A Week Arguing With a Vendor About a Tolerance Spec, and What It Taught Me About Rules Disputes

Spent most of a week back in the spring going back and forth with a vendor over a part that measured out of spec, and about halfway through the fourth email chain I had the distinctly unhelpful thought that this was, structurally, the exact same argument my game group has whenever a rulebook leaves something ambiguous.

Here’s the actual dispute, stripped of anything proprietary: a machined part came in that measured within tolerance by their gauge and out of tolerance by ours, and neither gauge was wrong, they were calibrated to slightly different reference standards that both looked reasonable on paper. Nobody was lying, nobody was incompetent, two systems that mostly agree had found the one edge case where they didn’t, and now two teams had to figure out whose interpretation actually governed the part going into production.

What finally got it resolved wasn’t some engineering insight, it was going back to the original spec document and finding the one sentence that actually settled which reference standard applied, buried in a section neither team had reread in months because everyone assumed they remembered what it said. Both sides had drifted slightly from the actual written spec toward their own team’s convention, confidently, without noticing it was drifting.

That’s exactly what happens at the table on a Tuesday when a rulebook’s ambiguous. Somebody remembers a rule one way from three games ago, somebody else remembers it differently, and both people are usually acting in good faith, they’ve just each drifted toward whatever interpretation made sense to them the last time it came up, and nobody’s gone back to actually reread the paragraph in question because everyone assumes their memory of it is fine. Marcus is annoyingly good at just pulling out his phone and finding the actual FAQ ruling instead of letting the table vote on vibes, and I used to find that mildly irritating before I realized it’s the exact discipline that would’ve saved my team three extra days of email at work.

I don’t think I’m actually saying anything profound here, just that the instinct to trust your memory of a written rule over the actual written rule is apparently a universal failure mode, whether the stakes are a board game or a supply contract. The fix is the same both places too: go back to the source document before you argue about what it says. Boring advice. Keeps being true anyway.