Yesterday I had my first meeting as a member of the CAPE-OPEN Interop SIG.
If you do not live in process simulation, CAPE-OPEN is the open standard that lets a thermodynamic package or a unit operation written for one simulator run inside another one, without a rewrite. A model I write can plug into a big commercial simulator like Aspen Plus, or into the free COCO, because they all speak the same interfaces.
CO-LaN, the non-profit behind the standard, runs a handful of these groups. The interoperability one does what the name says: making sure the pieces actually work together once someone loads them into a real simulator. That last part is where the trouble usually hides.
This matters to a lot of software, on the industrial side and in academia. A university group spends a year on a new property model, and if it only runs in the one in-house tool, it mostly dies in a paper. Interoperability is what carries that model into the simulators engineers actually use, and carries an industrial unit operation back the other way. Good interoperability is invisible, you notice it only when it breaks, and then everyone ends up writing the same glue code again, for the tenth time.
My position in all this is a bit particular, and I think it helps. I have few eggs, and they are spread across many baskets. I am not a large vendor with a platform to defend, and no product of mine depends on one particular simulator winning the market. That leaves me relatively independent, free to look at interoperability for what it is rather than for what it would do to my own roadmap.
There is a practical side too. I build CAPE-OPEN compliant software for my customers, so I know the standard from the inside, the good parts and the sharp edges. When a job calls for it, I also integrate the old-fashioned way, through a vendor's own API, like the User Added Subroutines in PRO/II. CAPE-OPEN is not the only way a piece of code talks to a simulator, and it helps no one to pretend otherwise.
I have also been at this for 25 years. That is long enough to have watched more than one clever design turn into a maintenance tax five years later. You start to feel where a choice will hurt down the line, and, if you are in the room early enough, you can say so before the concrete sets.
And the people. I already know most of them, some through years of close work, others more at a distance. Every one of them is a pleasure to work with. SIG work is volunteer work, it is time you give away, so doing it with people you like is a must.
First meeting done. Next on my desk: a round of QA, with a bit of automation so I stop repeating the same checks by hand.