[OSM-talk] What3words

Iván Sánchez Ortega ivan at sanchezortega.es
Wed Jul 13 10:40:01 UTC 2016


El Miércoles 13. julio 2016 12.10.29 Frederik Ramm escribió:
> [...] is a coordinate system, not an addressing system.

I think there's a thin line separating the both. But this is not obvious to us 
europeans, I think.

See south america: house addresses are the distance in meters to the start of 
the street/road. So your linear position is part of your address.

See gridded cities in north america: There's 1st ave, 2nd ave, 3rd ave, the 
first block is 1000, the second is 2000, etc. So actually your position within 
the grid is your address.

The main difference being, the reference line or grid is an actual feature in 
the ground (made of asphalt or concrete). And there's enough irregularities in 
those lines/grids that it makes sense to have that addressing info in OSM.


On the other hand, fully algorithmic systems (like W3F) would just add useless 
overhead to the OSM database. I have no doubt I would be labelled as a moron 
if I were to add W3F addresses to OSM nodes.


Now a smarter question would be: Does it make sense to have addresses based on 
*invisible*, *imaginary*, *non-surveyable* grids/lines/landmarks that you can 
only see when your cellphone has battery?

Does it make sense to have physical milestones/postboxes/streetsigns to enable 
an algorithmic system to work in the physical world (but not viceversa) ?

Would land records be based on the algorithmic or on the physical system?



That's the kind of questions that should be discussed, and not the "but you 
can pronounce them" or "but you can download a 10MB android library" 
b**lsh**t. For f**k sake.


-- 
Iván Sánchez Ortega <ivan at sanchezortega.es> <ivan at geonerd.org> 
<ivan at mazemap.no>



More information about the talk mailing list