[Legal-general] [Geowanking] Flickrs WOE stuff

[SteveC]

On 6 Nov 2008, at 19:37, Eric Wolf wrote:
> They do attribute OSM at the bottom of the page when they talk about  
> Baghdad. Of course, they sort of have to because OSM has the best  
> digital map of Baghdad on the planet.

So my personal definition in this context is to put the text on top of  
the map, like what's demanded by our commercial friends. I think that  
is respectful to the community IMHO but lets give the flickr guys time  
to respond, they're nice guys and it's surely just a little oversight  
or lack of clarity on OSM or my part.

> Aren't these polygons just vast derivative works of either OSM or
> TeleAtlas/NavTeq data? All those geocoded photos that were just placed
> on a map... derive the places from the map, right? Then if you derive
> the polygons from that set of points.. the polygons are just derived
> from the map. I thought those guys dont like you deriving polygonal
> datasets without paying them extra beer tokens?
>
> You could argue at all of the points owe beer tokens to the US  
> Military for flying the GPS birds.

No... the federal government can't claim copyright on anything it  
produces and you can't copyright a fact. But we're straying in to  
horrible IANAL territory here :)

> So I think the end result is that, unless someone managed to geotag  
> a photo with a non-real "fact", there can be no licensing issue. And  
> unlike some copyleft schemes, derivative works utilizing what can be  
> assumed "factual" in the source work would be considered original.  
> So the geolocations derived from copyrighted works presented as  
> databases of "facts" would be considered facts. And creating  
> polygons based on those facts would not be impacted by the original  
> copyright.

Um no.. so a collection of those derived points is no longer a fact,  
it's a collection of facts which is covered by either sweat of the  
brow or database directive. When I last spoke to a copyright lawyer  
about this, he was pretty clear that if I was to derive a point on a  
map that was okay but deriving the two endpoints of straight street  
and then by logic deriving the street vector... that was not cool. By  
a long way. Then deriving polygons and more vectors was even less cool.

> What would be REALLY cool is if someone mocked up photos of some of  
> the fake streets in the TeleAtlas data, posted them to Flickr and  
> geotagged them as "photos of this fake place".

No what would be really cool is if someone set up  
letsinfringethefrickingmap.com with subdomains  
points.letsinfringethefrickingmap.com and  
vector.letsinfringethefrickingmap.com and so on. Each one would start  
by inviting the public to derive vast sets of points, vectors and  
polygons from Google, Yahoo and MSFT maps. You wont get a cease and  
desist for allowing individuals to derive a single point I suspect.  
But you will for allowing a crowd to derive an entire street network  
from a map dataset. You just have to fill in about 20-30 use cases in  
between as sub-domains. So allow people to derive street intersection  
points but not the streets... that kind of thing.

Then just sit and wait for the cease and desist letters to come in,  
and you can 'map' out what they feel is legal and what isn't, since  
they're all so don't-wake-the-sleeping-giant on this and won't  
actually tell you clearly. You then can cross off the use cases that  
they don't like. Send a long personal email to everyone you know at  
the big three so they can't claim to be unaware of the existence of  
the sites.

> Of course, I'm also one of the first paleo-geos arguing for the  
> openness of OSM. Talk about beautiful - an "map" almost entirely  
> devoid of centralized direction and purpose.

Direction is one thing, purpose is entirely another. We have a clear  
purpose.

Best

Steve