Russ Nelson wrote:

>Anthony writes:
>  > On Fri, Nov 13, 2009 at 9:15 PM, Anthony <o...@inbox.org> wrote:
>  > > On Fri, Nov 13, 2009 at 6:57 PM, Dave Hansen <d...@sr71.net> wrote:
>  > >> We'll work on making sure that these data look good and I think some
>  > >> people have some plans on how to get these integrated a bit at a time.
>  > >
>  > > And hopefully in some places not at all.  In many locations there's
>  > > way more accurate data than TIGER.
>  >
>  > For examples of how bad TIGER data is, see
>  > http://maps.huge.info/geocoder/ and
>
>That URL doesn't work ... as an example of bad TIGER data.  I checked
>four addresses, and every one of them was correct within about 30 feet.
>Presumably your experience is different.  What addresses did you try?
>
>Some many locations TIGER data is quite good in location and name.  Of
>course, it has the structural problem of not knowing about bridges,
>but we can and are fixing that.

Try this address "4628 Barwick Drive, Fort Worth, TX" compared to my 
survey of that address 
http://www.openstreetmap.org/?lat=32.675599604845&lon=-97.3925864696503&zoom=18 
.

This demonstrates that while TIGER may put the block numbering start 
properly, the assumption in the interpolation that all possible numbers 
may exist in a numbering block can throw the computed location off 
significantly, which it often does. If a numbering block extends past an 
intersection, which it often does, it can put the address in the wrong 
physical block.

Unlike what some say, I think poor data, as the TIGER addressing data 
certainly is, is better than no data. It allows navigation apps to get to 
the approximate location, which is better than "address does not exist", 
in my opinion.

(Interesting aside: An OSM address search has Fort Worth located in the 
neighboring county, to the west. Guess we need to look into that. Maybe I 
shouldn't have been removing that TIGER county data from all the ways I've 
been "fixing".)

-- 
Randy


_______________________________________________
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

Reply via email to