by Serguey Shinder
Two teams owned two halves of a system that had to work together, and they did not get along. Not dramatically. There were no arguments and nothing was ever escalated. There were simply two groups with different priorities, different vocabularies for the same objects, and a shared interface between them that neither side considered fully theirs to fix.
I ended up in the middle of it more or less by accident, because I had worked on one side and then moved to the other, and I could hold both descriptions of the same thing in my head at the same time. For about a year, a meaningful portion of my job was translation. Not diplomacy. I was not smoothing anybody's feelings. I was saying things like "when they say record they mean the version after enrichment and you mean the one before", and then watching a two week disagreement dissolve in a minute.
None of this was in my job description, and I remember being uneasy about that. It did not look like engineering. There was no artifact at the end of the week I could point at. I was mildly anxious that I was drifting away from the kind of work that keeps a person employable.
I think now it was the most durable thing I did that year, and it took me a long time to understand why. The parts of my work that turn a known input into a known output have been getting steadily cheaper for my entire career, and that trend has accelerated in ways I did not predict. What has not got cheaper is noticing that two groups of intelligent people are using the same word for different things and that nobody in either room has spotted it.
That kind of knowledge is not written down anywhere, because writing it down would require already having seen it. It comes from having stood on both sides, from paying attention to how people talk rather than only to what they decide, and from being uncertain enough to ask what someone means at the exact moment you are fairly confident you already know.
I do not think it is a role that survives being formalised. The moment you appoint someone the translator, they lose the thing that made them useful, which was doing the actual work on both sides and having a stake in each. It stays informal, it stays a little invisible, and it is almost never the thing anyone gets promoted for.
But it has outlasted every framework I learned in the same period. The systems changed and the languages changed, and the gap between two teams who both sincerely believed they were talking about the same object turned out to be permanent, and permanently worth being able to see.
– Serguey Asael Shinder
Leave a Reply