I know that this is not really a Geoserver question, but I am so new at this that I don’t even know where to look. The promise of Geoserver and Mapbuilder is huge. What seems to be lacking is a full, easy-to-use set of data that compares to what Navteq is offering. I recently read an article where it said that Navteq’s dataset used 100,000 sources and that 75-80% of their budget was spent driving locations. I realize that it is not realistic to think of coming close to Navteq, but the little bit of investigation that I have done seems to indicate that most (all?) states have a wealth of data in shapefile format. I would think that the next logical open source step would be to put together standards for shapefiles (same metadata fields and values?) and converting each state to that standard. Is there something like that going on?
Am I wrong to think that this would be incredibly useful? Are there problems that I am not seeing?
Just curious.
-Al
Al Byers ha scritto:
I know that this is not really a Geoserver question, but I am so new at this that I don't even know where to look. The promise of Geoserver and Mapbuilder is huge. What seems to be lacking is a full, easy-to-use set of data that compares to what Navteq is offering. I recently read an article where it said that Navteq's dataset used 100,000 sources and that 75-80% of their budget was spent driving locations. I realize that it is not realistic to think of coming close to Navteq, but the little bit of investigation that I have done seems to indicate that most (all?) states have a wealth of data in shapefile format. I would think that the next logical open source step would be to put together standards for shapefiles (same metadata fields and values?) and converting each state to that standard. Is there something like that going on?
Am I wrong to think that this would be incredibly useful? Are there problems that I am not seeing?
As for something going on, look for OpenStreetMap, it's an alive and kicking effort to have some free carthography created from scratch
by people and their GPS.
As for doing what you suggest, I see the following problems:
* Data coming from different institutions will have different kind
of licenses. The fact that you can use get some dataset for free
does not mean you can use it as you see fit, usually there are some
(strict) limitations. See the VMAP0 world dataset for an example
of such a license.
* Accuracy mismatch. Some data will be good enough to be displayed
1:250.000, some other 1:1.000.000. Better not mix the two.
* Attribute set mismatch or layer meaning mismatch.
Most of the time you won't be able to find a common subset of
attributes for layers coming from different institutions, or
they look similar but aren't the same, or they are sometimes split
over multiple columns, sometimes merged into a text one.
Also, some data provider can give you a "transportation" layer,
some others give you separate layers.
In the end I believe the major issue is the licensing one: OpenStreetMap
is free of such problems because data is gathered directly from the
field free of legal limitations.
Oh, for a collection of free gis data, look here:
http://www.ing.unitn.it/~grass/docs/free_gis_data.html
http://freegis.org/database/?cat=1&_ZopeId=11823733A2-Bc1ovqhE
Cheers
Andrea
On 8/7/07, Andrea Aime <aaime@anonymised.com> wrote:
Al Byers ha scritto:
I know that this is not really a Geoserver question, but I am so new at
this that I don’t even know where to look. The promise of Geoserver and
Mapbuilder is huge. What seems to be lacking is a full, easy-to-use set
of data that compares to what Navteq is offering. I recently read an
article where it said that Navteq’s dataset used 100,000 sources and
that 75-80% of their budget was spent driving locations. I realize that
it is not realistic to think of coming close to Navteq, but the little
bit of investigation that I have done seems to indicate that most (all?)
states have a wealth of data in shapefile format. I would think that the
next logical open source step would be to put together standards for
shapefiles (same metadata fields and values?) and converting each state
to that standard. Is there something like that going on?
Am I wrong to think that this would be incredibly useful? Are there
problems that I am not seeing?
As for something going on, look for OpenStreetMap, it’s an alive and
kicking effort to have some free carthography created from scratch
by people and their GPS.
As for doing what you suggest, I see the following problems:
- Data coming from different institutions will have different kind
of licenses. The fact that you can use get some dataset for free
does not mean you can use it as you see fit, usually there are some
(strict) limitations. See the VMAP0 world dataset for an example
of such a license.
- Accuracy mismatch. Some data will be good enough to be displayed
1:250.000, some other 1:1.000.000. Better not mix the two.
- Attribute set mismatch or layer meaning mismatch.
Most of the time you won’t be able to find a common subset of
attributes for layers coming from different institutions, or
they look similar but aren’t the same, or they are sometimes split
over multiple columns, sometimes merged into a text one.
Also, some data provider can give you a “transportation” layer,
some others give you separate layers.
In the end I believe the major issue is the licensing one: OpenStreetMap
is free of such problems because data is gathered directly from the
field free of legal limitations.
Andrea,
I was thinking of the U.S. data that comes from the federal (USGS) and state and county governments. Do you think that data would be incumbered by licenses?
-Al
Oh, for a collection of free gis data, look here:
http://www.ing.unitn.it/~grass/docs/free_gis_data.html
http://freegis.org/database/?cat=1&_ZopeId=11823733A2-Bc1ovqhE
Cheers
Andrea
Al Byers ha scritto:
Andrea,
I was thinking of the U.S. data that comes from the federal (USGS) and state and county governments. Do you think that data would be incumbered by licenses?
I have no idea, I live in Italy 
Anyways, if there is any license the data provider will have to make
you accept it before downloading it. It may be that, given the Freedom
of Information Act, you get unencumbered data sets from your government
(lucky you), but as I said, you'll have to check. Maybe someone else knows more about this issue.
Cheers
Andrea