Way back in 1996, when I was still working at the University of Washington, I wrote a little web doo-dad called YASC: Yet Another Soundex Converter. Soundex is a coding system for surnames that was used by the US Census Bureau beginning in 1880. The idea is that similar-sounding names should correspond to the same Soundex code. So for example, both Mohr and Moore receive the same code (
M-600). Soundex is deeply flawed in many ways, but I guess it served its purpose.
When I first wrote YASC, most researchers found census records the old fashioned way. It was a two-step process: first find the family you were seeking in the microfilmed index for a given census year, then use the index information to track down the actual census entry (also on microfilm). The indices were all organized by Soundex code rather than surname. Many folks figured out Soundexes by hand, working through each surname letter-by-letter using the relatively simple rules. Others had software that could generate Soundex codes on their PC or Mac. A rare few used one of the handy web-based Soundex Converters. These were great, but in my eyes, all of these had the same fatal flaw: they could only encode one name at a time.
So YASC was born. I named it Yet Another Soundex Converter because even back then, there were already many others. But YASC filled a useful niche: it could convert a hundred names just as easily as one. I wrote the original version in a largely-forgotten language called UserTalk (part of a product called Frontier). It ran on the lowly Power Mac 8500 that served as our department’s web server. In 1999, I moved it to my own web site and rewrote it in Perl. With a few minor updates, that’s the same version that has been running on this site ever since. Until today.
Of course, the fact of the matter is that Soundex converters aren’t as useful as they once were. These days, almost no one uses the old Soundex indices, at least not at first. All of the US census records have been digitized and fully indexed. Finding a census entry is often as easy as typing in a name. Still, it’s sometimes useful to go back to the old indices to track down a hard-to-find entry.