In my area they often have facilities that serve as a "rescue mission" or temporary halfway houses. I'm not sure how I'd tag in cases where that's the primary purpose. There are also secular recovery homes, which I'm not sure how I'd tag...? (not sure they want the publicity anyway)

Cap'n,

I hope you're not using OSM for navigation :)

(I still don't believe the official stories about why DAFIF was removed.)

NAIP imagery is available by WMS, as described here:

http://seamless.usgs.gov/wms_services.php?layerid=17

you passed

Oh, I'm finding dozens, if not hundreds of hours worth of work down here in Alameda County correcting TIGER import data.

For what it's worth, the main map's Mapnik and Osmarender refresh time is *incredibly* quick now compared to what it was six or nine months ago; I usually see my edits show up in less than four hours. I didn't know about the above tricks to "force" a refresh, but haven't found it necessary.

(95% of my work is up around Fremont/Newark/Union City.)

I don't think I will be running mapnik on the windows box. I'm setting up on a big linux server at work that's sitting idle right now.

Until I have time/hardware to build another server, I'm just running it on my primary workstation, under vista64. I do not yet have a set workflow, but I hope to write about it in my osm wiki area once I do. Feel free to send me a PM about it.

I'm still reading the paper proposed as a rebuttal, at this link: http://www.geog.ucsb.edu/~good/papers/457.pdf Only partway through, but a good read.

Laura, I'm confused about some of your references to GPX; simply uploading raw GPX to the Public GPS Traces area does not mean that the map/database will be modified. Or are you suggesting that people are using JOSM to convert their GPX tracks into OSM ways and just uploading them without regard to the existing data? I would be surprised if that practice is widespread; I would expect that the overwhelming majority of people using JOSM know enough about OSM to understand the impact of their edits.

I've mapped a few racetracks that I use around here, but the other day I I ran into the question: What about private test-tracks? There's an auto plant a few miles from me (NUMMI, in Fremont CA) that has a test track, and I wasn't sure how to label it.

Somehow your coords got changed and lost the '-' so I ended up on the other side of the world :)

What do you mean "zoning"?

Your username is hilarious :) (Hi, Russ, too)

Yeah, I just moved down to Fremont (Niles) from Hayward, where I was doing the same. We are so lucky to have the USGS Urban Area layer, but I'm finding some surprises as I go out and learn my new hometown.

Oh, sheesh, you're right. I fuss with that all the time in my GIS software but didn't think about it in JOSM for some reason. Duh.

Have you been out there since the fire? How's it look?

I've done a little bit of work on the trails in that area, but recently have been cleaning up some horrific misalignments up in the Niles area. There's some really wacky stuff over there.

I'm kind of working my way outward from the segments of the SF Bay Trail that I've been riding and mapping.

Indeed, I misread the file size by three decimal places, but not the number of trackpoints. I am trying to find a compromise between laboriously hand-tracing the features on the Urban Area DOQ (which I won't have most places) and just dumping raw GPS data into the map. The high sampling rate is important for some of the complex footpaths and cycle paths I'm mapping. I may be able to split the tracks apart on either side of the very complex bits and simplify the straighter bits even more.

It does with Osmarender: highway=raceway renders with black borders, and {highway=track;track=racetrack} renders with red-and-white borders. The string 'race' doesn't show up in the mapnik templates at all, though.