Welcome to Feiba Peveli: Procedurally Generated Place Names from the 1820s

medium.com

28 points·polm23··5 comments

Article describes a scheme for encoding lat/long into pronounceable words, down to 2 figures past the decimal point, which is enough to get you about a kilometer square (at the equator, finer towards the poles), so appropriate for names of cities and towns but not individual houses.

Extending the scheme to longer words for more precision seems straightforward but is not discussed.

What's unique about this as compared to W3W or OLC or any other schemes is that the encoding is so simple it can be done in your head once you learn the table. I suspect that folks who dealt a lot with locations would learn to do so very quickly, somewhat like De Smet speculates that Soundex coding on driver's license numbers might enable bouncers to quickly suss out sloppy fakes where the DLN doesn't match the printed name: http://www.highprogrammer.com/alan/numbers/dl_us_shared.html

> Soundex coding on driver's license numbers might enable bouncers to quickly suss out sloppy fakes

Wouldn't corrupt bouncers likely be a good source of 3rd-party QC consultants for those in the fake ID industry?

seems like this would get out pretty quickly. might stop home fakes, but that's about it..

Sure, it becomes an arms race, but I think your garden-variety kid trying to sneak into clubs is A) not getting caught for this reason, and B) not being told what's faulty about their fake ID even when they do get caught.

Or maybe the faker-maker software packages already take this into account when making up a DLN. I have no idea.

Can we add this to the Geohashing app?

Coming soon to a whiteboard near you!

Reply on news.ycombinator.com