Abstract
This disclosure describes techniques for turning a latitude and longitude into a short
sequence of words, and back again, in which each axis is handled on its own. Latitude and
longitude are each quantised into a separate integer, for example 2^24 values per axis,
about 1.2 m × 2.4 m at the equator. Each integer goes through its own keyed full-range
permutation, such as a Feistel network or an affine map, and is then written as digits in
the radix of a word list. Every word therefore encodes a field of one axis only, and no value
that determines the whole location is ever computed. The default identifier has four words
from a 4,096-word list, two for latitude and two for longitude, with an optional check word
and version prefix. Further published variants add independent streams (altitude, floor,
time), self-describing variable lengths, per-axis Open Location Code digits, hierarchical
prefix words, precomputed stored triples, grid-component tokens, hashed and digit
intermediates, nested intervals, adaptive longitude, shared low-bit words and search-
decoded one-way mappings.
Creative Commons License
This work is licensed under a Creative Commons Attribution-Share Alike 4.0 License.
Recommended Citation
Boll, Joe, "GeoWords: Location-to-Words Identifiers Built from Separately Quantised, Separately Permuted Latitude and Longitude Axes", Technical Disclosure Commons, ()
https://www.tdcommons.org/dpubs_series/12737