[Talk-ca] silly borders

[James Ewen]

On Tue, Mar 31, 2009 at 8:47 AM, James Ewen <ve6srv at gmail.com> wrote:
> On Tue, Mar 31, 2009 at 7:41 AM, Brent Fraser <bfraser at geoanalytic.com> wrote:
>
>> And for those who like to parse latitude and longitude:
>>
>> http://www.internationalboundarycommission.org/products.html#coordinates
>
> Aha, the definitive source that I was asking about... Actual lat/long
> values of the monuments. One would be hard pressed to dispute this
> information. The IBC is the authority on the matter as far as I am
> concerned.

So, I checked out the monument at the SE corner of BC...

 M03140   MONUMENT 272         48 59 56.00    114 04 01.96

That's in the NAD27 datum, so I translated that to WGS84

48 59 55.60  114 04 05.53

Then converted it to DD.dddd

48.9988 114.0682

And looking at the corner of BC in Potlatch, highlighting the node,
and pressing L, I get:

49.9988 114.0685

That's pretty darned close...

Node 331773601 is even closer at 49.9988 114.0683, which might
actually be the monument location, and the corner of BC is computed
from that.

The end result, is that I would concur that the GeoBase borders are
much closer to the real world location than the manually input border,
or the USGS imports.

So, now we need to clean up the erroneous data.

The county outlines in the US are circular ways, the GeoBase ways are
not.Will it affect rendering by cutting up the US county circular
ways, and making them part of a combined way? Can I cut the GeoBase
ways at the Alberta, BC and US confluence, and then add in a bunch
more tags to the common borders? Can we tag the same way as a
border_type: state, and also border_type:international, then do
state:left, province:right, county:left, county:right, etc?

James
VE6SRV