7. Primitive, NonTopological Vector Profile

a.k.a. Primitive SDTS Vector Profile, SDTS Lite,

Clean Spaghetti Profile to SDTS, Primitive Transfer Standard,

Simple Spatial Form, DIGEST Spaghetti Profile

Intended Use --

A primitive, nontopological, vector profile to SDTS has been suggested as a way to transfer digital vector geospatial data that is not constrained to a planar graph and not structured using topological relationships. This type of data is sometimes referred to as "spaghetti" data and may use SDTS strings. Graphic, topological, and raster elements would not be allowed. One purpose of a "primitive" profile is to simplify SDTS for desktop systems and low end map data users, and make it work for "the little guy." All-ASCII physical encoding and limited-metadata requirements have also been mentioned as part of this primitive vector profile.

The USGS considered a nontopological vector profile at one time for the distribution of a proposed, but later rejected, elevation product of tagged vector contour (TVC) data. The USFS Cartographic Feature file (CFF) data might be appropriate for a nontopological vector profile.

Based on a request from Henry Tom (NIMA & ANSI), the SAIC contractors to NIMA proposed a Simple Spatial Form (Special Sub-Profile) of SDTS and DIGEST during a 1996 meeting on the harmonization of SDTS and VPF. This Simple Spatial Form would be composed of "clean" spaghetti data, which does not have explicit topological relationships but could be topologically structures quickly without needing a separate cleaning step to find and remove small gaps, overshoots, and slivers. The SAIC proposal included a one-to-one relationship between spatial primitives and basic features, thus keeping some of the feature-based flavor of VPF. Discussion of this simple, clean spaghetti, profile included suggestions for its use for NSDI Framework, for a Tri-Services CADD profile, and for OpenGIS "well known structures."

Primary Sponsors, Interested Groups, and Contacts --

USGS researchers in Reston and Woods Hole, in addition to individuals from the Virginia State Geologist's Office and the Dept. of Geology at the College of William and Mary were the original advocates for this profile. Several individuals at the 1996 meeting on SDTS and DIGEST also expressed interest in this concept, including representatives from N IMA, SAIC, Tri-Services, and OGC.

Information from the June 1996 Woods Hole meeting on this issue is available from Robin Fegeas of USGS. Notes from the September 1996 Reston meeting with NIMA and SAIC are available from Charles Hickman of USGS

Future Plans --

The requirements for an primitive, nontopological, vector profile to SDTS may overlap some of the requirements for a CADD profile, an object-based profile, or a DIGEST spaghetti profile. This should be investigated.

One purpose for a simple, vector, all-ASCII profile is to help avoid the complexity of SDTS (or VPF). Current development by USGS and contractors of improved, windows-based tools to read, write, and view SDTS and ISO 8211 will hide much of this complexity, and may lessen the need for an ASCII implementation (rather than ISO 8211) of SDTS, and remove some of the requirements for a simple profile. Richard Hogan from the FGDC states,

"What they [users] want is a complex solution that is invisible to them - powerful applications with 'point and click' interfaces - lossless transfers of complex spatial data that requires no user interaction or understanding of the underlying standards. ... Complex tools that help application developers 'hide' SDTS is [the right way to go]."

Several of these new tools will be presented at the September, 1997 SDTS work shop. Keeping a single (ISO-8211) physical implementation of SDTS will also help with the overall data exchange situation.