At first glance, two recently coined terms in the data industry seem ready for intuitive understanding: data mesh and data fabric. What a relief for an otherwise inscrutable lexicon. But it’s not to be.
The everyday English speaker would assume that “mesh” is an unrefined predecessor of “fabric.” Perhaps data fabric would be raw, and data mesh would be cooked. Salmon barely off the hook versus salmon barely off the grill.
Sadly, each term’s actual meaning is the opposite.
“Fabric's about sending raw transactions or raw data,” said Evan Levy, a longtime consultant who’s renowned in the data industry for helping executives grasp tech. “not send me a salmon dinner.’”
Most business people, he explained, don't need to look back into their data’s origins. They use the data they’re served. “I ordered a salmon dinner. I don't know how many people are on the other side cooking the salmon, fishing for the salmon, prepping and cleaning the salmon, and adding the green beans and baked potato.” He orders mesh, not fabric.
It’s as if “mesh” and “fabric” were swapped at birth. Why did it all come out so wrong? Just think, the data industry finally came up with names that seem to clue business people into their meaning and instead it all comes out backwards.
Tech terms aren’t usually made for lay people, Evan explains. They were meant for the data geeks. We can only guess why they find such puzzles satisfying.