[Talk-GB] OpenKent, OSM coverage estimation

[TimSC]

Hi all,

Some stats on OSM coverage of Kent. I tried to pair the records of KCC 
OpenKent with the OSM database. Assuming the KCC list is complete (which 
it is usually, but not entirely), we can estimate OSM's coverage in the 
area.

Schools: 618 of 915 (915 (67.54 %)
Pharmacies: 67 of 274 (274 (24.45 %)
Doctors: 33 of 286 (286 (11.54 %)
Libraries: 70 of 101 (101 (69.31 %)
Opticians: 12 of 170 (170 (7.06 %)
Hospitals: 24 of 33 (33 (72.73 %)

So, OSM is good on some features and poor on others. It seems "for 
profit" locations are not so well mapped, compared to public services.

My philosophy is that OSM omissions should be regarded as errors. With 
complete lists of addresses, we can go and find exact positions of these 
services. I am still unsure if this is compatible with the relicensing. 
This data is distributed under OGL (and sometimes OS OpenData too). Can 
LWG attempt to reduce the legal uncertainty of this, by a definitive 
statement?

Regards,

TimSC