Hi Brian, I think we talked this weekend, but if not, and you're here in SF, please stop by and say hi to me.
I have some small comments which appear inline, and them some larger discus at the end. On Sun, Jun 9, 2013 at 1:06 PM, Brian H Wilson <[email protected]> wrote: > I am working on getting the Benton GIS buildings data into the right shape > to do a mass import. > > I have not done an import before so I am looking for comments. That's great, and I'm hoping we can provide you useful feedback and help. > At this point I have done lots of processing for fun learning PostGIS but I > think the building footprint tags will be relatively minimal. > I also have address point data which I will be loading and think the more > advanced tagging (situs etc) belongs there. > > building=yes > attribution="Benton County, Oregon GIS services" > source="Buildings shapefile, 11 May 2013" source is good for inclusion on the changeset, so it's not needed on the objects. And in your case, you want the information from your attribution tags on the source tag, like: source = "Benson County Oregon GIS Services - Buildings Shapefile 2013-05-11" on the changeset. > I have names for every building on OSU campus but might exclude that area > because it's already got buildings in OpenStreetMap. Will just double check > that it's correct. This gets into some issues which I discuss below, but if some of the objects are already in OSM, then you should compare them and see which are of higher quality- this is a manual process. > Does anyone have an opinion on the minimal size I should pull in? My data > includes sheds and outbuildings that are very small. I was thinking of > cutting it off at 200 sq ft since buildings smaller than that do not require > permits. (My 6x8 greenhouse and 6x6 shed are in the database which is > derived from air photos) Does your building data data have the building type? building=shed or building=greenhouse are valid tags. > Comments in the Wiki say importings of parcels is "bad" but don't have any > comments as to WHY that's bad. (It's TBD) If I knew why I would update the > Wiki. There are a few reasons that parcels are not useful in OSM. Here are a few: 1. They're not surveyable Data in OSM should be ground surveyable. You should be able to look at some data that I've entered and make improvements to it. Parcel data isn't generally ground surveyable. 2. Parcel data changes often Parcel data changes quite frequently as owners change. OSM is a bad place for data that changes frequently. 3. Parcel data is not of general use A parcel doesn't tell you very much. Sometimes parcels correspond to addresses, but not always, for example. Now, let me ask you a few questions: 1. Can you point us to this data? Is it available online? One of the questions for me, for example, is the data license. 2. Are these buildings addressed? Buildings imported without addresses are of very limited value. So it's very important to have the addresses on the buildings themselves. If your address are on the buildings, then that's very useful. If not, and there's another address dataset, then we'd want to talk about a conflation process. 3. What existing data exists? You mentioned that there are buildings on the local university. It's quite common for a given town to have *some* editing, so your process needs to take that into account. Do you have a sense of how much existing data there is in your area? I also would encourage you to work with local community for a hybrid import, of the type that's being done in Seattle. I think a good first step would be to show the data source, and also the .osm file your process would create. - Serge _______________________________________________ Imports-us mailing list [email protected] http://lists.openstreetmap.org/listinfo/imports-us
