Recently FOAM have announced the first release of their product later this year - “a permission-less and autonomous network of radios that offer secure location services through time synchronization, independent of centralized sources like GPS” (source). Yet another side of FOAM’s endeavor has had more of my attention over time - FOAM Lite. FOAM Lite is the idea of using FOAM nodes to provide direct blockchain connectivity to low power devices over the LoRa radio standard. I think this a powerful idea that is a subset of general applicability of blockchain in industrial settings. In this essay I will first focus on blockchain as a communication protocol for industrial entities in general and then attempt to extend this argument onto FOAM Lite and sensor data.

In Clifford Simak’s 1953 novel “Ring Around the Sun” highly-evolved humans gain access to a series of pristine Earths situated in parallel dimensions. On one such Earth those few who have started colonizing these spaces have constructed factories producing extremely high quality items for every day use but without any human workers. The everlasting light bulbs and razor blades exported from those factories wreak havoc to the financial and social systems on our more mundane Earth and push humanity towards a new stage development.

Being an exploration of the strain that ever-accelerating automation has been putting on real-world structures, in the self-running extra-dimensional factories the book paints a vision of perfect automation, production of necessities devoid of the human hand.

It seems then that automation was on the mind of 1950s writer just as it is on our contemporary minds. Leaving the significant political implications of automation aside, the striking thing to me is our general failure to automate. Now, almost 70 years since the publication of the book, presidential campaigns are won and lost over promises to “bring jobs back” to poor areas in otherwise highly developed countries. (The irony of fighting for your constituency’s access to relatively un-sexy jobs as wealth inequality is at an all time high is not lost on me but that’s beyond the scope here). That there are jobs to “bring back” is what contrasts so starkly with the vision of the science fiction writer of mid-20th century. Why we are not further along the automation line is a complex question that I could not claim to have an answer to, yet the emergence of blockchain has brought to public scrutiny considerable inefficiencies in what might appear to be a well-oiled system of logistics and manufacturing.

What is blockchain good for?

Over the past couple of years more people started seeing how blockchain could be a non-obvious answer to long standing inefficiencies. Some of the people arguing for blockchain’s utility beyond monetary applications are Paul Brody (@pbrody) and John Wolpert (@jwolpert), both of IBM at some point but now leading EY Global Blockchain and Baseline Protocol respectively.

Their argument goes something like this:

Much of inter-company communication runs on FAX, PDFs and Excel spreadsheets. In today’s extremely complicated web of business relationships where, for example, parts for some products need to be sourced from hundreds of manufacturers world-wide this leads to a tremendous amount of inefficiency. Adjudication cause by miscommunication alone accounts for billions of dollars of waste. The reasons for this inefficiency are:

• The technological problem of lack of communication protocols and infrastructure to facilitate it.
• The political problem of lack of trust which prevents companies from adopting middlemen platforms to facilitate the communication.

The technological strength of blockchain

In order for Simak’s cross-dimensional factory to exist it needs to communicate with many entities: providers of electricity, delivery transports, sources for parts, raw materials, machinery within the factory itself and sensors. Lets use procurement of parts as an example. What would it take to automate? If you are a software architect you know how much time is spent designing and implementing communication protocols:

• Design specs for messages passed between entities.
• Account for failure scenarios where the other system is unreachable.
• Provision discovery method so that entities know how to find each other by IP or domain.
• Determine and implement an authentication method and grant proper permissions to counter-parties.
• Communicate above specs to maintainers of the other entity hoping that they implement it correctly.
• Provision servers to run your system with proper runtime.
• Work on points of diligence like monitoring.

Now, say, they talked to each other by sending transactions (for procurement requests for example) to a smart contract. Enter the smart contract as communication protocol:

• Always on.
• Built-in authentication.
• If blockchain accepts the message it is guaranteed to have been “delivered” as both entities share the same data which is also immutable.
• The blockchain is your infrastructure, just pay for the transactions.
• The smart-contract code is the protocol documentation.
• Private with zero-knowledge technology.
• Easily composable, akin to the “money legos” of the decentralized finance.

So for a procurement example the flow would look something like:

A minimal system to send a transaction to where the supplier can read it using a smart contract would look like:

One would still need to agree on some rules but the the process revolves around simply signing transaction to interact with a smart contract - well understood common primitive, PoC for which could be hacked in day. (This is, by the way, what Baseline Protocol aims to standardize. For a much more complex example take a look at an implementation of private auction protocol for IoT devices here.)

The political

Middlemen with network effects tend to grow and monopolize their domain. Social media platforms are a good example. And while one could probably relatively painlessly leave a social media platform, for a company to put essential parts of its business into the hands of some unpredictable rent-seeking behemoth is unacceptable. According to Paul Brody this is the reason why EY needs to look through tons of Excel spreadsheets every year. This would not be an issue if the middleman were to be a neutral public blockchain instead of a Facebook of supply chain.

The booming “decentralized finance” system has coined a new concept - money legos. It refers to unprecedented composability of financial instruments on-chain once some primitives such as ERC-20 tokens have been agreed upon. As such standards for industrial communication are developed we might expect the emergence of “industry legos”.

Sensors and the blockchain

A buzz-word in industrial circles in the past few years has been “industry 4.0”, which is described by IBM as being an industry "characterized by increasing automation and the employment of smart machines and smart factories, informed data helps to produce goods more efficiently and productively across the value chain.” In particular this supposed new industry consists of four building blocks:

• Interconnection - the ability of machines and system to communicate with each other and with people.
• Information transparency - the sharing of information that makes interconnection possible.
• Decentralized decisions - allowing systems to make decisions on their own.
• Technical assistance - use of machines for informing human decisions.

And I suspect that light bulbs are going off already in the heads of those readers who have been steeped in the blockchain ethos, as at least three of the characteristics above use the same parlance they are well familiar with.

Industry 4.0 has elevated network-connected devices and sensors in particular on a pedestal they have not enjoyed before.

The sensor architecture

The architecture around collecting data from sensors (and acting upon it) has some of the same issues that we have seen with general inter-system communication. To make this easier to illustrate lets take as an example Net Feasa, an Irish company that specializes in making sensors for the shipping industry. The sensors add trace-ability to cargo by relaying via radio (LoRa, LTE and Wifi, depending on where they are) information about container location and door access. (An interesting side is that some of their sensors are able to power themselves by scavenging environmental energy.)

Different environments where data about containers needs to be collected and the issues with data siloing that come with it, according to Net Feasa.

The issues with how such a system works are well-described in FOAM Lite docs and are quiet similar to issues outlined in general case but lets recap.

![The complexity of running sensors on a private network. Protocol between each layer needs to be designed for the cause. No industry legos in
there.

• If one is is looking to communicate with a sensor you need to either provide your own infrastructure for every step, which is very complicated, or rely on a 3rd party such as Net Feasa.
• By relying on a third party one gives up composability and entrusts data to a closed proprietary system. (If Net Feasa went bankrupt tomorrow then the whole endeavor of tracking containers would experience a major setback).
• If you would like to set infrastructure up yourself (for your own use or as a service provider) then all of the communication protocols need to be defined, both, for internal use and to interact with external systems.

While Net Feasa does solve data siloing issues across different shipment stage it itself is a silo. The blockchain in this case acts as an “anti-silo” - by not belonging to any one party it gives the user a more direct access or even full ownership of their data and encourages lego-like composability. For example, and lets get back to our extra-dimensional factory, a message from a sensor could directly trigger an order for delivery of a certain part, or the sensor data could be harvested for AI analysis, or trigger and emergency shut-off system all with a few lines of code instead of layers of infrastructure.

Aside: Scalability

“But Ethereum transactions cost tens of dollars!” One might say. And this is where Ethereum’s roll-up-centric roadmap shines. Roll-ups, using the main chain only for providing security, could be designed with task-specific traits, such as privacy, high data throughput, specialized VMs, etc. So not only scalability is unlikely to be a problem with roll-ups but they could be made to accommodate the specifics of application.

Final thoughts

We are very early and my words are full of futuristic excitement, yet they are hardly utopian. Focus in blockchain thought has been skewed heavily towards financial usage while overlooking a world of utility that is only now getting discovered. If one views blockchain as open and neutral infrastructure for communication and execution of logic, and a toolbox of standardized primitives then under the hype of value-exchange one discovers a friction-less layer for machine communication the futuristic power of which is bound transform how our industrialized world operates.