|
Background
The vision of a FLOSS land records management project designed specifically
for use in developing nations was initiated in late 2007 by the School of Survey-
ing, University of Otago, New Zealand in collaboration with the Land Tenure Group
of he Food and Agricultural Organization (FAO) of United Nations, Rome. Reviews
of FLOSS geospatial projects by Pieper (2007, 2008), Ramsey (2007) and Stei-
niger and Boucher (2009) report numerous well-supported (in terms of their asso-
ciated community of developers), mature and sophisticated projects that can con-
tribute to this effort. In addition, projects officially supported and sanctioned
under the auspices of the Open Geospatial Foundation (OSGeo)
26
add credibility
to the readiness of FLOSS geospatial tools to meet the needs of land records
management. Current FLOSS packages include database, geographic information
system (GIS)-based mapping applications, software development environments
and languages, Web servers, operating systems, and workflow designers and
servers, most of which have been assessed also by researchers at the University
of Otago.
In addition to the FLOSS geospatial projects under development, a large number
of open standards associated with the integration, sharing, storage and retrieval
of geospatial data in general and Web services in particular have emerged and are
now in widespread use (for more information, see the Open Geospatial Consor-
tium Web
27
). These standards are complemented by a relatively large body of re-
search on data models for land administration systems and the temporal and spa-
tial requirements of cadastral and registry records. Hence, it is timely and impor-
tant now to integrate these independent developments with the production of a
FLOSS solution to land records management. However, successful (i. e. sustainable)
open-source projects require the creation and organization of communities of de-
velopers and users. To this end, establishing a sound user base provides the mo-
tivation for continuing development of a FLOSS project. Within this user base
there is likely to be a group of curious and the committed users whom are also de-
velopers and whom have the need, desire and ability to produce software that can
satisfy their own particular computing needs.
The “power” users within a FLOSS community typically contribute the most in
terms of development effort and, simply stated, with a couple of exceptions the
land administration domain has not yet reached the stage where a cadre of power
user/developers has evolved and coalesced around a demonstrable FLOSS project
(Pieper, 2008). There are currently several isolated national projects where various
FLOSS components (such as PostgreSQL or MySQL for database management,
PostGIS for managing spatial geometries with Postgre tables, and uDig for desk-
26
Open Geospatial Foundation (OSGeo) – http://www.osgeo.org
27
Open Geospatial Consortium Web – http://www.ogc.org
42
top GIS and Web Map Services) are being deployed to manage land records (Pieper,
2007). However, there is no integration of effort and no common suite of land ad-
ministration functions yet developed for widespread use. Hence, this domain re-
mains embryonic and software applications, whether FLOSS or proprietary, tend to
be lead by external consultants. In order to avoid the well known pitfalls of not
looking beyond short term, project-based gains, the OSCAR project will adopt a
software development strategy that seeks to mentor local developers linked op-
erationally between the three potential national sites.
The local OSCAR teams will manage the development process within each
country adding locally relevant functions to a set of land records management
tools that will be built into the generic OSCAR core in the first stages of the project.
The
multi-national user base will allow for improved testing of the core code
(Mockus and Herbsleb, 2002), and beyond this the local developers will make de-
cisions about which functions, some of which may be programmed by developers
in other countries as plug-ins or add-ins, should be integrated into each build for
their particular circumstances. The OSCAR project specifically focuses on variation
in schema within the cadastral and land registry domain and how this may be
resolved, without increasing system complexity or requiring specialized code. The
approach will also integrate the storage and retrieval of historical data, the notion
of evolving requirements, the use of local languages, distributed use within a wide
area network, and process requirements within the core design. Some, but not
all of these building block concepts are discussed in the next section.
Do'stlaringiz bilan baham: | 
