[OSM-legal-talk] Proposed "Metadata"-Guideline

Mon Oct 12 21:43:31 UTC 2015

Steve

No, not addresses (though that would be nice, too, though. Wasn’t Foursquare trying to help with that?), but geocoding to administrative boundaries,as well as POI. We were interested in building a system that could be elastic enough to geocode to esoteric localities, such as “1 km north of Kendua, Bangladesh” or “Madan Pharmacy, Madan, BG” and provide options for retrieving a point, point with an uncertainty radius, or an actual json admin boundary, moved north 1 km. This would be something similar to Tulane’s Geolocate, but with access to the ever growing corpus of POI and features in OSM. Originally, we wanted to build something that could be used in a variety of use cases for augmenting medical data with location data based upon OSM extracts.

I’m aware of the PD sources for data, however, for instance, there does not currently exist a PD source for the locations of all villages (96,817) in Bangladesh (we’ve been provided a table of the village names and their unions, upazila, etc… by the BG Health Ministry, but without geometries of any type). There is, however, the potential for OSM to have most of these villages, especially given the efforts of World Bank, HOT… in supporting mapping in developing countries, in the future. We’d wanted to be able to identify villages from OSM for our data dashboards, but couldn’t determine whether mixing the OSM data with personal health information for cholera patients would cause an issue and so didn’t.

On the Damascus project, we were interested in doing network analysis on reports from informants on the ground about grey-outs of communication and power infrastructure, followed by military sweeps of neighborhoods, using OSM for street data and the neighborhood boundaries.

Neither of the projects was scrapped because we couldn’t use OSM for the project, but because we couldn’t determine IF WE COULD use OSM for our particular uses.

Additional question… Is it like the “Elephant in the room” that the humanitarian relief space disregards ODbL sharealike? How does OSM feel about the difference between potential users who want to operate within the parameters of the licensing and those who ignore it (arguably for noble cause, in both cases)? I’ve asked people in that space about how ODbL effects/hinders their efforts and the answers were the equivalent of shrugs and sly smiles. Unfortunately, our IRB seems a bit more skittish than that.

In F,L&T,
Stace Maples
Geospatial Manager
Stanford Geospatial Center
@mapninja
staceymaples at G+
Skype: stacey.maples
214.641.0920
Find GeoData: https://earthworks.stanford.edu<https://earthworks.stanford.edu/>
Get GeoHelp: https://gis.stanford.edu/

"I have a map of the United States... actual size.
It says, "Scale: 1 mile = 1 mile."
I spent last summer folding it."
-Steven Wright-

From: Steve Coast <steve at asklater.com<mailto:steve at asklater.com>>
Reply-To: "Licensing and other legal discussions." <legal-talk at openstreetmap.org<mailto:legal-talk at openstreetmap.org>>
Date: Monday, October 12, 2015 at 1:19 PM
To: "Licensing and other legal discussions." <legal-talk at openstreetmap.org<mailto:legal-talk at openstreetmap.org>>
Subject: Re: [OSM-legal-talk] Proposed "Metadata"-Guideline

Stace

Regarding your first email on this topic you said -

"to built a geocoding platform on Open Source software and Public Domain data that could be used to geocode research data”

Could you give an example of what the geocodable string would look like (just make one up)? Is it like “1 Alpha Street, Fooville” or is it more like “Smallville, AK” ?

If it’s address data - are you aware that OSM doesn’t actually contain much at all and thus can’t help you?

If it’s not address data - are you aware that there are multiple PD sources for non-address level geocoding?

Because either way I’m having trouble understanding why OSM is in the way to achieving what you’re trying to do?

Best

Steve

On Oct 12, 2015, at 2:08 PM, Mr. Stace D Maples <stacemaples at stanford.edu<mailto:stacemaples at stanford.edu>> wrote:

Thanks, Alex.

Clarity is exactly what is needed. Ambiguity = IRB Death. I’m going to be going through the OSM Licensing/Copyright Guidelines more closely over the next week and will comment outside this thread, if I have comments.

For the record, I hardly think solving things like diarrhoeal disease (2nd leading cause of death in children, globally) and tracking human rights abuses in repressive regimes are a 1% problems.

In F,L&T,
Stace Maples
Geospatial Manager
Stanford Geospatial Center
@mapninja
staceymaples at G+
Skype: stacey.maples
214.641.0920
Find GeoData: https://earthworks.stanford.edu<https://earthworks.stanford.edu/>
Get GeoHelp: https://gis.stanford.edu/

"I have a map of the United States... actual size.
It says, "Scale: 1 mile = 1 mile."
I spent last summer folding it."
-Steven Wright-

From: Alex Barth <alex at mapbox.com<mailto:alex at mapbox.com>>
Reply-To: "Licensing and other legal discussions." <legal-talk at openstreetmap.org<mailto:legal-talk at openstreetmap.org>>
Date: Monday, October 12, 2015 at 12:32 PM
To: "Licensing and other legal discussions." <legal-talk at openstreetmap.org<mailto:legal-talk at openstreetmap.org>>
Subject: Re: [OSM-legal-talk] Proposed "Metadata"-Guideline

On Fri, Oct 9, 2015 at 5:41 PM, Steve Coast <steve at asklater.com<mailto:steve at asklater.com>> wrote:

On Oct 9, 2015, at 3:22 PM, Alex Barth <alex at mapbox.com<mailto:alex at mapbox.com>> wrote:
On Fri, Oct 9, 2015 at 12:07 PM, Steve Coast <steve at asklater.com<mailto:steve at asklater.com>> wrote:
If you want all these rights, you can just pick up the phone and pay HERE or TomTom for them, they’d love to hear from you.

What's more interesting than sending people to HERE and TomTom is making them contributors to OpenStreetMap, no?

Absolutely, but at what cost?

OSM solved 95% or 99% of our problems. Should we fundamentally change OSM to claim the last 1% so someone can make slightly more money or complete an academic project? I don’t think that’s a worthwhile tradeoff. I’m super happy with the 99% we achieved already.

I'm very happy about what we have achieved too. I don't think we're solving 95% of our problems with OSM though.

"our problems" would of course need more definition and I'm running the risk here of misinterpreting what you said. I'm thinking about all the cases where OSM isn't used yet, all the mapping that isn't happing in OSM yet. OSM has the potential to fundamentally change how we capture and share knowledge about the world but we aren't anywhere near the full impact we should be having. 300,000 active mappers is impressive but the world is much bigger. At a time where the internet that was supposed to be Open is turning more and more into a closed game of big players and growth for OSM is linear - what's our plan? Fixing the license surely can't be the extent of our plan, but we need to be able to have a frank conversation about how licensing is hurting use cases and engagement on OSM, without second guessing people's intentions and without just showing them the door to TomTom and HERE. In that context I find comparing ODbL to Public Domain absolutely useful.

I think Stace's comments give a great glimpse into licensing pain points in the academic community in the US and the guideline Simon pulled together is going to fix some of the issues he's brought up. Having clarity how data linked to OSM does not extend the ODbL's share alike to that data should go a long way to address some of the concerns he raised.

_______________________________________________
legal-talk mailing list
legal-talk at openstreetmap.org<mailto:legal-talk at openstreetmap.org>
https://lists.openstreetmap.org/listinfo/legal-talk

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/legal-talk/attachments/20151012/c2a03df7/attachment-0001.html>