There are still some data inconsistencies in this file. See this thread for details : http://forum.geonames.org/gforum/posts/list/208.page#1143 The duplicates have been removed from the dump, but the problem, that some admin1 features are in the '00' region, still exists with some '00' adminCodes.
Your help in fixing these errors would be welcome.
As soon as these errors are fixed we will replace the admin1Codes.txt with what is now the admin1CodesASCII.txt.
Just wondering - is there a commercial arm to what you're doing here with Geonames? I ask because this is an incredibly useful tool that needs to be sustained, and I hope that there is/will be sufficient incentive to you (and any team you have) to keep doing it!
There are still some data inconsistencies in this file. See this thread for details : http://forum.geonames.org/gforum/posts/list/208.page#1143 The duplicates have been removed from the dump, but the problem, that some admin1 features are in the '00' region, still exists with some '00' adminCodes.
Your help in fixing these errors would be welcome.
As soon as these errors are fixed we will replace the admin1Codes.txt with what is now the admin1CodesASCII.txt.
Cheers
Marc
I have two problems with the admin1CodesASCII.txt version:
1. The US state names are inconsistent. They should match the versions that are in the admin1Codes.txt file.
2. District of Columbia is contained in the admin1Codes.txt file but not the admin1CodesASCII.txt file.
Geonames.org is aggregating information from many sources this causes a lot of inconsistencies like the one between the two admin1Code files. The admin1Codes.txt will sooner or later be replace by the admin1CodesASCII.txt. The first is from a separate source and the second is generated from the geonames database.
I have changed the feature code in the geonames data base for the DC to ADM1 (from ADMD). It should therefore be in the admin1CodesASCII file with the next dump (in a couple of hours)