Wilden sued several employees and their startup JDA, alleging trademark infringement of its numeric codes for pump replacement parts. JDA argued that the part numbers were functional in that they were formulated to communicate specific part information. “Wilden assigns part codes according to the following scheme: XX–YYYY–ZZ where XX identifies the size of the pump, YYYY identifies the part itself, and ZZ identifies the material from which the part is made.” According to JDA, the aftermarket industry used Wilden’s part codes for years without objection from Wilden (which would suggest at least lack of secondary meaning, if not genericity), using a prefix to identify the aftermarket manufacturer. Wilden argued that the use of a prefix was the market recognizing that the part numbers themselves identified Wilden products, but the court saw that as a fatal concession: if a prefix distinguishes the aftermarket product from a Wilden product, “then it can hardly be said that Defendants are guilty of trademark infringement.”
But more significantly, while the part numbers might represent Wilden products, they didn’t serve as source identifiers. Instead, it was the prefix that designated the source, while the numbers represented the product with which the part was compatible. “As relevant here, then, the part numbers are not source identifiers, but rather, compatibility indicators.” Wilden simply didn’t deal with the industry practice of using prefixes, which it did in its own aftermarket parts business when it used competitors’ part codes, stating: “For ease of ordering all you need to do is find the OEM part number and add an ‘S’ in front. To distinguish between an OEM part and a Spectrom part simply look for the logo ... on the part.” The part numbers weren’t source identifiers; rather, the prefixes were.