A critical problem with tokens and the industry-wide solution to solve it!
You are already familiar with tokens, whether you realize it or not
The topic of tokens can be mystifying when couched in the language of cryptography, blockchains, and that sort of technological talk. But understanding the value and purpose of tokens is actually quite straightforward — a token is just a representation of, or substitute for, something of real value. The value might be physical (a bar of gold), digital (a share of stock — most exchanges are now fully dematerialized), a promise (a seat on an airplane), a proof (college degree), an authorization (access to medical records) … there are many items of value that are represented by tokens today. In order for a token to be a valid substitute for the item of value it must be nearly impossible to fake, and it must have an owner (whether or not ownership can be passed is a function of a particular token — more on that later).
You already use tokens every day! If you own a house, the value of that house is represented by a property deed recorded in the clerk’s office. Paper currency represents the value of a specific amount of money. Checks (does anyone still write checks?) are another form of fiduciary token. The certified value of a diamond may be represented by a certificate from a gemological institute. UPC codes are an interesting form of token — a UPC represents a particular product type but can be reproduced across hundreds or even thousands of copies of that product (each UPC ‘token’ being equivalent in value — ‘fungible’). Seats on plans and in theaters are represented by tickets. Degrees and certificates represent demonstrated knowledge in particular areas. Your passwords are tokens that give you access to the value or personal information in your accounts.
What is the difference between a digitized and digital token?
Tokens are very familiar indeed. And many of them are, in fact, already digitized today. As mentioned earlier, dematerialized stock markets don’t issue certificates anymore — your ownership is a digital record that is no longer backed by a physical certificate. Many forms of physical tickets have been replaced by Q-codes on your smart device. But these digital representations are only that — representations. They do not encode behaviors. For example, if you want to sell your share of stock, you still have to work through a broker and a whole series of traditional transactions must be set into motion to match you up with a buyer, obtain the agreed-upon equivalent in funds, and move those funds to your account. Sometimes these processes are still very manual.
The true value of digital tokens is the ability to encode all of the behaviors and rules directly into the digital token itself in order to greatly simplify, unify, and streamline the use of these tokens. In such a world, for example, if I want to sell my share of stock I would invoke a ‘sell’ method on the stock token, which would then directly manage the work of exchanging my share for the agreed-upon equivalent in financial tokens (written in software called smart contracts) — which I could then exchange for fiat currency if I wished.
The goal here isn’t so much about removing intermediaries in transactions (although that can happen); it’s about having all parties in a token-based transaction share a mutualized, decentralized, and trusted infrastructure for the execution of that transaction.
It’s fairly straightforward to come up with a definition of digital tokens in this context. A digital token would reside on a shared, distributed ledger — the mutualized infrastructure. It doesn’t really matter if the underpinning of that ledger is a blockchain, a database, or something else as long as all parties agree that what they see on that ledger is truthful and immutable. Cryptography is the technology used to secure ownership, ensure authenticity, and prevent tampering with digital tokens — fraud-proofing is a requirement of tokens in general. And all of the behaviors and rules that govern what can be done with a token, and by who, are agreed-to in advance by all parties and baked into the shared decentralized ledger (via smart contracts).
Many digital tokens that fit this description exist today. But the world of tokens is far from optimal.
The token tower of Babel
The problem is there is no common language for defining the properties, behaviors, and control messages associated with tokens. Sure, there are emerging standards for how token contracts need to interact with their mutualized infrastructure — ERC-20, for example, is a well-known standard for describing the basic interfaces a token must have for running on the Ethereum network. But describing all of the myriad rules of behavior for ERC-20 tokens is totally up to the developer. As of April 16, 2019, there were more than 181,000 ERC-20-compatible tokens on the Ethereum mainnet. They all implement the six functions required by ERC-20, but nothing else is standardized. And that’s just the developer view; how do you help a businessperson, consumer, auditor, or a regulator, understand what these tokens are actually doing without a common language? The current language around tokens is, unfortunately, steeped in blockchain terminology — understandable since digital tokens first gained mass popularity on public blockchain networks, but this linkage catastrophically adds to the confusion that developers and business people attempting to develop tokens for their use case have when trying to parse the token ecosystem.
If you have no common means of defining and triggering token behaviors, how then can you begin to write contracts that trade tokens of different types produced by different development teams? Let’s say, for example, that I want to introduce a token for seating at a sporting event, and I’d like the purchase contract to be able to accept a hypothetical dollar token as payment. Not too much trouble to work that out. But what if I represent 10 such venues, each with its own independently developed seating token, and want to be able to accept 15 different payment tokens and cryptocurrencies from various platforms? Without common controls, each pairwise combination of venue and payment token would require a unique contract — that’s 150 contracts in this case. And if I want to be really flexible and accept combinations of payments for a seat, the number of contracts required can increase exponentially.
Without standard controls, even simple token exchanges can require an excessively large number of contracts. And this is holding up the advancement of digital tokens in the industry.
How do we fix this problem?
In looking across the vast array of digital tokens already defined for various platforms, we find three common characteristics shared by all:
• Digital tokens are comprised of properties and behaviors (use case and instructions on how the token is used) that are interacted with using standard control messages
• All tokens have a common set of base token properties and a collection of non-behavior related properties
• Tokens differentiate themselves in the market based on the desirability of their particular combination of behaviors and property sets in that market (its use case)
With these characteristics in mind, we can greatly advance the state of the art for digital tokens by creating a common taxonomy (universal language) that clearly defines and organizes these behaviors and properties in a standardized way.
And that is, very simply put, what the Token Taxonomy Initiative is all about. The TTI is an independent, member-led organization formed to develop a modular, industry-neutral, and technology-neutral Token Taxonomy Framework (TTF) (universal language) that will set the standard for tokenizing anything of value.
Since the TTI is a member-driven organization, the TTF is being designed by a diverse set of individuals representing both business and technical expertise based on their own real-world projects and requirements. The TTF is intended to represent the best that the market has to offer to define a token for any use case, and will drive business-level interoperability amongst tokens by being neutral to the underlying technical across different networks and platforms.
The TTF achieves a level of precision by allowing someone without technical expertise to use and contribute to a repository of artifacts (ingredients), formally described in JSON. These JSON artifacts can contain links to actual code examples in a variety of languages, as well as links to business descriptions that lay out the functioning of the artifact in plain language. The availability of this repository should greatly simplify the task of defining and implementing new tokens — and it is also a great educational tool for developers designing their first tokens.
How a token taxonomy framework works
To understand better how the TTF works, it’s helpful to take a deeper look at what makes a token a token. As already mentioned, not all tokens are the same but they break down into a set of standard behaviors, properties, and controls. There are hundreds of these elements, and more being defined all the time, but to give a few small examples:
Value. The value of tokens is usually described as being fungible (where two tokens of the same type are interchangeable because they have no difference in value) or non-fungible (the value identified by a token instance is unique). Paper currency, when created (in TTF language mintable‘mintable’) — is a good example of fungible tokens since any paper currency is indistinguishable value-wise from any other of the same denomination. Land titles are non-fungible; the title to my property is completely different from the title to yours. Some tokens can also be hybrids of the two; for example, my token to a general-admission movie theater is non-fungible in terms of its date, time, and location but fungible in terms of precisely which seat I purchased.
Properties and behaviors explain how a token reacts to a control message. If a token represents a barrel of oil, and that barrel is consumed, the behavior of that token would destroy the token upon verified receipt of consumption (in TTF language burnable).‘burnable’). Conversely, a paper currency used in exchange for its equivalent in coins (TTF language ‘sub-dividabledividable’) simply passes ownership and is not destroyed.
Transfer of ownership. Some tokens like paper currency are transferrable without restriction, others cannot be transferred at all (such as airplane tickets, for security reasons).
The TTF was initially populated by looking at a wide variety of existing tokens, decomposing them, and identifying those reusable parts. From this work a composition framework was built that could contain the various types of parts; the base set of token types was defined; the known set of behaviors identified; and all of these parts neatly categorized and loaded into the framework. By design, the TTF will never be complete: as new properties, behaviors, and controls for any use case are identified, they are loaded back into the TTF for future use.
If this all still sounds confusing, let me offer a very simple analogy — baking a cake. Doing so requires a bunch of ingredients — eggs, flour, sugar, etc. — and a recipe that explains how much of each to use, when to add them, and how to control the process. If you think about the TTF as containing the ingredients (called artifacts) and recipes (called templates) — you’ve got it! It’s your responsibility to gather the artifacts/ingredients, choose the template/recipe, then bake/code your token class. Individual tokens are instances of that class, just as slices of cake are servings from the whole!
If you want your cake to have slightly different properties, such as a little more sweetness, you would adjust the recipe accordingly. If you wanted to bake a carrot cake instead of a pumpkin pie, you would use a different set of ingredients and a different recipe. Same with tokens. You have complete control over which artifacts and templates you use to create your tokens for any use case.
A token-based economy can (literally) change the world
We know how to create digital tokens today. What we don’t do very well at all is deliver a means for tokens to seamlessly interact with each other, or provide a consistent model for defining tokens. A common taxonomy and framework for token development has the potential to bring amazing new efficiencies to a wide array of today’s business interactions. But it also could form the basis of a solution to some of the world’s most intractable problems.
How about this one:
Over 1 billion people worldwide are unable to prove their identity through any recognized means. As such, they are without the protection of law, and are unable to access basic services, participate as a citizen or voter, or transact in the modern economy. Most of those affected are children and adolescents, and many are refugees, forcibly displaced, or stateless persons.
A common language for tokens (TTF) will speed this process.
Digital tokens have the potential to mitigate some of the issues with digital identity — but only if all involved entities have the ability to understand and interact with these tokens. A standard token taxonomy such as the TTF can accelerate the ability of these entities to overcome issues such as technical design patterns and interoperability standards — paving the way towards development of a private, portable, persistent, and verifiable global identification system that empowers and protects all individuals.
Let’s get it done!
The first version of the TTF is now available. For any business to contribute to the TTF they would be asked to join our TTI Initiative. Being a member of the TTI ensures your ability to be at the cutting edge of the token-based economy — either as a consumer or produce of digital tokens. Visit The first version of the TTF is now available. For any business to contribute to the TTF they would be asked to join our TTI Initiative. Being a member of the TTI ensures your ability to be at the cutting edge of the token-based economy — either as a consumer or produce of digital tokens. Visit www.tokentaxonomy.org for more information. Or send email to for more information, or send email to info@tokentaxonomy.org.
