The Status of State-Wide GIS Parcel Mapping in New Jersey
This comes from Andy Rowan, Director of the New Jersey State Office of GIS (OGIS). This is his response to a listserv discussion of the Parcel Mapping Task Force of the New Jersey Geospatial Forum (NJGF), published with Andy's permission.
"OGIS will not be creating any parcel data, nor
are we undertaking to do any routine maintenance of parcels. What we
are going to do, *starting* in early 2008, is take in the latest
versions of each county's parcel data and contracting to have them
edgematched to align with the updated municipal boundaries we are
working on, and the attribute table structures normalized. (We don't
have a completion date because we don't even have a scope of work or a
contract yet.) We will then give these modified data sets back to the
counties for continued maintenance going forward. At that point we will
also be arranging processes with each county to keep our replicas of
their data up to date on an ongoing basis. Once all of this is in
place, we will publish public map services with this near-statewide data.
I say "near-statewide" because ... as the parcel task force has already documented ... no county-wide parcel layer exists for Essex or Middlesex.
This is our short-term plan. The long-term plan is for digital submission of tax maps, and automated generation of GIS data from those submissions. Implementation of that is on a time scale measured in years."
I say "near-statewide" because ... as the parcel task force has already documented ... no county-wide parcel layer exists for Essex or Middlesex.
This is our short-term plan. The long-term plan is for digital submission of tax maps, and automated generation of GIS data from those submissions. Implementation of that is on a time scale measured in years."
Wow,
I wish them luck.
The ICI Society in BC tried to do this with local government data, spent tonnes of money, and ended up with a static product that was not useful:
- they had adjusted data that was survey-level accurate in some cases in the name of defining common boundaries
- they took too long, and the amount of development in the meantime had made moving to the "adjusted" model too onerous
Things to think about:
- make the data model as simple as possible; only the fields that are required for local government business needs. Tie to data from other sources using lookup tables. Don't expect the local governments to maintain data they don't use.
- Instead of dictating boundaries en masse, work with neighbouring data custodians in small groups to work out boundaries that are legal and most accurate _in their current working bases_ before starting to take delivery of the source data for aggregation
- understand that taking the data away from the custodians, "fixing" it, and giving it back is engaging in an extremely high-risk option. Constant and meaningful communication with the custodians and a real commitment to timeliness would be critical for this to even have a chance of success.
Jason
Reply to this