[Talk-us] Comparing Tiger 2017 dataset with OSM in a automatedway.

Kevin Kenny kevin.b.kenny at gmail.com
Thu Oct 26 23:28:59 UTC 2017


On Thu, Oct 26, 2017 at 5:12 PM, Nathan Mills <nathan at nwacg.net> wrote:
> The tree cover issue is precisely why many states that have seasons have a
> recurrent leaf-off (sometimes even in IR) imaging program.
>
> Arkansas has their imagery, along with a raft of other open data, available
> on Geostor as a WMS service that should be usable in JOSM and also as
> downloadable data in your choice of extent and format. Oklahoma's is also
> available somewhere, but I'm not sure where. They lack a statewide data
> repository, so their data is scattered about various state and university
> sites.
>
> I know the wiki has a list of imagery sources, but I don't think it has any
> listed specifically as leaf off. Maybe someday I'll find some time to
> compile one.

New York definitely has leaves-off data, in 4-band R/G/B/NIR:

https://orthos.dhses.ny.gov/

although we also have some
significant amount of coniferous and mixed forest where tracing is
still difficult.  I would absolutely love it if this data source were available
for tracing in JOSM. I've emailed the data custodian repeatedly asking
for confirmation that this is an acceptable use, but never received a
reply.

Is the statement in the accompanying metadata,

Access_Constraints: Some imagery tiles are classified as sensitive due
to their content.

Use_Constraints:Use of sensitive imagery, if granted, is only for the
use specified in the request.

something on which we may rely? I was proceeding with a formal request
out of a superabundance of caution, and because I live in a jurisdiction
where County of Suffolk v First American is still good law. That said,
the State's official position (https://docs.dos.ny.gov/coog/ftext/f15695.htm)
is that its own open records are, to all intents and purposes,
in the public domain. If I were hosting the data on my own server,
I'd have no reservations at all, but I surely do not want to risk
the project in any way.

For that matter, do we also need further assurance that an official
who grants us permission is, in fact, authorized to grant it?
How far does our (justified!) official paranoia about external data
extend?



More information about the Talk-us mailing list