Interoperability & Standards In Web3

Open

I had mentioned in my last post that we should revive the slumbering Linux Foundation Decentralized Trust (#LFDT) Financial Market Special Interest Group. FMSIG. We have had many projects shepherded through the SIG, including a Standards Project. The Standards Project in FMSIG is woefully behind times. This post is about reviving that project, given the changed nature of innovation projects in Finance. This is also on the state of Standards and Interoperability in the Web3 world.

I attended an Interoperability Workshop put together by the LFDT team last week. The workshop featured a mix of participants: big investment banks, a bunch of vendors and others like me who are interested in the topic. I created the report on Interoperability released by ITU DCGI. Just to show that I have more than a nodding acquaintance with the topic. 

As a developer within MBS Risk for many years, Interoperability was a daily concern as it was a means of aggregating risk from mortgages and their hedges. Totally different kinds of instruments were unified in that risk. I also contributed to a security master project, built one for mortgages with data fed from Bloomberg, Intex and other sources.

Interoperability is important in TradFi because it increases liquidity and reduces systemic risk. Frictionless interoperability also reduces transaction costs, increasing the efficiency of markets. Now, Web3 proponents are coming to this realization as unhealthy arbitrage and loss of liquidity saps the vigor of decentralized networks with sudden unexplained price movements creating peril. When the race is on to get RWA into decentralized or decentralizable systems, the importance of Interoperability increases many fold. 

A major component of the lifeblood of Interoperability are standards. Standards make it possible to create interoperability. A. Have a definite idea of what instrument is being sent from silo to silo, traded, settled or held. B. To make sense of messages that fly across systems. 

For price-discovery, pre-trade, trading, settlement, custody, risk or just about any use case these standards can be roughly grouped into two categories. Standards connected to the dynamic definition of instruments and entities; standards connected to the Interchange of Data. Standards for bilateral contracts between buyer and seller can be thought to be a special kind of data interchange. In the dynamic category are prices which are much more dynamic and rife with errors and manipulation for crypto-currencies in particular.

For example, securities are defined using ISIN (internationally), CUSIP (in the US) and what we call the Bloomberg Name, now FIGI. Message formats such as Swift (basis for ISO20022) and FIX embody interchange standards. For contracts look to CDM. In bilateral agreements, the Master Agreement between counterparties covers much. All of these are pretty well defined. 90-95% of use cases are satisfied by this. The standard for swaps since they are one-off is covered by the CDM. CDM has been expanded recently to cover other security types. CDM is a lifecycle standard, one that covers the lifecycle of the contract (swaps are long-standing contracts with money flowing between parties multiple times). This capability is very useful for modeling all credit and fixed income instruments, such as bonds, MBS, ABS etc.

It surprised me that most of the Interoperability workshop participants, with few exceptions, had any knowledge of these standards. In spite of the existence of standards defining existing instruments as well as new instruments such as Tokens. There was a lot of hand-waving and frankly speaking dangerous attitudes such as, let us develop Interoperability and not wait nor look at any standards because there aren't any. A devil-may-care attitude masquerading as reason.

The people who say things like "Standards are great because there are so many of them" and "We will develop a new standard just like the rest of them." are taking an easy way out. Standards are being used every day (in spite of there being several on the same topic). Ease of use and regulation coalesce around a single or handful of standards as we saw in security definitions.  

Let not the existence of multiple standards for a single purpose confuse the user. Such has been the case in traditional markets since these standards developed over the years in an organic way. There are canonical methods for aligning these standards. We used to align BBG Name, CUSIP and ISIN for pricing which in turn fed margin calls. Real money flows depended on this. 

In terms of the new, there are a couple of standards covering security definitions and their dynamic nature. ISO has DTI. A draft version soon to be replaced by a final one in 2025. The arguments against the ISO standard are multiple, they are not open, just to get a copy of the standard costs CHF 65. This has not stopped the ISO standards for currencies and others, for example (the three-letter standard as in CHF) from being popular.

There is CoinMarketCap which lists all tokens and what seems like official records on every token listed there including the team, the whitepaper etc. Of course price information is pulled from exchanges around the world and after filtering and aggregating and averaging it is presented. As it is now owned by BNB, there can be questions about its integrity. Looking at data in exchange APIs, we can get an idea about what other than ephemeral data can be extracted from an exchange. A vast amount is scraped from social media and custom analytics are run on the token. Data about the token itself found in the About tab. Presumably this is the data that is invariant or just slowly changing. This is just a single data source, operating in a bespoke fashion.

For foundational standards such as Identity there is GLEIF. For others like Privacy, the messaging protocols have privacy embedded in them. For transactional privacy it will be using ZKP or other standards to prevent PII and transaction data being exposed on chain. As tooling becomes more standard, we will see more and easier adoption for non-specialists. Many chains are bringing in privacy into the chain. However privacy has to be balanced with the need to know.

So instead of hand-wringing and hand-waving let us work together on educating and yes, bringing to light a few standards that seem most promising in the new era of Digital Assets.

There is much to unpack in this post. This is stored with links on LinkedIn. Over the years, many of these (CDM, ISO20022, GLIEF for example) were presented by experts on the FMSIG or the IDSIG or both. We have lasted longer than some projects that came and went. The FMSIG is a constant in this stream of volatile ecosystem that surrounds crypto and RWA adoption.

We will keep this page updated to the latest with a short list of all standards that have been finalized relevant to Web3