On Tuesday, May 17, WatersTechnology is going to host its first in-person conference in the US since 2019. Here’s the agenda for the North American Financial Information Summit, aka Nafis. (FYI, if you work at a bank/broker or asset manager/hedge fund, it’s free to attend. Register here.)

The event will hit on an array of issues facing the capital markets, one of which is interoperability, a topic near and dear to my heart. So if you’ll permit me, dear reader, I’m going to write about it once more.

While I was away on vacation visiting family in Raleigh (from which, I now need a second vacation that doesn’t involve family), Reb Natale wrote about how the Fintech Open Source Foundation (Finos) was looking to fully open-source implementations of the FDC3 standard. Just in case you don’t know, FDC3—which stands for Financial Desktop Connectivity and Collaboration Consortium—is a set of codified specifications for writing APIs and for messaging that enables a trader’s desktop applications to interoperate and share information.

The term “interoperability” means different things to different people. For this discussion, I’ll specifically address interop at the desktop level, though I will also hit on interoperability between systems’ back-ends. 

Version 2.0 of FDC3 will likely be released later this year, so the standard is still evolving. But one of the key aspects of FDC3 as it exists today is this: end-users implementing it are majorly reliant on mainly three vendors who act as desktop agents—OpenFin, Cosaic and Glue42. These companies have jumped on the desktop interop trend, with OpenFin having created FDC3 and contributed it to Finos in 2018.

At the core of their offerings, these vendors offer Electron-based “containers” that allow different applications inside it to “talk” to one another. On top of those, they add proprietary bells and whistles like notification management, visualization tools, and streamlined workflows.

An end-user could try and implement the standard themselves, but there’s no real way to guarantee that the solution will be FDC3-compliant and work—resulting in wasted money and resources. The vendors’ containers are tried and tested, making them better equipped to implement FDC3-based interop, especially if you’re connecting, say, an internal order management system to a third-party analytics, charting or risk platform. For all intents and purposes, they take the heavy lifting of development off banks’ and asset managers’ internal developers, freeing them up to work on a litany of other projects.

The potential issue for some end-users, however, is that their interoperable environments become reliant on a third party, and they don’t have full control over them.

Finos and its executive director, Gab Columbro, are evangelists of open source. The group, which is part of the Linux Foundation, essentially does two things: it contributes, houses and manages purely open-source projects, and they create open standards, meaning anyone can collaborate on them. Finos is an entirely public forum. Non-members can dial into meetings and offer their contributions, which will be voted upon by the relevant working group and potentially codified and incorporated.

The open-source FDC3 implementation is a way for Finos to standardize an example of what FDC3 looks like in an end-user’s computer/desktop, should they want to implement it natively.

Here’s where I go down the conspiracy-theory rabbit hole: Doesn’t this essentially cut out the middlemen, meaning OpenFin, Cosaic and Glue42? Again, these vendors are great at creating containers and doing the heavy lifting, but an open-source implementation of the FDC3 standard potentially stands to democratize desktop interoperability as the industry knows it.

From Reb’s story: The new project will offer developers and adopters a neutral reference point and the full stack of FDC3 implementation, including a desktop agent, a locally installed container that exposes the API in Electron (a free and open-source software framework), and an application directory. … “Buy side and sell side want to collaborate more often and more directly. We have a standard that allows that—but having an actual piece of code that supports that standard and allows a faster adoption of [it] seems in the best interest of the industry,” Columbro said.

Now, it should be noted that Dan Schleifer, CEO of Cosaic, said that he doesn’t think the project will undercut the vendor community, noting that those bells and whistles mentioned before is where a company like Cosaic adds value. “FDC3 is baseline interoperability, and it should be an open standard. You shouldn’t have to buy my product or anybody else’s to do basic FDC3 work,” he told Reb.

And Columbro said that they’re not trying to cut out the vendors—in fact, these vendors can benefit because the ethos of open source is all about giving back. When end-users are able to develop their own cool, new tools and open-source them, that tide raises all ships.

It’s here, though, that I’ll address the fact that Cosaic and Glue42 are members of Finos, and OpenFin is not. While an executive at the company told Reb last year that it is a “committed supporter” of Finos, they also said that the membership benefits “are primarily marketing-related.” But even though OpenFin is not a member of Finos, some OpenFin executives have actively participated in recent meetings, according to metrics tracked on GitHub. To me, that says there’s still this push-pull situation between Finos and OpenFin.

So where am I going with this? Great question. There’s desktop interoperability and interoperability of back-ends. That means that when it comes to interoperability, there’s already a split. I suppose what I’m wondering is, should this open-source iteration of FDC3 prove effective with end-users, will this essentially fork FDC3? (And if you read that story from last year, some would argue that there’s already something of a fork between what OpenFin is doing and what those working in lockstep with Finos are doing.)

One more question that’s bouncing around in my head: OK, there’s desktop and back-end interoperability—maybe this open-sourcing of FDC3 can actually be something of a glue that eventually connects the two.

Recently, Melissa Fleur Afshar, our newest reporter, wrote about how Bloomberg has connected its Asset and Investment Manager (AIM) buy-side OMS with its Portfolio & Risk Analytics (Port) portfolio management system. Matthew York, global head of buy-side OMS and compliance at Bloomberg, said that he’s a fan of the FDC3 movement, but that the industry is only “skimming the surface of where interoperability will go.”

“Interoperability is, in my view, super important, but the challenge is assuming that you can only do desktop interoperability. The current trend is making everything Finos/FDC3-compliant, but that only looks at desktop interoperability,” he said. It’s on the back-end where interoperability needs to really take off, he said.

Connecting desktop to back-end interop is a concept not only being worked on by aforementioned vendors, but also companies like Vision57, established by former Fidessa exec, Steve Grob. So this isn’t happening in a vacuum—everyone seems to understand the importance of connecting these worlds of interop.

By open-sourcing FDC3 implementations, perhaps that will create new opportunities—a new path—toward desktop and back-end interop, and perhaps in that sandbox, everyone can play nicely and no one gets cut out (as long as their tech is solid).

Here’s why I find the space of desktop interoperability so fascinating: I’m impressed by the people involved in these projects. I admire Gab’s passion for helping companies through the use of open-source tools. Mazy Dar, CEO of OpenFin, was, in my opinion, something of a visionary when he really got the ball rolling on desktop interoperability. Schleifer and the folks at both Cosaic and at Glue42 are talented technologists coming up with interesting new tools. And, quite frankly, it’s kind of fun that there’s this element of frenemy (which is actually in Merriam-Webster), where these companies are working together, while also competing against one another.

But most importantly, as I’ve written about so often before, interop (both desktop and back-end) has been such an important evolution in the world of capital markets technology. With the rise of cloud, open source, and APIs, and as firms move away from monolithic, closed-off systems, interoperability will only become more important. So an open-sourced FDC3 is an important development to watch because it could prove to be an important evolution in this critically important field of technology development.

Think I’m losing the plot? Let me know: anthony.malakian@infopro-digital.com.

The image accompanying this article is “Kitchen Utensils with Leeks, Fish, and Eggs” by Jean-Siméon Chardin, courtesy of the Cleveland Museum of Art’s open-access program.