[Talk-us] What would you want done with TIGER 2010?
toby.murray at gmail.com
Tue Aug 24 06:18:39 BST 2010
I have been curious about this as well. In my opinion it would be
great to be able to update untouched TIGER ways with the new data. I
agree that this would likely have to be a county-by-county decision.
Could there be a tool like OWL but that would allow looking at a whole
county at a time to quickly determine if anything has been changed? It
would probably have to know about the abbreviation expansion bots and
ignore those (presumably the new data could be pre-expanded) and
possibly some of the other automated bots that have been run over the
Also, has there been any progress since the last import with regards
to handling county borders? The current data is an absolute nightmare
in that regard. Luckily here in Kansas there really isn't too much
demand for routing along all the tiny county roads that straddle the
border but still, from a "perfect world" perspective it would be great
to get that cleaned up.
On Mon, Aug 23, 2010 at 10:28 PM, Katie Filbert <filbertk at gmail.com> wrote:
> On Aug 23, 2010, at 11:05 PM, Antony Pegg <antthelimey at gmail.com> wrote:
> Hi Everyone,
> First time mailing, probably overdue, but I've been reading for a while.
> Got a question I'm hoping will spark some discussion:
> What would you like to see done (or NOT see done) with TIGER 2010 as regards
> OSM when it is released?
> Good to see you on the mailing list :)
> For tiger 2010, I think the data could be converted to osm format and made
> available to mappers
> The default though would be not import it and take it on a case by case
> basis (e.g. by county).
> If there are any local mappers in an area, they should make the decision.
> Situation could be that:
> 1) mappers already fixed up most of the road data
> 2) some local data source has been imported
> 3) the tiger data that is in osm, how it came to be (e.g. How census sourced
> it) differs from place to place. In case of Washington DC, I have done
> detailed comparison an tiger/osm matches up almost 100% perfect with dcgis
> street centerlines. I am examining the few spots where they are different
> and correcting it (based on usgs/dcgis ortho imagery). Certainly, census
> sourced the data from dcgis... the best source and happy with to leave it be
> in osm (even w/ option of importing dcgis data)
> For Montgomery County, MD and Arlington VA, I have had to "fix" tiger data.
> Arlingon, there was an offset. For Montgomery County, some was okay and some
> quite aweful.
> 4) could be there are no mappers in a place, tiger imports are untouched.
> Suppose we could do a comparison and decide on a county-by-county case
> Ultimately, a completely automatic import or replacement problem could end
> up making things worse, leaving artifacts and mistakes. Quality checking and
> some human involvement, especially by locals, is important.
> I'm very interested in the "why" behind any suggestions, and would love to
> hear what everyone has to say
> Content-Transfer-Encoding: quoted-printable
> Hi Everyone,<br><br>First time mailing, probably overdue, but I've been=
> Talk-us mailing list
> Talk-us at openstreetmap.org
More information about the Talk-us