OK, so I am neither a lawyer nor do I play one on TV. I did, however,  
work for two remote sensing companies in years past. That most likely  
colors my current opinion on the matter, but I don't consider myself a  
Google / DigitalGlobe / Navteq apologist.

Here is my naive world view (no pun intended) that helps me sleep at  
night:

Thing 1: Licensing
----------------------------
If I buy a digital camera and put the photos online, I get to choose  
if I give them away (public domain/creative commons/etc) or charge for  
them.

DigitalGlobe bought a big expensive digital camera and strapped it to  
a satellite (QuickBird and WorldView I, with WorldView II coming soon  
to an airspace near you).  DG (and Resource21 before them) is a for- 
profit enterprise, trying to make commercial remote sensing viable.  
There's no need to shed a tear for them, but DG is not sitting on an  
Uncle Scrooge McDuck-sized pile of money. R21 went out of business.  
This is not an oil company "windfall profits" situation -- both  
companies hemorrhaged money while I was there. It's simply not  
economically viable for them to give the satellite imagery away for  
free, nor is it a part of their articles of incorporation.

That said, the USGS offers up a country-wide (mostly panchromatic, but  
urban areas are multispectral) 2m mosaic. This is in the public  
domain. It is free for both commercial and personal use. See 
<http://terraserver-usa.com/ 
 >. Yes, the imagery is 10+ years old at this point. Yes, you might  
have to pay for distribution media (DVDs, Firewire drives, etc.), but  
the pixels are yours for free, no strings attached. Do with them as  
you'd like. See also Landsat 5 / 7 for low-res worldwide coverage.

So, we have two models -- commercial and the commons. Commercial data  
will always exist, but we can apply positive pressure to governmental  
agencies (Federal/State/County/City) to push as much geodata into the  
commons as possible. This applies to individual countries / the EU /  
etc as well. For world datasets, perhaps the UN, or perhaps a  
charitable entity like the Bill and Melinda Gates Foundation. Perhaps  
the WikiMedia Foundation. Perhaps the OGC. Perhaps the OSGeo. Perhaps  
the Illuminati... (wink)

Building awareness is a good thing. Building open standards is a good  
thing. Perhaps it's time to build a data commons as well. Part of the  
mandate could be providing pointers to existing data in the commons.  
Part of the mandate could be buying data and contributing to the  
commons. But assembling it all in one place with a clean, simple  
interface would be a good first step.

Trying to assemble a US-country-wide dataset is no picnic if you have  
to go state by state, county by county, etc. There are 3,141 counties  
in the US. I would love to see some federal mandates or guidelines in  
place to simplify the matter, but states-rights / "small gov't"  
activists might take issue with the idea. I hadn't considered the  
Freedom of Information Act until someone brought it up in this thread.  
Interestingly, geodata concerning wells is one of the 9 exceptions to  
the act. 
<http://en.wikipedia.org/wiki/Freedom_of_Information_Act_(United_States) 
 > I don't know if that helps or hurts the cause, or even if it sets  
any sort of legal precedent.

Bottom line: I don't think that there is a clear message in the US on  
geodata, whether by legislation or by the Supreme Court. So, who's  
going to fix that? I would love to see a "New Deal" like program that  
finally sends a clear and simple message -- we the people think that  
this data is valuable, and the multiplier effect of having out in the  
commons far outweighs the costs of collecting it.


Thing 2: Derived Work
---------------------------------
In both the USGS and DG imagery, we aren't looking at raw pixels off  
the sensor. The imagery is heavily altered to remove all sensor  
artifacts (orthographic rectification). At DG it was a pipe-dream for  
this to be a hands-off sausage factory exercise, but the reality is  
each image is (sometimes heavily) touched by humans at some point.  
Then there is the issue of all of the ground control points, DEMs,  
etc. Some of this additional raw material is in the public domain  
(SRTM, GTOPO 30), and some of it is privately, commercially collected.  
As interesting as it would be to see some sort of viral GPL-like  
licensing issue come into play here, I think that the reality is the  
end product has been substantially altered enough to qualify it for  
commercial protection.

So what, then, about derived works from the derived works? What if we  
start pulling shore line, road, river, etc. vectors off of these  
orthos? I don't know, but I think that the reasonable person would see  
that the accuracy of the road vectors is really due to the accuracy of  
the underlying pixels.

TeleAtlas, AFAIK, does feature extraction from orthos to build up  
their dataset. Navteq puts wheels on the road. (More on this in just a  
moment.)

Assuming that the underlying pixels are in the public domain, I would  
imagine that the case could be easily made that all of the work  
involved in doing the feature extraction and associating it with all  
of the non-spatial attributes (names, populations, speed limits,  
number of lanes, directionality, etc.) would qualify it as derived  
work -- one that could be commercially licensed or put out in the  
commons as well. You could hit the terraserver/USGS datasets for this  
path.

Navteq putting wheels on the road sounds to me like creating a dataset  
that they can license as they please. As does OpenStreetMap.

"New Deal" -- we the people feel that the data derived from President  
Obama's 2012 campaign promise of "a GPS in every car" -- stripped of  
any personal data that could be incriminating or a breach of privacy  
-- should also be in the public domain. This crowdsourced data  
represents both a net cost savings in gathering valuable geodata and a  
source of national pride.


Thing 3: Google Maps
--------------------------------
Google is a commercial entity, aggregating largely commercial data  
(from DigitalGlobe, Navteq, et al), and providing you a free (as in  
beer, not as in speech) platform to create mashups. Google does use  
public domain geodata in Google Maps, but it isn't virally licensed in  
a way that affects the commercial data. If you overlay data over the  
map, your data doesn't materially change the licensing of all of the  
foundation data. I'm pretty sure that you still "own" the location of  
your first kiss, and you can license it however you'd like. Tracing  
roads? Not so much.

What would make this issue much easier is if there was a Google-like  
app out in the commons, comprised of free data (as in speech). OSGeo,  
OSM, OGC, Wiki Foundation -- are you listening? Google has proven that  
it is possible. Who will pick up the challenge for the commons?

Long windedly yours,
Scott Davis
[EMAIL PROTECTED]




On Nov 15, 2008, at 9:53 PM, Christopher Schmidt wrote:

> On Sun, Nov 16, 2008 at 02:52:00PM +1030, stephen white wrote:
>> On 09/11/2008, at 8:02 PM, Ed Parsons wrote:
>>> As others have pointed out this is somewhat a grey area, however I
>>
>>
>> I don't see Google rushing to clarify this gray area, so I can only
>> think that this is the intended status quo. They want to see what
>> happens, while retaining a legal shotgun to blast away bits that they
>> don't like.
>>
>> So that seems quite clear to me. Google _do_ retain ownership over  
>> map
>> derivatives, but they're just not exercising legal responses
>> immediately. If you want to keep in the clear, do not use Google Maps
>> (or Microsoft Virtual Earth).
>>
>> As a side benefit, all those crappy mash-ups would go away! Yay! :)
>
> Nah. The mash-ups are using Google data on Google maps -- and Google
>
>  1. Has the rights to the data
>  2. Has reason to support the continued use of their data with *their*
>     maps.
>
> So what would go away would be things which use data created via  
> Google
> Maps but don't use it with the Google Maps API, presumably. Of course,
> all of this depends on how a legal team decides to pursue things, and
> any behavior that involved actually pursuing takedowns of things like
> Google Maps Mashups would probably be *social* suicide no matter how
> legally 'correct' it is. As such, it seems unlikely to be something  
> that
> would happen in the world we actually inhabit, as opposed to the  
> made up
> one we have to occasionally pretend to live in to cover our asses
> legally.
>
> Regards,
> -- 
> Christopher Schmidt
> MetaCarta
>
> _______________________________________________
> Geowanking mailing list
> [email protected]
> http://geowanking.org/mailman/listinfo/geowanking_geowanking.org


_______________________________________________
Geowanking mailing list
[email protected]
http://geowanking.org/mailman/listinfo/geowanking_geowanking.org

Reply via email to