[postgis-users] PostGIS case usages

[Andy Colson]

On 10/26/18 12:05 PM, Regina Obe wrote:
> Hey all.  So we've been in talks with our editor about having a 3rd Edition
> of PostGIS hopefully to be released around the same time as PostGIS 3.0.
> 
>   
> That said, if people can respond with what things they are currently using
> PostGIS for and also what hosting they are using for PostGIS, that would be
> helpful for us to get a better idea of focus points.
> 
> It'd be great if you posted on the list, but if you are shy or need your
> usage anonymized, you can write directly to me.
> 
> Thanks,
> Regina



To add to my write up here: https://postgis.net/2016/03/17/vanguard_appraisals/

Sometimes we get data w/out measurements, so when we are routing and lot sizing a map, sometimes I've gone low level to add measurements. (I break up each line of a parcel, measure it, and add an annotation.  All in sql, thousands of parcels all at once.)

The gis parcel data for a lot/parcel may not be correct when we get it, but once measured we can compare the gis data to the parcel data to know if we should remeasure when we visit the property.  It gives us a starting point, and helps organize and prepare the data.

I think the biggest thing is the easy of joining other data to the gis maps.  Wanna combine the gis data with parcel data and show res acres vs agland acres.  Easy.  Color the map with parcel zoning?  Tax districts?  Easy and Easy.

I have not used esri software very much, so maybe I'm wrong about it, but it seems painful to join maps and databases.  We've had to add view's on top of the database just so esri could work with it.  There are some file formats it works ok with, and some that are very painful.

Not PostGIS, anything your twisted mind can work up, it'll do.  I'd bet I can write a perl stored proc that'll download a webpage, joins that to a redis fdw, join that to a csv, join that to active directory, and join that to Tiger data.

Another thing we do on the maps is find all parcels 100 feet from this point.  Or 200 feet from this parcel.

We also do heat maps for land pricing.  Plot all the land rates for an area or subdivision to make sure nobody is too different.  Similar pricing of similar land is important.  Similar area <> similar land, but its a good start.

Exporting data is popular.  Maybe someone needs a csv dump of: parcelNumber, Address, Latitude, Longitude
We get data in different projections, but its easy to ST_Transform(ST_X(ST_Center(the_geom)))

Oh, tiger data.  We use that to supplement missing info.


We don't use cloud hosting.  We buy a commercial line and host from our office.

-Andy