[Imports] How good can an import be?

[Hillsman, Edward]

On Tue, Apr 6, Andy Allan <gravitystorm at gmail.com> wrote:

>[. . .] Throw a 20km radius around most people in the US, and
>you'll find they have a local population density similar to urban and
>rural areas in Europe. So why is OSM gaining traction less quickly in
>urban and suburban areas of the US? Country-wide population density is
>a distraction from whatever is really going on, and annoyingly is
>often used as an excuse for blanket-importing data even in urban and
>suburban areas.
>
>[. . .] I strongly believe
>importing data is detrimental to growing the community. Of course
>that's unprovable, but I believe there's only so many correlations
>before causation can be reasonably suspected.
>
There are many good-sized cities in the US with very few edits. I'm tempted to propose an experiment. What if "we" (figuratively: I mean the OSM Foundation or an authorized group) simply removed all of the original-and-unedited imported TIGER data from a few of these. Leave everything else in place. Then do some outreach/publicity in these cities and a few others of comparable size, situation, and mapping activity, and then observe the subsequent mapping rates. Of course, this would not be perfectly controlled--there would be nothing to stop people from outside the area going in and tracing imagery. Or to stop locals from importing local data. But some things such as shops, pubs, bicycle parking, businesses really do have to be mapped from on the ground, and they could be used as indicators, even if nonlocals still could visit and map (here in Florida, I've encountered data entered during a week or two, and only that, by people who edit frequently in Europe before and after--consistent with mapping while on a vacation here).

I think a contributing factor to the low levels of mapping in much of the US is that the built environment makes it unsafe. You can run a GPS unit while driving or bicycling to get a trace, but unless you have someone in the car with you to take notes, all you get is the trace of the road alignment. Dictation by the driver is dangerous (tried it for a few minutes, never again); ditto by the cyclist. You really have to get out and walk around to observe and take notes and photos, and a lot of our suburbia is dangerous to walk around in, because it was built for cars, not pedestrians. I'm talking about crossing 8-lane divided streets, or smaller streets that intersect them. For the record, I work with the TIGER data, I trace from imagery, I have overseen a few small local imports of bus stops, and I mount expeditions where I drive to areas, get out and photograph within the limits of what is walkably safe, move the car, repeat, and then match the photos and notes to what I've traced from the imagery, correcting as I go. I've not had much luck in interesting others here in mapping, but motivating others has never been my strong suit.

Ed Hillsman
USF
Tampa, FL USA