Reminds me of the guy that streamed a talking banana on Twitch, where viewers could make it say things. People submitted variations of the n-word and got him banned, and after trying to filter out all character combinations he could think of he wrote a phonetic filter. That apparently worked much better than trying to think of every permutation of characters that sounds like bad words.
Anything exposed to the open internet devolves into porn or racism or both unless active effort is made to prevent it. I'm reminded of https://en.wikipedia.org/wiki/Tay_(bot)
quite curious as to how Instagram managed to avoid this, despite launching with no moderation, (just Mike and Kevin), yet all you needed was an email to use the service...
They certainly have moderation now, and it's a constant exercise in boundary-pushing. I'd be interested in a "history of Instagram moderation", that would be a great piece of anthropology.
Was the no-links policy there from the start? I think that would have helped a lot. As it is, you're allowed basically one outbound link from your profile, so there are link-expander services which people use to link to more things.
The list included mike hawk (phonetically similar to 'my...'), so they are interested in phonetics, apparently, even for these usernames. The banana streamer has their stuff set up better than twitch, then.
Ellis Island in New York harbour (AFAIK) used to be the main immigration center for people arriving by ship from Europe, and is famous (among other things, I assume) for having originated some weirdly spelled names when the immigration officials who registered the newly arrived got it wrong.
Don't know if the GP meant that they had soundex, or that they invented it, but in any case it seems they would have needed it.
https://youtu.be/bJ5ppf0po3k?t=715