[Talk-us] State Open Data

Kevin Kenny kevin.b.kenny at gmail.com
Sun Aug 12 20:26:35 UTC 2018


On Sun, Aug 12, 2018 at 1:05 PM OSM Volunteer stevea
<steveaOSM at softworkers.com> wrote:
> I'm not an attorney, though were I to attempt to sharpen focus on these two replies, I'd say that in California, it's more like this:  data produced by state agencies (by our state government personnel "on the clock") publishing them as "produced by the state of California" cannot have onerous copyright terms/restrictions put upon them.  They simply "belong to the public."  (This is especially true of GIS data, as in the County of Santa Clara and Orange County/Sierra Club cases).
>
> So when you say "copyright...owned by the government," that is effectively equivalent to "copyright owned by the People of the state" because of California's Open Data laws and stare decisis (law determined by court precedence/findings).  Whether "public domain" is the correct legal term I'm not sure, but if there is a distinction between the legality of California-produced data and "the data are in the public domain" it is either very subtle or completely non-existent; I consider California-produced data "somewhere around, if not actually PD" and "fully ODbL-compatible" for OSM purposes.  So, (and I hope this dispels any confusion and answers your question, Pine), "created by the government" means they can't put "onerous copyright" on it, meaning it is effectively owned by the People for any purpose for which We see fit.

TL:DR: The closest answer to Clifford Snow's original question for New
York is http://gis.ny.gov/gisdata/inventories/details.cfm?DSID=932
which is virtually certain (the law, as always is muddy) to be
ODBL-compatible (and in fact, there is a colourable case that it is in
the Public Domain.) The digital raster quads available from
https://gis.ny.gov/gisdata/quads/ (these are State, not USGS!) are
also a potential data source for tracing, and again, aren't deeply
mired in the legal swamp.

Of course, as people like Frederik Ramm are quick to point out, no
imported data are truly safe! (In the US system, a deep-pocketed
plaintiff can simply bankrupt the defendant before the conclusion of a
civil suit, and the law is particularly murky in the
copyright-friendly Second Circuit, which comprises New York, Vermont
and Connecticut.)

----

I already sent Clifford Snow a reply in private email, but this
deserves to be elaborated more in public, given how the conversation
has turned.

I am not a lawyer either, but as a scholar I have had to learn some of
this stuff.

I live in the Second Circuit, and the situation with respect to State
and local government works is complicated here. The confusion stems
from a decision rendered by the Second Circuit in the case of _Suffolk
County v First American Real Estate Solutions_ (261 F. 3d 179 (2001):
https://openjurist.org/261/f3d/179/county-v-first). First American was
a real estate multiple listing service that had acquired Suffolk
County's tax maps under New York's Freedom of Information Law, and was
republishing them without license to its participating realtors.
Suffolk County, motivated by the desire for cost recovery for the
maintenance of the tax rolls, sued for copyright infringement. First
American moved to dismiss, on the grounds that the Freedom of
Information Law extinguished any copyright interest that Suffolk
County might have had in the product. The district court initially
denied the motion. On petition to reconsider, the district court
agreed with First American that the Freedom of Information Law
requires that Suffolk County may not use whatever copyright subsists
in the tax maps to restrict their republication.

The Second Circuit held that the Freedom of Information Law is fully
satisfied as long as the public has the right to inspect the
copyrighted records, and that FOIL does not extinguish the possibility
of a copyright claim. Since it was ruling on a motion to dismiss,
there was no ruling on the facts of the case, so the judicial opinion
did not reach the argument of whether the maps had sufficient
originality to survive a claim under the _Feist v Rural_ (499 U.S. 340
(1991) - https://en.wikipedia.org/wiki/Feist_Publications,_Inc.,_v._Rural_Telephone_Service_Co.)
standard. Suffolk's initial argument appears to have been crafted to
follow the 'sweat of the brow' standard. Nevertheless, the Second
Circuit accepted that the originality claim was sufficiently pleaded.
Nevertheless, the opinion recognized that 'items such as street
location and landmarks were "physical facts"-and thus not protected
elements" (_Suffolk_ at 24) and that in an earier case it had 'thus
focused on "the overall manner in which [the plaintiff] selected,
coordinated, and arranged the expressive elements in its map,
including color, to depict the map's factual content."' No deference
was accorded to a contrary advisory opinion from the New York State
Committee on Open Government.

The appellate decision remanded the matter to the district court for
further proceedings; the matter of whether the tax rolls could be
sufficiently original to meet the _Feist_ standard could not be
reached because the evidence had not been adduced at trial. The
parties then settled, leaving the matter under a legal cloud.

The New York State Committee on Open Government continues to assert
that the counties' treatment of GIS data as a profit center runs
contrary to their duties under New York law: it so stated in advisory
opinions to Nassau County in 2005
(https://docs.dos.ny.gov/coog/ftext/f15695.htm) and to New York State
Archives employees in 2007
(https://docs.dos.ny.gov/coog/ftext/f16494.htm). While these opinions
do not carry the weight of law, the open government laws enjoy fairly
broad bipartisan support, and it seems rather unlikely that the full
Legislature would overrule the committee, were such matters to be
brought to the floor. The state administrative agencies have in large
measure declined to assert copyright in their work product, and so
state in the metadata files (with 'None' under 'Access conditions',
and only an informal request to credit the source under
'Redistribution Conditions') For this reason, I've been confident in
using State agency data sources with only a note in the 'Contributors'
page on the Wiki.

To dispel any confusion surrounding copyrights supposedly owned by
localities, New York City in 2012 enacted Local Law 11
(http://legistar.council.nyc.gov/LegislationDetail.aspx?ID=649911&GUID=E650813B-B1E9-4E56-81BA-58261487DA4A),
a broad mandate to release data to the public without further
encumbrance on reproduction and publication. New York City datasets
thus may be safely assumed free to use, and subsequent legislation
(https://opendata.cityofnewyork.us/open-data-law/)  has only
strengthened the provisions of the law. - including a requirement that
any data compiled in response to a Freedom of Information demand be at
least considered for inclusion on the city's open data portal. The
existence of these laws is what emboldened me to PDF-scrape the city's
maps of the watershed recreation areas to include them in OSM.

In addition to New York City, fifteen counties so far have decided to
give up the struggle and open their books, so GIS data on at least the
tax rolls are available for the following:

Cayuga
Chautauqua
Cortland
Erie
Genesee
Greene
Lewis
NYC - Bronx
NYC - Kings (Brooklyn)
NYC - New York (Manhattan)
NYC - Queens
NYC - Richmond (Staten Island)
Ontario
Orange
Rensselaer
Sullivan
Tompkins
Ulster
Warren
Westchester

at the New York State GIS portal. The remaining counties, it would
appear, still cling to some hope of cost recovery for their deed
registries through copyright licensure. These counties' data must be
assumed to be off limits to us. I do personally feel free to use their
plats AMONG OTHER SOURCES to cross-check public parcel boundaries,
that's called 'research' rather than 'plagiarism'! Moreover, once the
several sources have been cross-checked and integrated, any creative
spark that the county may have ignited in its data has surely been
extinguished.

In addition, New York State makes freely available most of its data,
including cadastral data for all State-owned parcels, even in the
counties that restrict redistribution of their tax rolls.

To the best of my knowledge, the Second Circuit is the only one to
accord so little deference to open data laws and opinions of state
agencies in states that have them. I expect that eventually, this
difference will come before the Supreme Court as a circuit split.
Given the likely makeup of the Court for the next generation or so, I
don't expect that the ensuing decision will go well for the advocates
of open data; there is simply enough public opinion that government
policy should be informed more by cost recovery than by public
service.

This whole situation leaves New York's open data in rather a messy
situation. There is no 'one stop shopping' of uniform quality for the
highway data of the sort that Clifford envisions. Interstate, US, and
State highways, and the major local highways that are eligible for
Federal aid (and bear New York State Reference Route numbers) are
available as open data
http://gis.ny.gov/gisdata/inventories/details.cfm?DSID=1302.  In
addition, there is a comprehensive street database of sorts, but the
data are of much higher quality in the cooperating counties. I notice,
for instance, that the apartment complex where my daughter lives (in
nonparticipating Schenectady County), built about 3-5 years ago, is
not on the map. The non-participating areas are still fairly accurate
albeit outdated, and are at the very least better than TIGER. The data
set in question is
http://gis.ny.gov/gisdata/inventories/details.cfm?DSID=932. It is what
informs the New York State digital raster topographic maps that are
available at https://gis.ny.gov/gisdata/quads/ - which are more up to
date than the classic USGS quads while not having the incompleteness
issues of the newer USTopo series.

The county street centerline data, in the participating counties, is
generally quite good indeed, but care must be taken because the datums
are not always consistent.
http://gis.ny.gov/gisdata/inventories/results.cfm?themeIDs=24 indexes
the county data sets.

None of these data sets have very extensive coverage of
privately-maintained roads, or of the roads in the wilderness areas.
Many of these are obtainable from the DEC Roads and Trails
https://gis.ny.gov/gisdata/inventories/details.cfm?DSID=1167 and
Snowmobile Trails
https://gis.ny.gov/gisdata/inventories/details.cfm?DSID=427 data sets,
but I'm actually somewhat reluctant to mention these. The data quality
of these is highly variable. Some tracks and trails have been updated
quite recently. Others have apparently been digitized from maps of
inappropriately large scale - perhaps even statewide maps - and are
wildly out of alignment at useful scales in the field. I do render
them on my own maps because any data is better than no data
(particularly when looking for missing features to map in the field!)
but call them out with a distinctive magenta colour to warn myself
that they may not be accurate.

Consider this an approximate 'brain dump', and make of it what you will.



More information about the Talk-us mailing list