Interoperability in digital systems is important period.happens to be the latest and greatest recent breakthrough in that, so it applies equally as well. Interoperability in any software system has become the de facto standard. What makes software quality software, is that it works with other software. In software terms, we’ve been looking at the API economy, we’ve been throwing out statistics around mobile computing that every mobile application uses no less than 12 distinct mobile services and that’s all about interoperability. If you build a mobile application today that is going to make a payment and then text someone that the payment has been made, that means your mobile app has to interoperate with PayPal and interoperate with Twilio to send the text.
So first and foremost, interoperability for blockchain is “Don’t forget the last two decades in computer science and what we’ve achieved”, which is that every blockchain network needs to have a good API that is swaggerized and that has a well-documented interface. That will go a very long way to blockchain interoperability. Nothing new here folks or, I should say, we should do what we’ve been doing and blockchain networks will greatly benefit.
Maybe that’s the 80% solution to interoperability. The next 19% or so is covered through having standard data models. The most obvious is around digital identity, that networks will likely interoperate when the people, places, and assets can communicate more universally. Thehas specifications like DIDS, digital identifiers and project and project under are all looking at codifying these standards. I think that starts to take care of the remaining 19%. So, we’re at 99%, the last 1% I think is the more difficult one, which is ledger to ledger interoperability. Of course, all of the press talks about that 1%, it’s like “How do blockchains interoperate?” and they’re really talking about how does ledger A talk to ledger B.
I think that there are some use cases where that’s really important, and there’s projects like Project Quorum out there that are looking at more ledger to ledger interchanging. For the computer scientists out there, think about when the problem of two-phase commit had to be introduced when a database was moving data over to another database and how the databases had to be locked so that the asset could be moved and only moved once, not duplicated and not removed completely from both. I think that’s the harder computer science problem to solve. It’s been solved before but we’re going to focus on solving the first 99%, then, where those cases matter, solve that part as well. But interoperability is and will remain key.
How many networks will there be at the end? Will there be one network to rule them all? Probably not. Will there be a million networks? Probably not. I believe the number is someplace in between which implies interoperability.