devices to get a fuller picture of what’s needed and having more options on how to do it.

Firms that manufacture IoT products only get direct data from the customers that use their products. One device only ‘experiences’ a part of a customer’s life. To get a fuller understanding of a single customer’s life different devices need to share data, preferably devices that help the same customers in different ways.

The classic example is a shopping app on your phone that uses data from your refrigerator to suggest recipes based on what you actually have in stock. Or a firm that uses data from all the devices that it has sold in the past to give advice to specific customers, like Waze uses crowd sourced data on traffic jams to warn its users.

But really useful data can come from external sources.  

Just like good feedback and useful suggestions, external information is more likely to give you a new perspective.  So devices that share data with many other devices will tend to get more useful information and make better decisions.  Decisions that are either more personalised, because of complementary information about the same user, or decisions that are chosen from more options, because of information on more user situations.

However, the problem is that firms don’t like to share with other firms for commercial reasons. More importantly, firms can’t just share customer data with any firm that they want to because of data protection legislation. How would you feel if your data was shared with anyone and everyone?

Of course you need some physical way of sharing the data, like Temboo’s stack or a platform like EVRYTHNG. But you also need some rules too.

The problem is that sharing data makes for better user services but sharing data must be done appropriately.

I’ve blogged about this before here, here and here. But now I’ve figured out how to do it using new technologies like smart contracts and blockchains.

Here’s the solution: if you want to share your data, meaning sell it or even just release the value in it for free, then first you need to stop talking about ‘data’. Be more specific. Being very specific is the key to reducing data sharing sensitivities.

If you ask me whether you can share all my data with everyone, everywhere and let them use it as they choose and pass it on to whoever they want to, forever. Then I’m going to get a bit tense.

But if you tell me exactly how you want to use it, why you need to use it, who else might get a look at it and what they are entitled to do with – either the data or the products of the data – and how long this is for. Then I will be much more open to your suggestions.

I still won’t give you my medical data so that your algorithm can charge me a higher insurance premium. And I hate it when it feels like my data is being used to sense when I will pay more for dodgy science fiction books on my Kindle or more expensive airline tickets.

But generally speaking a highly specific definition use of will remove a lot of my reservations. And real anonymisation will remove all of them. Q: who owns this anonymous data? A: dunno.

So how can we apply this to the IoT one devices at a time?

It is relatively easy to run a loyalty card programme because it is centralised and largely automated. But the IoT needs to be decentralised. If every device asks you or its manufacturer for permissions and instructions every time it needs data – or when is asked to supply data – then the IoT will not get very sophisticated.

So how do we ensure highly specific data sharing? Well, we need to design smartness into the IoT. So let’s use smart contracts.

Smart contracts are a mixture of contract law and software. The rules that govern a relationship between two parties are embedded in the software that facilitates the same relationship. To put it another way, the contractual terms are embedded in the thing that they govern. So they control it to help ensure that agreements are kept.

Software is made of rules and instructions for use but contracts also need proofs of things like identity, ownership, endorsements, permissions and rights to do things. And they need to be cryptographically secured, decentralised and machine readable so that every device can automatically ‘negotiate’ with other devices as required.

Fortunately all this can be written into blockchains, the distributed and cryptographically encoded databases that underlie digital currencies. In principle, the data sharing rules for each device can be broadly pre-specified; for consuming third-party data as well for supplying data.

Smart contracts could automate a multitude of interactions between IoT devices and pre-specify what constitutes appropriate behaviour in different circumstances.  Automation is important because the more devices the better and the more granular the interaction then the smoother the service. Pre-specification is important because you do not know which devices your devices will meet.

Each IoT device has it’s own ‘umwelt’ – a self-world – and combining umwelts is the key to successful IoT services.

The self-world of an IoT device is made up of all that it senses and all that it can do. All its sensor data and all its capabilities to change the physical world. Each device has different sensors and effectors, so many diverse devices working together can help each other.

The self-world of a single IoT device is very limited, like the self-world of a tick

Jakob von Uexküll’s idea of an ‘umwelt’ is really useful for understanding the Internet of Things (IoT), either for building a single IoT device, or an IoT app, or for building a whole IoT ecosystem.

If you want to design a single IoT device or a single IoT app then you need to know how it fits into one or more ecosystems. It cannot exist on it’s own. An IoT device on its own is just a device. The potential of the IoT lies in the combined capabilities of many devices working together.

If you want to design an IoT ecosystem – maybe because you want to build a platform, a network or something to help lots of people – then you need to understand how the devices in the ecosystem can help each other.

Whether you are wondering how an individual device can work with other devices or how can many devices all work together, then the umwelt idea helps to answer both these questions.

What exactly is an umwelt?

An umwelt is the ‘self-world’ of a machine, a person or an animal. It is a combination of all that it senses by plus all that it can do to change it self-world.

For example, a female wood tick hangs in a bush waiting for a deer or other prey. When it senses the butyric acid produced by all mammals, it lets go of its perch. If it lands on some fur this impact is the input trigger that makes the tick scurry around. If it then senses a warm membrane, like skin, then that input triggers piecing and sucking actions.

The tick’s self-world has three input signals and three output action. Its sensors are so limited that any warm membrane will trigger piecing and sucking. A rubber sheet holding warm glycerine will give the same input signal and generate the same action as skin.

Humans can see that ticks usually pierce and suck mammals’ blood through skin but the ticks’ actual sensors and are more limited that what humans see and understand. Ticks have a very different umwelt to humans, just as every IoT device has its own special sensors and effectors.

Von Uexkull also helps us to understand how every IoT device has its own perspective. He wrote about the ‘magic journey’ of animals. For example, ticks can survive for many years just hanging in wait for the scent of butyric acid; some birds migrate each year from pole to pole; and some insects just move from one end of a cereal gran to another.

The umwelts of animals are different because of their different sensors and effectors. And what they sense and do is strung together into the very different journeys of their lives. So the magic journeys of animals are hugely different in terms of timescale and distance. This gives each species a very different perspective on the same events.

Devices and applications have different sensors and effectors – different umwelts.

What devices and applications sense is subjective and specific to each one. For example, a phone might have special information because it monitors a particular person’s physical activity levels. That phone has the right sensors – accelerometers and a clock – and only that one is in the right place at the right time to record that user.

Also, the things different devices can do are subjective and specific to them because they all have different capabilities. Phones use screens, audio speakers, vibrators and other effectors to influence you. Cars do this as well but they can also move you. Websites can inform and guide you. And rowing machines can simulate different water conditions or just say when your exercise time has finished.

Devices and applications have different sensors and effectors. So if devices work with other devices they can get access to different information and different ways to help a user.

The secret to successful IoT services

No single device has enough information to help you with anything but the simplest of problems. For example, apps are usually highly specific in what they are for and they usually need you to supply most of the information. And no single device has enough capabilities to guide you using multiple ‘touch points’ and to deal with most of the problems itself.

For example, Sat Navs and GPS apps are best when they integrate lots of data sources and they are indispensable when then they can actually do something about the different problems that come up. A sensible shopper commonly looks at several different retailers’ websites using several different devices, plus some in-store checking, in their shopping journey to buy a high value item.

Bigger problems are solved by solving smaller problems one after the other in a sort of ‘journey’. And each smaller problem requires different information and different capabilities. The more complicated the problem is then the more complicated is the service that it solves. But a single device is too limited on its own. It can only know about its own self-world and it can only change it’s own self-world in a small number of ways.

Real-word problems are complicated. The more that devices can combine their information-sensing capabilities and their abilities the change the real world, then the more sophisticated are the IoT services that they can jointly produce.

The IoT offers the potential for ‘personal Sat Navs’ that use information from a network of sources and that employ a variety of ways to smooth and guide each user’s journey. Journeys in shopping, travelling, education, recreation and work – IoT services can solve the collections of serial problems that we call life.

If you want to design a single IoT device or a single IoT app then think about what extra data you need and what extra capabilities would complement whatever your device or app can do on its own. How do these change along the course of each user’s journey?

If you want to design an IoT ecosystem then think about the mix of data and capabilities that you have access to. Do the devices that produce them work smoothly together and how do they combine to fit the needs of the different users? Including the needs of the devices themselves.

IoT paranoia: can your devices trust the other IoT devices that help them?

For the Internet of Things (IoT) to function well then lots of IoT devices need to work together properly. But how can these other devices be trusted? Blockchain technologies might be the answer.

Lego block and chain a

The idea of any device with a chip and a web connection working with any other device has lots of potential. At home your freezer could partner with your cooker to swap ideas for meals and precise cooking instructions that are based on what is actually in your freezer and how your cooker best heats up food.

At work your office devices could work together to help you and your colleagues.  Your car could be operated by your phone and vice versa. The Internet Of Everything means any device could potentially work very closely with any other device.

The trust problem

But what happens if one of these devices has been hacked? Maybe your phone has a virus. Or if you need your car to communicate with someone else’s car – to organise routes or a place to meet up – then maybe the other car will infect yours? So how can we make the IoT secure?

The problem of ‘trust’ is bigger than just avoiding infection. How can you trust a device that you do not own or control? Maybe a toll booth will charge you incorrectly, maybe a person will use a phone app to pretend to be someone else. Maybe an IoT device will just give bad information. The cause might not even be hacking, maybe just a software bug, human error or biases caused by differences in peoples’ taste or perspective.  How can all these be avoided?

The coordination problem

Also, whatever your IoT project is about, the issue is more than just about whether you can trust some device. There is also a coordination problem. The potential of an IoT ecosystem lies in the combined sensors and capabilities of many devices. For example, many cars sharing congestion information is much more valuable than the information from just two cars. So the problem is to get many devices to work appropriately with many other devices. Trust and appropriate behaviour need to be guaranteed for all the devices that work on a particular problem or service.

It all boils down to two problems. How can you trust your own devices not to be hacked and how can you trust devices that you do not control to do what they are supposed to do – in a joined-up way?

How can we be sure of the identities, past behaviours and current permissions of other people and devices? How can we coordinate many devices so that they work in a joined-up way? And how can our devices do this checking automatically?

Fortunately a similar problem has already come up with digital currencies like Bitcoin. Digital currencies also need to be trusted and they need many people to join-up in agreeing that a particular buyer owns the digital cash to be able to pay for something. Buyers want to choose from lots of sellers so all the sellers have to agree that they trust the currency – even though the amount of cash each buyer and seller owns changes each time someone buys something.

Cryptocurrencies, like Bitcoin, solve this problem by using a distributed ledger like blockchain . A distributed ledger is a database of transactions that is shared and checked across many computers. And transactions can be money transfers or they can be IoT devices sharing data with each other.

Solution: You are what you do and what you do can be recorded

Blockchain technology is much more than the foundation of Bitcoin. Transaction data can include peoples’ identities, devices’ identities or any other useful data such as how they behave. Recording transactions makes it possible to know the real identity of every device and what it has done in the past.

New transactions are cryptographically recorded into blocks and each block in the chain of blocks is cryptographically linked to the previous block to stop tampering. The data in each block is encoded and part of that code is based on the contents of the previous block. To successfully tamper with this record would mean hacking all the computers in the network simultaneously, whilst at the same time guessing how to decrypt each block and the links between them.

Blockchain technology preserves trust in three ways: multiple copies of a single blockchain are shared and continuously checked; the data in each block is encrypted and to decode one block you need to decode the preceding blocks.

So, the problem of IoT devices trusting each other could be solved by using a blockchain technology to encrypt recordings of past behaviours and current permissions. And the problem of coordinating many devices could be solved by sharing a single blockchain database across a network of devices.

Of course there are many distributed ledger technologies and many types of blockchains. My point is that technologies like blockchain solve the trust problem by continuously checking multiple copies of a database that is securely synced.  And automatically synchronising multiple copies of the same database is also a strong basis for coordinating multiple devices.

Trust and coordination are two things that IoT ecosystems will badly need. So blockchain technology looks like it will be a foundation of the IoT ecosystem and much more.

What is an IoT ecosystem and how does it work?

Internet of Things devices work better together, the more IoT devices that link up the better. So which devices and which apps should yours connect to?

Natural ecosystems can tell us ideas for building Internet of Things ecosystems

In the Internet of Things (IoT), the more devices that connect with each other then the more perspectives and capabilities there are to be shared. More data from different sensors and data suppliers; and more ways to change the real world. Like operating cars, home appliances and other machines or getting really useful options from screens or bots.

If you are a device manufacture or an app developer the problem then is: which devices and apps should yours connect to? If you can potentially link to any device and any app then which are most appropriate?

How do you avoid confusing your users when they use your product? How do you avoid confusing yourself? What to connect to is not a problem for the user. The device manufacture or the app developer needs to figure this one out. Just give users a simple list of high quality options that are personalised to their current situation.

Your product cannot link to every other device on the Internet. So which devices have the most useful perspectives and capabilities? You need a strategy that helps your product to be better at its purpose.

And don’t forget security.

Next, having chosen which other potential devices or apps should work with your product you then need to persuade their makers to partner with you. There might be an API to help you connect but close data sharing and brand associations needs discussions and agreements. And that means you need to get noticed, get taken seriously and get a mutually beneficial deal.

The prize is that the first products to build up their IoT ecosystem of partners will get more data and features to build into better services. As my son knows very well, a bigger and more varied pile of Lego bricks means he can build a more interesting spaceship or a more secretive secret base.

There is a lot of talk about business ecosystems and an ecosystem of IoT devices is a lovely thought in principle, but what actually is it and how do you build one? Looking at natural ecosystems might help us.

Natural ecosystems are glued together by ‘nutrient pathways’

The glue that binds together natural ecosystems, like rain forests, deserts and even a single puddle of water is their nutrient pathways.

What we think of as natural ecosystems are actually the ‘pathways’ that recycle scarce resources. The essence of natural ecosystems are nutrient flows along pathways which are based on the natural activities of many different organisms.

Whatever the ecosystem, the quality that makes a natural ecosystem stand out; the thing that makes people say ‘that collection of organisms and stuff is an ecosystem’ is how it moves resources around itself. Microbes, insects, larger animals and plants and other living things move resources around just by living their lives.

The animals, plants and other organisms can come and go, die off or just move to another ecosystem. The pathways need not be dependent any particular organism or even a single species. But the thing that makes an ecosystem appear to us as an ecosystem is the way it recycles scarce resources.

For example, rain forests actually have relatively few nutrients, the soils are very poor. When leaves fall to the ground they are broken up by tiny organisms. Then the nutrients are absorbed by fungi and quickly recycled back into the trees by their roots.

Recycling and reusing nutrients along specific pathways is what makes one natural ecosystem different to another. Different organisms have different ‘roles’ in the pathways and each role might be performed by several different species.

Business ecosystem pathways glue together the IoT ecosystem

If pathways that recycle scarce resources are the essence of ecosystems then what are the scarce resources that business ecosystems can recycle?

The scarcest resource for most businesses is customer knowledge. Customer knowledge about the situation any individual customer is in at the exact moment when they use your product; and knowledge about how all customers have used the product in different ways and in different situations.

Knowing the situation which an individual customer is in as they are using the product enables the product to be more responsive to the customer. And it enables the customer to get better advice and suggestions for using the product.

Learning about how all customers have used the product in different ways and in different situations helps a firm to improve the design of the product with software upgrades or with hardware redesigns. Or it helps to suggest solutions to common problems that customers find as they use the product. These solutions can even be suggested to customers by the product itself.

For example, Sat Navs make travel route suggestions and cooking apps make recipe suggestions. Knowing more about the bigger picture of the users life – the reason for the journey or the reason for the meal – would suggest more personalised options. Knowing what other users have chosen in similar situations would help generate more options as well as a more accurate link between a suggested option and a given situation.

This sort of information was scarce before devices connected to the Internet because the direct relationship with users was mainly with retailers rather than product manufacturers. Also, an Internet connection enables products to record how they are used and then to send this information back to their manufacturer.

Product usage information can be combined with information from different products and other information about users’ lives. A deep understanding of the wider situation that a product is used in helps it to be used more successfully.

The IoT technology stack is a good way of explaining how smart products can connect up and share data. But how do you build ecosystem’s pathways?

Building an IoT ecosystem by choosing devices to partner with

To start building your ecosystem, first ask ‘What customer knowledge do you need to make using your product more successful as it is used and also as you design and (re)design your Minimum Viable Product?’ Do this for every stage of your users’ journeys.

Next you need to choose the data suppliers who can share the data you need to manufacture this customer knowledge. The data suppliers who you partner with (the devices, apps and other sources) will be the components of your ecosystem pathways. The order in which they work together is the flow plan of the pathways.

And how do you persuade them to do it? Just explain to them how it all works using the logic behind your flow plan of ecosystem pathways. Your flow plan describes how each device or app plays its small part in the wider scheme of your ecosystem’s work just by doing its job.

Each device or app has a job to do, its role.  So your flow plan of ecosystem pathways is also the business model of why your new ecosystem will work.