#POSTGIS IRC Log - 2008-07-05

For logs after Feb 3, 2007, all times are GMT-8. Prior logs are GMT-9.
Back to Logs
00:31:42 vomjom: is it possible for me to convert an address to latitude, longitude with postgis and tiger/line?
00:32:42 vomjom: the tiger geocoder doesn't seem to support the 2007 shapefiles
04:58:40 darkblue_b: vomjom: there is a geocoder in PostGIS
05:00:04 darkblue_b: you are having problems once the data is loaded?
05:02:38 darkblue_b: Snow-Man: are things up to date?
06:36:18 jbronn: darkblue_b: yeah, there's a geocoder: http://svn.refractions.net/postgis/trunk/extras/tiger_geocoder/
06:36:19 sigq: Title: Revision 2825: /trunk/extras/tiger_geocoder ( at svn.refractions.net )
06:40:41 jbronn: vomjom: yes, the 2007 TIGER shapefiles were a complete overhaul by the Census Bureau, and they were only released this past March -- the geocoder hasn't been updated to support them yet. You can still use the 2006SE files ( http://tinyurl.com/tiger2006se )
06:40:42 sigq: Title: 2006 Second Edition TIGER/Line File Download Page ( at tinyurl.com )
06:42:39 darkblue_b: jbronn: ah
06:43:42 darkblue_b: I have a copy of the files.. and a script to load them, courtesy of the PostGIS mailing list
06:43:51 darkblue_b: but I havent gone through them in detail
06:45:47 jbronn: the 2007 shapefiles are awesome -- it's nice to have that data in shp format, makes it much more accessible
06:47:08 darkblue_b: I pulled a couple of counties..
06:47:20 darkblue_b: you've seen the shift on them?
06:47:53 darkblue_b: the streets dont line up in many urban counties.. by a significant margin
06:48:26 darkblue_b: asking around, I was told basically.. TIGER data is good for its referential completeness, but not for actual on the ground accuracy
06:49:52 jbronn: Hmm. I haven't noticed any problem w/Harris county ( Houston ) but I need to give them a closer look.
06:50:14 jbronn: I've always heard that TIGER's urban coverage is not very good.
06:50:20 jbronn: s/urban/rural
06:51:36 darkblue_b: yeah
06:51:52 darkblue_b: search for shift TIGER 2007 at the PostGIS mail list archive
06:52:11 darkblue_b: there is a picture in one of the posts.. that shows it well..
06:52:24 darkblue_b: Google and OpenStreetMaps are self-consistant
06:52:33 darkblue_b: while TIGER is off
06:53:12 darkblue_b: there is so much in that TIGER data set.. it seems like a specialty in itself
06:53:14 darkblue_b: ( ! )
06:58:25 jbronn: Yeah, regardless, it rocks that we get that stuff for free in the U.S. -- the Canadian and EU gov't data policies are much less progressive.
07:02:10 darkblue_b: absolutely
07:02:35 darkblue_b: I am bringing up a flaw.. but in the larger sense, you are most definitely right
07:02:55 darkblue_b: the commonwealth heritage of data access is part of why the colonies revolted!
07:03:12 darkblue_b: the EU is still strugglinh
07:03:15 darkblue_b: g
07:04:32 darkblue_b: I read in Ed Parsons Blog that while in on some INSPIRE talks, he suddenly got the impression that none of it required any public access at all
07:05:10 darkblue_b: one of them here... http://isde5.pbwiki.com/BlogRoll
07:05:11 sigq: Title: isde5 wiki / BlogRoll ( at isde5.pbwiki.com )
17:48:06 CIA-6: kneufeld * r2826 /trunk/doc/postgis.xml: fixed typo in a WKT example.
23:23:17 CIA-6: robe * r2827 /trunk/doc/postgis.xml: fixed ST_Dimension example, flipped order of conditions in WHERE on some examples ( in later versions of PostgreSQL since we don't have our costs right - the planner may choose to process in order of conditions