[Talk-us] Admin boundaries tied to roads

[Alan Mintz]

At 2010-04-23 07:47, Apollinaris Schoell wrote:
> > While I understand the mantra of TIGER=Bad because of the state of the 
> road data, this is not true for the boundary data. Most of the boundary 
> data comes directly from recorded surveys (something not available for 
> roads) and is not "bad data" for most of the United States. The rural 
> areas would be the one exception (mostly because they did not have 
> surveys converted to digital layers in 2000), but rural areas are also 
> highly likely to have realigned boundary roads that no longer correspond 
> to the original boundaries.
> >
>
>I can tell for sure that they are completely wrong in California. They are 
>not even close to USGS 24k, don't align with official county borders from 
>official sources and don't align with natural features, fences which are 
>sometimes visible on Yahoo.

I don't know about "completely". The parts of the Kern/LA/Orange/San 
Bernardino/Riverside/San Diego borders that I have surveyed are at least 
close to the signage at important points (admittedly a weak standard), but 
I've also gone hunting for detail in law in some spots and found that the 
borders were right as of their date of creation in the source data. I 
remember manually fixing a little bit of the OC/LA border in La Habra from 
some sort of change description - maybe something out the BAS project. What 
a pain that was.

Is anyone working on borders currently? Is the BAS a reasonable source?

--
Alan Mintz <Alan_Mintz+OSM at Earthlink.net>