3 ms·

Thanks for the link. Hash function "First and fourth letter" wins for english surnames: https://gist.github.com/bd9fabf91a5501b215c5 https://gist.github.com/bd

• lars512 16y ago
  There's a tradeoff between speed of lookup and ability to find things when you only partially recall the name. For those situations (the "tip-of-the-tongue" phenomenon), I'd pick either the first and final letters as the easiest to recall, or the first and second letters. The fourth letter won't be easy to retrieve unless you recall the name exactly.
• kragen 16y ago
  This is awesome! Thanks! The reason I used English corpus words was precisely that I did not have this surname frequency list handy. I don't think this is a list of English names, though. #18 is Garcia, and #19 is Martinez, and so on. It's a list of US names. A similar list for Argentina would probably be most useful for me at the moment.